18 From Plate to Plow: Agricultural Diversification

in India
P. K . J O S H I A N D A S H O K G U L A T I

The sustained growth of India's GDP at 5 to 6 percent per annum and of its per
capita income at about 3.5 percent per annum over the past two decades has
brought about major changes in consumption patterns across the country. Pref-
erence is shifting from basic staples toward high-value agriculture such as
fruits, vegetables, and dairy, poultry, and fishery products. The dietary mix and
flavor are changing fast on the plates and palates of Indian consumers.
Interestingly, this change is happening not only in the upper-income
brackets of the Indian population but even below the so-called poverty line. For
example, the National Sample Survey Organization (NSSO) estimates that the
per capita consumption of cereals by people below the poverty line declined by
10 percent over the period 1983 to 1999-2000. On the other hand, their con-
sumption of milk increased by 30 percent, of vegetables by 50 percent, of fruits
by 63 percent, and of meat, eggs, and fish by 100 percent over the same period
(Table 18.1). These percentage changes in the consumption of high-value agri-
cultural products in the poorest segments of the population point to a silent
revolution.
This revolution is also reflected in the rapidly rising exports of high-value
agriculture, especially fruits and fish, during the past two decades (Figure 18.1).
Given the perishable and high-value nature of these commodities, the growth
in their export has strong implications not only for producers but for financiers,
processors, exporters, and the retail chain industry, too. It also has repercussions
for the institutional innovations that are emerging to link the plate to the plow—
efforts to link changing consumer preferences and rising exports of high-value
agriculture to the production decisions of the growers.
What could be driving this silent revolution? Besides rising incomes,
changes in relative prices between cereals and high-value agricultural products,
increasing urbanization and infrastructure, and more open trade policies are be-
hind this changing scene (Kumar and Mathur 1996; Joshi et al. 2002; Kumar and
Mruthyunjaya 2002).
Changing relative prices are resulting from a mix of technology impacts
as well as changing demand pressures. While the green revolution (wheat and

384
 r~- cn s q 1-; cn
CO
Tf en Os >n d OS
cn OS

CN Tf cn s q cn
od Tf OS cn SO

5 so

o CD
o
o s 00 CN

3!
SO s o
o
Tf so cn d od
a §• i n <—i *—i OS <—<
os CD 60
OS
OS

6
o
^ Os Os C N «n oo
£ = cN ^ ) -st d ri
T O
CD 00
o

Tf OS CN i n Tf
CO
CN SO d SO i n
4P in T f in

^ ! cn o 00 p Os
cn so cN cn
5 Tf cn

t-; en CN Tf t~- 00
t~ i n SO Os Tf
so oo
CD

—< s os o o s q r~; o s
r~ r~ CN s o — < ini —^
Tf en

a
CO
ce
73

CO co
r3 <D 00
w g co 3 °
00
t— co . CD ^
8
B a g oo sa a^
S 3 S
r

o £ w > £ 2
386 P. K. Joshi and Ashok Gulati

FIGURE 18.1 Indian exports o f nontraditional crops, 1980s and 1990s

U.S.S million
1.200 I
• T F 1981-82
• TE 1991-92
900 - • TE 1999-2000

600 V
461

Eggs Processed Meat Fruits and Fish

fruits vegetables

SOURCE: Joshi et al. (2002).

rice) technology was running out of steam in the 1980s, technological and mar-
keting boosts were given to dairy products, fruits, vegetables, poultry, fish, and
so on. Thus, this high-value segment of agriculture, within and outside the crop
sector, started increasing its share. Sizable changes took place within the crop sec-
tor, with the share of foodgrains, the hallmark of food security, giving way to
high-value nonfoodgrains. From the triennium ending (TE) 1981-82 to TE
1998-99, the share of foodgrains in the value of the output of the crop sector
fell from 48 percent to 40 percent, while in terms of area it fell from 70 percent
to 65 percent over the same period (Joshi et al. 2002).
By the end of the 1990s, surpluses of cereals and storage costs were rising
rapidly and putting greater pressure on the fanning community and policy-
makers to explore possibilities for a more remunerative and viable alternative
production mix. Diversification of agriculture in favor of noncereals and high-
value commodities such as fruits, vegetables, milk, meat, eggs, fish, and so on
continues to offer such an alternative. These commodities are also emerging as
promising sources of income augmentation, employment generation, poverty
alleviation, and export promotion (Jha 1996; Chand 1996; Vyas 1996; Delgado
and Siamwalla 1999; Ryan and Spencer 2001; Joshi et al. 2002).
Therefore, it is important to diagnose the production-consumption linkages
in the context of agricultural diversification. This requires identification of the
driving forces that alter the production portfolio and the consumption basket.
It is important to understand how the production portfolio is being trans-
formed in response to changes in the consumption basket, in a scenario where
smallholders dominate Indian agriculture and the majority of consumers live in
rural areas. The available evidence shows that the primary production centers
of high-value commodities are largely concentrated with smallholders. There
 From Plate to Plow 387

is also evidence to show that small farm holders are relatively more efficient
in producing these commodities (Jha 2001). Unfortunately, due to their tiny
amounts of marketable surplus and lack of access to appropriate markets and
information, their transaction costs are too high. This does not permit them to
take full advantage of the changing scenario in domestic consumption patterns
and rising exports of high-value products. Therefore, it is imperative to estab-
lish strong and cost-effective linkages between plow and plate and to examine
the role of innovative institutional arrangements for integrating production and
consumption.
In the first section of this chapter, we trace the nature of agricultural di-
versification in some detail, then examine what is driving that trend. Thereafter
we probe consumption patterns and review the types of institutional arrange-
ments that are emerging between growers and processors or exporters in some
selected segments of Indian agriculture. In the final section we highlight some
policy implications.

Agricultural Diversification in India

Our Approach
We make a shift from the usual definitions of agricultural diversification to a
definition that emphasizes movement of resources from a low-value commod-
ity mix (mainly foodgrains) to a high-value commodity mix (vegetables, fruits,
livestock, fishery products, and so on). The information we use is collated from
1

three ongoing studies on (1) constraints and opportunities related to agricultural

diversification in south Asia, (2) diversification in food baskets, and (3) inno-
vative institutions for accelerating diversification on small farms. Our analysis
covers the decades of the 1980s and 1990s. A comparison between the two
decades is expected to provide some useful insight into the implications of eco-
nomic reform for agricultural diversification and the consumption basket.

The Nature, Speed, and Determinants of Diversification

Those promoting agricultural diversification in India are gradually choosing to
emphasize high-value crops and livestock activities that augment incomes
rather than resorting to a coping strategy to manage risk and uncertainty. Crops,
livestock, fisheries, and forestry constitute the core subsectors of agriculture.
The crop subsector is the principal source of income in agriculture, followed by
the livestock sector (Joshi et al. 2002). There is strong synergy in the crop and
livestock subsectors, which are complementary to one another. The fishery sub-
sector is prominent in the coastal areas, forestry in the hilly regions.

1. The usual definitions are (1) shift of resources from farm to nonfarm activities and (2) use
of resources in a larger mix o f diverse and complementary activities within agriculture.
388 P. K. Joshi andAshok Gulati

The share of the crop subsector in the agricultural GDP marginally de-
clined during the 1980s (from about 76.25 percent in TE 1981-82 to 73.65
percent in TE 1990-91), then slowly recovered during the 1990s (rising to 74.91
percent in TE 1998-99) (Joshi et al. 2002). This happened due to the normal
monsoon rains during most of the 1990s and the greater emphasis on and higher
level of production of horticultural crops. On the other hand, there was an in-
crease in the share of the livestock subsector during the 1980s, from about 18 per-
cent in TE 1981-82 to 23 percent in TE 1990-91. Later, though the value of
livestock during the 1990s nearly doubled, its share in the sector remained stag-
nant at 23 percent. This was because the value of the larger crop subsector in-
creased relatively higher than that of the livestock subsector and hence masked
the latter's performance. The same was true for the fishery subsector, the value
of which swelled by about 50 percent during the 1990s, though its share in agri-
cultural GDP declined marginally, to about 1 percent in TE 1998-99 from 1.35
percent in TE 1990-91.
DIVERSIFICATION WITHIN T H E CROP SUBSECTOR. The trends show that non-
foodgrain crops have gradually replaced foodgrain crops, with the area of the
former increasing from about 30 percent of the total crop area in TE 1981-82
to 35 percent in TE 1998-99. This trend was more pronounced in terms of value—
from about 52 percent to 60 percent (Joshi et al. 2002). Coarse cereals were
mainly replaced by nonfoodgrain crops such as oilseeds, fruits, vegetables,
spices, and sugarcane.
Cereals continued to dominate the foodgrain crops and accounted for more
than half (53 percent) of the gross cropped area in TE 1999-2000, compared to
59 percent in TE 1981-82. Crop diversity within the cereal sector declined dur-
ing the past two decades, much faster during the 1990s than in the 1980s. The
sown area and production of rice, wheat, and maize increased, while the sown
area of barley, millet, and sorghum decreased rapidly. The expansion of the area
sown in wheat and rice was mainly on account of the availability and large-scale
adoption of remunerative and stable technologies as well as favorable and as-
sured government policies on the prices and procurement of these crops. Maize,
on the other hand, emerged as an important crop mainly to meet the require-
ments of the booming poultry sector. The availability of improved maize hy-
brids, the flexibility of the maize growing season, and the diverse uses of maize
were responsible for its area expansion. The crop is also finding niches in non-
traditional areas (e.g., the southern part of India) and seasons (e.g., there is now
winter maize). Noncereals, including pulses, were gradually moving toward
nontraditional areas, too, and silently increasing.
A swift diversification of agriculture was noted in favor of oilseeds, vege-
tables, and fruits. Oilseed production jumped remarkably, from 18 million tons
(mt) in TE 1981-82 to 30 mt in TE 1991-92, and touched 40 mt in TE 1999-
2000. The annual compound growth rate of oilseed production was quite im-
pressive, at 5.35 percent during the 1980s, which slowed down to 2.31 percent
 From Plate to Plow 389

in the 1990s (Table 18.2). The sown area and production of the majority of
oilseed crops increased substantially during 1980s, while in the 1990s only
soybeans, coconuts, rapeseed, and seed cotton gained in area, and groundnuts,
sunflower seed, and linseed lost. The remarkable success in oilseeds was a re-
sult o f the Technology Mission on Oilseeds (TMO) launched by the Indian
government in 1986 to meet domestic demand and control imports of edible
oils. The mission encompasses a blend of improved technologies and favorable
policies to augment production. Despite the acclaimed success of oilseed pro-
duction, the country is not globally competitive in the edible oils sector and
faces a severe threat to its domestic oilseed producers, so the tariffs on edible
oils had to be raised from 15 percent to more than 70 percent during recent
years. Sustaining the success of the TMO will depend on how well technical ef-
ficiencies can be increased at the production and processing levels.
During the past two decades, the sown area and production of vegetables
increased considerably, with the subsector diversifying toward new areas, new
crops, and new seasons. India is the second-largest producer of vegetables, next
to China. Output grew at an annual rate of 2.53 percent during the 1980s, then
slowed to 1.99 percent per annum in the 1990s. Yield increase contributed sig-
nificantly to higher levels of production in the 1980s, while area expansion was
important in the 1990s. The vegetables sector is becoming strong in the urban
peripheries and emerging as an important source of income augmentation for
small farm holders in water-scarce regions due to the massive subsidies extended
by the government on water-saving devices (e.g., sprinkler and drip systems).
In addition, the watershed programs gave high priority to vegetable production
to enhance the efficiency of conservation of the scarce water in the rain-fed areas.
Fruit production (both fresh and dry) is gaining importance. Production
grew at an annual rate of 6.3 percent during the 1990s, rising from about 3 per-
cent in the previous decade (Table 18.2). A large share (approximately 60-65
percent) of increased fruit production in both decades was realized through pro-
ductivity gains. Indian desheri and alfonso mangos have an excellent export
market. Mango production increased by 67 percent in the 1990s, as did the pro-
duction of bananas, oranges, grapes, apples, papayas, and pineapples. Dry fruits
and spices also gained over the past two decades. Higher fruit production was
a result of the changing diet of the high-income group and, on the supply side,
of the government's initiatives in food processing. During the mid-1980s, a new
ministry on food processing was set up to strengthen agroprocessing, reduce
postharvest losses, and enhance value addition. Private sector participation in
fruit processing is growing, though slowly.
In summary, during the 1980s, oilseeds, fruits, and vegetables performed
impressively in all regions of India. While the TMO resorted to high levels of
protection to encourage the expansion of oilseed production, the government
opted for the establishment of the National Horticultural Board in 1984 to pro-
mote the integrated development of horticulture by coordinating, stimulating,
 CN©—< O c n CN © ©

O 00 i f oo oo m r - r - o\
CN © o s c n i n oo oo cn
0 0 © . - H C N © O O ©

T3
c n c N © O S C N f f r — -*
o u ^ ^ i n O O s S O s q c N
m r N n t H r n ' m o d i N

ca

g c n c N » n c N c N < n © c n
a
<
© o o o s — < i l - «n cn as os

© © ' — ; © ' - H — ; rS —

•o © c N ' - i r - r ~ c N © 0 ' *
§o 0 \ t t - - i O l O s t O c C | 0 0
—„ ^ *n ©^ ©^
oo
0 0

CM a rn \ o cn o f _T ^
CN cn
-o "3
©
a o
3 o
3 S
5 •§
o s t - - ' ^ - o o o o o o t - - s o ©
h - H o m c n ^ t ^ ^ r -
a c N o o © r - - s o o o t — t-~"*

<D t3
ED
00
a ©
_ <
r>j ED
^ © © •n o s t— ^t" s o C N
3 o <n oo t— s o cn C N *n i f so
c n r— s o O CN so cn so f~
JS o it- C N s o cn CN — —
o CN © CN CN
"3 os

I?
,<D a
CN
00
I
33 a
5 tfl U ™ u aa
g <U t5 .
CD

£1 5 42 O
43 l>
eo ' W Q,
 From Plate to Plow 391

and sustaining the production and processing of fruits and vegetables. To fur-
ther promote the horticulture sector for domestic and global markets, adequate
attention must be given to the development of infrastructure and effective qual-
ity control.
DIVERSIFICATION WITHIN T H E L I V E S T O C K SUBSECTOR. The livestock
subsector is also growing rapidly. Its share in the total value of agricultural out-
put is progressively rising (Birthal and Parthasarathy 2002). Milk accounted for
around 68 percent of the total value of livestock products during the past two
decades (Joshi et al. 2002). The remaining 32 percent of livestock products is
distributed over several items, such as meat, poultry, wool, and so on. Milk pro-
duction more than doubled, from 33 mt to 71 mt over TE 1981-82 to TE
1998-99, with an annual compound growth rate of about 4.62 percent. The
growth rate of milk production was much higher, at 5.23 percent in the 1980s
than in the 1990s (3.46 percent). Such a breakthrough was aided by the Oper-
ation Flood Program, which was launched to accelerate the progress and rapid
development of the dairy sector.
The meat and poultry subsectors registered good performance, increasing
from a low of 0.80 mt in TE 1982-83 to 2.73 mt in TE 1991-92 and finally to
4.41 mt in TE 1998-99, with an annual compound growth rate o f about 5.81
percent in the 1980s vis-a-vis 3.90 percent in the 1990s. Meat's share in the
value of livestock grew from 6.6 percent in 1982-83 to 8.4 percent in 1989-99
(Birthal and Parthasarathy 2002). The large increase in meat production over the
1980s was due partly to the severe drought in 1987 in most parts of the country.
A n acute shortage of green and dry fodder forced the disposal of less produc-
tive animals at a large scale. Poultry, too, flourished during the 1980s, with its
share in the value of livestock rising from 7.8 to 9.6 percent between 1982-83
and 1998-99 (Birthal and Parthasarathy 2002). The share of poultry and goat meat
in the total value of meat production went up from 66 percent to 77 percent.
Similarly, egg production increased by 8.46 percent annually in the 1980s as
against 4.60 percent annually in the 1990s. Unlike the dairy sector, the poultry
sector grew with the help of the private sector, which controls roughly 80 per-
cent of total production.
The future of the livestock sector is quite promising, because it still has a
huge potential to raise the production, consumption, and export of different live-
stock commodities. Meat production, for instance, is mostly confined to the
unorganized sector and is just waiting for modern slaughter facilities and the
development of cold chains to take off in a big way.
DIVERSIFICATION WITHIN T H E FISHERY SUBSECTOR. The fishery subsector
has seen a gradual shift from marine to inland fisheries. Traditionally, marine
fisheries, accounting for more than 75 percent of the total production in 1960-61,
used to dominate fish production. In recognition of the importance and poten-
tial of the fish sector in the inland areas, a greater impetus was accorded to in-
land fisheries. The share of marine fisheries in the sector's total production fell
392 P. K. Joshi and Ashok Gulati

to about 54 percent in TE 1999-2000, while that of inland fisheries rose to about

46 percent, from less than 25 percent in 1960-61. The annual compound growth
rate of inland fisheries was higher, at 6.54 percent, in the 1990s than in the
1980s, when it was at 5.27 percent. Marine fish production, which performed
poorly during the 1980s (0.12 percent), improved afterward (2.53 percent). The
potential of inland fisheries is still higher, with well-spread locations of rivers,
canals, and reservoirs.
The higher rate of growth in inland fisheries is mainly attributed to the
overwhelming progress in aquaculture, in both fresh and brackish waters. The
share of culture fisheries in the inland sector has gone up, from about 43 per-
cent in 1984-85 to a high level of about 84 percent in 1994-95 (Kumar, Joshi,
and Birthal 2001). The bulk of the growth in culture fisheries has come from
freshwater aquaculture (Krishnan and Birthal 2000). There is good scope to
expand the production of culture and other products in brackish-water areas,
because only 10 percent of the available brackish-water areas (12 million ha)
had been exploited by 1995-96 (IASRI 2001). The expansion of inland fish-
eries has also led to some negative externalities related to the degradation of
arable lands due to salinity.
The remarkable progress in fisheries was the outcome of a well-knit strat-
egy to augment production, enhance exports, and overcome the poverty of
fishermen. Several production- and development-oriented programs were im-
plemented in both marine and inland areas by the freshwater aquaculture, inte-
grated coastal aquaculture, coastal marine fisheries, and fish farmers' develop-
ment agencies established in freshwater areas and the brackish-water fish
farmers' development agencies in brackish-water areas. For better infra-
structure facilities, fisheries industrial estates were developed by grouping the
fishing villages.
The future of the fisheries sector is bright with the opening up of the econ-
omy. There is a promising export market for both marine and inland fish and
aquaculture products. In this context, sanitary and phytosanitary issues gain im-
portance. There is a need to focus on quality control, modernize the crafts used
in marine areas, and use the full potential of the inland fisheries.

Determinants of Crop Diversification

Several forces influence the nature and speed of agricultural diversification from
staple food to high-value commodities. Earlier evidence suggests that the pro-
cess of diversification out of staple food production is triggered by rapid tech-
nological change in agricultural production, improved rural infrastructure, and
diversification in food demand patterns (Pingali and Rosegrant 1995). These are
broadly classified as demand- and supply-side forces. The hypothesis in this
study is that demand-side forces, including per capita income and urbanization,
and supply-side forces, including infrastructure (markets and roads), technology
(relative profitability and risk in different commodities), resource endowments
 From Plate to Plow 393

(water and labor), and socioeconomic variables (pressure on land and literacy
rate), influence agricultural diversification.
The generalized least squares (GLS) technique with a fixed-effect model
is applied to examine how different forces have influenced crop diversification
in India. The analysis is based on the pooling of cross-section and time-series
information from major states (19 out of 28) in India for the period from
1980-81 to 1998-99. The GLS technique eliminates the effect of heteroscedas-
2

ticity that arises due to cross-section data and the autocorrelation that results
from time-series data. The following model is used to examine the determinants
of diversification:

D = f(tech, infr, prof, know, dema, rain).

The variables are defined as follows. The dependent variable, D is defined

in two ways: (1) as the Simpson index of diversity in the crop sector (SID ) and c

(2) as the index of the output values of horticultural commodities at constant

prices, with 1980-81 used as the base year. The results for the latter were found
to be statistically superior (Joshi et al. 2002) and are therefore used for discussion.
The independent variables are broadly grouped as follows: (1) technology
related (tech), (2) infrastructure related (infr), (3) profitability related (prof),
(4) resources and information related (know), (5) demand-side related (dema),
and (6) climate related (rain). To capture their effect, a few proxy variables are
used. For technology (tech), these include proportion o f area sown in high-
yielding varieties of foodgrain crops (percent), fertilizer use (kg per ha), pro-
portion of gross irrigated area to gross cultivated area (percent), mechanization
(number of tractors per 1,000 ha). For infrastructure (infr), the proxy variables
are market density (number of markets per 1,000 ha of gross cropped area) and
road length (square km per 1,000 ha of gross cropped area). The relative profit-
ability of high-value enterprises with cereals and other crops is the proxy for
the profitability-related variables (prof). Average size of landholding (ha) and
the proportion of small landholders in total holdings are used as proxies for
available resources and rural literacy (percent) for information-related variables
(know). Urbanization (percent urban population) and per capita income (rupees
per person) are used in the model as demand-side variables (dema). Annual
rainfall (mm) is used to define the climate-related variable (rain) in the model.
Different combinations o f independent variables have been tried to arrive
at the best-fit equations. Both linear and double-log equations have been used
and the best ones selected.
The study covers two decades divided into two periods—1980-81 to
1989-90 and 1990-91 to 1999-2000—for two reasons. First, historical evidence
shows that the impact of the green revolution gradually faded during the 1980s.

2. Nineteen states in the country are categorized as major and the remaining nine states as
small on the basis of geographical area, production, and population size.
 394 P. K. Joshi and Ashok Gulati

TABLE 18.3 Determinants of diversification in favor of horticultural commodities in India,

1980-81 to 1999-2000 (double-log estimates of generalized least squares)
Dependent variable: Index of gross value
of horticultural commodities at 1980-81 prices
Explanatory variables Equation 1 Equation 2 Equation 3
Irrigation - o .4575*** (0.0614) -0, 4697*** (0.0607) - 0 .5073*** (0.0564)
Relative profitability 0.,3549*** (0.04450) 0,.3329*** (0.0411) 0 .3152*** (0.0441)
Roads 0,,2873*** (0.0664) 0,,2843*** (0.0665) —
Markets 0.,1261* (0.0710) 0,,1870*** (0.0528) —
Rural literacy -0,,7976*** (0.1458) -0, g415*** (0.1419) -0, 5497*** (0.1389)
Small landholders 1.,1964*** (0.2283) 1.,2016*** (0.2285) 1,.6043*** (0.2002)
Urbanization 0.,1840 (0.1438) — 0,.3050*** (0.1094)
Income 0,,4892*** (0.0668) 0,.5082*** (0.0652) 0 .4671*** (0.0686)
Rainfall -0.,0583 (0.0422) -0,.0712* (0 .0411) - 0 .0949** (0.425)
Time dummy: 0.,8944*** (0.0700) 0,,8839*** (0.0696) 0,.8960*** (0.0722)
1981-90 = 0;
1991-99= 1
R-square 0.,7735 0..7722 0,,7572
Adjusted R-square 0.,7642 0,.7637 0,.7490
F-statistic 82, g2*** 90, 00*** 91 4Q***

NOTES: The figures in parentheses are standard errors o f the respective, coefficients.
*** Significant at the 1 percent level.
** Significant at the 5 percent level.
* Significant at the 10 percent level.

Second, the process of economic reforms started in the early 1990s, and India
bound itself to the World Trade Organization commitments, which will impact
agriculture. The data for the study were collected from various published
sources, especially the national statistical bulletins (CMIE 2001).
To examine the forces influencing diversification in favor of high-value
commodities, explanatory variables related to infrastructure development, tech-
nology adoption, relative profitability, resource endowments, and demand-side
factors including urbanization and income level were studied. The double-log
estimates of generalized least squares are given in Table 18.3.
To capture the effect of infrastructure development, two important vari-
ables, namely markets and roads, were included in the model. Both variables
yield positive and significant influence on the diversification of the crop sector.
Obviously, better markets and road networks induce diversification in favor of
horticulture, because they result in low marketing costs and easy and quick dis-
posal of commodities. They also reduce the risk of postharvest losses of perish-
able commodities.
 From Plate to Plow 395

Although technology was defined by area sown in high-yielding vari-

eties of cereals, irrigated area, and extent of mechanization, irrigated area
turned out to be the most significant and represented the technological ad-
vancement in the region. The regression coefficient of this variable shows a
negative relationship with diversification, which means that crop diversifica-
tion in favor of horticultural commodities is declining with increasing irrigated
area.
This suggests that crop diversification is more pronounced in rain-fed
areas, which are deprived of technological advancement in terms of irrigation.
These areas are characterized by low levels of resource endowment but with
abundant labor force and were bypassed during the green revolution.
The regression coefficient is significant and positive, because obviously
the higher profit of horticultural commodities would induce farmers to diver-
sify in their favor. The relative profitability of fruits is more than 8 times higher
than that of cereals, while the corresponding figure for vegetables is 4.8. How-
ever, while high profits from horticultural crops encourage their cultivation,
price and yield instability limit widespread cultivation. The high price variability
of fruits and vegetables is due to poor vertical linkages between production, mar-
keting, and processing. This calls for the development of appropriate institutional
arrangements for minimizing the price uncertainty, and more rigorous work
needs to be done in this area.
There is a positive relationship between the growth of horticultural com-
modities and the proportion of smallholders. This indicates that diversification
in favor of horticultural products is mostly confined to the smallholders because
it is labor intensive and generates a regular flow of income. However, absence
of appropriate markets and increase in supply may adversely affect the prices
and opportunities for higher income (Joshi et al. 2002).
Rainfall is another variable included in the model to assess the effect of
climate on crop diversification. This variable is negative and highly significant,
indicating that crop diversification is limited to higher-rainfall areas. Obviously,
high-rainfall areas specialize in rice, while farmers in medium- and low-rainfall
areas lean toward diversification to increase their income and minimize their
risk. Demand-side factors such as urbanization and per capita income also have
a positive and significant impact on crop diversification.
Our discussion suggests that assured markets and a good road network
could stimulate agricultural diversification in favor of high-value crops because
they help maximize profits and minimize the uncertainty of output prices.
Encouraging appropriate institutional arrangements for better markets through
cooperatives or contract farming would go a long way toward strengthening
farm-firm linkages. The role of technology also cannot be ignored. The high-
yielding and more stable genotypes of fruits and vegetables need to be propa-
gated by the development of a strong seed sector.
396 P. K. Joshi andAshok Gulati

Diversification of Dietary Patterns

There are apprehensions that a shift in the crop portfolio from foodgrain to
nonfoodgrain commodities may lead to an imbalance in dietary patterns and
adversely affect food security from the nutritional point of view. However, the
high-yielding nature of foodgrain crops has in fact improved availability, and
production trends reveal that the per capita daily availability of foodgrains in-
creased from 448.56 grams in TE 1981-82 to 475.4 grams in TE 1999-2000.
Similarly, the per capita daily availability of milk increased substantially, from
128 grams in 1980-81 to 214 grams in 1999-2000 (India, Ministry of Agricul-
ture, Agricultural Statistics at a Glance, 2002).
Income levels of consumers and urbanization are two important forces that
influence dietary consumption patterns as well. Earlier studies have shown that
diversification is strongly determined by these two demand forces (Joshi et al.
2002). As income increased and urbanization grew, diversification moved in
favor of high-value commodities. Evidence from the consumption basket shows
similar patterns across all income brackets. 3

Consumption Patterns and Income

The level of per capita consumption of all commodities is higher in the upper-
income group than in the lower-income group (Table 18.1). Consumption of
fruits, vegetables, milk, meat, eggs, and fish has increased in all the groups.
Even in the lower-income group, consumption of these commodities has in-
creased markedly. On the other hand, consumption of cereals declined during
the past two decades in all income groups, though the pace of decline of
4

cereal consumption was faster in the upper-income group. For example, per
capita consumption of cereals in the upper-income group declined from
194.3 kg in 1983 to 154.6 kg in 1999-2000. The corresponding figures for the
lower-income group were 147 kg and 132 kg. Contrary to the trend for cereals,
the per capita consumption of vegetables increased from 36 kg in 1983 to about
54 kg in 1999-2000, an increase of about 50 percent. Similarly, consumption
of meat, eggs, and fish doubled, from 1.9 kg in 1983 to 3.8 kg in 1999-2000.
The bottom income group, however, consumed smaller quantities of high-
value commodities, although their consumption of these commodities also
increased. This is an indication that the consumption baskets of poor as well as
rich consumers have changed, shifting in favor of noncereals. The diversifica-
tion of consumption patterns was a combined result of three different forces:
(1) income effect, (2) price effect, and (3) changes in tastes and preferences.

3. Those below the poverty line were classified as the lower (or bottom) income group, those
above the poverty line but below 150 percent of the poverty line as the middle-income group, and
those above 150 percent of the poverty line as the upper-income group.
4. This may be one of the reasons for the growing buffer stock of foodgrains in the late 1990s.
 From Plate to Plow 397

The fall in the relative prices of these commodities also contributed to the in-
crease in their consumption (Kumar and Mruthyunjaya 2002).

Consumption Patterns and Urbanization

Another important factor that determines consumption patterns is urbanization.

Earlier studies have shown that urbanization has strongly influenced the diver-
sification of agricultural production. The urbanization-induced production of
high-value commodities is rapidly increasing in the periurban areas (Parthasarthy
et al. 2002). The level of consumption of high-value commodities, namely
fruits, vegetables, milk, meat, eggs, and fish, has been much higher in the ur-
ban areas than in the rural areas. The share of cereals, pulses, and edible oils in
total expenditures increased from 41 percent in 1983 to 51 percent in 1999—
2000 in the urban areas. In rural areas their share in expenditures declined from
69 percent in 1983 to 61 percent in 1999-2000. The share of fruits, vegetables,
milk, meat, eggs, and fish in total expenditures increased from 41 percent in
1983 to 49 percent in 1999-2000 in the urban areas. The corresponding increase
in the rural areas was from 31 to 39 percent.
Per capita consumption of cereals has declined in both urban and rural ar-
eas, with the decline steeper in the latter. On the other hand, per capita con-
sumption of milk, vegetables, meat, eggs, and fish increased sharply in both
rural and urban areas. Per capita consumption of all commodities except cereals
was higher in urban areas than in rural areas. The shift in consumption patterns
in favor of high-value commodities did not have any adverse effect on calorie
intake. Per capita calorie intake increased marginally in rural areas, from 2,205
in 1983 to 2,332 in 1999-2000, while the corresponding increase was quite sub-
stantial in urban areas, from 1,972 in 1983 to 2,440 in 1999-2000.
In sum, the evidence we have presented reveals that income and urban-
ization are influencing the diversification of dietary patterns in favor of non-
foodgrain commodities and therefore toward the diversification of agricultural
production. A more favorable environment for diversification toward high-
value commodities will not only ease the pressure of storing huge surpluses of
rice and wheat but also accelerate the growth of the agricultural sector and agri-
cultural incomes.

Integration of Consumption and Production

Although the demand for noncereal commodities is growing fast, farmers, the
majority of whom are smallholders, are constrained by high transaction costs
resulting from (1) lack of access to markets, (2) limited marketable surplus,
and (3) the perishable nature of their products. Due to the rising demand for
high-value commodities, different forms of production-market integration are
slowly emerging in the food supply chain. These include (1) spot or open market
398 P. K. Joshi and Ashok Gulati

transactions, (2) agricultural cooperatives, and (3) contract farming. The features
of these modes of integration are now briefly discussed.

The Spot Market

Spot or open market transactions are traditional and common in developing
countries where the level of market uncertainty is high. In this model, the prices
are determined by the demand and supply of the commodity under transaction.
In India, the concept of spot or open markets has been initiated in some states
where producers and consumers transact business without any middlemen. Ryat
Bazaar in Karnataka and Apana Bazaar in Andhra Pradesh are examples of spot
or open markets for fruits and vegetables. These markets provide a forum in
which producers can deal directly with consumers. However, the uncertainty of
prices during excess supply periods and high transport costs still persist.

The Cooperative Model

The agricultural cooperative model has overcome the problems of spot markets.
In this model a group of producers with common interests own and manage
production and/or marketing to take advantage of economies of scale. Coop-
eratives enhance the bargaining power of the producers in input and output
markets, and by integrating input and output markets, they tend to reduce trans-
action costs. In India, one of the most successful models of a cooperative is in
the dairy sector, where the breakthrough is ascribed to the implementation of
the Operation Flood Program through the National Dairy Development Board
(NDDB), which developed a cooperative for procuring and marketing milk and
milk products. Encouraged by the success of dairy cooperatives, the NDDB has
diversified its product portfolio in recent years to include fruits, vegetables,
oilseeds, and plantation crops, which are sold in cooperatives developed along
the lines of dairy cooperatives. In 1985, the NDDB started under the banner o f
SAFAL to meet the growing demand for fruits and vegetables of the Delhi
metropolitan area. This is one of the largest public sector undertakings in the
marketing of fresh fruits and vegetables in the world. In 1996, the company es-
tablished an ultramodern 100 percent export-oriented fruit processing unit in
Mumbai. At present, there are 75 associations throughout the country with a
membership of 15,000 growers selling about 200 tons of fruits and vegetables
every day through its 275 retail outlets in and around Delhi. This model has
most benefited the smallholders in remote areas where markets for fruits and
vegetables were previously absent.

Contract Farming
The contract fanning model is relatively new in India. In this model, farmers
are contracted to produce the commodity desired by a firm. The firm controls
the production process without owning or operating the farms, while the farms
are assured procurement of the output at remunerative prices. India is witness-
 From Plate to Plow 399

ing a silent revolution in this form of mutually beneficial integration of farm,

firm, and consumer.
One successful example of contract farming is the multinational company
Nestle India Limited. Nestle entered into the dairy business in 1961 by collect-
ing a mere 540 kg of milk from 180 farmers in four villages and setting up a
milk plant at Moga in Punjab (National Dairy Development Board 2002). By
2002 the milk collection had grown to over 650 thousand kg/day from about
90,000 farmers in about 1,600 villages in Moga and the adjoining districts of
Ferozpur, Faridkot, Muktsar, and Ludhiana in Punjab. The company's success
resulted from developing effective backward and forward linkages. Most of the
milk comes from the small dairy producers. The company provides free veteri-
nary aid and extension, breeding services, information on fodder production
techniques, and so on for quality production.
Another successful example of the integration of production and market-
ing is witnessed in the poultry sector. The poultry industry grew mainly due to
the strong integration of poultry producers and firms. Several poultry firms have
entered into contract farming for the production, marketing, processing, and ex-
port of eggs and broiler chickens. The most important ones are Saguna Hatch-
eries Limited and Venkateshwara Hatcheries Limited (VHL). The latter is the
leading firm in the poultry sector and has been in operation since 1971. Initially
the firm was engaged in the breeding of chicks, the production of vaccines, and
their sale to poultry producers. Since the early 1990s, V H L has ventured into
contract broiler farming in the major poultry-producing states of Andhra
Pradesh, Karnataka, and Maharashtra. V H L has its own poultry breeding farm,
feed plant, vaccine manufacturing unit, and research laboratory and has devel-
oped innovative approaches to reduce transaction costs and enhance production
efficiency. The consequence of contract farming in the poultry sector has been
a marked increase in the production of eggs and broilers in these states.
To sum up, the integration of production and marketing is critical for high-
value commodities, because they are perishable and their markets are restricted.
To expand the scale of production, integrating production and marketing through
cooperatives or contract farming seems to be a prerequisite.

Conclusion

India's agricultural production portfolio is gradually diversifying in favor of

high-value commodities. In particular, the production of horticultural com-
modities, milk, meat, fish, and eggs has shown a remarkable increase during the
past two decades. These commodities are well suited to the needs as well as the
resources of small farm holders. Therefore, strengthening the horticulture, live-
stock, and fisheries subsectors would benefit them and significantly contribute
to enhancing farm income, offering employment opportunities in rural areas,
and meeting the food and nutritional needs of small farm holders.
400 P. K. Joshi andAshok Gulati

The consumption basket is also diversifying in favor of high-value com-

modities in both rural and urban areas and across both rich and poor consumers.
In the absence of appropriate integration of production, markets, and con-
sumption, the potential of high-value commodities is yet to be fully exploited.
Strengthening the integration of these three areas would promote the produc-
tion of high-value commodities because of their growing demand.
To cater to the demand for these commodities in metropolitan cities, a
few innovative institutional arrangements are gradually emerging in the form
of cooperatives and contract farming, which benefit producers, firms, and con-
sumers. Establishing strong farm-firm linkages strengthened both types of en-
tities and gave them the opportunity to benefit from the expanding domestic and
international markets. These kinds of arrangements need to be replicated so that
more small farm holders can share the benefits of the growing markets for high-
value commodities. This will have several macro-level benefits, including food
management and diversification of agriculture.
In terms of policy implications, the following points are worth noting. First,
greater attention needs to be given to the nongrain economy in terms of research
and development (R&D) expenditure and investment in marketing, storage, and
processing facilities. Although the share of the nongrain component in the total
value of agriculture is already more than half, that component does not receive
commensurate attention and resources. This distortion needs to be corrected.
Second, India has constrained its own potential by not changing the re-
strictive laws toward the development of a high-value chain in agriculture. Giv-
ing preference to cooperatives or public sector firms in the past restricted the
entry of big players into the private sector. This needs to change i f a revolution
in the value addition process is to be unleashed. A l l legal impediments that
restrict the entry of big private sector entities into marketing, storage, and pro-
cessing facilities need to be lifted. Further, retail chain stores with foreign di-
rect investments (FDIs) are still not permitted in accord with the recommenda-
tion of the latest government task force on FDIs.
Third, to ensure an efficient basis in the value addition process, it is nec-
essary to withdraw the preferential treatment given to cooperatives and public
sector concerns, often in the form of corporate tax exemptions, subsidized fi-
nance, and so on. This preferential treatment often drives out private sector in-
vestments. Further, major investments in retail chain stores, processing, and
storage facilities would emerge i f the legal environment was cleaned up and the
private sector was provided with a level playing field vis-a-vis cooperatives and
public sector concerns.
Fourth, it is essential to facilitate the emergence of vertical integration be-
tween farmers, processors, and retailers (farm-firm-fork linkages) in high-value
agriculture. India must graduate from producing raw commodities to adding
value and developing brand equity. A major role needs to be played by the pri-
 From Plate to Plow 401

vate sector, the sooner the better, although it is disheartening to see that it took
10 years to delicense the dairy and sugar industries. Many others in agro-
processing are still waiting for delicensing, including those involved in the pro-
cessing of groundnuts and mustard oilseeds, which are reserved for small-scale
industries. To ensure food safety, laws need to be duly enforced and sanitary
and phytosanitary standards adopted, and it would be desirable to promote large
processing facilities with state-of-the-art technology.
A l l these policy changes are basically in line with the emerging demand-
pull forces, and therefore are likely to be more sustainable. But they need to be
supplemented by some policy changes on the supply side, too. First, the land-
lease market should be freed and smaller cultivators helped to increase the sizes
of their operational holdings. They are efficient producers, but their transaction
costs need to be cut if Indian agriculture is to remain competitive internationally.
Second, investments will have to be stepped up in basic infrastructure,
such as roads and power, where the private sector is still reluctant to enter. Gov-
ernment programs on highways and rural roads are laudable, but reform in the
power sector is another story altogether. Major institutional and price reforms
are required in the power sector to plug the leakages, raise efficiency, and gen-
erate surpluses to plow back into investments. The cold storage chain, often
talked about as an important form of infrastructure for high-value agriculture,
cannot be developed without reforming the power sector in rural areas.
Third, the level of R&D expenditure in India, at less than 0.5 percent of
agricultural GDP, is way below the 1 percent level in most developing countries.
In biotechnology research, the record is even worse. The country is losing out
on a revolution in biotechnology that is waiting in the wings. I f the government
does not have ample resources for this, the private sector could be invited in on
a large scale. For this purpose, the government should put in place for biosafety
appropriate regulatory institutions that are transparent and time bound.
Fourth, the Indian agricultural credit scene is contradictory: commercial
banks are saddled with excess liquidity, while farmers are still relying on in-
formal sources of finance for almost 45 percent of their requirements and at
much higher rates of interest than are offered by the commercial banks. High-
value agriculture needs higher amounts of working capital and also has higher
risks. While schemes like the kisan credit cards (Gulati and Narayanan 2002a)
are a step in the right direction, facilitating credit through processors, input
5

dealers, and others who are vertically integrated with the fanners for the provi-

5. Under this scheme, established by the government in 1998-99, farmers are eligible for
production credit in the amount of Rs. 5,000 or more issued against a kisan card, which is valid for
three years but subject to an annual review. Credit limits are fixed depending on need related to crop
production, operational holdings, cropping pattern, and scale of finance. By January 2001, over
13.4 million kisan cards had been issued to farmers by cooperatives and regional rural banks.
402 P. K. Joshi and Ashok Gulati

sion of critical inputs or for processing their produce could increase the credit
flow to agriculture greatly. These dealers or processors could act as nonbank-
ing financial intermediaries, able to obtain refinancing from the banking sector
with a margin to cover the risk of default. Such a scheme could bring about a
revolution in the financing of agriculture, provided the government facilitates the
entry of such nonbanking financial institutions and frees up interest rates.