Article

Impact of Socio-economic Factors The Indian Economic Journal

1­–21
on Agricultural Prices: A Case © 2024 Indian Economic Association
Article reuse guidelines:
Study of Paddy Cultivation in India in.sagepub.com/journals-permissions-india
DOI: 10.1177/00194662241265466
journals.sagepub.com/home/iej

Avanindra Nath Thakur1

Abstract
As part of agricultural price policy of the country, the government announces minimum support
price (MSP) for selected crops every year based on the recommendations of the Commission for
Agricultural Costs and Prices (CACP). Since large part of procurement of paddy happens though the
private procurement agencies, implementation of MSP remains extremely limited. Further, as far as
government-regulated agencies are concerned, procurement remained concentrated in a few states;
therefore uniformity of price even in the regulated markets is far from being a reality. Thus, the aver-
age price received by farmers is largely the outcome of intra-regional socio-economic dynamics. Even
within a particular region, not all farmers at any point in time receive same price for their produce.
In this context, the article, based on the analysis of unit level data of NSS 77th round, argues that
while differences in socio-economic status based on land ownership, caste, religion, education, gen-
der, etc., which are indicative of differential well-being conditions of households, are also instrumen-
tal in inducing some other relative deprivation in terms of bargaining power in the economy. Which
to a large extent remains instrumental in the determination of final prices of the produce for farmers.
Such deprivation, though prevalent in both regulated and non-regulated markets, is evidently more
pronounced in non-regulated markets. Since access to regulated markets is not independent of the
socio-economic status of the farmers, both exclusion and discrimination based on socio-economic
status seem prevalent in the realisation of final price by the farmers, particularly in case of paddy
cultivation in India.

Keywords
Agricultural prices, public procurement, minimum support price, discrimination, exclusion

JEL Codes: Q11, Q13, Q130

I. Introduction
Price received by farmers for their agricultural produce in India is far from being solely determined by
demand and supply conditions. The agriculture markets in India have largely been segmented across

1Jindal School of Government and Public Policy, O.P. Jindal Global University, Sonipat, Haryana, India

Corresponding author:
Avanindra Nath Thakur, Jindal School of Government and Public Policy, O.P. Jindal Global University, Sonipat, Haryana 131001,
India.
E-mails: [email protected]; [email protected]
2 The Indian Economic Journal

regions and states. Because of large transportation costs with respect to total price of agricultural goods
and being perishable in nature, it is also difficult to transport agricultural commodities far across regions
having relatively large price differentials. Further, as part of agricultural price policy of the country, the
government announces minimum support price (MSP) for selected crops every year based on the recom-
mendations of the Commission for Agricultural Costs and Prices (CACP). The main purpose of MSP-
based price is to ensure some basic level of assured income for farmers as well as to provide incentives
to increase productivity of certain crops (Acharya, 1997; Sen & Bhatia, 2004). Though MSP is only the
minimum floor price, no private procurement agencies are bound by such MSP regulations (Singh &
Bhogal, 2021). Since large part of procurement of paddy happens though the private procurement agen-
cies, implementation of MSP remains extremely limited. Further, as far as government-regulated agen-
cies are concerned, procurement remained concentrated in a few states (Narayanamoorthy, 2021);
therefore, uniformity of price even in the regulated markets is far from being a reality.
Thus, the average price received by farmers is largely the outcome of intra-regional socio-economic
dynamics (Thakur, 2023). Even within a particular region, not all farmers at any point in time receive
same price for their produce. This is possible for variety of reasons. First, there are different types of
procurement agencies, and therefore buyer–seller relationship remains far from being the same across all
procurement agencies. Based on National Sample Survey data (GOI, 2019), all procurement agencies
can be clubbed into two broad categories, namely ‘regulated’ and ‘non-regulated’. The regulated agen-
cies include all types of agencies, which are largely guided by agricultural price policies of the country
with respect to offering at least ‘MSP’ to farmers. The ‘Agricultural Produce Marketing Committees’
(APMC), ‘cooperatives’, ‘government agencies’ and ‘farmers’ produce organisations’ come under price-
regulated markets.1 While all other types of procurement agencies such as ‘input dealer’, ‘local markets’,
‘local processing units’, ‘contract farming’ agencies and ‘other’ agencies come under the category ‘non-
regulated’ markets, their operations and prices offered do not come under any regulation. The ‘non-reg-
ulated’ market prices are primarily based on one-to-one negotiations of farmers with the respective
procurement agencies; therefore, final price received by farmers is largely the function of individual
bargaining power of a particular farmer, thus bound to vary across farmers. Even in the ‘regulated mar-
kets’, where largely price regulations in terms of offering MSP are applicable, there is a scope of bargain-
ing. First, MSP regulation for such markets is only expected to ensure the minimum floor price (MSP);
the actual price realised by farmers (over and above the MSP) depends on individual bargaining power
of farmers. Further, there is a provision of price differential in terms of qualitative differences of various
produce in the regulated markets. This is where the possibility of bargaining with such agencies becomes
more prominent. Sometimes, on the question of ‘quality’, farmers receive even lower than the MSP price
for their produce. Furthermore, since in most cases the local price of agricultural produce has been lower
than the MSP (Thakur, 2023), and it is not easy for farmers to sell in the regulated market, consequently,
prevalence of corruption is also possible, which in turn is reflected in the difference between official
documented prices and realised price by farmers. Since this article focuses on price realised by farmers
as reported in NSS 77th round survey, the possibility of such scenario cannot be denied. Thus, there is
scope of bargaining is regulated markets as well.
Second source of price differentials comes from temporal differences in price. Usually, there is downward
movement of price of agricultural produce at the time of harvest, and the price recovers only after some dura-
tion. Further, in case of bumper crop, such fall in the prices at the time of harvest becomes even more severe
(Deshpande, 1996). Farmers who are capable enough to hold their selling for some time might receive rela-
tively higher prices for their produce. However, holding sales is extremely difficult for small and medium
farmers as their entire expenditure depends on income received by selling the produce. In absence of any other
source of finance or income, holding produce for relatively longer period is far from being an easy option for
Thakur 3

farmers. Thus, the general price of agricultural produce, as realised by farmers, largely depends on the bar-
gaining power of individual farmers, which in turn depends on their relative socio-economic status.

Identification of Socio-economic Factors

The article is intended to explore the relationship between various socio-economic indicators of farmers
and price realised for paddy in both regulated and non-regulated markets. In other words, the article tries
to explain to what extent differences in socio-economic status of farmers are instrumental in observed
price differentials of paddy across farmers. As mentioned earlier, despite the provision of MSP, the actual
price realised by the farmers is largely the outcome of individual-level bargaining within the market
mechanism. Within the market mechanism, economic discrimination based on social groups has not
been rare. In fact, discrimination based on religion, caste and gender has been evident in the functioning
of both private and government institutional arrangements (Thorat & Newman, 2007). In this article,
four broad categories of caste, such as scheduled caste (SC), scheduled tribes (ST), other backward class
and other castes, are considered for having possible impact on bargaining for prices. Similarly, major
religious groups such as Hindu, Muslim, Christian, Sikhs and other religions are also tested for having
any influence on price received by any farmers. Both male and female categories are also included in the
analysis in the same light. In fact, as argued by Agarwal (1997), the role of social norms and social per-
ception has been important in the determination of bargaining power in the market, and such norms are
well observed in rural India, particularly with respect to gender division of the society. Thus, one expects
the difference in price realisation across gender in rural areas. Farmers with different age groups are also
considered as part of social status affecting the price bargaining in rural areas. Inclusion of age is based
on the fact that farmers in relatively young age cohorts are supposed to be more active and enthusiastic,
and therefore are expected to be more participative in the market process. However, for relatively old age
cohort, the role of experience becomes more important in the overall bargaining for final prices in rural
areas. Education has always been an important factor responsible for social empowerment and instru-
mental in the determination of freedom of ‘functioning’ and, therefore, remains foundation of ‘capabil-
ity’ development of any individual (Sen, 2002). Therefore, undoubtedly, the level of education is expected
to play an important role in the realisation of final prices for any farmers and is therefore included in the
analysis. In other words, one expects that better education level provides better understanding of the
market mechanism as well as the workings of public procurement system, thus giving higher edge in the
bargaining process. Similarly, agricultural training and awareness about MSP are also expected to play
similar role in this regard and are therefore also explored in the article. Further, to understand the impact
of class on bargaining power, farmers are divided in different land cultivation categories. In fact, Kumar
et al. (2022) find very strong positive relationship between farm size and average price received by farm-
ers in states like UP, Bihar, Jharkhand, Odisha and West Bengal. Besides, Reddy (2015) finds strong
positive relationship between farm size and profitability, and therefore differences in price realisation of
paddy are expected to play an important role behind such relationships. So, this article includes different
categories of farm size to understand how at the national level, land size is instrumental in affecting price
realisation by farmers. Access to finance or liquidity at hand during the farming process is also an impor-
tant factor affecting price realisation. As argued in the forgoing, price of the agricultural produce has
usually been lower during the harvest period and is likely to be higher once the harvest is over. In fact,
Darekar and Reddy (2017), while forecasting the future price of paddy, observed the presence of both
non-stationarity and seasonality in price trends. Consequently, farmers with enough liquidity at the time
of harvest can hold their produce for some time and might receive higher price for their products. Higher
financial risk, often associated with drought-like situations or lack of access to institutional finance,
4 The Indian Economic Journal

often leads to weakening of bargaining power of farmers (Ranjan, 2017). Thus, having bank account,
crop insurance, access to institutional loans [Kisan Credit Card (KCC)], etc., not only helps farmers to
maintain the liquidity at the time of harvest or providing with confidence in the time of crises but also
indicates financial literacy and therefore likely to improve bargaining power. Further, non-institutional
loans with high rates of interest are largely a liability for farmers and are expected to reduce bargaining
power. Sometimes, these non-institutional loans are taken from the input dealer or local money lenders
on the condition of their procurement rights over the produce at predetermined discounted prices (Path,
2008). Thus, non-institutional loans are taken as a separate factor affecting the price realisation of farm-
ers. Type of holding such as ownership of land or land under tenancy puts farmers under completely
different platform. Under tenancy holdings, farmers are expected to have lack of autonomy to participate
in the market process and are also subjected to pressure of paying rent of the land immediately after the
harvest. Reddy (2015) finds clear trends of lower profitability on rented lands compared to owned land,
even accommodating for rent paid. This indicates that tenant farmers are not witnessing same bargaining
power in the market compared to landowners. Further, most of the government procurement agencies
(mainly rural cooperatives) procure food grain with the record of land ownership, and therefore, those
with only rented land remained out of the institutional procurement process (Thakur, 2023). Further,
individual or joint ownership of lands is also instrumental in determining the decision-making power of
the farmers and is thus expected to affect the final realisation of prices, therefore identified as part of the
independent variable in the analysis. Thakur (2023) argues that one of the major reasons for state-level
differences in average price of paddy has been the level of public procurement in any state. States with
higher procurement are not only offering higher prices in the regulated market but also pull general mar-
ket prices upwards in that state. Thus, in this article, whether a farmer is selling their produce in the regu-
lated market or in a state with higher procurement ratio is also taken as independent variables to show
the impact of public procurement on individual farmer’s realisation of prices. Membership of any farm-
er’s organisation is also expected to endow any farmers with higher bargaining of price as part of the
collective bargaining process, thus included in the study.

Data, Methodologies and Description of Variables

The analysis is based on the unit-level data analysis of NSS 77th rounds on ‘Situation Assessment of
Agricultural Households’. The ‘farmers’ are identified based on farm units.2 Further, to understand the
relationship between various socio-economic factors and average price received by farmers, some basic
econometric tools such as ordinary least square (OLS) regression and logistic regressions are done. On
the basis of the discussion in the foregoing section, Table 1 describes the list of variables included in the
article. Table 1 shows the distribution of variables in the sample within the broad group. Besides, it also
explores the descriptive statistics3 of price received by farmers across each subgroup of variables.
Table 1 shows very clearly that with the variation of socio-economic factors, average values of price
realised change visibly. Such results reflect that socio-economic status of a farmers plays a key role in the
realisation of final price and therefore provides a rationale for this study. More precisely, range, mean and
median values as well as standard deviations of prices received by farmers are visibly different across
socio-economic factors identified in the model. Median values of prices across all variables are reported to
be lower than the mean value of prices, which indicates that higher percentage of farmers are receiving
lower than average mean price across all socio-economic groups of farmers.4 Besides, it is also clear that
the higher average prices are more often associated with larger standard deviations. So, the socio-economic
categories of farmers reflecting higher average prices are also showing higher variation in the price within
the group. Similar results are also evident in Thakur (2023) in case of state-level variation in prices.
Thakur 5

Table 1. The Details of Variable with Select Descriptive Statistics.

Percentage Distribution Average Median Minimum Maximum Standard
Variable (Within Broad Group) Price Price Price Price Deviation
Social group
Scheduled caste (SC) 16.3 15.2 14.5 7 70 3.63
Scheduled tribe (ST) 15.7 15.9 15 6 100 5.02
OBC 43.5 15.8 15 5 75 3.72
Others 24.5 16.0 15 0 83.3 4.69
Religious groups
Hindu 84.5 15.8 15 0 80 4.02
Muslim 10.7 14.2 13.8 0 83.3 2.81
Sikh 1.46 19.6 17.5 10 35 5.36
Christian 2.42 18.9 16 6.1 100 7.96
Other religion 0.9 17.3 15 8.33 80 6.83
Gender
Male 91.4 15.7 15 0 100 4.11
Female 8.6 16.2 15 7.86 83.3 5.0
Age group
Below 40 years 23.1 15.5 15 7 100 3.97
Between 40 and 60 years 55.3 15.8 15 0 83.3 4.14
Above 60 years 21.6 15.9 15 5 80 4.55
Operational land size
Up to 1 acre 56 15.3 14.5 0 100 3.92
1–2 acres 25 15.7 15 0 83.3 4.07
2–4 acres 13.7 16.6 15.3 5.36 75 4.58
4–10 acres 4.55 17.9 17 7 80 4.92
Above 10 acres 0.7 19.8 17.5 12 55 7.07
Educational status
Not educated 36.45 15.5 15 6 83.3 3.94
Up to primary 39.8 15.7 15 5 100 4.16
Between primary and 17.7 16.3 15 0 70 4.58
higher secondary
Above higher secondary 6.05 16.2 15 6 61.1 4.55
Whether attended agricultural training
Yes 1.5 15.8 15 8 50 4.12
No 98.5 15.7 15 0 100 4.19
Awareness about minimum support price (MSP)
Yes 36.8 16.1 15 0 100 3.89
No 63.2 15.5 14.5 0 83.3 4.34
(Table 1 continued)
6 The Indian Economic Journal

(Table 1 continued)

Percentage Distribution Average Median Minimum Maximum Standard

Variable (Within Broad Group) Price Price Price Price Deviation
Whether having bank account
Yes 97.86 15.8 15 0 100 4.19
No 2.14 15.0 14 8 75 4.34
Whether having institutional loans (Kisan Credit Card)
Yes 17.3 17.1 15.3 5.36 83.3 5.04
No 82.7 15.5 15 0 100 3.40
Whether having non-institutional loans
Yes 20.3 15.7 15 0 70 3.40
No 79.7 15.8 15 0 100 4.37
Whether having crop insurance (Fasal Bima)
Yes 6 18.2 17.5 7.5 60.2 4.86
No 94 15.6 15 0 100 4.10
Whether selling in the regulated market
Yes 9.3 19.7 17.7 9.42 50 4.31
No 90.7 15.3 14.5 0 100 3.96
States with more than 50 per cent procurement production ratio
Yes 25.6 17.0 16.4 0 62.2 4.57
No 74.4 15.3 14.5 0 100 3.96
Membership of farmer’s organisation
Yes 3.8 16.7 16 6.1 75 4.61
No 96.2 15.7 15 0 100 4.17
Type of holding
Ownership 75.1 15.9 15 0 55 4.36
Tenant 24.1 15.1 14.5 5 100 3.42
Type of farming operation
Individual 95.8 15.7 15 0 100 4.18
Joint 4.2 15.7 15 7 70 3.76
Source: Authors calculations from unit level data of NSS 77th round.

II. The Model

The following models try to understand the influence of the various socio-economic factors on the final price
received by the farmers. Model 1 is a linear OLS model in which rate received by the farmers is a dependent
variable. Model 2 is a logit model taking whether a farmers received at least MSP5 while selling their produce
in markets as a dependent variable. So here MSP is a binary variable that takes value 1 if a farmer gets at least
MSP for their produce, and if not, the value is 0. Both the models are run on same set of socio-economic vari-
ables as independent variables, the list of which is provided in the regression description in Table 2 (descrip-
tive statistics of independent variables are mentioned in Table 1). Both the models are highly significant, and
Thakur 7

Table 2. Regression Table for Linear and Logit Models.

Dependent Variable
Rate OLS (Robust) MSP MSP
Coefficient Logistic Odd Ratio of Logit Regression
Model 1 Model 2 Model 2
(1) (2) (3)
WST –0.651*** –0.509*** 0.6
(0.148) (0.109)
WSC –0.700*** –0.552*** 0.58
(0.147) (0.101)
WOBC –0.401*** –0.148* 0.86
(0.109) (0.08)
Wkcc 0.915*** 0.224*** 1.25
(0.12) (0.079)
Wfbima 0.676*** 0.520*** 1.68
(0.19) (0.119)
Worg_mem 0.172 0.243 1.28
(0.178) (0.15)
Wmsp_aw –0.423*** –0.173** 0.84
(0.084) (0.071)
Old_A 0.128 –0.095 0.91
(0.122) (0.096)
Middle_A 0.121 –0.041 0.96
(0.097) (0.079)
Wbacc 0.16 0.580** 1.79
(0.273) (0.252)
Wmus –1.258*** –1.167*** 0.31
(0.105) (0.137)
Wsikh 2.087*** 1.335*** 3.8
(0.657) (0.2)
Wchrist 3.562*** 1.298*** 3.66
(0.273) (0.118)
Wothr_rel 2.045*** 1.222*** 3.39
(0.435) (0.224)
Wreg_mkt 3.392*** 2.418*** 11.22
(0.178) (0.104)
State_hproc 0.678*** 0.523*** 1.69
(0.103) (0.077)
(Table 2 continued)
8 The Indian Economic Journal

(Table 2 continued)

Dependent Variable
Rate OLS (Robust) MSP MSP
Coefficient Logistic Odd Ratio of Logit Regression
Model 1 Model 2 Model 2
(1) (2) (3)
Wmale –0.602*** –0.278** 0.76
(0.168) (0.115)
Wprimary 0.205** 0.223*** 1.25
(0.089) (0.076)
Whsecond 0.405*** 0.356*** 1.43
(0.122) (0.091)
Whigh 0.407** 0.331** 1.39
(0.186) (0.148)
Wtrain –0.360 –0.274 0.76
(0.243) (0.254)
Windi 0.088 0.107 1.11
(0.166) (0.152)
Wown 0.549*** 0.581*** 1.79
(0.082) (0.083)
Wnoninsti –0.319*** –0.082 0.92
(0.091) (0.075)
Area_2 –0.014 0.042 1.04
(0.094) (0.078)
Area_4 0.279** 0.267*** 1.31
(0.116) (0.093)
Area_10 0.656*** 0.286** 1.33
(0.197) (0.117)
Area_10A 1.815** 0.632** 1.88
(0.876) (0.28)
Constant 15.205*** –2.621*** 0.07
(0.365) (0.319)
Observations 28,585 28,585
R2 0.156
Adjusted R2 0.155
Log-likelihood –11,125.720
Akaike inf. crit. 22,309.43
Residual std. error 169.890 (df = 28,556)
F-statistic 187.997*** (df = 28;28556) 51.02*** (Wald)
Source: Authors calculations from unit level data of NSS 77th round.
Thakur 9

Figure 1. Coefficients Plots of Linear and Logit Models.

Source: Authors calculations from unit level data of NSS 77th round.

most of the variables taken in the study are reasonably significant. For the purpose of generalisation of regres-
sion results, both the regression exercises used NSS sample weights as frequency weights. The description of
variables and their impact are discussed thematically in following sections.

Procurement Agencies and Average Price

What is essentially coming from the models is that the farmers selling their produce in the regulated
markets are getting much higher prices. In fact, odds of getting at least MSP have been extremely high
10 The Indian Economic Journal

in case of regulated markets. Numerically, those farmers who are selling their produce in the regulated
markets are getting on an average ₹3.4 higher prices than that of non-regulated markets. The odd ratio
for selling in the regulated markets has been enormously higher (11.22) than the case for non-regulated
markets. In other words, not only do regulated markets ensure significantly higher prices compared to
the non-regulated markets, but largely they get higher than the MSP price for their produce. Figure 2
shows that the average price received by farmers in the non-regulated markets has been very low (15.33)
compared to regulated markets (19.7). Further, more than 75 per cent of farmers selling in the regulated
markets are getting at least MSP for their produce, while in the non-regulated markets such percentage
is even less than 20. Therefore, selling in the regulated markets has undoubtedly been the most important
factor behind farmers getting MSP prices.

State with High Government Procurement

It is evident that farmers received higher prices in the regulated market. With higher presence of regulate
market, farmers are expected to receive higher prices on two accounts. First, as regulated markets offer
higher prices compared to non-regulated markets, therefore more farmers selling in the regulated mar-
kets, the more they are getting relatively higher prices. Second, with higher procurement of foodgrain by
state agencies at relatively higher prices also pushes general market prices in the upward direction
(Thakur, 2023). In the model, states with government procurement6 of more than half of the total reported

Figure 2. Distribution of Price Received by Farmers across Different Types of Markets.

Source: Authors calculations from unit level data of NSS 77th round.
Thakur 11

production are having positive impact on average price received by farmers. From the model, it is con-
firmed that even after controlling for procurement agencies, the average price received by farmers has
been higher in states with higher government procurement. Thus, spill-over effect of government pro-
curement in terms of raising average prices in any state is largely confirmed in the models. The same
results hold for farmers receiving at least MSP in a particular state. The odds ratio of getting MSP has
been nearly 1.7 for states with higher government procurement compared to other states. Thus, govern-
ment procurement is instrumental in raising the average price received by farmers in two ways: First,
directly through offering higher prices compared to that in non-regulated markets and second through
putting upward pressure on prices in the non-regulated markets.

Size Class of Operational Land

In the model, land ownership is categorised in five categories: Farm size between 0 and 1 acre (Area_1),
1 and 2 acres (Area_2), 2 and 4 acres (Area_4), 4 and 10 acres (Area_10) and above 10 acres (Area_10A).
The categories of farmers, that is, up to one acre of land size, have been taken as the base, and the
impacts are assessed relative to this class of farmers. The results show that in terms of receiving price
there is hardly any difference between farmers lying between 0 and 1 acre and between 1 and 2 acres. In
other words, both linear and logit models confirm that the prices received by both these categories of
farmers are not significantly different and largely their probability of getting MSP remained broadly the
same. However, the actual difference in this context starts with higher categories of farmers. The farmers
under the third category, that is, land size between 2 and 4 acres, on average received significantly higher
price than the lowest two categories of farmers. The coefficient becomes larger as the size of the farm
increases. For the largest category of farmers (with land 10 acres and above), the coefficient is nearly 1.8
for linear models. Further, the logit models confirm that the probability of getting at least MSP for this
category of farmers is nearly twice that of first two categories of farmers. In fact, such trends reflect
direct and strong relationship between land size and average price received by farmers, and the relation-
ship is significant in spite of the controlling for regulated and non-regulated markets.
It is very clear from Figure 3 that larger farmers are receiving higher prices of their produce in both
markets. However, the fundamental difference between these two markets has been that in the regulated
markets, all categories of farmers are, on average, receiving higher prices not only with respect to non-
regulated markets but also higher than the MSP for all categories. In fact, in non-regulated markets, only
the largest farmers are receiving average price higher than MSP. Therefore, land size has been one of the
most important factors explaining price differentials across farmers.

Caste and Price Differentials

Caste-based exclusion with respect to social relations in India has been a historical process, and it is argued
that being excluded from social relations can very much lead to other forms of socio-economic depriva-
tions, therefore further restricting livelihood opportunities (Sen, 2000). Such historically induced depriva-
tions like restrictions on access to land, labour, capital, credit, education and other inputs and services
necessary for commercial activity create gaps between different communities in order to provide capacities
to participate equally in the markets processes (Thorat & Newman, 2007). Deshpande (2011) argued that
while caste identity has been instrumental in differential economic outcomes, the role of markets to over-
come such limitations has been restricted; thus, she suggests active interventions in the market and non-
market institutions by the state in the form of well-articulated affirmative action policies. In this context,
12 The Indian Economic Journal

Figure 3. Land Size and Average Price Received by Farmers across Non-regulated and Regulated Markets.
Source: Authors calculations from unit level data of NSS 77th round.

this section explores to what extent the caste system alone plays an important role in the determination of
price received by any farmers. In other words, how far caste system in India has been instrumental in the
determination of bargaining power of any farmers in the agricultural markets, and what has been the role
played by the public action in terms of government procurement in this regard. It is also argued by many
that primary reasons for caste-based discrimination is routed in the economic disparity; therefore, the mod-
els control for many such economic and other related factors potentially affecting the socio-economic status
of the households and therefore influencing bargaining power of a farmer vis-a-vis selling agencies.
The linear models show that with respect to ‘other caste’ category of households, all caste groups such
as other backward castes (OBC), SC and ST received lower average price irrespective of procurement
agencies. The difference between average price received with respect to ‘other caste’ category has been
highest for both SC and ST compared to OBC. Similar results are shown in the logit models (model 2).
The probability of getting MSP is lower for all categories of households with respect to ‘other caste’
categories households; however, these probabilities are much lower for both SC and ST categories of
households. In fact, the odds ratio for both ST and SC categories of households has been around 0.6,
Thakur 13

Figure 4. The Average Price Received by Farmers across Different Caste and Class Groups in Both Regulated
and Non-regulated Markets.
Source: Authors calculations from unit level data of NSS 77th round.

which indicates that the odds of getting MSP are only little over half for both SC and ST categories of
households compared to ‘other caste’ category of households. This has extremely important implications
for variety of reasons. Such caste-based differentiation in terms of receiving average prices has been
strongly present despite controlling for land size of the households. In other words, within each category
of land size, both SC and ST categories of farmers are getting significantly lower price compared to other
caste categories (Table 4). Such difference also exists in lesser degree in case of OBC farmers. It is also
sometimes argued that it is primarily the awareness about MSP and differences in basic education levels
that are instrumental in lowering the bargaining power of any individual; however, in both models, both
awareness about MSP and education level are controlled for. This clearly indicates the presence of some
sort of discrimination across various caste categories in terms of bargaining power of any farmers. Some
possible explanations for the existence of such discrimination might be the caste-based nexus between
buyers and sellers. Since such discrimination is also true after controlling for type of markets and could
also be visible in regulated markets as well. One may think of the existence of caste-based alliance
between farmers and those representing or operating on behalf of regulated market agencies. However,
this article is not intended to explore the possible reasons for such discrimination, which needs a com-
plete understanding of caste representation in the various procurement agencies under study.
14 The Indian Economic Journal

Figure 5. Percentage of Farmers Selling in the Regulated Markets across Different Caste and Class Categories.
Source: Authors calculations from unit level data of NSS 77th round.
Note: OBC, other backward caste; SC, scheduled caste; ST, scheduled tribe.

Figure 4 indicates that for all caste categories, average price received has been substantially higher in
case of regulated markets. This is true for all land size categories of households. A few observations here
are worth noting. First, the gap between average prices received by ‘other category’ households and
‘OBC’ households has been lower than that for both ‘ST’ and ‘SC’ categories of households. Second, for
small land possession categories, ‘ST’ category households are better than ‘SC’ households in terms of
receiving average price of paddy in both regulated as well as non-regulated markets. Since ‘ST’ popula-
tions are largely concentrated in a few states and some very specific provisions are made in state-level
policies with regard to government procurements in such states, therefore, at all-India level, a better situ-
ation of ‘ST’ households compared to that of ‘SC’7 households are not unexpected. Third, such caste-
based gaps in terms of receiving average prices of their produce become nearly insignificant in the
regulated markets. And finally, all caste categories of households are getting on average higher than MSP
price for their produce in regulated markets. It is also evident that the change in the average price
received by ‘SC’, ‘ST’ and ‘OBC’ categories improved more remarkably compared to ‘other’ caste
groups in regulated markets compared to non-regulated markets. Notably, in the non-regulated markets,
the ‘ST’ category of households received more or less stable prices across different sizes of land hold-
ings. So, improvement in the average price received in this category remained lower compared to all
other categories with the increase of land holdings. For largest category of land size, the improvement in
average price received by farmers has been highest for ‘SC’ category. Further, it is not only that ‘SC’
Thakur 15

Figure 6. The Average Price Received by Farmers across Various Religious Categories in Both Regulated and
Non-regulated Markets.
Source: Authors calculations from unit level data of NSS 77th round.

category households largely received lower prices in the regulated markets compared to all other catego-
ries of households, but the access to the regulated markets for them remained relatively limited as well.
Figure 5 shows that within all land size categories of households, the percentage of households selling in
the regulated markets has been lowest in ‘SC’ caste group. So, there seems to be some kind of exclusion
as well as discrimination prevailing in the regulated procurement agencies as well.

Religion and Price

The difference in average price across religions seem extremely significant. The results show that with
respect to the Hindu religion, the average prices received by Muslim community have been significantly
lower. In fact, such differences have been more pronounced in case of receiving MSP. The probability of
receiving at least MSP price has been much lower for Muslim community compared to Hindu. In fact, the
odds ratio for Muslim community receiving at least MSP has been as low as less than one-third that of the
case for Hindu community. The relative deprivation in socio-economic status of Muslims compared to
other religious communities is well recognised in Indian Context (GOI, 2006; Islam 2019, etc.). However,
the difference in the context of price received by farmers is important as land ownership and education
status are controlled in the model. Thus, it seems that the gap between Muslims and Hindus is largely
16 The Indian Economic Journal

appearing due to religious networking that might exist between buyer and seller communities. However, in
contrast, all other religious communities like Sikhs, Christians and other religious communities like Jains,
etc., are not only receiving higher prices compared to Hindu communities, but their probabilities of getting
MSP have been remarkably higher as well. For other religions, the distribution of population is largely
concentrated in some specific states, and therefore state-specific factors could be responsible for such dif-
ferences.8 For instance, the Christian community is the majority in states like Nagaland, Mizoram,
Meghalaya, Arunachal Pradesh, etc., where terrain is largely hilly and cultivation costs of paddy are higher;
therefore, there is a possibility of higher general price level. Sikhs are a majority community in Punjab, and
Punjab has its own socio-economic-political structure, so state-specific factors might be instrumental in
Punjab as well. Jains are small community largely involved in business, so they might have some advan-
tages with regard to bargaining power in the markets. However, the case of Muslims is not the same as the
other religions. The distribution of Muslim communities is largely spread over majority of states in the
country, and therefore, differences in prices between Muslims and other religions might not be totally
explained through state-specific or some other external factors. Evidently, in the non-regulated markets, the
average price received by Muslim cultivators has not only substantial lower than other religious communi-
ties, but on average, their average prices have also remained much lower than MSP. Such trend is there for
all categories of farmers. In fact, in the regulated markets as well, except one category of farmers (areas
between 2 and 4 acres of land), all Muslim cultivators received substantially lower than the average prices
received by other religious communities including Hindus. Further, in the regulated markets, even larger
Muslim farmers with the cultivation of 4 or more acres of land are not only receiving lower average prices
than other religious communities, but their average has been significantly lower than MSP as well. Thus,
the gap between all religious communities and Muslims has been even substantially higher both in terms of
average prices and probabilities of receiving at least MSP. Thus, existence of such strong and significant
gaps between average prices received by farmers of different religious communities, even after controlling
for land size, education level and training status, indicates that some levels of religion-based favour/dis-
crimination are there in the markets for paddy in the country.

Status of Education and Training

The models also test whether education and agricultural training are having positive impact on the bargain-
ing power of the farmers in terms of receiving higher prices in general. As expected, the models show that
farmers with primary education are in general receiving higher prices compared to illiterate farmers. The
magnitude of the coefficients becomes even larger with secondary and above education level. Though
farmers with secondary or higher education are not much different in this regard. This confirms that the
bargaining power of farmers increases with the education levels till the secondary level and after that it
remains more or less stable. The result is largely intuitive, as with the basic education level farmers under-
stand MSP and basics of markets structure so they bargain for higher prices. However, higher education is
hardly expected to enable farmers in this regard any better. More surprising results in models are related to
the impact of agricultural training provided by various government agencies on the average price received
by farmers. Both awareness about MSP and agricultural training have largely negative impact on the aver-
age prices received by farmers. While the relation between agricultural training and price received is not
significant, awareness about MSP is having negative and significant impact on the average prices. The
results are to some extent surprising, and possible explanation lies in the fact that large part of such impact
could only be explained through the level of education already included in the models and there is no room
for the impact of such training and awareness beyond the degree of education. In fact, more details analysis
Thakur 17

regarding the same could only be done by evaluating the quality and focus of such training and therefore is
beyond the scope of this study. Nevertheless, negative impact of awareness about MSP on agricultural
prices is something puzzling and therefore needs further analysis.

Type of Holding
In models, two kinds of holdings are included. First, whether the farm is individual owned or has joint
ownership. It is largely expected that it is easier to take decisions with individual ownership and there-
fore might be influencing the price received by cultivator households. However, the models show that
there is no effect of type of ownership in terms of individual ownership or joint ownership on average
price received by farmers. However, in terms of owned land or rented land, the difference in price is
significant and positive. So, farmers with owned land are on average receiving higher prices for their
produce. This is quite expected as well. In case of rented land, the pressure for the tenant farmers is
always there to pay the rent first and on time. So, it is difficult for them to wait for prices to rise or to
hoard till they get reasonable price for their product. Further, their margin of profit is very low as they
have to pay rent on their land; therefore, it might be difficult for tenant to wait for long in order to get
higher prices. Therefore, with same level of cultivation, tenant farmers are likely to receive lesser prices
compared to owned land, keeping other things constant.

Financial Indicators
In the models, four major indicators are taken as part of financial indicators possibly affecting prices reali-
sation by farmers. First, among them has been the farmers having bank accounts. With Prime Minister Jan
Dhan Yojana9 (PMJDY), the proportion of farmers having bank accounts has increased significantly, and
this is supposedly the first basic step towards financial inclusion. It is expected that with better financial
integration and financial literacy, the ease to transacting in the economy improves and therefore farmers can
get higher bargaining power in the markets. Further, in case of official transactions, its leaves no room for
financial mediators and direct transfer of money is possible. The first model shows that, as far as bargaining
power is concerned, there is hardly any impact of just having bank account. The coefficient in the linear
model is not significant. Model 2 shows that the probability of getting at least MSP is significantly higher
for farmers with bank account. This is also largely expected as in case of regulated market transactions, the
money transfer happens largely through bank account. So, with bank account, farmers become eligible to
sell in the regulated markets and their probabilities of getting at least MSP are higher. Therefore, bank
account helps farmers more in terms of receiving at least MSP for their produce and hardly having signifi-
cant linear relationship with prices. The second indicator here in models is KCC.10 Since KCC is formal
subsidised loan given to the farmers largely for the purchase of inputs or to meet other farm-related expen-
diture, it is expected that with more liquidity in hands, farmers bargaining power with various procurement
agencies might be higher or farmers, in case of lower price at the time of harvest, could wait for the price
to rise without being cash drained. As expected, the coefficient of KCC on average price received by farm-
ers is positive and significant. In fact, model 2 also confirms that the probability of receiving at least MSP
is higher for farmers having KCC. Thus, access to institutional credit in terms of KCC helps farmers not
only to receive higher prices but also to get at least MSP for their produce. However, other indicators within
this category have been indebtedness from informal sources. Though loans from informal sources are also,
in some sense, access to credit and liquidity, they are indeed a costly source of fund, with rate of interest on
such loans are extremely higher. The models confirm that there is a negative and significant relationship
18 The Indian Economic Journal

between non-institutional loans and average price received by farmers. Such negative relationship can be
explained by the fact that with such a high rate of interest, any farmers would not wait for price of their
produce to rise and then sell; rather, they would prefer to sell their produce at the prevailing rate in order to
get rid of such loans bearing extremely high rate of interest. Thus, on the one hand, institutional loans give
some empowerment to farmers in terms of better bargaining power, therefore receiving higher price in
general; on the other hand, loans from non-institutional reduce their bargaining power, and therefore, on
average, farmers receive lower average price for their produce with such loan. Lastly, crop insurance has
also been taken as one of the indicators of financial services. Both models show a positive and significant
coefficient of crop insurance as part of Pradhan Mantri Fasal Bima Yojana (PMFBY).11 So, farmers with
insurance coverage are more secure; therefore, it is possible to have more confidence in bargaining. Further,
those who are eligible and avail the scheme are financially more aware and therefore naturally have better
bargaining power in the economy. In fact, similar explanation is valid in case of higher probabilities of
farmers getting at least MSP with crop insurance compared to group of farmers without crop insurance.
Further, crop insurance also gives some sense of security to farmers against possible crop loss; therefore,
such securities might be the reason for having more confidence in the overall bargaining power.

Age and Gender

The role of experience in agricultural operations is well acknowledged, and in this context age factors
are expected to play an important role. However, the models show that neither the young generations
with active age nor the experienced farmers under middle and old age are having important differences
in terms of getting higher prices for their farm products. Thus, as far as bargaining power is concerned,
both in terms of receiving higher prices and to get at least MSP, there is hardly any role of age and related
experience. Coefficients of both middle age and old age binary variables are not significant. However,
more surprisingly, gender plays an important role in affecting average price for farm products. Farmers
with male head received lower average price for their farm products compared to their female counter-
part. Coefficients are both negative and significant in both linear as well as logit models, confirming that
farmers with female heads are not only receiving higher prices but also their probabilities of receiving at
least MSP have been higher compared to their male counterparts. This seems a bit surprising as in rural
areas there are only a few farmers with female head, and the entire system of procurement has been male
biased. There might be some regional or state-specific factors responsible for such differences. On a
positive note, one may think of more formal and sincere approach being followed by female farmers
compared to their male part as a possible explanation for such trends.

Organisational Membership
Contrary to expectations, the results show that the membership of farmers’ organisation has hardly played
any role in influencing prices received by farmers. The coefficients of binary variable ‘whether member of
a farmer’s organisation’ are not significant in both models. In fact, more importantly, logit models also
confirm that there is no significant impact of membership of farmer’s organisation in terms of receiving at
least MSP. It implies that either the farmers’ organisations are not targeting price as an actionable goal or
the entire organisation has been relatively weak in terms of its collective bargaining strength.
Thakur 19

Summary and Concluding Remarks

It is evident from the analysis that the socio-economic status of the households plays an important role
in the realisation of price of the agricultural produce. Upper caste and large landholders are the groups
having highest bargaining power in the economy, and therefore, they are not only receiving higher prices
for their produce in the local markets but are also successful in getting higher prices in the regulated
markets as well. Surprisingly, religion also plays a significant role in the determination of final price. In
fact, relative disadvantage of Muslim community in the bargaining process, both in regulated and non-
regulated markets, indicates some degree of exclusion and discrimination for Muslim compared to all
other religious communities. The impact of education seems much more important in enhancing the rela-
tive bargaining power the farmer compared to general awareness about the MSP, agricultural training of
farmers and any age-related experience of farmers. However, a surprising element of the result is that
farm households with female head are more likely to receive higher price for their produce, and their
chances of receiving at least MSP are also higher than their male counterparts. Since the proportion of
farmers with female head has been very small and concentrated in a few pockets, any concluding state-
ment would not be justifiable without doing deeper study. As far as type of landholding is concerned,
there is hardly any difference found between farmers with joint ownership or individual ownership in
terms of average price realised by farmers; however, there is a significant difference lies between farmers
with ownership holdings and with tenancy. Farmers with ownership holdings reflected relatively higher
bargaining power in both regulated and non-regulated markets and therefore received significantly
higher prices compared to tenants. Further, their probability of receiving at least MSP has also been
much higher compared to tenant farmers. Some financial indicators such as access to crop insurance,
access to institutional credits like KCC, and having bank account to some extent affect the confidence of
the farmers; and they also provide some ease in terms of access to regulated markets and therefore affect
average prices in a positive manner. However, non-institutional loans with relatively high rates of inter-
est are largely been the liability for farmers and therefore weaken the bargaining power of the concerned
farmers. Further, farmers’ organisation have hardly been helping farmers to get higher price or even to
provide access to regulated markets for selling their produce. So, one suspects weak organisational struc-
ture and less collective bargaining power of such organisations.
Thus, while differences in socio-economic status in itself indicate differential well-being conditions
of households, they are also instrumental in inducing some other relative deprivation in terms of bargain-
ing power in the economy. Such deprivation, though prevalent in both regulated and non-regulated
markets, is evidently more pronounced in non-regulated markets. Since access to regulated markets is
not independent of the socio-economic status of the farmers, both exclusion and discrimination based on
socio-economic status seem prevalent in the realisation of final price by the farmers, particularly in case
of paddy cultivation in India.

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship and/or publication of
this article.

Funding
The author received no financial support for the research, authorship and/or publication of this article.
20 The Indian Economic Journal

Notes
1. NSS 77th round categorises procurement agencies as local markets, input dealers, cooperatives, government
agencies, local processing units, contract farming and other agencies. This article categorise all mentioned
agencies in two categories as ‘regulated and non-regulated markets’ as defined in the text.
2. In the analysis, gross farms are taken as the unit of analysis. For instance, if the same farmland is cultivated
twice, it is counted two times. However, since only paddy cultivation is taken for the study, such repetitions are
very limited as a proportion of total farmers.
3. All values other than ‘range of price’ are weighted values where NSS sample weights are applied while
calculating the averages.
4. For detail description of distribution of price, see Thakur (2023).
5. The MSP for the year 2018–2019 was 17.5 for normal variety of paddy and 17.7 for higher variety of paddy.
However, the study takes the conservative values 17.5 as MSP for that year, and the results are based on this
value.
6. States such as Haryana, Chhattisgarh, Punjab, Kerala, Andhra Pradesh, Telangana and Odisha procured more
than half of the total production during 2018–2019. For details, see Thakur (2022).
7. As ‘SC’ households are spread over large part of the country and therefore comes under general provisions
applied in majority of the states.
8. For details of religion-based distribution of population, see https://www.thehindu.com/news/national/religious-
communities-census-2011-what-the-numbers-say/article7582284.ece.
9. PMJDY is a National Mission on Financial Inclusion encompassing an integrated approach to bring about
comprehensive financial inclusion of all households in the country. The plan envisages universal access to
banking facilities, with at least one basic banking account for every household, financial literacy, access to
credit, insurance and pension facility. For details, see https://pmjdy.gov.in/scheme.
10. The KCC scheme was introduced in 1998 for issue of KCCs to farmers on the basis of their holdings, for
uniform adoption by the banks so that farmers may use them to readily purchase agriculture inputs such as
seeds, fertilisers and pesticides, and draw cash for their production needs. The scheme was further extended
for the investment credit requirement of farmers, namely allied and non-farm activities, in the year 2004. The
scheme was further revisited in 2012 by a working group under the chairmanship of Shri T. M. Bhasin, CMD,
Indian Bank, with a view to simplify the scheme and facilitate issue of Electronic KCCs. The scheme provides
broad guidelines to banks for operationalising the KCC scheme. Implementing banks will have the discretion
to adopt the same to suit institution/location-specific requirements. For details, see https://www.rbi.org.in/
commonman/English/Scripts/Notification.aspx?Id=2311.
11. PMFBY scheme was launched in India by Ministry of Agriculture & Farmers’ Welfare, New Delhi, from Kharif
2016 season onwards. National Insurance Company started participating in PMFBY from Rabi 2016 onwards
and covered eight states and two union territories during the past five seasons, namely Rabi 2016–2017, Kharif
and Rabi 2017 and Kharif and Rabi 2018 covering 7,027,637 farmers.

ORCID iD
Avanindra Nath Thakur https://orcid.org/0000-0002-2914-0256

References
Acharya S. S. (1997). Agricultural price policy and development: Some facts and emerging issues. Indian Journal
of Agricultural Economics, 52(1), 1–47.
Agarwal, B. (1997). “Bargaining” and gender relations: Within and beyond the household. Feminist Economics,
3(1), 1–51.
Darekar, A., & Reddy, A. (2017). Forecasting of common paddy prices in India. Journal of Rice Research, 10(1),
71–75.
Thakur 21

Deshpande, A. (2011). The grammar of caste: Economic discrimination in contemporary India. Oxford University
Press.
Deshpande, R. S. (1996). Demand and supply of agricultural commodities: A review. Indian Journal of Agricultural
Economics, 51(1–2), 270–287.
GOI. (2006). Prime Minister’s high-level committee. Cabinet secretariat. Government of India. https://minorityaf-
fairs.gov.in/WriteReadData/RTF1984/7830578798.pdf
GOI. (2019). Situation assessment of agricultural households and land and livestock holdings of households in rural
India, 2019, NSS 77th round.
Islam, M. (2019). Indian Muslim(s) after liberalization. Oxford University Press.
Kumar, A., Sonkar, V. K., & Bathla, S. (2022). Does public procurement benefit paddy farmers? Evidence from
Eastern India. Economic & Political Weekly, 57 (21), 42–48.
Narayanamoorthy, A. (2021). Farm income in India myths and realities. Oxford University Press.
Ranjan, R. (2017). Challenges to farm produce marketing: A model of bargaining between farmers and middlemen
under risk. Journal of Agricultural and Resource Economics, 42(3), 386–405.
Path, V. B. (2008). Agricultural indebtedness: Crisis and revival. Economic & Political Weekly, 43(5), 47–52.
Reddy, A. A. (2015). Regional disparities in profitability of rice production: Where small farmers stand? Indian
Journal of Agricultural Economics, 70(3), 260–271.
Sen, A., & Bhatia, M. S. (2004). Cost of cultivation and farm income in India. Academic Foundation.
Sen, A. (2000). Social exclusion: Concept, application, and scrutiny (Working Paper, Social Development Paper No
1). Asian Development Bank.
Sen, A. (2002). Development as freedom. Anchor Books.
Singh, S., & Bhogal, S. (2021). MSP in a changing agricultural policy environment. Economic & Political Weekly,
56(3), 12–18.
Thakur, A. N. (2023). Public procurement, land ownership and price variation across states: A case of paddy cultiva-
tion in India. Agrarian South: Journal of Political Economy, 12(3), 319–351.
Thorat, S., & Newman, K. S. (2007). Caste and economic discrimination: Causes, consequences and remedies.
Economic & Political Weekly, 42(41), 4121–4124.