Growth in high-value agriculture in Asia

and the emergence of vertical links with farmers

Ashok Gulati1

Nicholas Minot2

Chris Delgado3

Saswati Bora4

December 2005

Paper presented at the workshop

“Linking Small-Scale Producers to Markets: Old and New Challenges”

The World Bank, 15 December 2005

1. Director, Markets, Trade, and Institutions Division, International Food Policy Research Institute. 2. Research Fellow, Markets, Trade, and Institutions Division, International Food Policy Research Institute. 3. Director, ILRI-IFPRI Joint Program on Livestock Market Opportunities and Senior Research Fellow,

Markets, Trade, and Institutions Division, International Food Policy Research Institute. 4. Senior Research Assistant, Markets, Trade, and Institutions Division, International Food Policy Research Institute.

Note: This paper is a revised version of one presented at the Symposium “Toward High-Value Agriculture and Vertical Coordination: Implications for Smallholders” at the National Agricultural Sciences Centre, Pusa, New Delhi, 7 March 2005.

Table of Contents

1 Introduction ............................................................................................................................. 1 1.1 Background..................................................................................................................... 1 1.2 Objectives ....................................................................................................................... 2

2 Factors behind growth of high-value agriculture .................................................................... 2 2.1 Income growth................................................................................................................ 3 2.2 Demographic factors ...................................................................................................... 4 2.3 Outward-looking trade policies ...................................................................................... 5 2.4 Foreign investment ......................................................................................................... 7

3 Shift in composition of food demand...................................................................................... 8 3.1 High but declining share of food in household budgets ................................................. 8 3.2 Rising share of high-value foods in food consumption .................................................. 9 3.3 Growth in export demand for high-value agricultural commodities ............................ 11

4 Growth in production of high-value agricultural commodities............................................. 13 5 Consolidation and vertical coordination in food marketing.................................................. 15

5.1 Consolidation and growth in the retail food sector....................................................... 16 5.2 Food processing consolidation ..................................................................................... 20 5.3 Emerging forms of farmer-buyer vertical coordination................................................ 24

6 Vertical coordination of high-value agriculture and smallholders ........................................ 27 6.1 What contract farming does for small-scale farmers.................................................... 27 6.2 Impact of contract farming of animal products on smallholders in Asia...................... 32 6.3 Impact of supermarket growth on smallholders ........................................................... 36

7 Summary and discussion....................................................................................................... 38 7.1 Summary ...................................................................................................................... 38 7.2 Implications for policy ................................................................................................. 40

References...................................................................................................................................... 43

2

1 Introduction

1.1 Background

Throughout the developing world, the relative importance of grains and other starchy staple crops

is declining, while that of high-value agricultural commodities is increasing. High-value agricultural

goods are generally defined as agricultural goods with a high economic value per kilogram, per hectare,

or per calorie, including fruits, vegetables, meat, eggs, milk, and fish1. This transformation of the

agricultural sector has profound effects on the nature of agricultural supply channels, the opportunities for

small farmers, and the role of public policy and investment. In particular, the growth in high-value

agriculture implies a greater need for close linkages between farmers, processors, traders, and retailers to

coordinate supply and demand. Examples of institutions for vertical coordination of agricultural supply

channels include grades and standards, price information services, inspection and certification services,

contract farming, farmer cooperatives, professional associations, and vertical integration2.

The growth of high-value agriculture, the development of institutions for vertical coordination,

and other structural changes in agricultural supply channels present both opportunities and challenges for

small farmers in developing countries. They create opportunities for small farmers to raise their income

by participating in the growing markets for high-value agricultural commodities. At the same time, the

changes pose challenges to small farmers because high-value agricultural commodities often involve

higher costs of production and greater production and marketing risk. Vertical linkages between farmers

and buyers can serve to overcome these obstacles, but in some cases buyers decide small farmers cannot

satisfy new demands from consumers for quality and food safety, leading to the exclusion of these

farmers from supply chains. These trends raise new issues for policymakers who wish to promote pro-

poor agricultural growth.

This transformation has been called a “silent revolution,” inviting comparison with the Green

Revolution of the 1960s and 1970s. Although both represent a significant shift in the patterns of

agricultural production, the causal factors are quite different. The Green Revolution was driven by

technological innovation, namely the development of high-yielding varieties of rice and wheat,

particularly in Asia and Latin America. In contrast, the current restructuring of the agricultural sector is

driven not by supply factors but by shifts in consumer demand, both domestic and international. Given 1 Other crops could be also considered high-value commodities such as spices, flowers, medicinal plants,

many industrial crops, and even crops that yield illegal drugs. The motivation for focusing on fruits, vegetables, animal products, and fish is that they are widely grown and face growing demand due to shifts in consumption patterns.

2 Vertical integration refers to the situation where different activities within the marketing channel (such as processing and exporting) are carried out by the same firm.

1

that the transformation is demand-led, it is important to understand the source of these changes in food

demand (Delgado, 1999).

1.2 Objectives

The objective of this paper is to describe the growth of high-value agriculture, its direct

implications for the restructuring of the agricultural supply chain, and its indirect implications for the role

of small farmers. More specifically, the paper attempts to address five questions:

• What factors are driving the shift in food consumption toward high-value agricultural commodities?

• What is the rate and pattern of this shift in food consumption?

• How have the patterns of agricultural production in the region changed in response to new demand patterns?

• What are the implications of the shift toward high-value agriculture for agricultural marketing channels?

• What are the implications of these changes for public policy and investment priorities?

We focus on three countries in South Asia (India, Bangladesh, and Pakistan), four countries in

Southeast Asia (Thailand, Vietnam, Indonesia, and the Philippines), and China. These countries are the

most populous countries in their respective regions and represent a range of different development

strategies and levels of development. In terms of commodities, the paper focuses on fruits and vegetables,

meat, eggs, milk, and fish. Among the broader category of high-value agricultural commodities, these are

some of the more widely grown products, so trends in these commodities have broad effects on the

marketing channels and the opportunities of small farmers in the region.

2 Factors behind growth of high-value agriculture

What is causing the growth in high-value agriculture in developing countries? On the one hand,

there is a rising domestic demand for high-value food commodities, driven by rising incomes,

urbanization, and perhaps changing preferences. At the same time, trade liberalization has opened export

markets in other countries where high-income consumers demand fruits, vegetables, animal products, and

fish. And finally, market reforms have (to varying degrees) allowed more foreign direct investment in

developing countries, introducing more competition in food processing and retailing sectors, as well as

allowing foreign companies to organize production for export. In this section, we explore the trends in

2

income, urbanization, trade liberalization, and foreign direct investment to help explain the growth in

high-value agriculture.

2.1 Income growth

One of the most consistent patterns in economics is Engel’s Law, which states that as incomes

rise, the share of expenditure allocated to food tends to decline. For example, the share of household

budgets allocated to food declines from over 65 percent in the least development countries to less than 10

percent in most industrialized countries (the patterns in South and Southeast Asia are described in Section

3). A less-well-known pattern is Bennet’s Law, which states that as income rises, the share of the food

budget allocated to starchy staples declines relative to more expensive sources of calories. Thus, the level

of per capita income is an important determinant of the composition of food expenditure, and the rate of

growth of income is a key factor in determining the pace of change in food consumption patterns.

Table 1 shows the level of real per capita gross domestic product (GDP) and the growth rate over

1990-2002. Thailand has the highest per capita GDP by a significant margin, followed by the Philippines,

Indonesia, and China. Vietnam and the three South Asian countries have lower levels of per capita GDP,

all close to the range of US$ 400-500 in 2002.

Table 1. Trends in per capita income and per capita income growth

GDP per capita (1995 US$)

Annual growth rate

(percent) 1990 1996 2002 1990-2002 South Asia Bangladesh 278 325 396 3.0% India 324 402 493 3.6% Pakistan 448 507 518 1.2% SE Asia Indonesia 777 1,113 1,060 2.6% Philippines 1,091 1,122 1,209 0.9% Thailand 1,997 3,015 3,000 3.5% Vietnam 211 305 413 5.7% China 350 630 944 8.6%

Source: World Bank (2004).

China and Vietnam experienced the most rapid rates of per capita GDP growth over the period

1990-2002, 8.6 percent and 5.7 percent, respectively. Bangladesh, India, and Thailand achieved healthy

growth rates of more than 3 percent per year. Average per capita growth rates in Pakistan and the

Philippines were the lowest, hovering around 1 percent per year. These averages hide the large shocks

experienced by Thailand, Indonesia, and (to a lesser degree) the Philippines as a result of the Asian

3

financial crisis. Although growth has returned to all three, Thailand and Indonesia are only now returning

to their pre-crisis level of per capita GDP.

2.2 Demographic factors

Demographic factors also affect the growth of high-value agriculture. For example, urbanization

affects the composition of food consumption. The connection is less well-documented than the effect of

income on food consumption patterns, but several studies indicate that urban and rural household food

consumption habits differ, even after holding income and other household characteristics constant. One

study found that, after controlling for income and prices, urbanization is associated with lower rice

demand in India, Bangladesh, Pakistan, Indonesia, and Thailand3 (Huang and David, 1983). A study of

food demand in Vietnam indicates that urban household spend more on meat, fish, and sugar and less on

rice than rural households, even after controlling for income and household characteristics (Minot et al,

2003). These changes are presumably related to the greater variety of food available and perhaps the

higher opportunity cost of time of household members.

Table 2 shows that the percentage of the population living in urban areas has increased over the

period 1980-2002 in all seven countries. The Philippines is by far the most urbanized, with over half its

population in urban areas, followed by Indonesia and Pakistan. Thailand is the least urbanized. Although

cross-country comparisons are difficult due to different definitions, the example of Thailand should serve

as a note of caution against assuming that there is a close relationship between income level and the level

of urbanization.

Table 2. Urbanization and population growth

Urban population (as % of total population)

Population growth rate

1980 2002 1990-2000 South Asia Bangladesh 15 26 2.3% India 23 28 1.9% Pakistan 28 34 2.5% SE Asia Indonesia 22 43 1.5% Philippines 37 60 2.2% Thailand 17 20 1.1% Vietnam 19 25 1.7% China 20 38 1.0% Source: World Bank (2004). Note: The definition of the urban population varies across countries, so it is difficult to compare levels of urbanization across countries.

3 In China and the Philippines, there was no statistically signficant effect of urbanization on rice demand.

4

In addition, population growth increases the total demand for all food, including high-value

agricultural commodities and staples. In addition, it may affect the composition of food demand to the

extent that growth is greater in some regions than others. As shown in Table 2, the annual population

growth rate over 1990-2000 among the eight countries varies from 1.0 percent in China to 2.5 percent in

Pakistan.

2.3 Outward-looking trade policies

Policies to promote trade, including lower tariff barriers, market-determined exchange rates, and

deregulation of international trade, have created opportunities for developing countries to export

agricultural commodities, both to high-income countries and to other developing countries. The lowering

of import barriers in developed countries has probably facilitated the growth of high-value exports such as

fish and seafood products. But perhaps more important is the fact that developing countries themselves

have reduce import tariffs and moved toward market-oriented exchange rates, which increase the

incentives to export. Since high-value agricultural commodities and processed foods represent a larger

share of the food budget of high-income consumers, it is natural that, as farmers in developing countries

shift from meeting domestic demand to meeting international demand, they also shift production from

staple crops toward high-value agricultural commodities.

It should be mentioned, however, that trade liberalization is a two-edged sword when it comes to

high-value agriculture. In some cases, trade liberalization makes local farmers more exposed to

competition from imported high-value agricultural commodities.

Table 3 and Table 4 show two imperfect but widely-used measures of the degree of trade

liberalization. Table 3 gives the weighted average tariff rate around 1990 and around 2001 for all goods

and for primary products. Six of the seven countries have significantly reduced average taxes on

imported goods over the 1990s. In fact, all six reduced the mean import tariff by more than one half over

this period. Focusing on primary products, again six out of seven countries reduced the weighted average

tariff over the 1990s. It is worth noting that the levels of import protection are generally higher in the

three South Asian countries than in China and the four Southeast Asian countries.

Table 4 shows the value of merchandise trade (exports plus imports) as a percentage of gross

domestic product and the value of agricultural trade as a percentage of agricultural GDP. In seven of the

eight countries, the overall trade ratio increased between 1990 and 2002 (it declined slightly in Pakistan).

In the case of agricultural trade, the ratios are much smaller, which is not surprising given the large

number of semi-subsistence farmers in these countries and the fact that agriculture commodities tend to

5

have low value-bulk ratios. Six of the eight countries show an increase in agricultural trade ratios, while

the ratios declined in Indonesia and China. It is worth noting that China was the only country for which

the weighted mean tariff of primary products increased over the 1990s. The highest agricultural trade

ratios are found in Vietnam and Thailand, while India has the lowest ratio. The low agricultural trade

ratio in India is related to the relatively high level of general and primary-product tariff protection, as well

as the large size of the country.

Table 3. Measures of trade liberalization

Weighted mean tariff rate as a % of all products

Weighted mean tariff rate as a % of primary products

Period of comparison

Before (percent)

After (percent)

Before (percent)

After (percent)

South Asia Bangladesh 1989-2002 131 23 54 20 India 1990-2001 50 21 25 23 Pakistan 1995-2002 45 15 24 11 SE Asia Indonesia 1989-2001 12 4 6 2 Philippines 1988-2002 21 3 18 5 Thailand 1989-2001 32 9 24 5 Vietnam 1994-2001 13 17 47 21 China 1992-2001 35 13 14 19 Source: World Bank (2004). Note: Weighted mean tariff is the average of effectively applied rates or most favored nation rates weighted by the product import shares of each country. Primary products are commodities classified in SITC sections 0-4 plus division 68 (nonferrous metals).

Table 4. Measures of trade openness

Trade in goods as a percentage of GDP

Trade in agriculture as a percentage of agriculture GDP

1990 2002 1990 2002 South Asia Bangladesh 18 29 10 14 India 13 21 4 8 Pakistan 38 36 20 35 SE Asia Indonesia 48 51 23 19 Philippines 48 92 27 36 Thailand 66 106 67 98 Vietnam 80 101 39 43 China 32 49 21 16

Source: World Bank (2004) and FAOSTAT. Note: Trade refers to the sum of the value of merchandise imports and merchandise exports. In this table, agriculture excludes fishery products.

6

2.4 Foreign investment

Another factor which has stimulated the transformation of agricultural production toward high-

value agriculture is foreign direct investment (FDI). Foreign investment in the food sector of developing

countries is rarely targeted at direct agricultural production, except in the case of capital-intensive and

technically complex crops that can be grown on a large scale such as bananas, pineapple, and flowers.

Rather foreign investment is usually focused on food processing, animal feed, exporting, and (more

recently) food retailing. The entrance of foreign companies into the agricultural sector puts competitive

pressure on local agribusiness companies, but it has the potential of reducing margins through

competition and/or creating new markets, which generally creates new opportunities for farmers.

Foreign direct investment can promotes the growth of high-value agriculture in one of three ways.

First, FDI in the export sector may serve to link farmers in developing countries with high-value export

markets, particularly those in the home country of the company. This is particularly relevant in the case

of the export of fresh produce and fish, where foreign-market expertise is required to meet food safety and

quality standards. Second, FDI in the processing sector may create a new market for high-value

agricultural commodities by preserving perishable goods and supplying the processed item to high-

income markets. Third, to the extent that foreign companies use their expertise and scale of operations to

reduce marketing margins in the processing and/or retail sector, they may reduce the price and increase

the domestic demand for high-value agricultural commodities.

Table 5 show that net inflows of foreign direct investment over the 1990s. China receives almost

ten times more net foreign investment than the other seven countries combined. Net FDI flows have

increased dramatically in some countries, but the trends are erratic. The most dramatic rise in FDI is in

China and India, where it has increased more than ten-fold since the early 1990s. Similarly, FDI in

Vietnam has grown almost ten-fold over this period. In spite of large annual fluctuations, FDI inflows in

Bangladesh, Pakistan, and the Philippines are at least twice as large as in the early 1990s. Indonesia’s net

inflow grew rapidly until the Asian financial crisis of 1997-98, but flows have been negative since then.

Thailand was also seriously affected by the crisis: although FDI net inflows remained positive, they have

declined sharply since 1998. China was affected by the financial crisis, but, after dropping in the late

1990s, net FDI has since surpassed its pre-crisis levels.

Foreign direct investment in the food processing sector is only a small portion of the total, less

than 5 percent in most countries. Although data on food processing FDI are not widely available, a study

by the U.S. Department of Agriculture reveals that foreign direct investment by the United States in the

food processing sector has increased from US$ 9.3 billion in 1985 to US$ 32.4 billion in 1995, an annual

increase of more than 13 percent. Of the global FDI from the U.S. in the food processing sector, almost

half has been invested in the European Union and about 20 percent in Asia and Africa. Looking at the

7

sectoral composition of this FDI, investment in beverages and preserved fruits and vegetables has grown

the fastest over 1985-95.

Table 5. Net inflows of foreign direct investment (millions US$)

India Vietnam Bangladesh Indonesia Pakistan Philippines Thailand China 1990 237 180 3 1,093 245 530 2,444 3,487 1991 74 375 1 1,482 258 544 2,014 4,366 1992 277 474 4 1,777 337 228 2,113 11,156 1993 550 926 14 2,004 349 1,238 1,804 27,515 1994 973 1,945 11 2,109 421 1,591 1,366 33,787 1995 2,144 1,780 2 4,346 723 1,478 2,068 35,849 1996 2,426 2,395 14 6,194 922 1,517 2,336 40,180 1997 3,577 2,220 139 4,677 716 1,222 3,895 44,237 1998 2,635 1,671 190 -356 506 2,287 7,315 43,751 1999 2,169 1,412 180 -2,745 532 1,725 6,103 38,753 2000 2,657 1,298 280 -4,550 308 1,345 3,366 38,399 2001 4,334 1,300 79 -3,278 383 982 3,820 44,241 2002 3,030 1,400 47 -1,513 823 1,111 900 49,308

Source: World Bank, 2004.

Thus, foreign direct investment has grown rapidly in South Asia and Southeast Asia, although the

Asian financial crisis sharply reduced investment in Indonesia and Thailand. Data on foreign direct

investment in food processing are scarce, but it appears to be growing rapidly.

3 Shift in composition of food demand

Income growth and urbanization are changing the composition of domestic food demand in

developing countries, reducing the share of household budgets spent on food while increasing the share of

the food budget allocated to high-value commodities such as meat, milk, fish, and fruits and vegetables.

At the same time, more open trade policies have increased the export demand for some types of

agricultural commodity, particularly high-value commodities. In this section, we describe these changes

in the structure of domestic demand in South and Southeast Asia, as well as changes in export demand for

agricultural commodities.

3.1 High but declining share of food in household budgets

As described above, rising incomes in South and Southeast Asian countries are reducing the share

of household budgets allocated to food. Table 6 shows changes in the food share in six Asian countries.

Three patterns can be observed in the table. First, the food share is substantially higher in rural areas than

in urban areas. This is consistent with Engel’s Law and the fact that urban incomes are higher than rural

8

incomes. Second, the food share is declining both in urban and rural areas in each country, the only

exception being the urban areas of Indonesia where the food share increased slightly between 1990 and

2002. This is presumably the result of the Asian financial crisis of 1997-98 which reduced urban incomes

Table 6. Share of household budgets allocated to food

Share of household budgets spent on food (percent)

Country Year Rural Urban National Bangladesh 1995-96 62 46 58 2000 59 45 55 India 1987-88 64 56 1994-95 61 53 1998 61 50 57 1999-2000 59 48 Pakistan 1998-99 55 41 49 2001-2002 54 39 48 Indonesia 1980 74 60 69 1990 67 51 60 2002 67 53 59 Vietnam 1993 70 58 66 1998 59 45 53 China 1980 62 57 61 1990 59 50 55 2000 48 38 46

Source: For Bangladesh, Report of the Household Income and Expenditure Survey, 2000. For India, National Sample Survey Organization, India. For Pakistan, Household Integrated Economic Survey, 1998-99 and 2001-02. For Indonesia, Central Bureau of Statistics. For Vietnam, the Vietnam Living Standards Survey 1993 and 1998 (GSO, 1998 and 2001). For China, National Bureau of Statistics, China Statistical Yearbook, various years.

more than rural incomes. Third, the decline in food shares was particularly rapid in Vietnam, where it

dropped 13 percentage points in five years, and China. This is consistent with the fact that China and

Vietnam experienced the fastest income growth rates among the countries shown in the table.

3.2 Rising share of high-value foods in food consumption

At the same time that the share of food in household budgets is declining, the composition of food

budgets is changing. In particular, as incomes rise, there is a shift from grains and other starchy staple

crops (such as cassava and sweet potatoes) to meat, milk, eggs, fish, fruits, and vegetables. Table 7 shows

the changes in per capita consumption of selected foods over the period 1990-2000. In most of the seven

countries considered here, per capita grain consumption increased very slowly (Bangladesh, the

Philippines, and Thailand) or decreased slightly (China, India, and Pakistan). Only in Indonesia and

Vietnam did the annual growth rate in per capita grain consumption exceed 0.2 percent, the highest being

1.2 percent in Vietnam

9

In contrast, per capita vegetable demand grew fairly quickly (above 2 percent per year) in five of

the eight countries and above 4 percent in two countries (Vietnam and China). Fruit demand appears to

have grown somewhat more slowly, but the growth rate still exceeded that of grains in seven of the eight

countries. Bangladesh is the exception, where fruit consumption appears to have declined, perhaps as a

result of market reforms and improved technology in the rice sector which reduced the relative price of

rice.

Table 7. Changes in consumption of selected foods over 1990-2000 (kg/person/year)

Bangladesh India Pakistan Indonesia Philippines Thailand Vietnam China Cereals 1990 177.7 159.9 151.8 185.8 48.0 121.3 165.2 207.7 2000 181.3 153.1 151.2 202.9 48.4 123.7 185.8 182.3 Annual

Growth 0.2% -0.4% 0.0% 0.9% 0.1% 0.2% 1.2% -1.3%

Vegetables 1990 12.0 53.4 27.4 22.0 22.0 38.3 45.9 98.9 2000 12.0 65.9 34.1 30.4 22.1 40.4 74.3 224.5 Annual

Growth 0.1% 2.1% 2.2% 3.3% 0.0% 0.5% 4.9% 8.5%

Fruit 1990 11.3 28.2 33.5 29.3 34.6 88.5 43.0 16.5 2000 9.7 37.5 35.1 35.6 35.5 91.7 50.6 43.0 Annual

growth -1.5% 2.9% 0.5% 1.9% 0.2% 0.3% 1.7% 10.0%

Milk 1990 13.6 53.9 113.2 4.1 7.0 13.7 1.3 5.9 2000 13.9 64.9 152.8 7.2 8.1 22.4 4.7 9.6 Annual

Growth 0.2% 1.9% 3.0% 5.9% 1.5% 5.0% 13.5% 5.0%

Meat 1990 2.8 4.6 11.9 8.0 6.1 21.4 16.0 25.9 2000 3.1 5.0 12.1 8.3 9.6 24.8 24.4 50.1 Annual

growth 1.0% 0.9% 0.2% 0.4% 4.7% 1.5% 4.3% 6.8%

Eggs 1990 0.6 1.2 1.7 2.1 1.9 10.5 1.2 6.4 2000 1.0 1.4 2.1 3.0 2.2 10.1 2.2 16.2 Annual

Growth 4.6% 1.9% 1.9% 3.7% 1.6% -0.4% 5.8% 9.7%

Fish 1990 7.4 3.8 2.0 14.8 12.2 20.9 13.2 11.5 2000 11.7 4.7 2.3 20.3 10.6 30.6 19.0 25.7 Annual

growth 4.7% 2.0% 1.6% 3.2% -1.4% 3.9% 3.7% 8.4%

Source: FAO Food Balance Database.

Milk demand experienced some of the highest annual growth rates: 13 percent in Vietnam,

Pakistan, and 5-6 percent in Indonesia, Thailand, and China. Per capita demand for meat grew very

rapidly (over 4 percent annually) in China, the Philippines and Vietnam and more modestly in Thailand,

Bangladesh, and India. With the exception of Thailand, where demand is high but stagnant, annual

growth in the demand for eggs ranged from 1.6 percent in the Philippines to over 4 percent in China,

10

Bangladesh, and Vietnam. Similarly, the growth in demand for fish and seafood was over 3 percent per

year in five of the seven countries under consideration.

It is worth noting that the two countries with the highest GDP growth rates, China and Vietnam,

have also experienced some of the highest growth rates in per capita demand for high-value agricultural

commodities. The relationship between income and high-value food consumption can be seen by looking

at the patterns across countries as well. For example, Thailand, with the highest income among the seven

countries, also has the highest levels of consumption of fruit, meat, eggs, and fish. Bangladesh, with one

of the lowest incomes, has relatively low levels of consumption of many of the high-value foods.

Regional factors, such as local prices and preferences, also play a role in determining high-value food

consumption patterns. For example, milk consumption is much higher in South Asia than in Southeast

Asia, while fish consumption is greater in Southeast Asia. It is worth mentioning that Table 7

describes food consumption patterns in terms of quantities, but as income rises, household purchase more

expensive items within each category. For example, they may shift from cabbage and onions to green

beans, from domestic litchi to imported apples, from whole chickens to boneless cuts, and from raw milk

to pasteurized milk. Thus, as incomes rise, the expenditure on each category of high-value food rises

more quickly than the quantities.

Another aspect of the shift toward higher-value food is the growing demand for prepared or semi-

prepared foods. Among urban households, particularly higher-income households, there is a trend toward

ready-to-cook and ready-to-eat foods, including pre-cut vegetables, de-boned meat, and filleted fish.

Food consumed outside the household at restaurants, fast food establishments, and street stalls is another

trend in urban areas. As income rise and women join the work force, the opportunity cost of the time

spent cooking and shopping rises, making these choices more attractive.

3.3 Growth in export demand for high-value agricultural commodities

The opportunities faced by farmers in developing countries are increasingly affected not just by

the composition of domestic demand but by that of export demand. As shown in Table 8, the growth in

agricultural and fishery exports in the eight countries has been substantial: 4.8 percent per year over 1990-

2000. But the export demand for high-value agricultural commodities has increased even more rapidly.

By far the largest category of high-value agricultural exports is fishery products. Fish and seafood

exports from these eight countries grew from US$ 8.8 billion to US$ 17 billion, representing an annual

growth rate of 6.9 percent. In seven of the eight countries, the growth rate was over 4 percent per year.

Five of these countries (China, Thailand, India, Indonesia, and Vietnam) now export more than US$ 1

billion per year in fish and seafood products.

11

Fruit and vegetables are the second largest category of high-value agricultural exports. The total

value of fruit and vegetable exports from the eight countries grew at 5.6 percent per year over 1990-2000,

surpassing US$ 5 billion. Furthermore, these exports increased by more than 4.8 percent per year in

every country except Bangladesh. India and Vietnam experienced annual export growth rates of over 9

percent.

Table 8. Changes in exports of selected foods over 1990-2000 (US$ million/year)

Bangladesh India Pakistan Indonesia Philippines Thailand Vietnam China Agricultural products (including fishery products) 1990 325 3,320 1,015 2,975 1,575 6,681 832 12,748 2000 470 6,005 1,173 5,753 1,917 10,087 3,719 17,841 Annual growth 3.7% 6.1% 1.5% 6.8% 2.0% 4.2% 16.2% 3.4% Fruits & vegetables 1990 9 134 51 74 355 582 31 1,726 2000 11 323 106 162 575 940 79 2,899 Annual growth 2.3% 9.2% 7.7% 8.1% 4.9% 4.9% 9.8% 5.3% Dairy and eggs 1990 0 3 - 17 0 25 4 58 2000 0 45 2 75 13 45 4 86 Annual growth - 33.1% - 16.3% 46.6% 5.9% 0.3% 4.0% Meat products 1990 5 79 1 14 1 314 29 1,483 2000 0 325 7 13 2 782 119 1,257 Annual growth -35.2% 15.2% 23.3% - 6.0% 9.5% 15.4% -1.6% Fishery products 1990 167 468 101 1,109 419 2,321 185 3,997 2000 371 1,483 160 1,831 484 4,472 1,702 6,624 Annual growth 8.3% 12.2% 4.8% 5.1% 1.5% 6.8% 24.9% 5.2% High-value agricultural exports as a % of agricultural exports 1990 56% 21% 15% 41% 49% 49% 30% 57% 2000 81% 36% 23% 36% 56% 62% 51% 61% Source: FAO Agricultural Trade Database. Note: For the purpose of this table, fruits and vegetables are defined more narrowly that the FAO category, as we exclude sugar crops, pulses, and starch root crops such as cassava and sweet potato. The agricultural exports are defined broadly to include the sum of agricultural exports, as defined by the FAO, and fishery product exports.

Meat product exports from these countries were smaller (US$ 2.5 billion in 2000) and grew more

slowly (2.7 percent), mainly because China’s meat product exports declined over the decade. Excluding

China, meat product exports from the other seven countries expanded at 10.9 percent per year.

Finally, dairy and egg exports are relatively small, US$ 270 million in 2000, but grew at 9.7

percent per year over the 1990s. The growth was concentrated in four countries (India, Indonesia, the

Philippines, and Thailand), which experienced growth rates between 6 and 46 percent per year, In all four

countries, dairy exports dominate this category, but egg exports are significant in India and Thailand.

12

The last two lines of Table 8 show the share of high-value agricultural exports in total agricultural

exports. The percentage increased substantially over the 1990s in seven of the eight countries (the

percentage declined slightly in Indonesia). For the eight countries as a whole, high-value agricultural

exports increased from 47 to 53 percent of the total.

Another change in the composition of export demand is a growth in the share of agricultural

exports that are processed. Table 9 shows the growth in processed food exports between 1980 and 1999

for six of the eight countries. The annual growth rate in processed food exports is greater than 5 percent

in every country, and it exceeds 10 percent in three countries: Thailand, Bangladesh, and Indonesia. In

every case, the growth rate of processed food exports is greater than that of agricultural exports in general,

implying that processed food exports represent a rising percentage of the total.

Table 9. Changes in exports of processed foods over 1980-1999 (US$ million/year)

Bangladesh India Pakistan Indonesia Philippines Thailand Processed food exports (US$ million)

1980 46 768 102 723 1631 826 1999 350 2,376 305 3,947 1,650 6,611 Annual export growth (%) Processed food 15.1% 8.4% 6.9% 14.6% 5.2% 17.0% Agriculture 6.7% 7.3% 3.7% 9.0% 4.4% 10.9%

Source: Athukorala and Jayasuriya (2003), based on United Nations trade data (Series D).

4 Growth in production of high-value agricultural commodities

Largely in response to the growth in domestic consumption and, to a lesser degree, export

opportunities, production of high-value agricultural commodities has grown more quickly than that of

traditional grain crops. Grain production in the eight countries under consideration grew about 1.3

percent per year in volume over the 1990s. This rate is slightly below the annual rate of population

growth for the eight countries (1.5 percent). The highest rate of growth in grain production was in

Vietnam, which has gone from being a chronic rice importer in the 1980s to one of the three largest rice

exporters in the world in the 1990s. The relatively high grain production growth in Thailand (3.7 percent)

is also linked to rice exports.

By contrast, the production of high-value agricultural commodities has grown rapidly in many

countries. Fruit and vegetable production in the eight countries has grown 7.7 percent per year in volume

over the 1990s. China represents a large and growing share of Asian fruit and vegetable output. It grew

over 10 percent per year over the 1990s, reaching about two-thirds of the output of the eight countries

13

combined. But fruit and vegetable production growth is not limited to China; it grew at more than 3

percent per year in India, Pakistan, Indonesia, and Vietnam as well.

Milk production has grown at 4.6 percent annually in the eight countries under consideration.

India, Pakistan, and China are the dominant producers in the region, and all three have production growth

rates above 4 percent per year. Thailand is a minor producer but output grew at almost 15 percent

annually over the 1990s.

Table 10. Growth in production of grains and high-value agricultural commodities

Bangladesh India Pakistan Indonesia Philippines Thailand Vietnam China Grains (million mt) 1990 28 194 21 52 15 21 20 404 2000 40 235 30 62 17 31 35 407 Annual growth

3.6% 1.9% 3.8% 1.7% 1.4% 3.7% 5.7% 0.1%

Fruits & vegetables (million mt) 1990 3 76 7 10 13 9 7 149 2000 3 117 10 15 15 11 11 393 Annual Growth

1.7% 4.3% 3.8% 4.1% 2.1% 2.1% 4.7% 10.2%

Milk (thousand mt) 1990 1,594 53,678 14,723 598 20 130 60 7,037 2000 2,135 80,830 25,566 787 10 520 84 12,374 Annual Growth

3.0% 4.2% 5.7% 2.8% -6.5% 14.8% 3.5% 5.8%

Eggs (thousand mt) 1990 86 1,161 225 484 373 725 97 8,175 2000 159 1,749 351 783 518 807 185 22,826 Annual Growth

6.4% 4.2% 4.6% 4.9% 3.4% 1.1% 6.7% 10.8%

Meat products (thousand mt) 1990 308 3,929 1,325 1,448 1,089 1,323 1,079 30,421 2000 429 5,304 1,751 1,695 1,873 1,889 1,982 63,177 Annual growth

3.4% 3.0% 2.8% 1.6% 5.6% 3.6% 6.3% 7.6%

Fishery products (thousand mt) 1990 846 3,800 479 3,022 2,209 2,790 939 14,818 2000 1,661 5,609 627 4,909 2,291 3,736 1,961 43,069 Annual growth

7.0% 4.0% 2.7% 5.0% 0.4% 3.0% 7.6% 11.3%

Source: FAOStat.

Egg output grew at an impressive 9.2 percent per year in the eight countries over the 1990s.

Much of this was driven by the high growth rates and large volume of egg production in China, which

accounts for 83 percent of the total output for the eight countries. However, egg production grew rapidly

in most of the other countries as well, exceeded 4 percent per year in five of the other seven countries.

14

The patterns in meat production are quite similar to those in egg production. The overall growth

rate is quite high, 6.7 percent, driven largely by rapid growth in China which represents more than three

quarters of meat output in the eight countries. Meat production in the other countries grew more slowly,

but it still grew at more than 3 percent per year in five of the seven countries. Only in Pakistan and

Indonesia was meat output growth less than 3 percent.

Finally, the volume of fishery production grew at 8.3 percent per year over the 1990s, led by

China which accounted for about two-thirds of the output in 2000. The production growth rate was at

least 3 percent per year in six of the eight countries. Only in the Philippines did fishery production grow

more slowly than the population.

As mentioned earlier, growth in the output of high-value agricultural commodities is driven by

growth in domestic demand and growth in export demand. Which factor is more important? In general,

the growth in domestic demand for food is much more important than export demand in stimulating the

growth in output of high-value agricultural commodities. For example, in China, fishery exports doubled

over the 1990s, but this increase represents just 8 percent of the total increase in production over the

decade. The vast majority of the increase in production was to serve the growing domestic demand for

fish and seafood.

5 Consolidation and vertical coordination in food marketing

Growth in consumption and production of high-value agriculture commodities in Asia has been

accompanied by changes in the food supply chains linking the two. Changing consumption patterns

towards perishable high-value products embody changes in the characteristics of the products demanded,

in addition to increases in quantities demanded. Product attributes such as food safety, convenience, and

perceived organoleptic qualities become more important and are compensated by price premia relative to

traditional items. The new demands require changes in marketing infrastructure such as cold chains, and

better management of market information along the chain to deal with the risk of product spoilage before

final sale. Increasing exposure to risks from selling bad food may not be very damaging to commerce in

the wet market, where expectations are low and the rule is “caveat emptor”. But they are potentially

catastrophic to a branded supermarket, and therefore require increased control of risks along the chain.

New forms of retail chain and large-format stores such as supermarkets and their associated procurement

and distribution infrastructure have risen to fill these needs. The entry of private players from outside the

traditional food retailing sector and foreign direct investment by existing globalized supermarket chains

have also facilitated the consolidation of Asian retail chains in response to new consumer demand.

15

As elsewhere in the world, there has been a struggle in Asia for market power (measured as the

share of final value captured by each agent along the value chain) between food processors and food

retailers. In some cases the two have merged, where supermarkets sell their own brands. For the most

part, however, they remain separate, and the relative market power of each depends largely on the

production characteristics of specific goods. Processors tend to be in the ascendant with regard to highly

processed items requiring multiple inputs benefiting most from product differentiation, such as packaged

sauces. Vertically-integrated retail chains tend to be in the ascendant where processing involves value-

addition to a single well-defined commodity, as in the case of Tilapia fillets, packaged chicken legs, or

selected premium vegetables. Because of the need for control when handling perishable, quality-sensitive

products, the increased importance to profits of securing a steady throughput of products of consistent

quality, and in an effort to acquire increased market power, both supermarkets and agro-processors seek

privileged supply relationships from farmers. This presents both new threats and new opportunities to

farmers, particularly as the share of high-value commodities in total consumption of agricultural products

rises.

In sum, rapidly changing food demand patterns worldwide, driven by income growth,

urbanization, population growth and cultural change are inducing fundamental changes in the way food

gets from the farm to the table in Asia, and from Asian farms to non-Asian tables. These changed

circumstances are inducing change sin the industrial organization of agriculture in Asia. Changes are

showing up the most rapidly for the most perishable commodities with the highest income elasticities of

demand, such as fish, meat, eggs, and milk, but are also increasingly affecting higher value fruits and

vegetables. These changes are changing the relevant actors in the food sector, and have implications—

both positive and negative—for the traditional smallholder farmers that still constitute the bulk of Asia’s

population. Understanding how these changes affect the rural and urban poor requires working backward

from change in urban demand.

5.1 Consolidation and growth in the retail food sector

The retail food sector in developing countries has undergone considerable restructuring over the

past 15 years. One marker of change has been the growth in supermarkets and hypermarkets. In Latin

America, supermarkets have doubled their share of retail food sales during the 1990s in many countries,

now accounting for over 50 percent of retail sales in Brazil, Argentina, Chile, and Costa Rica (Reardon

and Berdegue, 2002). The number and sales of supermarkets in Africa are growing rapidly in Africa as

well, although outside South Africa their market share is still quite small (Witherspoon and Reardon,

2003). The reasons for their growth and diffusion are similar to those factors behind the growth of high-

16

value agriculture: income growth, urbanization, and the liberalization of foreign investment regulations.

In addition, the trend is also driven by increasing consumer interest in one-stop-shopping and food safety.

The growth of supermarkets is affecting food marketing channels and production patterns,

particularly for high-value foods. To ensure quality and consistent supply of perishable goods,

supermarkets are pushing the food marketing system toward more vertical coordination. At the same time,

supermarkets by stocking a variety of fresh and processed food products are stimulating agricultural

diversification.

Growth and consolidation of the modern food retail sector

Supermarkets and other modern retail food stores4 have grown rapidly in Asia. In 1990, China

had one supermarket. By 2002, there were 53,000 supermarkets and convenience stores (Hu et al, 2004).

Sales among these stores have grown from US$ 1 billion in 1995 to US$ 55 billion in 2002, or 77 percent

annually. In recent years, sales growth among Chinese supermarkets “slowed” to 30-40 percent per year

(Hu et al, 2004). In Thailand, annual growth in the number of modern food outlets was 11 percent in

2001-2002 (USDA, 2002). In the Philippines, the number of supermarkets has increased from 496 in

1994 to 3989 in 2001, a 30 percent annual growth (Digal and Concepcion, 2004). In Bangladesh, there

are 30 supermarkets today, all of which opened since 1999 (USDA, 2004). Indonesia has seen the

number of supermarkets and hypermarkets grow from 237 in 1989 to 1400 in 2002, though much of this

growth occurred before the Asian financial crisis of 1997-98 (Chowdhury et al, 2004).

The importance of supermarkets and hypermarkets in the total value of retail food sales varies

widely across Asian countries. In Thailand and the Philippines, supermarkets and hypermarkets

accounted for over half of retail food sales (USDA, 2002 and Digal and Concepcion, 2004). This is

consistent with the fact that Thailand has the highest income of the eight countries considered here and

the Philippines have the second-highest income and a high rate of urbanization. In Indonesia, these

modern retail outlets are estimated to represent 25 percent of retail food sales. In contrast, the share is

about 10 percent in Pakistan, less than 5 percent in India and Bangladesh, and 30 percent of urban food

sales in China.

Initially, supermarkets tend to be located only in the largest cities, catering to high-income

consumers. This is currently the case in Pakistan, Bangladesh, and Vietnam. As the number of

supermarkets and their market share increases, they spread to secondary cities and towns, as they have in

Thailand and are beginning to do in China. As part of this process, supermarkets also begin to cater to

4 For convenience, we defined supermarkets broadly to include hypermarkets, convenience stores, and

other modern retail outlets, although definitions vary from one country to another.

17

middle- and lower-income urban consumers, although it is likely that supermarket customers still have

incomes above the national average (Chowdhury et al, 2004; USDA, 2002).

Another trend is the change in product mix. Initially, supermarkets focus on non-perishable

goods such as rice, sugar, cooking oil, and noodles, with only small areas devoted to fresh produce, meat,

and fish. Consumers continue to rely on wet markets for these goods. As supermarkets expand and

acquire experience, they become more aggressive in sales of fresh produce (Reardon and Berdegue, 2002).

In China, fresh produce sales in supermarkets were negligible before 1995. By 2001, 20 percent of fresh

products in big and medium cities were sold through supermarket chains (Fang, 2002).

Causes of the retail food restructuring

The rise of supermarkets in Asia (as elsewhere) is partly driven by rising per capita income. The

importance of supermarkets is greater in higher income countries such as Thailand and the Philippines

than in Vietnam or Bangladesh. Furthermore, the growth in supermarkets seems to be related to

economic growth, both being highest in China. Finally, supermarket expansion slowed in Indonesia

following the Asian financial crisis, as consumers returned to traditional markets during the crisis (USDA,

2002 and Chowdhury et al, 2004).

It is likely that supermarket growth is also related to the pace of urbanization, since they are

limited to urban areas throughout the region. It is worth noting that, although incomes are lower in the

Philippines than in Thailand, the share of supermarkets in retail food sales is similar, perhaps due to the

higher level of urbanization in the Philippines (see Table 1).

Third, liberalization of foreign direct investment has contributed to the growth of supermarkets.

The growth of supermarkets in China began in the early 1990s, but took off after 1995 when rules on

foreign investment were relaxed. In Thailand, seven of the ten largest chains have foreign investment. In

Indonesia, foreign investment regulations were liberalized in 1998, and the share of supermarkets in food

retail sales rose from 6 percent in 1997 to 20 percent in 2001 (Chowdhury et al, 2004). India has

relatively tight regulations on foreign investment in the retail food sector. Although supermarket chains

are growing, particularly in the south, the organized food retail sector still accounts for less than 10

percent of food sales. In Pakistan, there is no foreign investment in food retailing. In 1998, the sector

was dominated by Utility Stores Corporation, a state-owned enterprise with 715 stores. About half have

since been closed in an attempt to reduce losses (SDPI, 2004).

Perhaps unique among Asian countries, China is using various policy instruments to accelerate

the transition from traditional stores and wet markets to supermarkets in order to address food safety

concerns and enhance tax collection. Based on the 2002-2008 plan of the Agricultural Products

Marketing of Beijing, there are plans to increase the share of fresh produce sold in supermarkets and

18

neighborhood markets to 90 percent by 2008, while reducing morning markets and wet markets to 10

percent (Bi et al, 2004)

Consequences of retail food consolidation

One consequence of the growth of supermarkets in Asia is increasing competitive pressure on

traditional retail outlets. For example, in Thailand, the total number of modern outlets grew at a rate of

10.6 percent from 2001-2002 while traditional outlets declined by 14.9 percent in the same period.

(USDA, 2002). In Indonesia, hypermarkets grew at a rate of 20 percent in 2002 while independent

grocers grew at 8.5 percent. (USDA, 2003). Even among supermarkets, there is a tendency to increase in

size.

Table 11. Structure of the retail food sector

Growth in supermarket outlets (%)

Country Year

Number of super-

markets

Share of super-

markets in total food sales (%) Period

Annual growth rate Source of information

South Asia Bangladesh 2004 30 1% 1999-2004 97% USDA, 2004 India 2000 2% 2003-2008 24-49% Chengappa, forthcoming Pakistan 2000 800 10% SDPI, 2004 SE Asia Indonesia 2003 1307 25% 1989-2002 15% Shaymal et al, 2004; USDA, 2003 Philippines 1995 3989 68% 1994-2001 30% Digal and Concepcion, 2004 Thailand 2004 600 54% 2001-2002 11% USDA, 2002 Vietnam 2003 <70 <2% Tam, 2004 China 2003 37,000 30%(urban) 1995-2002 36% Hu et al, 2004 Note: Supermarkets are defined to include convenience stores, hypermarkets, department stores, and large discount stores, though definitions vary from country to country. The India growth rate refers to a projection by EuroMonitor. Growth in supermarket sales is generally greater than growth in the number outlets since the average size tends to increase over time.

Another consequence of the growth of supermarkets is change in the procurement channels,

especially for fresh high-value products. Small chains and independent supermarkets often procure from

wholesalers and wet markets. But when supermarket chains reach a certain size, they generally establish

centralized food distribution centers that supply all stores in the chain. This vertical integration in the

wholesaling function allows them to standardize quality, improve bargaining power, and achieve

economies of scale in distribution. In addition, they usually adopt a list of preferred suppliers who are

known to be able to produce consistently the quantity and quality demanded by the supermarket chain.

The need to standardize quality (particularly if the chain offers store brands) leads to the development of

detailed private standards, most importantly for fresh fruits and vegetables, meat, and fish. The

19

procurement system is more demanding than the ones used traditionally by wholesalers and retailers.

Thus, the trend has been to move towards contracts with dedicated suppliers to reduce the transaction

costs of bargaining as well as reducing risks, wastage and guaranteeing food safety and quality control

(see Hu et al, 2004; Chowdhury et al, 2004; and Digal and Concepcion, 2004).

Supermarkets have started setting food standards, moving away from informal standards to

formalized private standards based on quality and food safety. This is partly a response to consumer

demand and partly a reaction to the lack of success of public standards. In some countries, there are

public standards, but where foreign companies have entered the supply chain, the standards become more

stringent. For example, in some countries in Southeast Asia, Carrefour has entered the market and is

demanding stricter standards then public programs such as the Carrefour Quality Certification program

(Hu et al, 2004).

The stringent quality control standards imposed by supermarkets are being led by consumer

demand, combined with a fear of damage to the reputation of the retailer. With more education and media

exposure to health and safety aspects of food products, consumers are demanding better packaging,

certification and quality control. Large storage facilities and bulk merchandising give supermarkets an

edge over smaller shops selling fresh and processed food. Quality standards are shifting the procurement

methods of supermarkets which have shifted to dedicated suppliers that can ensure strict quality standards.

5.2 Food processing consolidation

Food processing can be defined as the transformation of agricultural commodities as part of their

preparation for human consumption, encompassing relatively simple activities such as cleaning, grading

and storage as well as more involved transformation such as milling, canning and freezing. Food

processing is extremely important in rural development as it provides new outlets for agricultural products,

raises income, generates employment and reduces wastage.

The food processing industry in most countries reflects the changes in income and consumption

patterns. As income rises, the share of food expenditure declines and consumption patterns change from

staples to high-value food commodities. As discussed in Section 3, when incomes rise, households tend

to buy more processed food because they save time in preparation. Furthermore, as income rises,

households pay more attention to food safety issues and prefer to buy branded, labeled, packaged

products that they can trust in terms of quality.

Food processing industry also evolves as consumption patterns evolve. Low-income countries,

where the diet primarily consists of staples, have a low level of food processing. But as incomes increase

and diet diversifies, food processing becomes more important. As Table 12 shows, value addition in food

20

processing is positively correlated with GDP per capita, at least for the three countries for which

comparable data are available. As expenditure on processed food keeps increasing, processing becomes

more complicated with more attention to packaging, labeling, and branding.

Table 12.: Value addition in Food Processing

Value addition in food processing (%)

GDP per capita (1995 US$) 2002

India 7 493 China 23 944 Philippines 45 1209

Source: Government of India (2005) and World Bank (2004).

The development of food processing sector assumes significant importance due to the growth of

high-value products. The seasonality and perishability of high-value products demand that these products

be processed as swiftly as possible as storage for a long period is not possible and processing can avoid

wastage and shrinkage. Thus, the emerging trend of demand-driven growth in high-value agriculture has

to be accompanied side-by-side by the development of the food processing sector. Countries which have

neglected the development of food processing sector are paying a heavy price. This is highlighted in the

case of India and Thailand.

Although India is one of the leading producers of grains, milk, fruits and vegetables, livestock,

and fish, the share of output that is processed is low by international standards. It is estimated that the

processed share of fruits and vegetables is just 2 percent, meat and poultry 2 percent, milk by way of

modern dairies at 14 percent, fish at 4 percent and bulk meat de-boning at 21 percent. (Government of

India, 2005). In contrast, Thailand is one of the world’s top 10 exporters of processed poultry. The food

processing industry has grown more than 20 percent annually between 1980 and the late 1990s (USDA,

2002)

As shown in Table 13, food processing industries account for 13 percent of domestic

manufacturing in Thailand as compared to 18 percent in India. But it accounts for 4 percent of GDP and

generates total export value of over US$10 billion, the second largest exporter of processed food among

developing countries after Brazil. Also, most of the food processing firms in Thailand are small

enterprises (88 percent of all registered processing plants) of less than 10 million bahts. About 80 percent

of their raw materials are locally procured, and hence food processing is seen as a major revenue earner as

it does not depend on imported inputs (APO, 2001). The case of Thailand shows how a food processing

industry, based on small enterprises and procuring locally, can be a major revenue earner for the country.

21

Table 13. Comparison of food processing in Thailand and India

Year Percentage of total domestic manufacturing

Percentage of GDP

Total value of exports (US$ billion)

Thailand 1999 13 4 10 India 1998/99 18 2 3 Sources: TDRI (1999), Gulati et al (1994), Srinivesan (2000), and Athukorala (1998).

The importance of processed food as an export revenue earner has gained importance in recent

years. Many countries have exploited this as a new source of export growth. As Table 14 shows,

processed food exports have been growing faster than primary and agricultural product exports in all

countries studied. Bangladesh, Indonesia and Thailand have exhibited annual growth rate of 15 percent or

more. There seems to be a positive correlation between high income growth and exports of processed

food with the exception of Bangladesh and Philippines. In spite of being the poorest country among the

ones studied, Bangladesh has performed better than most countries. Its 15 percent growth rate is double

the developing countries average and higher than its manufacturing exports. This is related to the

importance of fishery product exports in Bangladesh, which are processed. Philippines, on the other hand,

has a surprising low rate of processed food exports.

The growth of the domestic food processing market is affected by tariff and non-tariff barriers.

Thailand protects its domestic processing industry through the use of tariff and non-tariff barriers. Duties

on imported consumer-ready food products range from 40-50 percent, highest among ASEAN countries5

(USTR, 2002) High tariffs are often kept on processed food products to keep value addition in the

domestic market, while low tariffs are kept on raw materials for processing. Effective rates of protection

(EPR) are a tool to measure the effects of commodity tariffs on the nominal tariff of a processed product.

For cookies, positive ERPs (tariff escalation) has sheltered the EU cookie industry while negative ERPs

in the United States and Japan, due to high protection on sugar, may be a reason for increased investment

by these countries in offshore production facilities. Tariff escalation is prevalent for China when

comparing tariffs on carcasses and fresh and frozen boneless beef with those on beef cattle. (US ITC,

2001)

5 It is difficult to know whether the tariffs are the cause of growth in the food processing sector, by

offering it protection from imports, or the effect of this growth, since it creates a political lobby for protection. Clearly, there is little political pressure for high tariffs on products not produced in the country.

22

Table 14. Growth of processed food exports

Annual Compound Growth 1980-1999 Country Processed

food Primary products

Agricultural Products

Manufacturing

South Asia Bangladesh 15.1 0.3 6.7 11.7 India 8.4 6.5 7.3 11.3 Pakistan 6.9 3.1 3.7 9.4 SE Asia Indonesia 14.6 10.1 9 21.6 Philippines 5.2 4.3 4.4 15.7 Thailand 17.0 9.6 10.9 20.9 All developing countries 8.7

Source: Athukorala and Jayasuriya (2003) based on UN trade series(Series D). Note: Data not available for China.

Tariffs on dairy and sugar products in developed countries tend to be extremely high due to the

existence of tariff rate quotas with small quota quantities and extremely high over-quota rates. Tariffs on

beef, poultry, pork, eggs, and certain types of processed fruits and vegetables are generally very high.

However, there are regional variations with certain large markets like India and Pakistan imposing tariffs

of over 35 percent in meat products, while others like Indonesia have applied rates of 5 percent or less.

Tariffs on processed fruits and vegetables are over 20 percent for India and China. In some countries like

in the Middle East, shelf-life standards has been cited as an important barrier to imported processed fruits

and vegetables. These products are given a shelf-life of 12 months and must have 50 percent of their

shelf-life remaining upon entry. However, most processed fruits and vegetables are produced once a year

from the seasonal fresh products and such restrictions prevent year-round distribution of these products.

(US ITC, 2001)

Processed foods also face a variety of non-tariff barriers, such as export certification and

registration, labeling, traceability, food standards, intellectual property rights, customs procedures, and

sanitary and phyto-sanitary (SPS) restrictions. For example, Thailand requires manufacturing information

such as the points of ingredient addition, technical information on the packaging material, specifications

on the finished product, certificates attesting to the product being free of genetically modified organisms

(GMO), as well as samples of the product and packaging. The products also need to be re-registered each

time packaging or simple ingredient is changed. (US ITC, 2001)

Foreign direct investment (FDI) is an important component of food processing. At one level, the

entry of FDI and consolidation of retail industry has influenced processed products. Supermarkets are

stocking fresh as well as processed food products, thus generating higher demand for processed food. At

23

another level, FDI in the processing sector has generated technology transfer and productivity growth.

The relationship between exports and FDI is ambiguous. In some cases, companies will simultaneously

export and invest in foreign markets. In other cases, where trade barriers are high, investments might

replace exports.

5.3 Emerging forms of farmer-buyer vertical coordination

A key aspect of markets for high-value agricultural commodities is that large price premia can be

secured by sellers for products that are both superior, and which the buyers also believe to be superior.

Being of high quality is not enough; buyers have to believe it in the store and be ready to pay more than

they would for otherwise similar, but lower quality items. A second distinguishing fact of high value

agriculture is that it is often difficult to distinguish by simple inspection what the key quality attributes of

the item are at the time of sale.

Many of the characteristics sought in the high value food market relate to issues such as food

safety (as in microbial content, pesticide residues, etc.), fat content, and other taste issues. The desirable

properties in each of these areas are typically hard to see. This is especially true in the animal-source

food area where often the first sign that the milk is bad is getting sick, or where the quality of meat is only

evident once the animal is slaughtered and the dish prepared. The quality of vegetables may be easier to

see at sale, if only because destructive sampling by the buyer is more likely to be an option. Even so,

pesticide residues are hard to observe. One estimate of pesticide use on eggplant for the Delhi market is

that each fruit has been sprayed 80 times prior to final consumption (Lumpkin, 2005).

The key insight is that quality or safety are the result of a long process by which a commodity is

produced and handled from soil to fork. Having confidence (exhibited by a willingness to pay more) is a

function in having confidence that the right procedures were followed, and this can only be believed if the

buyer has enough information about how the product was produced and handled. Essentially, this requires

branding the product in some fashion that is thought by consumers to assure that the right procedures

were followed.

Without the means to see the desired characteristics at sale, consumers of high value agricultural

commodities are typically unwilling to pay price premia for quality (regardless of the actual quality). An

important incentive for vertical coordination of the supply chain is that both farmers and buyers lose if

consumers do not have confidence in the superior quality of the final product. This situation arises

because the buyer cannot know what the producer and handlers along the production and marketing chain

know about the product.

24

In economic jargon, there are transactions costs that arise from the differences between buyers

and sellers in what they know about what is being sold. These transactions costs are typically hard to

observe, but are very real. In the Philippines, for example, it was found that smallholder contract farmers

and large farmers gained exactly the same price per kg liveweight for pigs sold on the open market in

Manila—the public had equal confidence in their output. However, otherwise identical smallholders

selling otherwise identical pigs were paid on average 8 percent less per kg liveweight for pigs in the same

markets (Costales et al, 2003). Similarly, large scale milk operations in India and elsewhere in the

developing world typically sell milk of a given fat content at a slight premium over smallholder sales

(Delgado et al., 2003). Small-scale farmers lose because they cannot be rewarded consistently in the

market place for better quality. Buyers lose because they cannot consistently get the quality and

reliability that they need.

Thus the crux of the issue: high value agricultural commodities that are perishable are inherently

quality-sensitive and subject to high transactions costs, particularly in the case of smallholder production.

These transactions costs arise from asymmetries of information between buyers and sellers and the nature

of the predominant agricultural production systems in Asia. They are difficult to observe, but are

devastatingly real. They are additional to the high marketing costs that arise when infrastructure is poor.

Because of high transactions costs in the high value agriculture sector, institutional forms of vertical

coordination are key to giving both buyers and sellers a better deal. The integrating institutions distribute

knowledge about the product more evenly between buyers and sellers along the marketing channel.

Because both sides win, there is enough value-added to make everyone better off.

Both South and Southeast Asia have witnessed the rise of arrangements for vertical coordination

of primary production of high-value items with input suppliers and processing/exporting firms during the

last 20 years. Input suppliers like seed companies and feedmillers have typically promoted profit and

risk-sharing relationships with farmers. Contract farming can be defined as an agreement between a

series of farmers and a retailing, processing and/or input supply firm for the production and supply of

agricultural products under forward agreements, frequently at a predetermined price, in return for the

purchaser providing production support. The latter often includes quality inputs given on credit and

technical advice (Eaton and Shepherd, 2001). Typical contract farming schemes in animal production

involve feedmillers who supply young animals, feeds, veterinary medicine, and extension advice on credit

to farmers who provide holding sheds, dispose of waste, and provide all required labor, water and

electricity. Major production decisions are made by the integrating firm. Processors get involved in

contract farming when they need a more reliable supply of raw materials than they can easily get

otherwise.

25

Milk is a specialized case of contract farming, where dairy coops process the milk for sale in

addition to facilitating farming access to inputs and extension. What makes milk in many places different

from most other contract farming schemes is that the integrator is typically cooperatively owned, although

this distinction blurs in the case of large-scale dairy processing of the public variety. While seed

companies have occasionally become involved in contract farming of vegetables, typically contract

farming of high value crops is carried out under the leadership of processing and exporting firms where

quality control throughout the production process is critical to achieving a quality grade and quality can

be monitored along the chain. Cut flowers provide a good example, and fruit and vegetables for

industrially-processed foods are another. Agro-processors and retail chains that need to be re-assured

about the quality of their raw materials find it costly to monitor the quality of what they buy, particularly

when they are buying from many smallholders. Whether dealing with animal-source foods or high value

crops, integrating firms typically use a variety of institutional arrangements for obtaining reliable supplies

of raw materials of consistent quality for processing.

Each form of arrangement embodies a different way to share the risks, cost, and benefits of high-

value commodity supply chains. At one extreme, vertically-integrated corporate farming typically

involves a processor or exporter who finds it expedient to produce the basic raw material themselves

without having to deal with semi-independent production units. Plantation crops such as tea, rubber,

coconuts and sweet bananas are typical commodities on such holdings. Because of the nature of these

commodities and how they are grown, the per unit costs of supervision to obtain quality output are low.

Often these sorts of activities can be carried out independently of processing, but typically only on large

scale farms that can secure the inputs and extension advice on a cost effective basis per unit of output.

Simply producing the raw materials oneself is often not a good option for processors. The monitoring

and supervision costs of large firms in ensuring that employees use only the right inputs at the right time

in the right way can be steep, particularly as the employee does not directly benefit at the margin from

doing a better-then-average job.

On the other hand, contract farming arrangements are typically observed for commodities that

require considerable close monitoring in production, have characteristics that are hard to ascertain on an

individual basis at sale, require specific quality inputs for quality outputs, have high requirements in terms

of producers credits, and embody a substantial degree of market risk, defined as a highly fluctuating

producer price across time. Integrating firms are also able to expand supply quickly without having to

invest in land or buildings. They have much lower adjustment costs if demand should fall, as the loss is

born by the contractor. Contract farming also is attractive to industrial firms if it reduces their potential

liabilities, such as in waste disposal from animal agriculture, which becomes the problem of the

contracting farmers. Finally, certain agro-processing or retailing firms may simply be unable to get

26

reliable supplies of agricultural produce from large scale farms, whether independent or integrated. In this

case, they need to deal with smallholders, but also to become involved in the production process in a way

that it produces what they need.

The attractiveness of contract farming schemes to integrators is mitigated by the cost of having to

deal with myriad small suppliers, particularly if the latter are widely dispersed and of small individual

size. There is always the risk that if market prices rise, farmers will be tempted to not honor contracts that

they feel are less favorable. Finally, integrators share in the production risk of individual small farmers,

which they would not if they only procured final product in the open market. In catastrophic cases, such

as the bird culls in Southeast Asia mandated by health measures in regard to Avian Influenza outbreaks,

this can be a crushing burden that leads to integrator withdrawal in favor of production methods with a

higher degree of direct control.

Typical contract farming commodities are poultry (broilers in particular), pigs, milk, certain high

quality fruits and vegetables for processors, and to a lesser extent inputs to industrial processing that

require close producer quality supervision such as coffee, tea, cocoa, and sugar. In Thailand, for example,

virtually all commercially produced broilers are produced under contract, whereas the corresponding

figure for the Philippines is 80 percent (Delgado et al, 2003). In India, roughly 11 percent of milk was

produced within the public cooperative system in 2001, but a higher share would be correct if contracts

with the emerging private sector dairies are included after 1991, and perhaps the majority of production if

informal contracts between informal sector milk traders (dudhyas) and producers are included (Sharma

et al., 2003). Under some circumstances, contract farming can represent an attractive short run

opportunity for smallholder producers, and even offer their best chance to remain involved with high-

value agricultural production over time.

6 Vertical coordination of high-value agriculture and smallholders

6.1 What contract farming does for small-scale farmers

The analysis in section 4 above shows that the only part of agriculture in developing countries

that will continue to grow significantly faster then population in the next twenty years is the high value

sector. The implications for the vast mass of smallholder farmers in Asia are sobering: to significantly

improve their incomes per capita over the next twenty years, they must either be part of the shift to high-

value agricultural production or increase the share of income they get from non-agricultural sources.

Furthermore, the analysis in the preceding section suggests that unless smallholders become vertically

integrated with processors and retailers, they will increasingly have difficulties in participating in

increasingly more demanding high-value markets. Finally, even if markets worked well in every sense,

27

many poor rural people are faced with such poor infrastructure that they would have trouble taking

advantage of new urban and international market opportunities under the best of conditions. To quote

Torero and Gulati (2004) , they must overcome a “real access gap” of being able to cost effectively

transport their produce, before being able to address a “market efficiency gap” that revolves around being

competitive with better organized, better informed, better capitalized and larger scale producers.

Two instruments appear critical to break this deadlock for the smallholders: physical

infrastructure (such as information technology, roads, and ports) that connects smallholders to markets,

and a set of accompanying institutions that reduce marketing risk and transaction costs in the process of

exchange between producers and consumers. Appropriate policies of investment in infrastructure need to

go together with well-functioning market institutions, to take advantage of market opportunities to sustain

increased agricultural output and raise rural incomes. This is a critically important for smallholders in

countries recently experiencing market liberalization. Even if adequate hard infrastructure exists, farmers

capture little of the value that they create when market information and markets themselves are not

accessible to the smallholders.

Previous conventional wisdom had it that institutions would improve as a consequence of

individuals’ self interest, and therefore take care of the transaction cost problems arising from information

asymmetries (Torero and Gulati 2004). The reality is in that the presence of coordination failure,

innovation failure, and authority failure, the necessary institutional solutions to overcome high

transactions costs facing smallholders fail to emerge. The high risks of production and cycles of over-

supply and price depression create financial risks throughout the distribution chain; these inhibit

investment and access to capital. Monopolistic practices, corruption, and excessive regulation also add to

the burden of the rural marketplace. The high costs, risks, and “friction” in high-value agricultural

markets prevent these markets from achieving sufficient scale for efficiency and similarly prevent the

low-cost and reliable supply of production inputs such as seed, fertilizer, and other goods to farmers. Very

poor farmers also lack the political empowerment, market knowledge, and business knowledge to address

these market roadblocks.

Thus, poor rural farmers typically lack the capacity to improve and influence the markets upon

which their lives depend. But some of these assets can be developed through effective organization,

technical training, and means for assembly and communication. Pro-poor market institutions are needed

to reduce transaction costs, manage risk, build social capital, enable collective action, and redress missing

markets. The necessary institutional infrastructure to facilitate market exchange is a critically important

area in countries recently experiencing the shortfalls of market liberalization with regard to smallholder

agriculture. When market information and markets themselves are not accessible to the rural poor,

farmers capture little of the value that they create, demand and supply are highly unstable, and

28

distribution costs for rurally produced goods are very high. Small farmers in Asia in particular tend to be

subject to a specific set of marketing problems.

First, traditional smallholder farmers in Asia typically receive relatively low prices for their

produce. This stems both from abnormally high—on a world scale—margins between the farm-gate and

retail price, but also from low market trust and reputation typically accorded undifferentiated smallholder

output when true quality is not known to the buyer at the time of sale. With respect to margins, farmers in

India receive only 20 to 30 percent of the retail price of fruits and vegetables, compared with 50 percent

or more in the United States. (US Dept. of Commerce, 2001). Institutional arrangements such as contract

farming can reduce the number of intermediaries, wastage, transaction costs and market risks. With

respect to market trust and reputation, which a large firm approaches through branding, smallholders are

in a disadvantageous position. They do not have a sufficient sales volume to differentiate the product of

individual producers from each other. Sales of sub-standard goods by other smallholders rebound on

them.

Even when it is possible for smallholders to band together to give a geographical brand to their

product (i.e. Central Gujarat milk), it is not helpful unless a mechanism is in place to credibly ensure that

bad product is not included, and to gradually improve the quality of existing product. Performing this

market function requires some form of collective action on the part of producers and a form of

governance that translate the discipline of the market into enforceable incentives for compliance with

norms. Much of the practical implementation of quality improvement revolves around improving the

quality of inputs used and optimizing production and handling practices. In effect, credible certification

of output quality revolves around credible certification that only the right inputs and procedures were use

din production and handling. Contract farming is the private sector solution to accomplishing these

functions in a way that distributes costs, benefits, and risk sin a manner to maintain incentives for all side

to participate. In the animal products sectors, where purchased variable inputs such as young animals and

feed are typically 70 percent of the farm-gate price of the output, input supply firms naturally tend to

provide the integrating function of contract farming. Transactions costs apply to inputs as well as outputs.

Small farmers often are ill-equipped to know the true quality of the animal genetics and feed resources

that they buy, compared to larger farmers than either mill their own feed as most in countries such as the

Philippines do (Costales et al, 2003), or enforce better compliance with standards from suppliers.

Improved inputs are combined with better practices to embody new technology for production.

Contract farming schemes are typically associated with significant improvements of productivity of

contract farmers compared to otherwise similar independent farmers, particularly in the case of small-

scale farming (Delgado et al, 2003). This observation is not limited to livestock enterprises. The Pepsi

project, a joint venture among Pepsico, Voltas and Punjab Agro Industries Corporation approved in 1988

29

by the Government of India in the State of Punjab, set up the biggest tomato paste plant in Asia, with the

capacity to process 650 tons of tomato a day. It contracted hundreds of tomato farmers. It introduced the

technology of deep chiseling, and new methods of transplantation such as shovel techniques and bed-head

planting, in addition to the introduction of new seed varieties. The technological innovations introduced in

contract farming increased productivity and reduced costs. Within three years of operation, tomato yields

increased from 7.5 to 20 tons per acre. The harvesting season for tomatoes was extended from 25 to 70

days and the company also successfully initiated the winter cultivation of tomato in Punjab, with the help

of green house technology dissemination (Sukhpal, 2004).

Market risk in terms of fluctuating prices is another problem of great concern to smallholders in

the high-value area. The short-run price elasticities of demand and supply for perishable products can be

rather inelastic, leading to considerable day-to-day price instability for these commodities. Eggs and

broilers in southern India whose daily prices fluctuate as much as 10 percent are an example in a sector

where average profit margins are 4 percent (Mehta et al. 2003). Whether on a fee or contract farming

basis, the returns to the contract farming enterprise are likely to fluctuate less than for independent

farmers. Another factor is that in some localities in Asia inputs such as feed are taxed. Integrated

operations escape this tax through accounting transfers of feed to contractors that do not count as sales.

Inordinately high market risk can be substantially mitigated by improved methods of sharing

relevant market information in a vertically coordinated framework. One such initiative is the e-choupal

initiative in India, organized by the Indian Tobacco Company’s (ITC). E-choupal connects 3.1 million

farmers from 29,500 villages in 6 states in India through Internet kiosks running on solar-charged

batteries and connected by satellite links. At the e-choupal sites, farmers can a) obtain information on

commodity prices, weather, and news; b) search for customized knowledge on farm and risk

management; c) purchase inputs and other products, and d) sell their crops to ITC centers or the local

market. E-choupal has been used to source a range of agricultural commodities like foodgrains, oilseeds,

coffee, and aquaculture and market a variety of goods and services like agri-inputs, consumer goods,

insurance and market research. The new “e-chain” registered transactions of US$ 100 million in 2003-04

and has reduced transaction costs for a typical soybean farmer from Rs 705 to Rs 335 per metric ton.

(Sivakumar 2004) Farmers selling through e-choupal realize at least 2.5 percent higher price for their

crops than they would receive through the government auction system because of lower transaction costs.

At the same time, procurement costs for ITC are also reduced by 2.5 percent as they save on commission

paid to traders. The system provides direct market access to farmers and it is estimated that their

incremental income is over 20 percent. (ITC case summary by IFC)

Vertical coordination is also an essential way to lower the transaction costs of lenders to supply

credit to small rural producers, by helping ensure that the capital is used as intended by the lender and in a

30

way that ensures repayment. Typical contract farming schemes provide inputs on credit, thus providing

the farmer with an important additional resource. Market interest rates for Asian smallholders are

typically very high, if they exist at all. In the Philippines, for example, small-scale pig farmers could

borrow at private banks for 24% per annum in 2001, whereas large-scale farmers could often borrow at

12%. Credit provided within a contract farming scheme is more likely to be repaid as the integrator has

better control over the final disposition of output. In the animal sectors at least, empirical analysis of field

date in Asia consistently point to the role of credit in allowing entry of smallholders in high-value

agricultural markets (Delgado et al, 2003). Since improved production practices are critical to achieving

quality, contract farming schemes typically are associated with a much higher incidence of farm visits by

technicians that is independent smallholder farming. The contract farming scheme basically imposes a

package of practices, technology and inputs that it then monitors the use of (Tiongco and Delgado, 2005).

Beyond producer credit, vertical coordination also helps mobilize substantial new private sector

capital for investment in rural infrastructure and extension. Indian telecom giant Bharti Enterprises

recently signed a joint venture of US$ 50 million with de Rothschilds to promote agro-exports from India.

As part of the venture, sourcing and contract farming activities will be carried out in six states for fruits

and vegetables for exports to global markets. This venture will then extend to processing activities. To

promote this, the venture is using innovative technology to do infrastructural development in the entire

supply chain through the creation of storage facilities, processing plants and cold transportation

capabilities. There are also plans to establish a research centre and model farm which will work towards

identification and adoption of conventional and emerging technologies, especially on hybrid seeds and

agro-farming techniques, and promote their usage on the field. (Bharati , 2004).

Because it accomplishes these functions from start to finish, contract farming schemes are in a

position to credibly certify the quality of output. They can do this by directly marketing items raised by

contracting farmers themselves, or else by branding that farmers output, which is then sold as such on the

open market. Both forms exist in Asia, with Venkateshwara Hatcheries broiler operations being an

example of direct marketing (Mehta et al. 2003), and the Soro Soro Ibaba (Swine) Cooperative in the

Philippines being an example of branding for sale by the farmer on the open market (Costales et al, 2003).

Finally, where direct procurement from the farmer is practiced, as is typical for broiler sin Thailand, both

fee-based and price-guarantee schemes are used for farmer incentives. Fees are a per unit of product

return for the farmer’s labor, land, buildings, water and electricity. Price guarantees increase the

incentive to farmers to cut costs, but greatly increase the burden on integrators to monitor production

practices and input usage (Poapagsakorn et al. 2003).

31

6.2 Impact of contract farming of animal products on smallholders in Asia

Although contract farming of vegetables is now relatively common in Asia, there is more in-

depth empirical evidence available with regard to the older practice of contract farming for animal

products, which has been present in both South and Southeast Asia since at least the 1970s (Tiongco and

Delgado, 2005). A major IFPRI-FAO research study on comparing profit efficiency of small versus large

independent livestock producers and smallholder contract farmers was carried out with national

collaborating institutions in Brazil, India, the Philippines, and Thailand in 2001-2003 (Delgado et al.

2003). The main focus of these studies was to assess the ability of independent and contractor

smallholders to compete effectively against the competition from large scales producers. The metric of

competitiveness used emphasized a necessary condition and a sufficient condition (provided the necessary

condition was met) for smallholders remaining competitive. The necessary condition was that they were

able to secure average unit profits that were at least as high as those of the commercial sector. Given the

low volume of production that most smallholders have to rely on for their livelihood, they would be

driven pout of business if high volume producers could afford to cut their average unit profits

significantly below those of smallholders. If smallholders meet the necessary condition, the sufficient

condition for them remaining competitive, other things equal, is that they are at least as efficient in

securing profits from a given kit of farm resource sand facing a given set of input prices as are larger

farms.

The main empirical findings based on farm surveys of independent and contract farmers of

different scales in the Asian cases are summarized in Tables 15, 16, and 17 below. For broilers, most

production in the Philippines and almost all in Thailand is through contract farming. The latter is

becoming more common in India as well, but is less well established than in Southeast Asia. Philippines

contractors were predominantly of the fee type, and smaller and larger size contract farms did equally

well per unit of output, and in both cases did significantly better than did either small-scale or large-scale

independent producers. This situation probably reflects distortions in the Philippines concentrate feed

market that de facto give tariff preferences to integrators over independents, and also reflects the market

power of the few providers of branded day-old-chicks, who are also the integrators. In Thailand, small-

scale fee contractors did a little better than small-scale forward contractors and independents, but this

relationship was strengthened considerably at larger scales. In India, where contract farming of broilers is

not as well established, the opposite relationship is very much evident. Independents did substantially

better than contractors in the survey year, even when all inputs were properly costed across both sides of

the relationship. These relative findings would change depending on whether prices rose or fell during

the survey period, since contractors are not affected by price changes once the production period has

begun but independents are. The picture for swine producers in the two Southeast Asian countries, shown

32

in Table 16, is quite different from the picture for broilers. Here independents did substantially better

than contractors on the whole. This was due in part to strengthening during the year of pork prices.

Table 15. Average profit of broilers by farm size, independents versus contracts in selected Asian cases (2002, US $ per kg of liveweight)

Farm Size

Smallholder <10,000 Large/Commercial >=10,000

Independent Contract Independent Contract India 0.11 0.01 0.09 0.03 Philippines 0.03 0.08 0.02 0.08

Forward Contract &

Independent

Per-bird Fee Contract

Forward Contract &

Independent

Per-bird Fee Contract

Thailand 0.02 0.03 0.06 0.04

Source: Compiled in Tiongco and Delgado (2005) from R. Mehta, et al.. (2003); A. Costales, et al. (2003); and N. Poapongsakorn, et al. (2003)..

The picture for the relative efficiency of broiler producers in India, the Philippines and Thailand

is shown in Table 17. In all cases except for smallholder producers in India, where there is no significant

difference in profit efficiency of independent and contract farmers, contract broiler farmers are

significantly more efficient at producing profits than are independent farms. Certainly in the Southeast

Asian case and probably eventually in India, there does not appear to be much chance for continued

independent production of broilers. Although the showing of smallholder independents versus large

scale independents in 2002 was quite robust according to the unit profit and profit efficiency criteria set

out above, the future of small scale broiler operations and is very much in doubt in Southeast Asia due to

the measure being put in place to control Avian Influenza, which very much favor large-scale contained

operations in closed housing with electrical cooling.

33

Table 16. Average profit of swine by farm size, independents versus contracts in selected Asian cases (2002, US $ per kg of liveweight)

Farm Size

Philippines Smallholder <100 head Large/Commercial >=500 head

Independent Contract Independent Contract 0.52 0.04 0.38 0.05

Smallholder <500 head Large/Commercial >=1000 head Independent Contract Independent Contract

Thailand

0.28 0.27 0.36 0.04

Source: Compiled by Tiongco and Delgado (2005) from R. Mehta (2003); A. Costales, et al. (2003); and N. Poapongsakorn, et al., (2003).

Table 17. Relative profit efficiency of independents versus contracts in selected Asian cases (2002, as percentage of performance of best farms in country case)

Farm Size

Smallholder <10,000 Large/Commercial >=10,000

Independent Contract Independent Contract India 41 40 24 66 Philippines 33 72 45 75

Contract < 5,000

Contract 5-10,000

Contract < 10-15,000

Contract >20,000

Thailand

78 84 81 85

Source: Compiled by Tiongco and Delgado (2005) from R. Mehta, et al.(2003); A. Costales, et al. (2003);and N. Poapongsakorn, et al. (2003).

In a different study, Birthal, Joshi and Gulati (forthcoming) quantified measurable marketing

costs incurred by contract and non-contract producers of milk, broilers and vegetables in India by

choosing one firm in each of these commodities. These included Nestle India Limited – a multinational

firm for milk and milk products, Venkateshwara Hatcheries Limited (VHL) – a private sector domestic

firm engaged in contract broiler farming, and SAFAL, a subsidiary of the parastatal Mother Dairy, which

sources fruits and vegetables through producers’ associations. It was found that contract farming attained

substantially higher net profit than non-contract farming, as suggested by the Southeast Asian work. Table

18 shows that both production and marketing costs were lower for contract farming. The share of

marketing cost in total cost for non-contract farmers was 20 percent for milk and 21 percent for

vegetables, but it was only 2 percent in both cases for contract farmers.

34

Table 18. Production and transaction cost of milk, broiler and vegetable production in contract and non-contract farming (Rs/ton)

Contract farming Non-contract farming Commodity Production

cost Transaction

cost Total cost Production

cost Transaction

cost Total cost

Milk 5,586 100 5,686 5,728 1,442 7,170 Broiler* 808 38 846 27,322 90 27,412 Vegetable** 1,485 35 1,520 1,630 437 2,067 Note: For broiler, the firm provides free chicks, feed and medicines to the contract farmers. Vegetable costs refer to spinach Source: Birthal, Joshi and Gulati, 2005.

As discussed above, contract farming leads to a sharing of risks between the producer and firm.

This is shown by Birthal, Joshi and Gulati (2005) in table 19, which illustrates the fluctuations of yield

and profit across seasons for contractors and independents. Table 19 shows the coefficients of variation

(CVs) of profit of contract farmers to be almost stable over different parts of the year, but they are high

with sharp fluctuations for non-contract farmers. Since there was not much difference in CV of yield,

price volatility was the main reason for high variability in profits of independents.

Table 19. Variation in yield and profit of broiler in contract and non-contract farming in India across seasons (coefficient of variation in percent)

CV of broiler yield CV of net profit Season Contract farmers

Non-contract farmers

Contract farmers

Non-contract farmers

January-February 10 8 22 65 March-April 8 16 20 137 May-June 5 22 22 296 July-August 20 21 20 270 September-October 9 7 26 107 November-December 8 7 26 49 Note: The coefficient of variation (CV) is calculated as the standard deviation divided by the sample mean. One cycle completes in 38 days. CVs are computed within cycles across sample. Source: Birthal, Joshi and Gulati, 2005.

Conventional wisdom suggests that other things equal, agro-processors will find it more

advantageous to deal with a smaller number of larger suppliers of raw materials than with a larger number

of smaller suppliers. It is therefore interesting that Birthal, Joshi and Gulati (forthcoming) observe that

firms were finding it more convenient to contract with smallholders and their associations due to: (i) a

lower risk for overall supply in the event of crop failure of one or few farmers; (ii) More flexible

production portfolios of smallholders, which would help them to quickly respond to consumers’ changing

preferences; and (iii) Smallholder can ensure better quality as they are seemingly more likely to strictly

comply with the mandated production practices of firms; and (iv) A low marketable surplus of

smallholders increases their dependency on the firm for profit maximization. Furthermore, apprehensions

35

about contract farming leading to exploitation of farmers were shown to be unfounded in at least one year,

as contract farmers in the year of the survey were offered a higher price than the prevailing market price.

Vegetable contract farmers received 8 percent higher prices on average, and milk producers received 4

percent more.

6.3 Impact of supermarket growth on smallholders

The growth of supermarkets with their heightened concern regarding food quality, consistent

volumes, and food safety represent a threat and an opportunity for small farmers. It is a threat in the sense

that food safety and quality control are barriers to the entry of smallholders in the supply chain. For

example, in China producers need to have their production environment sampled and checked, provide

production records and inspection reports in order to be certified as “green food” grower. Producers of

“green food” can get a margin five times larger selling to supermarkets, so supermarkets signing contracts

with large producers with these certifications can not only ensure quality control but also make greater

profits (Bi et al, 2004). In Philippines, for vegetables, only professional suppliers of small to medium

scale operations maintain their place in the supply chain. Small producers who managed to supply

hygienic vegetables found it difficult to maintain this business link and eventually dropped out. The

barriers to integration of smallholders in this chain have been countered in some countries by the

formation of cooperatives, contract farming and producers’ association which supply directly or through

some intermediaries to modern retailers.

However, supermarkets also represent an opportunity for small farmers in that supermarkets

know the product requirements of high-income consumers and have the incentive to transmit this

information to the farmer through mechanisms of vertical coordination. Thus, potentially supermarkets

offer access to relatively high-income consumers and assistance in meeting their requirements. In

practice, supermarkets rarely buy directly from small farmers, with or without contracts, but rather

procure goods through commissioned agents or assemblers. Depending on the production characteristics

of the crop and the distribution of farmers by size of farm, these assemblers may or may not choose to

work with small farmers.

The preponderance of smallholders in South and Southeast Asia makes their inclusion in the

changing retail structure especially important. The average size of land holdings is around 1.6 hectares in

South Asia and Southeast Asia. Farms of less than 2 hectares in size account for 88 percent of the

operated area in Indonesia and 81 percent in India. Farms are even smaller in Bangladesh and Vietnam,

where over three-quarters of the farms are less than one hectare in size (see Table 20).

36

Table 20. Structure of farm size in South and Southeast Asia

Bangla-desh

(1996)

India (1995-1996)

Pakistan (2000)

Indonesia (1993)

Philippines (1991)

Thailand (1993)

Vietnam (1994)

Percentage of farms <1 ha 79 62 36 71 37 20 88 1-2 ha 18 19 22 17 28 23 9 > 2 ha 2 20 42 18 35 57 3 Percentage of total farm area <1 ha 41 17 7 30 7 3 N.A. 1-2 ha 42 19 11 25 16 9 N.A. > 2 ha 17 64 82 45 77 88 N.A.

Source: Bangladesh: Census of Agriculture, 1996, Agricultural Sample Survey 1997; India: Agriculture Census Division; Nepal: Agriculture Census 1991; Pakistan: Pakistan 2000 Agricultural census (only private firms); Indonesia, Philippines, Thailand and Vietnam data taken from the Supplement to the Report on the 1990 World Census of Agriculture, FAO 2001. Note: In the case of Bangladesh, the categories are <1 ha, 1-3 ha, and >3 ha.

The participation of smallholders and their integration in the market chain, thus, assumes great

significance in South and Southeast Asia. The growth of large format stores, shrinking of wetmarkets and

traditional retail outlets, and the change in the procurement systems have given rise to the fear that

smallholders will be excluded from the coordinated supply chains. While supermarkets are ideal for

serving the growing urban demand for food, the change it is bringing about in the supply chain can

disadvantage small farmers due to several reasons:

• Smallholders cannot keep up with the stringent food safety and quality control requirements, • Supermarkets prefer large suppliers as they get volume discounts, concessions and promotion

support, • Small farmers have eased out of supplying directly to supermarkets because of long credit terms

usually for 15-30 days. • Farmers are usually too poor and cannot wait that long for payment, • Small-scale producers cannot provide standardized products, • Smallholders cannot afford the entrance and shelf fee requirements of supermarkets; and • Problem of market access due to lack of information and familiarity with the system.

For supermarkets reducing transaction costs, ensuring quality of output and avoiding supply

fluctuations are of utmost significance. Lowering transaction costs requires less levels and frequency of

transaction, thus modern retail chains have started relying on consolidators. This reliance as well as the

practice of passing any possible costs to consolidators makes it more difficult for smallholders to

penetrate the system. Smallholders who have managed to link up with the chain are either individually

equipped or have joined farmer groups or cooperatives.

However, new forms of vertical linkages, especially in Southeast Asia, are allowing smallholders

to participate in the supply chain. The dominance of smallholders in the regions make their inclusion

necessary and vertically coordinated supply chains are incorporating smallholders as well as lowering of

37

transaction costs and market risks for both small farmers and retail chains. Small and medium enterprises

(SMEs) have begun to expand by building production base and contracting farmers as their suppliers and

there are successful cases of producers associations and farmer’s professional associations gaining

bargaining power by acting together. At least three farmers’ groups have begun to collectivize efforts and

sell directly to retailers in Manila, including fast food chains (Digal and Concepcion, 2004). Processing

enterprises and suppliers are building their own production base or providing technical assistance to

contract farmers. Zheijang plums association in China is a farmers’ professional association comprising of

big producers, companies, small farmers and research institutes. It set up product standards for all farmers

and provided information on variety, production, and inputs of members. Technical assistance is provided

by universities, extensions services, and research institutes, who are also members of this association (Bi

et al, 2004).

7 Summary and discussion

7.1 Summary

In this paper, we examine the causes and consequences of the shift toward high-value agricultural

commodities, focusing on the case of eight Asian countries: Bangladesh, India, Pakistan, Indonesia, the

Philippines, Thailand, Vietnam, and China. Of particular interest is the effect of this shift on vertical

coordination within food marketing channels and its implications for small-scale farmers in the region.

Unlike the Green Revolution, which was driven by technology (higher-yielding varieties of grain), the

current transformation appears to be driven by changes in food consumption patterns.

What are the causes of the shift toward high-value agriculture? We identify four key factors.

First, rapid economic growth in many Asia countries has allowed consumers to shift from grains and

other starchy staples to higher-value foods such as fruits, vegetables, eggs, dairy, meat, and fish. Second,

urbanization accentuates the shift toward high-value foods by changing life-styles, increasing exposure to

media, and expanding the availability of high-value foods. Third, the change toward more outward-

looking trade policies have affected production patterns by creating new export opportunities, as well as

influencing food consumption patterns by increasing access to imported foods. Fourth, many countries in

the region have removed restrictions on foreign direct investment, which has affected the food marketing

channels, particularly in food processing and (more recently) the retail food sector.

How have food demand patterns changed? We use household survey data and the FAO food

balance database to demonstrate the shift in food consumption patterns in South Asia, Southeast Asia, and

China. While per capita grain consumption is either falling or growing at less than 1 percent per year in

the eight countries under consideration, per capita consumption of high-value foods is generally growing

38

at 2-10 percent per year, depending on the country and commodity. In addition, consumers are shifting

toward prepared and semi-prepared foods, including food consumed outside the household.

In addition to domestic food demand, export food demand has experienced similar changes.

Exports of high-value agricultural commodities, particularly fishery products and fruits and vegetables are

growing at 3-10 percent per year in most countries, faster than agricultural exports in general. Thus, the

percentage contribution of high-value agricultural commodities to overall agricultural exports is rising,

surpassing 50 percent in five of the eight countries considered.

What is the impact of these shifts on agricultural production patterns? The growth in grain

production is relatively slow (except in Vietnam, which has become a major rice exporter). In contrast,

the annual growth rates of most of the high-value agricultural commodities is 3-8 percent. We estimate

that the vast majority of this increased production is to meet increasing domestic demand rather than

export demand.

What is the impact of all these trends on the food marketing system in Asia? First, income

growth, urbanization, and deregulation of foreign investment have resulted in a dramatic increase in

supermarkets and other modern forms of food retailers. Annual growth rates vary between 10 and 90

percent, while the share of food sold through supermarkets varies from less than 5 percent in Bangladesh

and Vietnam to more than 50 percent in Thailand and the Philippines. Second, foreign direct investment

and growing demand for processed foods has increased the overall size of the food processing sector as

well as the scale of individual processing plants. Processed food exports have grown more quickly than

overall agricultural exports in spite of a range of tariff and non-tariff barriers on them. Third, various

institutional arrangements linking farmers to other stages of the marketing channel have become more

important over time. The emergence of contract farming and other forms of vertical coordination are a

response to 1) the rising share of perishable high-value foods being marketed, 2) the increasing scale of

processors and retailers which implies the need for a more organized procurement system, and 3) the

increasing demand by consumers for food safety and very specific quality attributes which are difficult to

ensure without some form of vertical coordination.

What are the implications of these changes in food marketing for small-scale farmers? High-

value agriculture is growing rapidly and offers small-scale farmers the opportunity to participate in a

relatively profitable activity. In order to participate, however, they need infrastructural development that

connects them to markets and urban centers and they need institutions that link them to high-value food

marketing channels. Contract farming can benefit farmers by providing them with specialized inputs,

technical assistance, credit, and an assured market, thus solving a number of problems small farmers

typically have in producing new high-value commodities. Empirical studies indicate that contract farmers

may enjoy higher profits (though the evidence is mixed), greater production efficiency, and more stable

39

incomes than independent farmers growing the same crops. The benefit to buyers is ensuring a reliable

supply of a product that may not otherwise be available on open markets, particularly perishable products,

specialized crops, or ones that are new to the area. The larger the buyer, the more important it is to

establish procurement systems. For example, large supermarket chains generally establish lists of

preferred suppliers and set private standards for the products they purchase. The decision whether to

source from small farmers or large farmers is based on the nature of the product, the skills and resources

of local farmers, and the land ownership patterns. In many Asian countries, the overwhelming

predominance of smallholders means that supermarkets and other buyers are forced to work with small

farmers.

7.2 Implications for policy

What are the implications for policy of the growth of high-value agriculture and the emergence of

marketing channels with vertical linkages? First, it is critically important that smallholders be able to

participate in these high-value agricultural sectors. High-value agricultural sectors (including livestock,

dairy products, fish, fruits, vegetables, and spices) were the only parts of Asian agriculture that grew

significantly faster than population in the last 25 years. The implication is that that for the broad mass of

population in Asia—which is still predominantly rural and agriculturally-oriented in most countries—to

increase their incomes from agriculture, they must gain greater access to these growing high-value

agricultural markets. The policy requirement is to identify critical areas for trade, marketing, capital

market, and regulatory reforms that can facilitate the integration of small-scale Asian farmers into rapidly

growing and increasingly more global markets.

Second, in order to increase the participation of small farmers in high-value marketing chains, the

first step is to remove policies that unnecessarily impede smallholder participation. Concessionary land

leases, tax exemptions, and infrastructure subsidies that are sometimes offered to large-scale vertically-

integrated plantation-processors need to be examined sceptically. Regulations, credit policy, and tax

policy should be evaluated to ensure there are no unwarranted impediments to small-scale farmers

participating in high-value agricultural sectors.

Third, the development of marketing infrastructure is critical to improve the access of small-scale

farmers to growing high-value agricultural markets. Roads, reliable electrical supply, and

telecommunications reduce the marketing risks of high-value agricultural commodity production. The

government must select infrastructure projects wisely, however. Some public investments, such as cold

storage facilities and food processing plants, do not qualify as public goods and will often be a waste of

public resources.

40

Fourth, agricultural policy and the legal infrastructure should create an enabling environment for

the experimentation with various forms of vertical coordination, including cooperatives, farmer

associations, electronic markets, and contract farming. Often contract farming can be facilitated by some

type of public-private partnership. For example, agricultural officials may work with processors to

develop a plan in which the extension services provide farmer training and mediation services in

preparation for the creation of a new contract farming scheme. Appropriate codes of contract

enforcement suitable for implementation in Asian smallholder situations are necessary, as are the

governance structures that give smallholders recourse in dealing with big industrial players. Informed

policies and a conducive regulatory environment also increase the incentives for agro-processors to use

the produce of small-scale farmers as inputs, and improve their capacity to meet the product attributes

required in a rapidly modernizing agricultural marketplace.

Fifth, greater attention must be given to policies and programs to support the development of

agro-industry and food processing in particular. They have, in cases discussed above, raised the income

of the rural poor through the development of value-added activities, institutions, and agro-food based

rural industrialization. Agro-processing is labour intensive and generates higher value added than

unprocessed agriculture products. Since it is usually located in rural areas, it generally benefits the rural

poor by increasing their incomes. Such increases in income will play a significant role in poverty

reduction, sustainable growth and food security in developing countries. However, agro-industry must

not be supported with policies that artificially lower agricultural prices (such as barriers to raw material

exports), since this would be counterproductive in terms of raising rural incomes. It is also questionable

whether blocking the entrance of international agro-industrial firms (including processors and retailers)

promotes the long-run development of the sector. Certainly, farmers gain from greater competition

among processors and retailers.

Sixth, agricultural policies must adapt to the transformation of agriculture from a sector of

relatively homogeneous commodities to one of highly differentiated and quality-sensitive products.

Governments need to support and develop institutions to address issues of food safety and food quality,

both domestically and internationally. The demand for food, safety, convenience and quality is rapidly

rising within Asian cities, and is mirrored by the rapid development of supermarkets and the food

processing sectors. Grades, standards, and methods of certification will become increasingly important.

On the international level, trade disputes over food safety and quality standards are likely to increase,

particularly if the Doha round of trade negotiations is successful in reducing agricultural tariffs.

Furthermore, disputes will increasingly arise among developing Asian countries, as well as with the

industrialized countries. Governments must develop the expertise to participate in international

negotiations on SPS issues and other types of non-tariff barriers.

41

The “silent revolution” toward high-value agricultural commodities is primarily driven by

changes in the demand patterns of hundreds of millions of Asian consumers whose incomes and access to

information are growing. The latter are not easily changeable by policy decrees. Furthermore, meeting

the challenges will require the mobilization of vast amounts of capital, effort, and entrepreneurship. The

latter will require policy changes. However, failure to meet these needs will waste one of the best

opportunities for driving substantial improvement in rural livelihoods on a wide-scaled basis.

42

References

APO (Asian Productivity Organization). 2001 “Rural-Based Food Processing Industry,” Thailand

Country Paper, Tokyo, Japan.

Athukorala P. and Jayasuriya S. 2003. “Food safety issues, trade and WTO rules: A developing country

perspective” World Economy, Vol 26, No. 9, Blackwell Publishing

Athukorala, P. 1998. Trade Policy Issues in Asian Development, London: Routledge

Bharati. 2004. “Bharti Enterprises & de Rothschilds announce a 50:50 JV for export of fresh fruits and

vegetables”, Press Release

Bi X., Dong X, Huang J., Hu D. and Rozelle S. 2004 “Regoverning Markets: Securing small holder

producer participation in restructured national and agri-food system. China Country Report”,

International Institute for Environment and Development. Accessed from

www.regoverningmarkets.org, February 15, 2005.

Birthal, P., Gulati A. and Joshi P.K. 2005. “Vertical Coordination in High-Value Food Commodities:

Implications for Smallholders” MTID Discussioin Paper No. 85. International Food Policy

Research Institute, Washington, DC.

Bolling, H. 1998. “U.S. foreign direct investment in the global food processed foods industries.”

Appendix 4 in Burfisher, M. and E. Jones. 1998. Regional Trade Agreements and U.S.

Agriculture. Agricultural Economics Report No. 771. U.S. Department of Agriculture.

Washington, D.C.

Camargo Barros, G. S., S. D. Zen, M. R. Piedade Bacchi, S. H. Galvão de Miranda, C. Narrod, and M.

Tiongco. 2003. “Policy, Technical, and Environmental Determinants and Implications of the

Scaling–Up of Swine, Broiler, Layer and Milk Production in Brazil” Annex V, Final Report of

IFPRI–FAO Livestock Industrialization Project: Phase II. International Food Policy Research

Institute, Washington, D.C.

Central Bureau of Statistics, Indonesia. 1980, 1990 and 2002.

Chengappa, P.G., Achoth L., Mukherjee A., Reddy B.M and Ravi P.C. Forthcoming. “Evolution of Food

Retail Chains in India” International Food Policy Research Institute, Washington, DC

Chowdhury S, Gulati A., and Gumbira Sa’id E. 2004. “High-value products, supermarkets and vertical

arrangements in Indonesia.” MTID Discussion Paper. International Food Policy Research

Institute. Washington, D.C.

43

Costales, A.C., C. Delgado, M.A. O. Catelo, M. Tiongco, A. Chatterjee, A. delos Reyes, and C. Narrod.

2003. “Policy, Technical, and Environmental Determinants and Implications of the Scaling–Up of

Broiler and Swine Production in The Philippines” Annex I, Final Report of IFPRI–FAO

Livestock Industrialization Project: Phase II. International Food Policy Research Institute,

Washington, D.C.

Delgado C., Rosegrant M., Steinfeld H., Ehui S. and Courbois C. 1999 “Livestock to 2020: The Next

Food Revolution.” 2020 Vision Discussion Paper. IFPRI-FAO-ILRI. Food, Agriculture, and the

Environment Discussion Paper 28, Washington, DC

Delgado, C., C. Narrod, and M. Tiongco. 2003. Policy, Technical and Environmental Determinants and

Implications of the Scaling-up of Livestock Production in Four Fast-Growing Developing

Countries: A Synthesis. Final Report of IFPRI–FAO Livestock Industrialization Project: Phase II.

International Food Policy Research Institute, Washington, D.C.

Delgado, C., Wada, N., Rosegrant, M., Meijer S., and Ahmed M. 2003. Fish to 2020: Supply and

Demand in Changing Global Markets, International Food Policy Research Institute, Washington,

DC and WorldFish Center, Penang, Malaysia

Digal L and Concepcion S. 2004. “Regoverning Markets: Securing small holder producer participation in

restructured national and agri-food system, The case of The Philippines” International Institute

for Environment and Development. Accessed from www.regoverningmarkets.org, February 15,

2005.

Eaton, C. and A.W. Shepherd. 2001. “Contract Farming: Partnership for Growth.” FAO Agricultural

Services Bulletin No. 145. FAO, Rome, Italy.

Fang X. 2002. “The Development of Fresh Agri-Food Supply Chain”

FAO (Food and Agriculture Organization of the United Nations). 2005. FAO Statistics database.

http://faostat.fao.org/?language=EN Accessed January and February 2005.

FAS USDA (Foreign Agricultural Service, U.S. Department of Agriculture). 1997. “Thailand:

Marketplace and Gateway to Indochina”, Washington, DC

General Statistical Office (GSO). 1998. Major social and economic information obtained from the large-

scale surveys in the period of 1990-1996. Statistical Publishing House. Hanoi, Vietnam.

General Statistical Office (GSO). 2001. Vietnam Living Standards Survey 1997-1998. Statistical

Publishing House. Hanoi, Vietnam.

GOI (Government of India). 2005. Food Processing Policy 2005, Ministry f Food Processing Industries,

Government of India, New Delhi, India

Government of Bangladesh. Agricultural Sample Survey. 1997. Bangladesh

44

Government of Nepal. Agriculture Census, 1991. Nepal

Government of Pakistan. Household Integrated Economic Survey, Pakistan. 1998-99 and 2001-02.

Government of Pakistan. Pakistan Agricultural Census 2000.

Gulati A., Sharma A., Sharma K., Das S. and Chhabra V. 1994. “Export Competitiveness of Selected

Agricultural Commodities” National Council of Applied Economic Research, New Delhi, India

Hu D., Reardon T., Rozelle S., Timmer P. and Wang H. 2004. “The Emergence of Supermarkets with

Chinese Characteristics: Challenges and Opportunities for China’s Agricultural Development,”

Development Policy Review 22 (5).

Huang, J. and C. David. 1993. “Demand for cereal grains in Asia: The effect of urbanization.”

Agricultural Economics Vol. 8. pages 107-124.

ITC case summary by IFC. 2004. “India’s e-choupals: A private sector approach to link farmers to

markets”

Lumpkin, T. 2005. Presentation at the CGIAR Science Council Priority Setting Meeting for High Value

Agriculture, Rome, Italy

Mehta R., R.G. Nambiar, C. Delgado, and S. Subrahmanyam. 2003. “Policy, Technical, and

Environmental Determinants and Implications of the Scaling–Up of Broiler and Egg Production

in India” Annex II, Final Report of IFPRI–FAO Livestock Industrialization Project: Phase II.

International Food Policy Research Institute, Washington, D.C.

Minot N. 1998. “Competitiveness of Food Processing in Viet Nam: A Study of the Rice, Coffee, Seafood

and Fruits and Vegetables Sub-Sectors,” Report prepared for UNIDO. Hanoi.

Minot, N., Epprecht, M., Anh, T, and Trung, L. 2003. Income diversification and poverty in the Northern

Uplands of Vietnam. Report prepared for the Japan Bank for International Cooperation. Hanoi,

Vietnam.

National Bureau of Statistics, China Statistical Yearbook. 1980, 1990 and 2000.

National Sample Survey Organization (NSSO). India. 1988, 1995, 1998 and 2000

Poapongsakorn N., V. NaRanong, C. Delgado, C. Narrod, P. Siriprapanukul, N. Srianant, P. Goolchai, S.

Ruangchan, S. Methrsuraruk, T. Jittreekhun, N. Chalermpao, M. Tiongco, and B. Suwankiri.

2003. “Policy, Technical, and Environmental Determinants and Implications of the Scaling–Up of

Swine, Broiler, Layer and Milk Production in Thailand” Annex IV, Final Report of IFPRI–FAO

Livestock Industrialization Project: Phase II. International Food Policy Research Institute,

Washington, D.C.

Reardon T. and Berdegue J.A. 2002. “The Rapid Rise of Supermarkets in Latin America: Challenges and

Opportunities for Development” Development Policy Review, Vol 20, No 4, Blackwell Publishing

Report of the Household Income and Expenditure Survey. Bangladesh. 2000.

45

SDPI (Sustainable Development Policy Institute). 2004. “Regoverning Markets: Securing small holder

producer participation in restructured national and agri-food systems. Pakistan” Accessed from

www.regoverningmarkets.org, February 15, 2005.

Sharma, V.P., S. Staal, C. Delgado, and R.V. Singh. 2003. “Policy, Technical, and Environmental

Determinants and Implications of the Scaling–Up of Milk Production in India” Annex III,

Research Report of IFPRI–FAO Livestock Industrialization Project: Phase II. International Food

Policy Research Institute, Washington, D.C.

Sivakumar S. 2004. Indian Agribusiness and ITC e-choupal, DuPont NSM

Srinivasan, T.N. 2000. Eight Lectures on India’s Economic Reforms, Delhi and Oxford University Press

(Chapter 2)

Sukhpal S. 2004 “Crisis and Diversification in Punjab Agriculture: Role of state and agribusiness”,

Economic and Political Weekly, December 2004

Supplement to the Report on the 1990 World Census of Agriculture 2001. Food and Agriculture

Organization, Rome, Italy

Tam P. 2004. “Regoverning Markets: Securing small holder producer participation in restructured

national and agri-food systems, The Case of Vietnam.” Accessed from

www.regoverningmarkets.org, February 15, 2005.

TDRI (Thailand Development Research Institute). 1999. Thailand Economic Information Kit, Bangkok

Tiongco, M. and C. Delgado. 2004. “Issues and Pitfalls of Contract Farming for Smallholder Livestock

Production.” For submission as MTID Discussion Paper.

Torero, M. and Gulati A. 2004. “Connecting Small Holders to Markets: Role of Infrastructure and

Institutions” International Food Policy Research Institute, Washington, DC

US Dept of Commerce. 2001. “Prospects for Food Processing Industry, India”, International Market

Insight, US & Foreign Commercial Service and US Department of State, Washington, DC

USDA (United States Department of Agriculture). 2002. “Thailand Retail Food Sector Report” GAIN

Report. Foreign Agricultural Service, U.S. Department of Agriculture. Washington, D.C.

USDA. 2003. “Indonesia Retail Food Sector Report” GAIN Report, Foreign Agricultural Service, U.S.

Department of Agriculture. Washington, D.C.

USDA. 2004. “Bangladesh Retail Food Sector Report” 2004, GAIN Report. Foreign Agricultural Service,

U.S. Department of Agriculture. Washington, D.C.

US ITC (United States International Trade Commission). 2001. “Processed Foods and Beverages: A

Description of Tariff and Non-tariff Barriers for Major Products and Their Impact on Trade”,

Report on Investigation No. 332-421 under section 332 (g) of the Tariff Act of 1930, USITC

Volume 1: Executive Summary and Chapters 1-15, Publication 3455, Washington, DC

46

USTR (United States Trade Representative), 2002. “Foreign Trade Barriers in Thailand” 2002 National

Trade Estimate Report on Foreign Trade Barriers, Office of the Unites States Trade

Representative, Washington, D.C,

Vietnam Living Standards Survey, Vietnam. 1993 and 1998.

Wiboonponse A. and Sriboonchitta S. 2004. “Regoverning Markets: Securing small holder producer

participation in restructured national and agri-food systems, The Case of Thailand” ”,

International Institute for Environment and Development

Witherspoon D. and Reardon T. 2003. “The Rise of Supermarkets in Africa: Implications for Agrifood

Systems and the Rural Poor” Development Policy Review, Vol 21, Issue 3 Blackwell Publishing

World Bank. 2004. World Development Indicators. CD-ROM database. The World Bank. Washington,

D.C.

47