Figure 1.1

EVOLUTION OF AGRI-FOOD TRADE, 1995–2018 (COUNTRIES CLASSIFIED IN GROUPS BY INCOME LEVEL)

Back

The growing trend peaked with the food price crisis in 2007–2008 and was abruptly interrupted by the financial crisis in 2008 and the global recession that followed. Although trade recovered in 2010 and 2011 and commodity prices surged again, the slowdown in the global economy, especially in emerging economies such as the People’s Republic of China, affected both trade and commodity prices significantly.¹ Since 2014, the decline in the value of agri-food trade has been mainly due to falling commodity prices and exchange rate fluctuations,^2,3 with growth rates partly rebounding between 2016 and 2018.

While high-income countries account for most of agri-food trade in value terms, emerging economies and developing countries increasingly participate in global markets (Figure 1.1, Panel A). Since the beginning of the new millennium, upper and lower middle-income countries together have increased their share in global agri-food exports from about 25 percent in 2001 to 36 percent in 2018. During the same period, the share of low-income countries in total agri-food trade remained almost unchanged at around 1.1 percent.

From 2008 onwards, with the slowdown of the global economy, growth of agri-food exports and imports has been sluggish as compared to 1995–2007, especially in high-income countries, whose economies were relatively more affected by the financial crisis (Figure 1.1, Panel B). Low-income countries, many of which export to high-income countries’ markets, were also affected through the slowdown in demand in these markets and declining commodity prices. Exports and imports of upper and lower middle-income countries continued growing rapidly between 2009 and 2011 and have only stalled since then.

Throughout the period 1995–2018, high-income countries as a group showed higher agri-food imports than exports, while the group of upper and lower middle-income countries was in a net exporting position. Imports of the group of low-income countries were slightly higher than their exports between 1995 and 2000, followed by a significant deepening of their net importing position until 2011, which has stabilized since then.

Trade in agricultural commodities and food

The larger part of agri-food trade is made up of trade in processed products from the food sector (Figure 1.2). Between 1995 and 2000, the share of food in total agri-food exports remained stable, exhibiting an increase thereafter, from around 70 percent in 2000 to 76 percent in 2018 (Figure 1.2, Panel A). During the 1995–2018 period, food exports grew faster at an average annual rate of 3.4 percent, while those of agricultural commodities increased at an average annual rate of 1.9 percent.

Figure 1.2

TRADE IN FOOD AND AGRICULTURAL COMMODITIES

Back

Globally, most food is traded by high-income countries, which account for an equal share of food exports and imports. All country income groups import, on average, more food products relative to imports of agricultural commodities (Figure 1.2, Panel B). Upper and lower middle-income countries export more food than they import, pointing out to a well-developed and export-orientated processing industry, on average. Low-income countries’ exports are characterized by a larger share of agricultural commodities, as they specialize in the production of raw materials and their food industry is relatively less developed.

There are pronounced differences in the export orientations of countries. While countries in Europe and Central Asia, and East Asia and the Pacific tend to trade with other countries in the same region, countries in South Asia, Latin America and the Caribbean, sub-Saharan Africa, North America, and the Middle East and North Africa are more globally oriented in their trade (see Box 1.1).

BOX 1.1

REGIONAL AGRI-FOOD TRADE

Countries can be more oriented towards trading within their region or trading globally, and the strength of this orientation can vary by sector and commodity (Figure 1.3).

Figure 1.3

SHARES OF INTRA-REGIONAL AND INTER-REGIONAL TRADE

Back

The majority of agricultural commodities are not traded within the region they are produced but exported to other regions. Approximately 90 percent of exports of agricultural commodities from sub-Saharan Africa and Latin America and the Caribbean are destined for other regions, where they often serve as inputs in the food industry (see Part 2). Only in East Asia and the Pacific and in Europe and Central Asia, most of the agricultural exports remain within the region.

Food is traded more often intra-regionally than agricultural commodities, suggesting that food processing facilities are, in general, located close to the consumers. Only in East Asia and the Pacific, intra-regional food exports are about the same as the intra-regional exports of agricultural commodities (60 percent). In South Asia and in Europe and Central Asia, the shares of intra-regional trade of food (10 percent in South Asia and 75 percent in Europe and Central Asia) are lower than those of agricultural commodities (approximately 15 percent in South Asia and 90 percent in Europe and Central Asia).

The general geographic pattern, however, holds across the two sectors. Some regions invest heavily in intra-regional trade (East Asia and the Pacific and Europe and Central Asia), and others tend to export globally (such as South Asia and Latin America and the Caribbean).

In some regions, there is a much stronger differentiation. Sub-Saharan Africa, for example, exports agricultural commodities to other regions, but exports of food are relatively more pronounced within the region.

In both sectors, food and agriculture, the share of intra-regional exports in total exports increased over time (1995–2018) in four of the seven regions (South Asia, sub-Saharan Africa, North America, and the Middle East and North Africa). This share decreased in Latin America and the Caribbean, East Asia, and Europe and Central Asia.

In Latin America and the Caribbean and in sub-Saharan Africa, the share of intra-regional imports of agricultural commodities is higher than that of intra-regional exports, while the other regions tend to source agricultural commodities more globally compared to the regional distribution of their exports. The same is true for food imports (except for Latin America and the Caribbean and for Europe and Central Asia). In most of the regions, the share of intra-regional imports increased over time.

Reflecting the general slowdown of growth in agri-food trade, growth of both intra-regional and inter-regional trade was much faster in the period 1995–2007 than in 2008–2018.

Significant potential for increasing intra-African trade is likely to come from the full implementation of the African Continental Free Trade Area (AfCFTA), with agri-food trade being projected to increase between 20 and 30 percent in 2040 compared to a baseline without the AfCFTA (see also Box 2.6 on the role of regional trade agreements).^4,5

SOURCES: ECA. 2018; ECA & TradeMark East Africa. 2020.^4,5

Back

Trade by food aggregates

Figure 1.4

CHANGE IN EXPORTS AND IMPORTS BY FOOD AGGREGATE, 1995–2018 (COUNTRIES CLASSIFIED IN GROUPS BY INCOME LEVEL)

Back

Starting from low levels, exports in middle- and low-income countries increased especially in the aggregates of fruit and vegetables (a fourfold increase in lower middle-income countries and a threefold increase in low-income countries); processed food (a threefold increase in lower middle-income countries and a sixfold increase in low-income countries); dairy and eggs (where exports grew around five times across low- and middle-income countries); and fats and oils (approximately a fivefold increase in lower middle-income countries and a threefold increase in low-income countries). Upper middle-income countries significantly increased their exports of dairy and eggs and of grains (by more than five and four times between 1995 and 2018, respectively).

Following Bennett’s law – which proposes that as incomes rise, people eat relatively fewer starchy staple foods and more nutrient-dense meats, oils, sugars, fruit and vegetables⁶ – low- and middle-income countries significantly increased their imports of higher value products such as meat and fish, fruit and vegetables, and processed food (Figure 1.4, Panel B).

Which foods countries trade depends on a multitude of factors, including their comparative advantage in production and consumer preferences. In agriculture, the product-mix is often determined by resource endowments and natural conditions such as climate. Many grains, for example, are mainly produced in temperate zones, while a large variety of fruit and vegetables can be produced in warmer climates. Trade shifts products from surplus to deficit regions, which is reflected in regional trade patterns. Countries which have a comparative advantage in the production of grains also feature relatively higher shares of these products in their exports. Countries where conditions favour the production of fruit and vegetables are characterized by higher shares of these products in their total exports (Figure 1.5). Equivalently, countries which are comparatively less advantaged in the production of cereals or fruits are more dependent on imports of these products (Figure 1.6).

Figure 1.5

SHARE OF EXPORTS OF SELECTED FOOD AGGREGATES IN TOTAL AGRI-FOOD EXPORTS, AVERAGE 2016–2018

Back

Figure 1.6

IMPORT DEPENDENCY FOR SELECTED FOOD AGGREGATES, AVERAGE 2015–2017

Back

Trade patterns emerging along differences in comparative advantage are also reflected at the country level (Figure 1.7). Brazil, for example, an emerging economy (upper middle-income) and major agricultural exporter, almost quadrupled its exports (in real terms) since 1995. Brazil saw a particularly strong increase in its exports of grains, meat and fish, and sugar and cocoa. At the same time, Brazil’s imports remained almost unchanged.

Figure 1.7

AGRICULTURAL EXPORTS AND IMPORTS: BRAZIL, VIET NAM, NEPAL AND UGANDA BY FOOD AGGREGATE

Back

Viet Nam, a lower middle-income country, increased both exports and imports since the beginning of this century. Among the food aggregates that exhibited a major increase in exports are meat and fish, and fruit and vegetables. Imports of grains and fruit and vegetables also increased (Figure 1.7).

Nepal – a landlocked low-income country – is characterized by difficult conditions for agricultural production and low integration in global markets, mainly due to its location in the Himalayas. However, since 1995, Nepal has slowly increased the value of its exports, as well as changed their composition (Figure 1.7). Although fats and oils made up a large part of exports in the late 1990s, improved processing capacity in the new millennium helped to significantly increase exports of processed food and tea and spices. Food imports increased from almost a negligible level in 1995 to more than USD 1 billion in 2018 (measured in 2015 prices), consisting mainly of grains, fruit and vegetables, and processed food.

Uganda, also a landlocked low-income country, shows a different growth path in terms of trade. The country is one of the ten biggest coffee producers globally, and coffee makes up around 35 percent of its total agri-food exports. Between 1995 and 2018, besides an increase in coffee exports, Uganda managed to also significantly increase its exports of grains, sugar and cocoa, and fruit and vegetables. During the same period, Uganda increased its imports of fats and oils, grains, and processed food (Figure 1.7).

International trade gives rise to a globalized economy and, by connecting food demand and supply across the world, allows countries to expand their markets. Together with agro-climatic conditions, how much and what countries trade are shaped by four main drivers which are closely related and, at the same time, also identify economic development. Rising incomes, population growth and demographic changes, technological advances, and policies all drive the growth and composition of international trade.

Population growth and demographic changes together with rising incomes affect overall food demand and dietary patterns, which in turn lead to adaptations in production, markets and trade, facilitated by technology. The globalization process is characterized by increasingly open markets, promoted through reductions in trade policy barriers, but also by technological progress, which results in lower transport costs, improved communication and thus increased commercialization. All of these drivers affect food supply, demand and trade simultaneously and through various channels.

Income growth

In general, trade is affected by income and, at the same time, can be one of the determinants of economic growth as it promotes efficiency gains and technology spillovers. Nevertheless, the relationship between trade and income is controversial. Between 1995 and 2018 – a period characterized by increasingly open markets and more trade – income growth across countries suggests that globalization only partly promoted convergence. Income growth rates in lower and upper middle-income countries were much higher than those of the high-income countries, indicating that during the period 1995–2018 these country groups were catching up with advanced economies. However, income growth has been slow for low-income countries, suggesting a lack of convergence and an expanding income gap.

The 2008 financial crisis also affected income growth. High-income countries, with more leveraged financial systems and credit expansion, were disproportionally affected by the financial crisis and suffered larger downward revisions to their economic activity (Figure 1.8, Panel A).^7,8 Upper middle-income countries also experienced a slowdown in income growth between 2008 and 2018, but at a significantly lesser extent. At the same time, a broader set of developing countries, lower middle- and low-income countries with limited integration in global financial markets, were less affected by the 2008 crisis.⁹ These income trends are also broadly reflected on agri-food trade (see Figure 1.1, Panel A).

Figure 1.8

INCOME DYNAMICS AND GROWTH IN FOOD CONSUMPTION (COUNTRIES CLASSIFIED IN GROUPS BY INCOME LEVEL)

Back

The nutrition transition is also reflected in Bennett’s law; as people become wealthier, they switch from simple starchy plant-dominated diets to more varied foods that include a broader range of fruit and vegetables and animal-sourced protein.⁶

The aggregate data clearly reflect the stages of the nutrition transition (Figure 1.8, Panel B). In low-income countries, increasing incomes per capita are associated with rising caloric consumption per capita. With accelerating income growth, this effect becomes stronger in the lower middle-income countries. In upper middle-income countries, the effect is already slowing down, and in high-income countries, income growth is only weakly associated with growth in caloric consumption.

Dietary changes in line with Bennett’s law can also be observed at the aggregate level. With increasing incomes, the share of cereals in per capita food consumption declined in low- and middle-income countries between 1995 and 2017 (Figure 1.9, upper Panel). High-income countries appear to have completed the nutrition transition with almost no change in cereals consumption.

Figure 1.9

AVERAGE CHANGE IN THE SHARE OF CALORIES AVAILABLE FOR CONSUMPTION PER CAPITA BY MAIN FOOD AGGREGATES, 1995–2017 (PERCENT)

Back

However, the intake of sugars, as a share of the daily diet, increased by more than half in low-income countries, compared to a 5 percent increase in lower middle-income countries. High- and upper middle-income countries exhibited a slight decline in sugar intake. The consumption of fruit and vegetables, meat, and fats and oils increased in all country income groups, especially in upper and lower middle-income countries. Low-income countries saw a strong increase in the consumption of dairy products.

CHILE
A woman shopping at a grocery store.

©iStock.com/Hispanolistic

Similar dietary shifts have been observed in Asia along with rapid economic growth, urbanization and globalization during the period 1961–2011.^12,13 More recently, in sub-Saharan Africa, economic growth has brought about changes in food consumption, away from cereals, roots and tubers and towards fish, meat, eggs, dairy products, fruit and vegetables, along with a general shift to more processed foods.¹⁴

While changing consumption patterns along the nutrition transition are evident at this aggregate level, there is more heterogeneity at the national level where diet evolution also depends on preferences, the distribution of income and the level of development (Figure 1.9, lower Panel). For example, in emerging economies such as Brazil and many developing countries including Nepal and Viet Nam, income growth resulted in significant declines in the share of cereals in per capita food consumption. However, in Uganda, the share of cereals in per capita food consumption increased; there, unlike other countries in the region in which maize dominates diets, staples consist of a variety of foods including cassava, sweet potatoes and matooke. In countries exhibiting accelerated economic growth such as Viet Nam, dietary shifts evolved faster.

Diets changed both at the urban and rural levels.^14,15 Nevertheless, the shift away from cereals to more energy-dense foods has been found to be greater at the urban level, although rural areas appear to be rapidly converging driven by income growth and food system changes.¹⁵

The linkages between average income and consumption may mask important trends in food demand that are related to the distribution of both income and calories between wealthy and poor population groups.

In fact, the emergence of a middle class in many developing countries has been identified as the most significant factor driving not only the demand for food but also its composition, leading to changes in food procurement systems (see Box 1.4 on vertical integration).^14,16,17,18

The rise of an urban middle class in Africa, for example, resulted in an increase in calories consumed overall and a higher demand for processed foods, meat, fruit and vegetables.^14,16 Middle-class consumers are also more likely to shop in supermarkets, or other types of convenience stores, and spend a higher share of their income eating out.^10,16

The dietary shifts spurred by income growth also affect trade. Increasing consumption of meat and fish, fruit and vegetables, and processed food are reflected in growing imports of these products, especially in emerging and developing economies (see Figure 1.4, Panel B).

At the time of writing this report, the outbreak of a novel coronavirus has been affecting global agri-food value chains, incomes and demand for food. The rapid spread of COVID-19 throughout the 2019–2020 winter forced hard choices on global policy-makers. As many countries implemented necessary social distancing practices in response to the pandemic, an unprecedented and multifaceted crisis unfolded.

Many countries faced multiple challenges in public health, the economy and food security, which interacted in complex ways.¹⁹ The threat COVID-19 poses to food security due to income loss is a cause of great concern to the progress made in the reduction of the prevalence of undernourishment over recent decades. Box 1.2 discusses the immediate effects of the pandemic on global trade, value chains and food security.

Population growth and demographic changes

The interaction between population growth and demographic changes impacts food demand, trade and markets in important ways. While population growth drives the demand for and trade of food in terms of volume, demographic changes affect its composition.

Population growth is associated with increasing trade across countries. If the pace of population growth differs between regions, trade will likely move food from regions where population growth is slower to regions where it is relatively faster. For example, fast population growth in countries characterized by low agricultural productivity per capita, which may also be negatively affected by climate change, will lead to increased imports. Long-term population trends exhibit strong growth in Asia; while that growth has started slowing down, the population is projected to peak at 5.3 billion people around 2050 (Figure 1.10, Panel A). The population in Africa is projected to continue growing strongly up to 2.5 billion people in 2050, posing significant challenges to agriculture. Populations in Latin America and the Caribbean, North America and Oceania are projected to grow slowly, while those in Europe are likely to contract by 2050.

Figure 1.10

POPULATION GROWTH AND DEMOGRAPHIC CHANGES

Back

Urbanization is associated with considerable changes in lifestyle and is a key driver of changes in consumption patterns and the transformation of food systems. As societies become urbanized and consumers live farther away from where primary agricultural production takes place, the demand for food that can be easily stored and transported strengthens, which in turn gives rise to increased food processing.²⁹ Urban consumers also tend to have relatively higher incomes which strengthens demand for a wider variety of foods. Their lifestyles allow less time for food preparation, which results in higher consumption of processed foods and more frequent meals away from home.³⁰

Urbanization is also linked with better modes of transportation and car ownership, access to refrigeration, and exposure to advertising.¹⁷ These promote access to new and evolving retail channels for food and strengthen the demand for higher-value products, including fruit and vegetables and processed foods. Car ownership in urban Zambia, for example, significantly increases retail purchases in supermarkets, which tend to stock and sell relatively more processed foods.¹⁶

While North America, Latin America and the Caribbean, and Europe and Central Asia already have highly urbanized populations (Figure 1.10, Panel B), urbanization rates between 1995 and 2018 grew fastest in East Asia and the Pacific. In the relatively less urbanized regions of sub-Saharan Africa and South Asia, the share of urban population also increased, but at a relatively lower pace. Urbanization is occurring at a more rapid pace in the developing world than it did, for example, in the United States of America and Europe. It took nine decades for the share of the urban population in the total population to increase from 40 to 75 percent in the United States of America; however, this threshold was surpassed in less than three decades in Brazil and the Republic of Korea.¹⁷

Technological progress, trade costs and trade policies

Technological progress has led to improvements in infrastructure and logistics and thus lowered transportation costs. It has also contributed to declining communication costs, which also influence trade and promote the global integration of value chains. By helping reduce inefficiencies in value chains, technological progress may also contribute to more sustainable food system outcomes.^31,32

Figure 1.11

TRADE AND COMMUNICATION COSTS

Back

The impact of trade costs on agricultural trade can be significant. A study analysing the effect of overall trade costs – including costs related to tariff and non-tariff policy barriers, freight, information, currency and legal and regulatory procedures – on agricultural trade, found that a 1 percent reduction in aggregate trade costs could increase global trade volumes by 2–2.5 percent.³⁴

At the same time, technological progress has revolutionized communication by reducing its costs and facilitating trade (see also Part 4 for an in-depth discussion of digital technology impacts on markets). For example, analysis of the impact of communication costs on bilateral trade suggests that halving the importer’s calling price leads to a 42.5 percent increase in aggregate bilateral trade.³¹ Such impacts were shown to be one-third larger on trade in differentiated products – which requires better information and greater coordination between traders – than on trade in homogeneous products.

Indeed, digital technology improvements and the associated decline in communication costs are seen as a major driver of global value chains, making coordination across different production stages in different geographic locations possible.³⁵

The internet is also having a significant effect on trade, allowing firms to communicate and market their products across borders with lower costs. Since the 1990s, the level of internet adoption has increased dramatically – today, it is estimated that around 54 percent of the global population have access to the internet (see Part 4). Higher internet adoption rates are also shown to have positive effects on trade – on average, a 10 percent increase in the exporter’s internet adoption rate can bring about nearly a 2 percent increase in bilateral exports.³⁶

There are marked differences in this effect depending on whether the internet is better adopted by the exporter or the importer. Nevertheless, high adoption rates by both trading partners can result in significant increases in the volume of trade but also in the number of products traded, as better communication can improve matching.

While costs for fixed broadband internet access have already been low in developed regions, they significantly declined between 2008 and 2017 in developing countries, contributing towards closing the digital gap across the world (Figure 1.11, Panel B). Nevertheless, although access to the internet is important for international trade, quality in terms of bandwidth and better speed is crucial. A study on the relative effects of internet subscriptions (reflecting the size of internet adoption) and bandwidth (reflecting its quality) suggests that a 1 percent increase in average data speed per subscription results in 0.5 percent increase in bilateral trade, while a similar increase in subscription rates brings about an increase of 0.3 percent.³⁸ Such a differentiated effect underlines the need to focus on improving the quality of digital infrastructure in the developing world, where bandwidth speed can be diverse across countries and subscriptions.

Declining trade costs also result from trade policies. A wave of opening to trade since the General Agreement on Tariffs and Trade (GATT) and the establishment of the WTO in 1995, as well as the proliferation of regional trade agreements, has reduced tariffs and trade-distorting domestic support, and improved the mutual recognition of non-tariff measures (NTMs).

Figure 1.12

AGRICULTURAL APPLIED TARIFF RATES, 1995–2018 (COUNTRIES CLASSIFIED IN GROUPS BY INCOME LEVEL)

Back

Tariffs can vary significantly across foods and agricultural commodities (Figure 1.12, Panel B). In high-income countries, average applied tariffs are relatively low on coffee and tea, fats and oils, and fruit and vegetables. However, on average, high-income countries impose considerably higher tariffs on imports of grains, dairy products and eggs. On average, low- and middle-income countries have much higher tariffs. They impose the highest tariffs on imports of processed food, followed by sugar and cocoa, and dairy and eggs. Tariffs in low- and middle-income countries are also relatively high on imports of fruit and vegetables, and meat and fish. For these countries, the lowest level of tariffs is imposed on grains.

Although the impact of digital technology on trade has led many observers to suggest that, in today’s environment, trade policies are relatively unimportant, recent analysis suggests that tariffs do matter, especially in the context of global value chains.³⁹ Although fragmented and vertically coordinated production across different countries is often seen as a result of technological progress, tariff reductions have had a strong impact on the emergence of global value chains by significantly reducing the trade costs of products that cross borders multiple times during the production process.⁴⁰

While tariff reductions have played a significant role in decreasing trade costs and stimulating agri-food trade, trade is also regulated by a myriad of NTMs. NTMs in agriculture include technical barriers to trade that reflect technical regulations and standards, and sanitary and phytosanitary (SPS) measures that ensure food safety.^41,42

In fact, NTMs are much more important in agriculture than in most other sectors, and their effects on trade can be much stronger than those of tariffs.⁴³ SPS measures tend to be more stringent in high-income than in middle- and low-income countries.⁴⁴ However, the effects of NTMs on trade can be mixed; food standards can be trade-enhancing, as well as trade-impeding, depending on the measures, products and countries involved.^44,45,46

The growth in high-value exports, such as fruit and vegetables, from developing countries has been accompanied by increasing attention to food safety standards – typically SPS measures – in the markets of developed economies.^47,48 While many food safety standards were initially imposed to meet the requirements in lucrative import markets, consumer awareness for food safety has also gained momentum in developing countries.^48,49 Food safety was, for example, identified as the most important sustainability attribute for rice consumers in Nigeria⁵⁰ and has become a societal issue that has received considerable attention in Viet Nam.⁵¹

Food standards can be public or private. Governments impose, for example, maximum residue limits (MRLs) on pesticides to denote the highest level of a pesticide residue legally tolerated in food. To minimize barriers to trade that might arise from divergent national regulations, global standard-setting bodies such as the Codex Alimentarius Joint FAO/WHO (World Health Organization) International Food Standards Programme aim at harmonizing standards at international level (see Box 1.3).

While many measures are enforced through public standards, increasingly globalized value chains have also led to a proliferation of private standards. These relate to product attributes such as quality grading, residue levels, traceability and branding, as well as process attributes such as organic production and animal welfare.⁴²

Private standards often complement public regulation, for example, by referring to sustainability attributes such as environmental protection or ethical sourcing. Moreover, private standards may also fill the gap created by missing public regulation or enforce more stringent requirements than foreseen in national regulations. This is often the case for food safety and food quality standards, in particular when large retailers require a certain, constantly reliable quality of a product. In these cases, private standards could become a barrier to participation in global value chains for farmers and processors that cannot easily comply with them.^42,52 Standards and sustainability certification schemes are also discussed in the contexts of global value chains in Part 2, smallholder farmer integration in Part 3, and digital technology applications on traceability in Part 4.

In order to further reduce trade costs by simplifying and harmonizing customs procedures and export and import processes, the WTO Trade Facilitation Agreement (TFA) entered into force in 2017 (see Box 2.6 in Part 2).

The same trends that cause shifts in trade patterns and dietary habits lead to profound changes also in food markets and value chains. Urbanization, in particular, furthers dietary changes induced by income growth and spurs transformations in food value chains and the retail sector.

As people move to cities and consumers live farther away from where food is produced, food retail becomes more important. Urban, and increasingly also rural, food retail has evolved since the beginning of the twentieth century.⁵⁴ Traditionally, retail consisted of stalls in wet markets (traditional and roadside markets) and small stand-alone shops, such as grocery stores and local kiosks.^16,54 Supermarkets started to emerge in the 1920s–1940s in the United States of America and Western Europe and in the 1980s–1990s in many developing countries.

Supermarkets initially only offered dry goods. Once procurement and storage improved, they also penetrated perishable food markets. Due to their ability to offer a great diversity of products through economies of scope, supermarkets have nearly fully captured food retail in developed countries and have acquired a rapidly growing share in developing countries.⁵⁴

By 2018, sales of leading supermarket chains increased between twofold and sixfold in countries in Asia and in Latin America and the Caribbean, regions where supermarket sales were already relatively high in 2002. More than tenfold increases were reported for countries in which supermarkets started to appear only around the beginning of the twenty-first century.⁵⁵

A survey of 475 urban households in Lusaka, the capital of Zambia, showed that consumers used both traditional and different modern retailers. About 73 percent of households frequented supermarkets, with the use of these modern retailers increasing considerably from the lowest to the highest income population groups. Traditional wet markets were also frequented by 73 percent of the households with almost no difference between income population groups. The use of grocery stores and roadside markets, however, decreased with rising household income. While modern retailers were usually frequented once a week to make larger purchases, traditional retailers such as wet markets, but also smaller grocery stores, roadside markets and local kiosks, were attended several times during the week to buy additional foods. In this survey, on average, about 42 percent of household food expenditure was made for purchases from modern retailers.¹⁶

While urbanization is the main driver, changes in the food retail sector are shaped by many factors. In Ghana, an inventory of supermarkets and processed products in eight major urban centres showed only modest supermarket growth, despite rapid urbanization and increasing household incomes.⁵⁶ The share of supermarkets in a cross-section of 42 countries at all stages of development was found to also increase with income, openness to inward foreign direct investment and female labour force participation.⁵⁷

Foreign direct investment (FDI) into retail, food processing, restaurants and fast food chains has risen rapidly since the 1980s, originating mainly from transnational food companies targeting markets in low- and middle-income countries. In fact, FDI appears to have proved more effective than trade in generating sales of processed foods in these countries.⁵⁸

Motivated by rapidly expanding populations and less developed retail markets, investments of European grocery retailers in East Asia, for example, peaked in the late 1990s. However, this intense initial phase of investment in the region was followed by a phase of divestment. As a combination of increased regulation in these new markets, growing indigenous competition and firm-level reassessments of global activities, most of these retailers have now divested from individual markets or even exited the region. Most of the exits involved the acquisition of the operation by a local or regional operator, while others were transferred between developed country retailers.⁵⁹

At the beginning of the new millennium, e-commerce started to emerge, adding to the transformation of the food retail sector.¹⁰ E-commerce giants, such as Amazon and Alibaba, combined and extended the advantages of economies of scale and scope that previously supermarkets had had over traditional retail outlets. However, unlike supermarkets at that time, e-commerce businesses further reduce transaction costs of consumers by allowing ordering online and delivering the product to their home.

E-commerce’s major drawback is that consumers have no direct observation of the food products. Recently, supermarket chains started adding e-commerce facilities and home-delivery, capitalizing on the fact that consumers are familiar with their products based on previous visits. At the same time, e-commerce firms moved to strategic alliances to include supermarkets in their platforms or add physical outlets to their portfolio, such as Amazon’s acquisition of Whole Foods and Alibaba’s of part of Auchan and RT Mart chains in the People’s Republic of China.⁵⁴ Especially in Asian countries such as the People’s Republic of China, Japan and the Republic of Korea, a phenomenal rise of e-commerce comprising grocery and meal delivery is observed.^60,61

However, food e-commerce, while growing rapidly, remains small around the globe with the United States of America and the People’s Republic of China experiencing the largest growth.^62,63,64 Although it is difficult to assess e-commerce penetration in food markets due to the lack of comprehensive data, available assessments indicate that the share of e-commerce in food and beverages sales is less than 1 percent of total food expenditures in Asia, Europe and North America. This contrasts with the rate of penetration of e-commerce for other goods which averages 80 percent in the United States of America and almost 60 percent in the People’s Republic of China.⁶³ The relative bulk of food items, their comparably low price per unit and the logistical challenges of the cold chain have so far prevented the emergence of food as a major category in online retailing.^10,62 These factors are expected to rein in further growth in the market share for food e-commerce, with supermarkets (and supermarket-like outlets) continuing to dominate and, particularly in Africa, expanding.

The transformation of food retail was paralleled by changes in the food service sector, such as shifts from small-scale independent restaurant outlets to fast-food restaurant and café chains. As with supermarkets, the transformation in food services was much faster in low- and middle-income countries than in the pioneering high-income countries. While product and process innovations were initially developed in high-income countries, they later diffused easily as multinational firms undertook FDI in search of profitable new markets. Local food services chains emerged and proliferated to serve lower income consumers and emerging middle-class markets. In the United States of America, the share of calories from food purchased to be consumed away from home was 17 percent in 1977 and 34 percent in 2011.^65,66 Gross sales of leading multinational food services companies in Asia tripled from 2008 to 2018.⁵⁵

Economic growth, urbanization, technological progress and globalization shape dietary changes and affect agricultural production. Increasingly affluent consumers and growing demand for processed and higher quality food spur changes in retail and distribution sectors, as well as in food industries. These trends give rise to demand for more standardized, higher quality and larger amounts of agricultural production from farmers.

Along the development path, these transformations generally evolve over three stages and are driven by private firms seeking profits through innovations, based on new technologies, new business practices and new products.^55,67,68

At the initial stage of the transformation, traditional value chains are short, with farmers often selling their produce directly to end-consumers or to small-scale traders and processors. Very little of the value added comes from off-farm activities such as processing or distribution. Markets are typically characterized by spot transactions without contracts and formal standards.⁶⁸

With increasing urbanization, people move away from rural areas and primary agricultural production and with higher incomes demand more processed and higher quality food. In this transitional phase, many micro, small and medium-sized enterprises evolve in food retail, distribution and processing. In response to rising consumer awareness, public and private quality and safety standards emerge. Spot markets still dominate, but vertical integration and coordination through contracting begins to evolve (see Box 1.4).^55,68

BOX 1.4

Vertical integration and coordination in value chains

A typical modern food value chain includes several stages. First, input suppliers provide seed, fertilizers and other inputs to farmers who produce agricultural commodities that are either sold to wholesalers or become inputs to processors. From the processors, food reaches the consumer through several stages, including distributors, wholesalers and retailers (see Figure 1.13).

Figure 1.13

STYLIZED FOOD VALUE CHAIN

Back

The multiple stages of a value chain to produce final food products may be controlled by one, a few or many firms or individuals.⁵⁵ Modern value chains are typically characterized by coordination between farmers and processors or traders, and between processors and retailers.⁶⁷ Firms can vertically integrate or coordinate through a variety of arrangements. These arrangements can be informal or contractual and involve intensive vertical coordination that can extend to common ownership. Value chain coordination can be initiated by downstream buyers, such as supermarkets and food processors, while others are initiated by upstream suppliers including farmers or farmer cooperatives. The arrangements can involve two collaborating parties at successive stages in the value chain or include more complex structures linking multiple stages in the chain based on multi-stakeholder agreements and partnerships.⁵⁵

Procurement by the retail and processing sectors is prone to concentration and integration processes leading to fewer and larger firms along the value chain. These are seen as more efficient than smaller firms as they can leverage economies of scale and scope. Large-scale firms in different segments of the value chain may also facilitate each other’s growth and evolve together. Supermarket chains, for example, tend to source from large distributors and processors to reduce transaction costs and ensure the compliance with private standards. When supermarket chains enter new countries, large logistic and wholesale multinationals as well as processors often follow.⁶⁸ At the same time, market concentration in food value chains raises concerns related to the emergence of market power (see Part 2 for a discussion on competition issues).

Supermarket chains in developing regions have been shifting away from sourcing products from traditional wholesalers and wholesale markets and rely, as much as possible, on specialized, dedicated wholesalers who assemble, grade and sort foods in compliance with the supermarket chain’s standards. Current supermarket procurement systems in developing countries are often based on three pillars: (1) specialized procurement agents such as “specialized/dedicated wholesalers” and independent distribution agents; (2) centralized procurement through distribution centres owned by the supermarket chains; and (3) assured and consistent supply through “preferred suppliers”, which can be farmers, farmer cooperatives or processors directly without other intermediaries (see Part 3 for arrangements that integrate farmers into value chains).^68,69,70

The exact procurement system can vary from country to country. In some countries, wholesalers have also vertically integrated into retail and compete with supermarkets.⁶⁹ In Botswana, for example, the retail sector consists of retailers owned by wholesale groups; main supermarket chains which own distribution centres; and independent retailers. The vertically integrated wholesalers source their products from suppliers directly and/or from independent distribution agents. Supermarket chains source from their own distribution centres, from wholesalers, from independent distribution agents and/or from suppliers directly. In Zambia, however, independent retailers source from traders and wholesalers, whereas supermarket chains primarily source from suppliers directly.⁶⁹

SOURCES: Adapted from Barrett et al. 2019; McCullough et al. 2008; Reardon et al. 2019; das Nair. 2018; Reardon et al. 2008.^{55,67,68,69,70}

Back

As value chains and market volumes expand, economies of scale and specialization in food retail and processing set in. Large retailers, such as supermarkets, rise and value chains become increasingly vertically integrated and coordinated, marking the change to modern value chains (see also Box 2.2 in Part 2). Consumers and the food industry increasingly demand quality and safety standards.^68,71

While the transformation from traditional to modern value chains was initiated with the Industrial Revolution, and took almost a century in North America and Western Europe, in many developing regions it set in later and has been much faster.¹⁷ This transformation process started in the 1980s in parts of East Asia (excluding the People’s Republic of China) and larger countries in South America (such as Brazil); it continued in the 1990s in Central America and parts of South America (for example in Chile, Colombia and Mexico), parts of Southeast Asia and South Africa. In the 2000s, emerging economies in Asia (such as the People’s Republic of China, India and Viet Nam) and other South American countries (including Peru and the Plurinational State of Bolivia) followed. The process also started in Southern Africa (Zambia), East Africa (Kenya) and West Africa (Ghana, Nigeria and Senegal) in the 2000s.⁶⁸

The pace of transformation differs by commodity, with grains value chains often transforming first, followed by animal products, and fresh fruit and vegetables, often leading to the co-existence of traditional, transitional and modern value chains in many developing countries.^17,68

Figure 1.14

SHARE OF VALUE ADDED OF AGRICULTURE AND OF FOOD IN TOTAL AGRI-FOOD VALUE ADDED BY INCOME, 2017

Back

The largest welfare impacts from trade and the transformation of agricultural and food markets are likely to accrue to food consumers. Productivity increases, in conjunction with more trade and competition, bring about increases in the availability of safe and nutritious food and drive its price down, resulting in improvements in access to food. For many people, this process results in improved food security and better diets, since it increases access to foods rich in micronutrients such as fruits, vegetables and animal-sourced foods.

At the same time, globalization, the rise of urban lifestyles, and the associated transformations in food production and food value chains are seen by some analysts as contributing factors in the shift towards less healthy diets, and in the increasing prevalence of overnutrition and obesity in many parts of the world (see Box 1.5).^11,15,16,72 In many low- and middle-income countries, overnutrition and obesity coexist with undernourishment and micronutrient deficiencies, denoting a “triple burden” of malnutrition.

Part 2 further explores the economic and health impacts of global value chains on consumers, as well as linkages with inequality and environmental impacts. The integration of smallholder farmers into modern markets and their inclusion in modern value chains is further discussed in Part 3.

Global value chains in agri-food markets

Since 1995, international trade in agricultural commodities and food more than doubled in real terms (see Figure 1.1 in Part 1). However, focusing on the value of trade measured through gross exports alone could mask important developments in global markets.

Over time, firms have increasingly used international trade to leverage specialization and comparative advantage by “unbundling” the production process into stages and identifying the least-cost location for each one. This resulted in production processes that run across borders, giving rise to global value chains (GVCs) – production chains that encompass at least three countries. GVCs have been a typical feature in manufacturing and services. Indeed, about half of global goods and services trade now takes place through GVCs.¹

GVCs are present in food and agriculture. This report estimates that about one-third of agri-food exports are traded within GVCs. Seeds and fertilizers, primary agricultural commodities (such as grains), processed and intermediate products (such as soybean oil or milk powder), but also services and industrial inputs are exchanged between different stages of production that run across multiple countries.

Although relatively new as a topic of analysis, GVCs are based on the fundamental concepts of comparative advantage and specialization in production that have their origins in the classical economic theory of the eighteenth and nineteenth centuries.^2,3 Viewing international trade from a GVC perspective helps to understand how trade contributes to value added that is generated in a country. GVC analysis allows to decompose the value of gross exports into the value that has been imported – and then used in production for export purposes – and the value that is added or generated domestically (see Box 2.1 for further explanation of the key terminology and Box 2.2 for an example).

BOX 2.2

Global value chain in action: orange juice – from the tree to the bottle

Orange-based beverages are among the most popular around the world. Of all the oranges produced worldwide, 20 percent of the total is sold as a whole fruit, while the remainder is used for making extracts and juice. The leading orange producers in the world are Brazil (accounting for about 30 percent of the world’s production) and the United States of America (accounting for about 10 percent of the world’s production). Over 90 percent of the oranges produced in the United States of America go to juice making.

The illustration below displays that the companies involved in producing orange-based beverages compete with each other and are also complementary. Companies in Brazil focus on processing by crushing domestically produced oranges and exporting the juice extract for further processing and distribution. Companies in the United States of America import the juice extract from Brazil and process it together with the US-produced orange juice extract to produce orange juice-based soft drinks. These are partly consumed domestically and partly exported to other countries, for example the People’s Republic of China.

In the GVC of these orange-based drinks, the value of exports to the People’s Republic of China is made up of value added from Brazil and the United States of America. For Brazil, the juice extract exports reflect domestic value added. For the United States of America, which imports the juice extract as an input, it reflects foreign value added. At the same time, the processing industry in the United States of America adds value to this input by further processing it – this reflects domestic value added for the United States of America. In this value chain, the forward linkages of GVC participation of the agriculture sector in Brazil are represented in the example by the exports of the processing industry in the United States of America to Peoples’ Republic of China. The United States of America has backward linkages (imports of the juice extract from Brazil) and forward linkages (exports of orange drinks to the People’s Republic of China). The total GVC participation level of the orange-based beverages sector in the United States of America is the sum of the foreign value added from Brazil (backward linkages), plus the value added generated in the United States of America that flows through to the People’s Republic of China (forward linkages).

SOURCE: Azevedo & Chaddad. 2006.¹³

Back

This emergence of GVCs is driven by declining transport costs and lower trade barriers, such as import tariffs, both of which gave rise to globalization. These trends have made the slicing and spreading of production processes across countries even more attractive. Technological advances and the rise of information and communications technology (ICT) have made coordination across countries cheaper, further promoting GVCs (see also the discussion on trade and communication costs in Part 1).

International trade can improve resource allocation and contribute to economic efficiency, boosting income growth and productivity in trading partners.^4,5,6 Furthermore, an emerging literature suggests that GVC-related trade has a stronger positive impact on productivity and income per capita compared to bilateral non-GVC trade.⁷ Participation in GVCs may enable greater competitiveness, better inclusion in trade and investment flows, and improved access to technology and knowledge, all of which help to upgrade towards higher value-added activities.

Export-oriented agriculture, boosted by GVCs, can provide on-farm and off-farm job opportunities. Scaling up agricultural production increases jobs within the sector. However, the enhanced production also implies increased demand for inputs, which can have a knock-on effect on employment in, for example, the seed and fertilizer industries, as well as in the relatively labour-intensive transport and commercial services sectors.

Developing countries, including sub-Saharan African countries, are also actively participating in agri-food GVCs.¹¹ GVCs can allow farmers and firms in developing countries to participate in and benefit from export activities, as it is easier to penetrate the global market when production slices are thinner and more specific. But the benefits of GVC participation are not automatic, and there is a large degree of heterogeneity. For example, although trade is expected to boost economic growth, many developing countries experienced an increase in inequality as they became more exposed to open markets, often due to the lack of complementary policies and investments and transferrable skills in sectors that were impacted the most by trade reforms.¹²

Recent trends in agri-food GVCs

Figure 2.1

GROSS EXPORTS AT GLOBAL LEVEL AND GVC PARTICIPATION, 1995–2015

Back

Agricultural commodities are a basic input in food and beverages but also in other sectors, and thus, GVC participation in agriculture is mostly through forward linkages (Figure 2.1, Panel A). On average, a significant share of agricultural production is linked to GVCs through exports, leading to high forward linkages (22 percent of the value of gross exports). The backward linkages of agriculture reflect imports of inputs, such as seeds and fertilizers, as well as significant use of services in the production process (such as quality controls, logistics, storage and financial services). At the global level, as exports cross borders, these backward linkages give rise to double-counting of value added and make up a relatively small part – about 12 percent – of the total value of gross exports (at the country level, backward linkages reflect foreign value added; see Boxes 2.3 and 2.4). The larger part of the value of agriculture’s exports, about 88 percent, reflects domestic value added, that is value generated by land and labour, production factors that are not traded internationally. A proportion of this domestic value added can become part of GVCs downstream through forward linkages.

The food and beverages sector (which includes all processed products) is more in the middle or at the end of the value chain. Globally, its GVC participation rate is comparable to that of agriculture (on average, 33 percent; see Figure 2.1, Panel B). However, food and beverages entail a larger share of backward linkages in production compared to agriculture (about 22 percent) and relatively fewer forward linkages (11 percent). This is because the sector uses domestic and imported agricultural commodities but also inputs from other sectors on a large scale. When these are imported, it leads to a significant level of foreign value added embedded in exports. Therefore, at the global level, a significant part of gross exports are backward-linked and thus double-counted. Part of the forward linkages of the food and beverages sector concerns exports of lightly processed products, such as orange juice extract, that can be used by the food industry of another country and processed further before they are re-exported (see Box 2.2 for an example).

Global exports of food and beverages are approximately twice as large as those of agricultural commodities, and, in absolute terms, the rapid increase in their value between 2002 and 2008 is remarkable (see also the discussion on the evolution of trade in Part 1). The increasing share of backward-linked GVC exports also shows that the trend in total gross exports is not all newly created value added.

The financial crisis and the slowdown in economic activity affected all trade. However, the decline in trade may be due in part to a structural change in the trade-GDP relationship. It may have resulted from a slowing pace of international vertical coordination due to the economic slowdown that is evident in the evolution of GVCs.¹⁸

Figure 2.2

GVC PARTICIPATION RATES IN AGRICULTURE IN 2015

Back

In general, low-income countries tend to have low backward linkages, as they primarily specialize in producing and exporting agricultural commodities. Their forward linkages vary substantially, depending on a range of factors, including geography. Nepal, for example, has relatively low backward and forward linkages, as it mostly trades with India rather than with the global market (Figure 2.3). Middle-income countries can exhibit a range of GVC participation patterns.

Figure 2.3

FORWARD AND BACKWARD GVC LINKAGES IN 2015 (COUNTRIES CLASSIFIED IN GROUPS BY INCOME LEVEL)

Back

Ghana, a lower middle-income country, enters GVCs with substantially higher forward GVC linkages (Figure 2.3 and Box 2.3). In contrast, in Viet Nam, another lower middle-income country, GVC participation is quite extensive, mainly through backward linkages (Figure 2.3 and Box 2.4).

BOX 2.3

EXAMPLE OF A COUNTRY WITH UNEVEN GVC LINKAGES: GHANA

Ghana’s GVC participation differs between agriculture and food and beverages. Agriculture is characterized by rapidly expanding export volumes and a high GVC participation, but the food and beverages industry is less developed (Figure 2.4, Panels A and B).

Figure 2.4

GROSS EXPORTS AND GVC PARTICIPATION IN GHANA

Back

As Ghana exports mostly unprocessed cocoa, it has very high forward linkages from its agricultural sector to the rest of the world (Figure 2.4, Panel C); similarly, Ghana’s food exports are largely lightly processed cocoa products and hence have low backward linkages to other economies. Together this leads to Ghana’s status as a large exporter with strong GVC linkages in agriculture but with relatively weaker GVC linkages in food and beverages.

SOURCES: FAO analysis by Dellink et al. 2020; AfDB, OECD & UNDP. 2014.^16,20

Back

BOX 2.4

Example of a country with strong GVC linkages: Viet Nam

GVC participation in Viet Nam is quite extensive and exemplifies its international orientation, particularly of the food and beverage industry (Figure 2.5).

Figure 2.5

GROSS EXPORTS AND GVC PARTICIPATION IN VIET NAM

Back

Trade liberalization and international economic integration have brought about significant contributions to export expansion, economic growth, employment creation and welfare improvement in Viet Nam, especially after 2000. Viet Nam benefited throughout the period from several bilateral trade agreements, its membership in the Association of Southeast Asian Nations (ASEAN) and the Free Trade Agreements that this regional trade bloc has signed. The country experienced persistently increasing participation rates in both sectors in the 2000s after the Asian crisis (Figure 2.5, Panels A and B). These positive impacts are, in large part, brought about by greater capital inflows. The extensive backward linkages in the food and beverages industry suggest that Viet Nam has specialized in processing imported basic inputs from its regional neighbours (Figure 2.4, Panel C).

SOURCES: FAO analysis by Dellink et al. 2020; Auffret. 2003; EU Commission. 2018.^16,21,22

Back

In Brazil, an upper middle-income country, GVC participation rates remain below the global average for both agriculture and food and beverages. Its forward linkages are significantly lower than those of Ghana, as most trade is bilateral – for example, with the United States of America due to trade agreements – and not through GVCs (Figure 2.3).

Some high-income countries – mainly in Europe – enter agricultural GVCs combining high backward with high forward linkages. Germany, a high-income country, is an example of an economy with high export intensities and significant GVC participation. Other high-income countries tend to have more backward linkages but relatively fewer forward ones (Figure 2.3).

GVC participation and value added across economic sectors

The relationship between international trade and economic growth is complex. Nonetheless, there is widespread empirical evidence that, in the long term, trade promotes growth and development. In the short term, every country has a relative comparative advantage in some goods and services, and all countries could potentially gain when engaging in trade. In the long term, these efficiency gains together with technological spillovers and the transmission of knowledge, which are stimulated by trade, can generate dynamic benefits in terms of higher productivity and innovation, leading to economic growth. The relationship between trade and economic growth runs both ways, as at the same time, economic growth, by strengthening demand, also boosts international trade.

Figure 2.6

RELATIONSHIP BETWEEN GROWTH IN VALUE ADDED AND GROWTH IN GVC PARTICIPATION BETWEEN 1995 AND 2015 (COUNTRIES CLASSIFIED IN GROUPS BY INCOME LEVEL)

Back

Nevertheless, several empirical studies, which are based on aggregated data from all economic sectors, have found significant causal effects of GVC participation on value added for middle- and high-income countries and negligible impacts for low-income countries. More specifically, the analysis suggests that increased backward linkages (through increased imports of foreign value added) have not led to economic growth in some low-income developing countries; these countries are characterized by low skill-sets and thus low capacity to learn and absorb knowledge to apply technological advances that otherwise could be diffused and promote growth.²³ This relationship between GVC participation and growth depends on capabilities to both adapt to production processes and innovate. For example, labour force education and skills, regulations that promote businesses, and investments in research and development all reflect a country’s ability to enter GVCs effectively.

Most studies that analyse the impact of GVCs participation on economic growth consider the economy as a whole. Indeed, GVCs link economic activities over different sectors and across countries. Within a country, a significant share of the growth in value added in agriculture comes from its linkages to other economic sectors. Increased GVC exports by the food and beverages sector, and other sectors that use inputs from agriculture, can further promote agriculture’s participation in global trade and create value added. GVCs also link economic sectors across countries. Globally, in food and beverages GVC exports, agriculture makes up 20 percent of the foreign value added.

In this way, GVCs, through technology and knowledge spillovers, can generate benefits for the whole economy and for other countries. Exports of agricultural commodities and food and beverages contain value added that is created by a range of economic sectors that supply inputs, such as fertilizers, energy and services. A significant part of foreign value added in agri-food exports, globally, is provided by the services sector – in 2015, for agriculture and food and beverages, services made up 42 percent and 38 percent of the foreign value added embedded in GVC exports, respectively.¹⁶

In addition, a significant share of imported inputs (on average, 22 percent in 2015) is provided by the chemicals and raw materials sector (which also includes petroleum). This large share partially reflects the globalization of the fertilizers and pesticides markets. The share of manufacturing (including machinery) in foreign value added is also sizable in both agriculture and food and beverages, amounting to 19 percent and 16 percent, respectively.

Such linkages also shape the strength of the relationship between GVC participation and economic growth together with the country’s capacity to effectively absorb technology and knowledge. Factors such as the structure of the economy, geography, the size of the domestic market, the level of development, but also crucially government policy settings matter. The impact of increased GVC participation is likely to vary depending on policies that promote the mobility of production factors, especially labour, and on conditions that allow economic activity to expand, such as investments in human capital through better skills, improved infrastructure and effective regulation.

GVC participation and agricultural labour value added

Analysis also suggests that GVC trade results in increased labour value added or productivity per capita.²⁴ The main mechanism for this lies on how value chains unbundle the production process – allowing farms and firms to leverage additional options for comparative advantage – promoting stronger competition and better access to capital and knowledge. For example, with adequate skills, backward linkages can work as a channel for the transmission of improved technology leading to better farm practices and improved labour productivity.

GVCs can represent an opportunity for supporting the ongoing transformation of food and agriculture in developing countries and spurring a shift from a low productivity to a more commercialized and productive agricultural sector with stronger backward and forward linkages to the domestic economy and the global market.²⁵

Figure 2.7

EFFECT OF A 1 PERCENT CHANGE IN GVC PARTICIPATION ON AGRICULTURAL VALUE ADDED PER WORKER

Back

Greater GVC participation of the food and beverages sector is also estimated to have a positive effect on agricultural value added per worker, with an average effect at around 0.08 percent. This is due to the strong links between agriculture and the food industry: agricultural commodities that are produced domestically and then processed and exported by the food and beverages sector through GVCs can raise productivity in agriculture.

The estimates also suggest that GVC participation can have sustained long-term impacts on labour productivity in agriculture. An increase of 1 percent in GVC participation continues to add to agricultural labour productivity even after two years, although this long-term impact diminishes moderately as time passes – the effect of agricultural GVC participation weakens to 0.10 percent after two years.

Additional analysis for the shorter period of 2009 to 2015 provides evidence that the effect of GVC participation on agricultural labour productivity was not a temporary feature of the high-growth economic boom at the beginning of the century but remained during the period of much weaker growth after 2008. This suggests that, during times of fast growth, processors and retailers procure from many farms, but the least productive of these drop out of the global value chain when growth stagnates. This sorting process, where only the more productive farms remain connected to global markets, could result in a higher average impact on productivity (Figure 2.7).

Over the past four decades, international trade negotiations in the General Agreement on Tariffs and Trade, and subsequently under the WTO, have contributed to opening up global markets. Import tariffs on agricultural commodities and food products have declined since the implementation of the WTO Agreement on Agriculture in 1995–1996 (see Figure 1.12, Part 1). Many developing countries have initiated policy reforms to reduce trade barriers and engage in international trade.

Opening global markets can bring benefits to all trading partners and can create important spillover effects through the transmission of technology and the transfer of know-how. Opening markets is more likely to result in significant benefits if complemented by other policies that underpin competitiveness, such as measures that improve governance and infrastructure, upgrade skills, remove rigidities in labour markets and facilitate the reallocation of labour between sectors. However, there are concerns about the short-term effects of opening trade, especially the impacts on income distribution and inequality.^12,35,36

To reap the benefits of GVC participation for economic growth, appropriate trade policies on both the import and export sides are critical. Opening to trade and removing market-distorting policies could enhance the unbundling of production processes internationally, and thereby promote GVC participation. Through various mechanisms, opening to trade stimulates economic activities in general and can facilitate food system transformation, including the emergence of a domestic food industry (see Part 1).

BOX 2.5

Analysing policies to promote GVC participation: effects by policy measure and returns to land, labour and capital

A computable general equilibrium model – a model of the global economy including agriculture and the food sector – is used to simulate the effects of different policies on GVC participation. The simulation exercise considers a policy package that includes the removal of all import tariffs and export restrictions in all sectors of the economy, as well as the removal of domestic subsidies and taxes on agricultural commodities, food and beverages products, and land inputs. As the model is a stylized representation of the economies involved, these results should be interpreted with care: the mechanisms and direction of impacts matter more than the size of the effects.

The policy changes have immediate impacts on gross agri-food exports and through those on value added and GVCs. While the overall effects on GVC participation are positive, the impact can differ by region, policy measure and production factor.

PROJECTED EFFECTS ON GROSS EXPORTS BY POLICY MEASURE

From a GVC perspective, tariff and non-tariff barriers – including those to trade in services – are viewed as important instruments in determining domestic value added. However, the strength of the effects on agri-food GVCs can differ by measure and economic sector to which these measures are applied.

In most regions, phasing out trade barriers in agriculture and food and beverages is projected to be more important than reducing domestic support distortions. Removing trade barriers in sectors other than food and agriculture also brings about impacts on agri-food exports (see Figure 2.8).

Figure 2.8

PROJECTED EFFECTS OF REMOVING DIFFERENT POLICY MEASURES ON GROSS AGRI-FOOD EXPORTS, PERCENT CHANGES

Back

Removing tariffs on agricultural commodities and food products across all countries and regions is projected to result in increases in agri-food exports. This also implies better opportunities for GVC participation in the form of foreign value added in production for exports (backward linkages), as well as increased exports of intermediate products for foreign processing and re-exporting (forward linkages).

The removal of domestic support to agriculture has little effect on agri-food exports compared to trade liberalization.

In contrast, removing trade barriers in sectors other than food and agriculture, leads to adjustments that benefit agri-food exports from some regions at the expense of exports from other parts of the world. For example, in Africa, phasing out non agri-food trade barriers has a positive effect on the economy of the region, through improved terms of trade that result in increases in the relative competitive position of exports of all sectors, including those from food and agriculture. African agri-food exporters can thus reap a larger share of the global market vis-à-vis their competitors.^* However, in Asia and Europe, this competitive position deteriorates, worsening export prospects for food and agriculture.

PROJECTED EFFECTS ON RETURNS BY PRODUCTION FACTOR

On average, trade and GVC participation can have a positive effect on agricultural income, both in terms of domestic value added and the share that accrues to labour.³⁸ Especially in developing countries, increased GVC participation could create more jobs for unskilled workers. Indeed, increased GVC participation through the removal of trade barriers and distortive policies is projected to lead to a relatively large increase in the demand for unskilled labour in regions where average income per capita is relatively low. In developed countries and regions, the results suggest that the benefits will accrue to both skilled and unskilled labour (Figure 2.9).^** Nevertheless, a key question relates to when low-skilled farmers and agricultural workers can reap such benefits, as GVC-based trade tends to place stringent requirements on production that necessitate specific skills and capabilities.

Figure 2.9

PROJECTED EFFECTS OF OPENING TO TRADE ON EXPORTED AGRI-FOOD VALUE ADDED BY PRODUCTION FACTOR, PERCENT CHANGES

Back

Land and capital are generally also projected to increase their contribution to exported value added.^*** The exception is Europe, where removing domestic taxes and subsidies related to land is projected to result in a decline in value added.^****

Back

Opening to trade and GVC participation

In agriculture and food and beverages, the removal of all trade barriers and market distortions is projected to increase GVC participation and value added through both backward and forward linkages in all regions.

Figure 2.10

PROJECTED EFFECTS OF OPENING TO TRADE ON GVC PARTICIPATION, PERCENT CHANGES

Back

In food and beverages, both domestic and foreign value added also increase across all regions, but, for some regions, backward linkages (through foreign value added) are not as important as domestic value added as in the case of agriculture (Figure 2.10, Panel B). This reflects different strategies to leverage the opening up of global trade. Some countries may increase GVC participation through increases in the use of domestic inputs, and therefore higher domestic value added. In other countries, the food and beverages sector may choose to expand exports by increasing imports of agricultural commodities, which reflect higher foreign value added.

Across regions, there are significant differences in the projected gains from opening to trade. This is because the model simulation results depend on the size of the shock (the initial import tariff level matters), the size of the global market for specific commodities (the sector matters), the size of the country (the size of the domestic economy matters), and the specific specialization and comparative advantage patterns of the country (the mix of exported commodities also matters). For example, North America has lower levels of tariff barriers and distortive measures compared to most other regions and thus is projected to gain less from removing them.

Trade policies and the transformation of the domestic agri-food sector

Figure 2.11

PROJECTED EFFECTS OF OPENING TO TRADE ON DIRECTLY AND INDIRECTLY EXPORTED AGRICULTURAL AND FOOD VALUE ADDED

Back

BOX 2.8

Emerging food processing sectors in developing countries

While research on agricultural development has traditionally focused on the sector’s integration in global markets, transformations in the middle segments (processing, logistics and wholesale) of agri-food value chains in developing countries have until recently been studied less.⁵⁰

Following trends in developed economies and fueled by domestic and international private sector-led investment, this transformation often started with a proliferation of small and medium-sized midstream enterprises and was followed by a process of consolidation and concentration. Today, midstream segments can form 30 to 40 percent of the value added in food value chains in developing countries.⁵⁰ For example, in Bangladesh, the People’s Republic of China and India, the share of the midstream segments in total marketing margins in rice value chains was found to average around 32 percent, while it was estimated at 42 percent for potato value chains.⁵¹

In West Africa, the food processing sector is the largest manufacturing sub-sector in terms of employment. It accounts for only 5 percent of employment in the total agri-food economy but represents an average of 30 percent of total secondary sector employment. In Niger and Nigeria, food processing accounts for almost 50 percent of all manufacturing activities (Figure 2.12), with many of these jobs being in micro, small and medium-sized enterprises in the informal economy.⁵²

Figure 2.12

IMPORTANCE OF FOOD PROCESSING FOR EMPLOYMENT – WEST AFRICA AND SELECTED COUNTRIES (SHARE IN TOTAL MANUFACTURING SECTOR LABOUR)

Back

Due to its nature in the midstream, the food processing sector creates strong domestic and international forward and backward linkages with agriculture and other non-agriculture activities. However, while the demand for food processing activities is expected to grow further in many developing countries, the growth of large-scale industrial processors is often constrained by an unreliable supply of local raw materials of consistent quality, often resulting in reliance on imported commodity inputs.⁵²

To improve the reliable and steady supply of agricultural commodities, food processors have started shifting from sourcing from spot markets to engaging in more formal contracts with farms. However, contract farming linked to processors seems to be emergent only in a few commodity categories (see Part 3). To ensure the quality of agricultural commodities, public food standards have been increasingly complemented by private standards (see Part 1).⁵⁰

SOURCES: Reardon. 2015; Reardon. et al. 2012; Allen et al. 2018.^50,51,52

Back

On average, in the food and beverages sector, open markets can stimulate expansion in both directions, but with a stronger impact on directly exported value added. The value added that is generated and exported by the food and beverages sector itself also reflects the more downstream nature of the sector (Figure 2.11, Panel B).

The effects for Oceania are large in percentage terms, but as the region makes up for less than 10 percent of global trade in both sectors, this increase masks low initial trade levels.

Trade policy and domestic support implications

Globally, open markets can spur economic activity and promote trade and GVC participation. Reducing trade barriers can lead to both more imports of inputs to agriculture and increased exports of agricultural commodities for processing in other countries. At the same time, the food and beverages sector can import more inputs from agriculture abroad and increase its exports for further processing and subsequently for final consumption in partner countries.

However, on average, a large part of agricultural production is projected to be used by the domestic food industry (see Figure 2.11, Panel A). This means that the food industry will export domestic value added from agriculture. Thus, lowering trade barriers could imply a proliferation of global value chains, but also the potential for developing a domestic food industry. Such a path is in line with the interaction of economic growth and the transformation of value chains in food and agriculture (see Part 1). Along the development path, the food industry grows, while the relative contribution of agriculture to total agri-food value added declines (see Figure 1.14 in Part 1). The stronger linkages of agriculture to the domestic but also to foreign food industries can stimulate further labour productivity growth, thus adding to economic growth.

The share of trade that flows through GVCs, in conjunction with their effects on productivity and economic growth can strengthen the argument for lower trade barriers and open markets. As GVCs slice the production process across countries, they combine the comparative advantage of many firms across many countries and thus provide an important entry point to international trade. With GVCs where production stages are thinner and more specialized, it is easier to penetrate the global market.

Analysing trade through a GVC perspective also reveals that the costs implied by trade barriers can be high.^1,53 The increased fragmentation of production across borders means that tariffs are incurred multiple times along the value chain. As inputs and intermediate products cross borders many times, tariffs are being applied to the full value of exports, including on the amount in tariffs paid previously. This can have significant knock-on effects on all trading partners in the global value chain. In addition, uncertainty about trade policies can also be amplified through GVCs, as firms are more reluctant to make further investments in new or existing relationships with foreign suppliers.

As tariffs cascade through GVCs, this makes their impact larger and, when a commodity is exported to be processed and subsequently re-imported in the country of origin, detrimental. Furthermore, as GVCs strengthen the trade linkages between countries, domestic value added creation is affected not only by domestic trade measures, but also by policies of other countries. Tariffs imposed in the destination market can have ripple effects on the production activities which are linked to the GVC and which span across different countries.⁵⁴

The benefits of tariff reductions are thus larger when a significant part of the agri-food trade takes place through GVCs. This can entail a shift in policy focus from import substitution policies and protection of the domestic producers through tariffs to providing incentives to increase domestic economic activity through enhanced exports and integration in the global market.^55,56 As an increasing share of global trade takes place among emerging and developing economies, and this share is projected to expand further, such a strategy can only be successful if implemented in as many countries as possible, rather than relying on increasing access to markets in developed countries alone.

GVCs, but also the evolution of food and agriculture more broadly, have led to increased requirements on technology, capital and labour skills for producing food and beverages (see Box 2.5). Linkages to GVCs can be facilitated by promoting the transformation and development of the domestic sectors. The development of competitive agriculture and food industry requires policies that provide incentives for the uptake of novel technologies, enhance skills and capabilities, and facilitate the cooperation between public and private actors.^38,55 At the same time, while open markets are generally conducive to economic growth, they may have various effects on environmental, social and health outcomes. Both positive and negative effects can be amplified through GVCs.

Domestic policy conditions need to be coherent with the trade policy framework to take advantage of the opportunities arising from increased GVC participation. Policy-makers should aim at creating an environment where both food and agriculture can best leverage their comparative advantage and be competitive in agri-food GVCs.³⁸

The debate on the economic gains from trade is enriched by a discussion on its impacts on the environment, inequality, and, specifically for food trade, concerns about health and nutrition. International trade, as all economic activities, can support sustainable practices, encourage unsustainable ones, and generate a range of environmental and social outcomes (Box 2.9). Global value chains can strengthen the effects on sustainable outcomes, as they support closer links between the different actors than looser forms of trade do. On the one hand, both positive and negative effects can be amplified through GVCs, especially under open trade. On the other hand, knowledge and technology spillovers, that can be leveraged in GVCs, can address the trade-offs between the various economic, environmental and social objectives.

Trade, GVCs and the environment

Global value chains that are coherent with sustainable development objectives can spread sustainable technologies and practices and, at the same time, promote productivity and income growth across countries. Increased GVC participation can propagate the positive impacts of environmental regulations across borders and contribute to sustainable development. For example, export-oriented firms in a country may adhere more strongly to sustainability regulations and use cleaner technologies than typical domestic firms, either to ensure that public standards in the importing country are met or because of private standards imposed by the downstream partners in a GVC.

Trade policies that facilitate regulatory harmonization and uphold high sustainability standards throughout GVCs can prevent regulatory arbitrage by multinationals that can easily move parts of the production chain across borders. For example, the provisional terms of the new European Union–Mercosur trade agreement directly link tariff elimination to animal welfare standards.⁵⁷

Environmental supply chain management to reduce environmental impacts, pollution and waste is especially relevant in the age of global value chains.^59,60 This includes green logistics management with reductions in emissions, waste and pollution from logistics activities; sustainable transport options through alternative transport modes and more sustainably refrigerated trucks; and reduced packaging and use of recycled packaging materials. Due to the global nature of many value chains, international coordination is essential, as is attention to environmental impacts that cannot easily be attributed to a country, such as those from international shipping and aviation.⁶¹

Private standards can also be an effective tool to make GVCs more sustainable. There are clear environmental and social benefits to adherence to the requirements of sustainability certification schemes (see below and Part 3).⁶⁸ The share of agricultural production affected by sustainability standards is growing rapidly; as of 2015, more than 50 million hectares were certified as organic, and the sustainability certified areas of cotton, bananas, cocoa and tea more than doubled between 2011 and 2015.

Trade, GVCs and inequality

The sharp increase in developing countries’ participation in trade and the emergence of global value chains coincided with a significant decline in extreme poverty worldwide.⁶⁹ Open markets are often seen as a tool for growth, but they are not a mechanism to reduce inequality.⁷⁰ Indeed, as globalization progressed, inequalities in wealth and income widened within many countries.⁷²

A recent analysis of the impacts of eliminating tariffs on agricultural products across 54 low- and middle-income countries pointed to increases in both income and inequality.⁷¹ The results suggest that, on average, liberalizing agricultural trade would increase household incomes. At the same time, eliminating import tariffs was found to have highly heterogeneous impacts across countries, and within countries across households. In 37 out of the 54 countries, the top 20 percent of the richest households would gain more from liberalization than the bottom 20 percent, thus exacerbating relative inequality even when all household groups gain in absolute terms.

For example, in Viet Nam, the income of richer households increased, on average, by 2.7 percent, while that of the poorer ones by 1 percent. Such differential impacts depend on household characteristics, such as different consumption patterns and income structures, but also call for complementary policies and actions.

The impacts of increased agricultural trade through GVCs on inequality may be more pronounced, as the associated new technologies and innovative processes require higher skills. GVC-based trade may therefore, to some extent, undercut the opportunities of developing countries to reap comparative advantage that is based on low-skilled labour.⁷² In the past, a number of economies in Southeast Asia experienced rapid growth and a shift to low-cost export-orientated manufacturing by leveraging regional and global value chains and low-skilled labour; this led to increased productivity and higher wages, leading these countries to a middle-income status. Recent evidence from analysing manufacturing-related GVCs across 58 countries suggests that although GVC participation led to productivity increases, it did not result in employment growth.⁷³ This could be related to the fact that manufacturing has become increasingly capital-intensive.

When the focus is on agriculture rather than manufacturing, the requirements on capital and high labour skills are perhaps mitigated. But even agri-food GVCs put a premium on skilled labour, farm size and access to credit. Not all farmers in developing countries have the skills and means required to adopt the farming practices, standards and logistic targets set by downstream GVC partners.

If access to GVCs is only feasible for higher-skilled and large-scale farmers, relative social inequality may increase, despite increases in average income. Agricultural commercialization that is sometimes induced by GVCs may marginalize poor smallholder farmers that cannot meet stringent requirements, even if average agricultural productivity is increased and those that do participate in GVCs reap economic gains. Given these distributional concerns, it is essential to address the market failures that prevent poor farmers from access to lucrative markets (see Part 3 for a discussion on farmers’ participation in value chains).

In general, access to water and energy gives people – and especially women who tend to spend more time gathering water and fuel – the opportunity to use their time productively rather than on addressing basic needs. Ensuring that all children have access to education and adults to lifelong learning is essential. The high skill requirements associated with GVCs and with sustainable production methods can be a major stimulus for improving the linkages between educational goals and decent work objectives. More modern production technology that is often required for integration in GVCs can also make food and agriculture more appealing for educated younger generations to remain in rural areas and contribute to vibrant rural economies.

Trade, GVCs, food security and nutrition

Generally, GVCs could be a significant source of socio-upgrading opportunities.⁷⁴ Participation in agri-food GVCs can improve the food security of smallholder farmers by promoting productivity, which in turn can increase rural incomes, reduce rural poverty and foster pro-growth opportunities (see also Part 3).⁷⁵ Positive spillovers, especially through productivity increases, on domestic food markets can also contribute towards food security for all.^76,77 Such gains can enable people to buy more food (thus increasing energy intake), to buy more diverse foods (thus increasing dietary diversity and possibly quality), or to invest in sanitation and healthcare (which are crucial determinants of nutritional outcomes, especially in children).⁷⁸ The trade-offs involved are, however, complex, and there are significant differences across regions and between markets.

When specific actions are taken, global value chains could also help contribute towards reducing malnutrition. Interventions may include fortifying processed food with specific micronutrients (such as folic acid and iron) that may otherwise be lost in food processing or that more generally are not consumed regularly or in sufficient quantities by the poorest. Well-functioning GVCs based on improved cold chain technology can allow more trade of fruit and vegetables, that would otherwise spoil during transport. They can thus increase diet diversity for consumers in countries that do not have a comparative advantage in the production of fruit and vegetables. Lastly, packaging and nutritional labelling could also lead to enhanced demand for more nutritious foods and possibly reduce the demand for energy-dense foods.

However, the increased availability of processed foods has led to concerns about the contribution of trade and GVCs to overnutrition and obesity. Urbanization and changing lifestyles, as well as the increased prevalence of households where both women and men have paid jobs, have resulted in greater consumption of processed foods. Evidence from Mexico points to significant increases in the share of energy consumed from highly processed foods in urban households with higher income, with a highly educated family head, and in which both men and women participate in the labour market.⁷⁹

The transformation of agricultural and food markets is the result of a combination of factors such as higher incomes, urbanization and the nutrition transition. It spurred a high penetration of supermarkets at the retail level and the introduction of stringent food quality and safety standards. The demand for differentiated products and the implementation of public and private standards across countries result in increasingly complex global value chains. At the same time, the proliferation of GVCs, especially the fragmentation of production processes across countries, requires strong vertical coordination and governance within the chains that often raise concerns over containing market power if national competition policies differ.

Standards and GVC access

Technical standards

There are many reasons why certification and standards are used throughout the various stages of GVCs. Governments use public regulations and standards to ensure the health, safety, and environmental and social quality of the agricultural commodities and food that enter their markets. These standards are regulated by WTO Agreements, such as the Technical Barriers to Trade (TBT) Agreement and the Sanitary and Phytosanitary (SPS) Agreement. The TBT Agreement covers product standards, technical regulations and conformity assessment procedures, and provides disciplines to ensure that imported products are treated equally with “like products” of national origin. The SPS Agreement serves to guarantee food safety and animal and plant health regulations.

Since along a GVC, the stringency of public standards differs across countries, standards are also imposed by private firms to ensure that they will be able to sell their final product in a given market. Downstream retail firms need mid- and upstream producers to adhere to the standards of the country where final consumption takes place.

Compliance with private standards may also be required to ensure that inputs can be used for the intended purpose in the downstream stages of the value chain. For instance, wheat protein content determines its suitability for a variety of purposes. Private standards may also be used by firms – typically vertically integrated – as a marketing tool.⁸⁷ They sometimes allow firms to differentiate products and increase market shares. The complexity in production processes associated with GVCs, as well as corporate branding and marketing strategies, has increased interest in third-party certification schemes to provide assurance of the compliance with private standards along the value chains.

At the farm level, analyses have found that compliance with private standards can have positive effects on productivity, exports and employment. In Kenya, for instance, incomes increased after farmers adopted the quality standards demanded by their international buyers, and these firms supported better traceability of the product throughout the entire supplier network.⁸⁹ The empirical evidence remains mixed, and many smallholder farmers may not have the capacity to supply food complying with stringent standards (see Part 3 for a discussion on compliance with private standards within the context of contract farming).^76–78

Voluntary sustainability certification schemes

Sustainability certification schemes and standards comprise voluntary norms adopted by businesses; they aim to address non-economic dimensions of sustainability and can promote social and environmental outcomes. Such voluntary sustainability standards specify requirements for production methods in terms of, for example: the respect for basic human rights; workers health and safety; paying farmers a fair price for their produce; and various farm practices that can better manage natural resources and reduce negative environmental impacts.

Sustainability standards are gaining importance in global markets, especially for high-value products with established links to global value chains. Growing consumer demand for sustainability certified products has resulted in increases in the share of agricultural land under sustainability certification. A relatively large share of tropical commodities cultivated in developing countries, such as coffee, cocoa, tea, palm oil and cotton, is certified. In 2015, over 50 million hectares were certified as organic, representing 1.1 percent of agricultural land worldwide. The RSPO-certified palm oil accounts for 0.07 percent of the global agricultural area. About one-quarter of the global coffee and cocoa areas are certified through standards developed by both non-governmental organizations (NGOs) and the private sector.⁶⁸

GVCs, with their effective vertical coordination mechanisms, hold significant opportunities to apply sustainability standards and align global markets with sustainable development outcomes. The proliferation of certification schemes is in part a response to increased consumer awareness over sustainability concerns, particularly in high-income countries, but increasingly also in emerging and developing countries. For example, a label or certification of standard-compliant production by Fairtrade, Organic or Rainforest Alliance addresses environmental and social concerns; it provides information to consumers to make decisions responsibly on what to buy in accordance with their preferences and social beliefs. There are several issues of concern for consumers, including food safety, environmental sustainability, and social norms such as child labour, gender equality and producers’ welfare. ⁹³

Standards and challenges to access GVCs

Certification schemes and standards in international markets can deliver positive impacts but also present challenges for small-scale processors and farmers, who often lack the technical and financial capacity needed to comply with complex and stringent requirements. This can induce retailers and firms downstream to reduce sourcing from small-scale suppliers. Transaction costs for monitoring compliance with standards may be very high in the case of sourcing from smallholder farmers.⁹³

In many developing countries, other obstacles can threaten production that is compatible with international standards, including weak regulatory institutions, poorly designed and implemented sanitary and phytosanitary regulations, and inadequate transportation, power and water infrastructure.⁹⁴ Consequently, the inclusion of smallholder farmers in value chains that have sustainable certification may only be feasible with external support from development programmes, public-private partnerships, non-governmental organizations, or collective action.

The cost of uncoordinated standards across countries can be amplified within GVCs, much more so than in bilateral non-GVC trade, as compliance needs to be coordinated at each stage of production and for each market ultimately supplied. Compliance can require firms to make costly investments in duplicate production processes, specific packaging and labelling or to undertake multiple certification processes for the same product. These compliance costs are particularly acute for small and medium-sized enterprises (SMEs) and smallholder farmers and are a major obstacle to their GVC participation.⁹⁵

Policies to facilitate and support compliance with international standards and to harmonize standards and certification can foster growth in GVC participation. International regulatory cooperation and the convergence of quality and safety standards may alleviate the burden of compliance and enhance firms’ participation in global markets (see Part 1).²⁵ International initiatives for sustainable business practices, such as the United Nations Global Compact, can also be crucial for addressing sustainable development issues. However, their voluntary nature may, to some extent, hamper progress when trade-offs between economic, social and environmental objectives reflect significant asymmetries between private and public gains.

Market power, competition and the distribution of GVC benefits

The transformation of agricultural and food markets in recent decades has also brought significant changes in market structures and market power for the various actors.⁹⁶ The dominance of supermarkets in food retailing and the importance of a relatively small number of large multinational food companies has also contributed to increased vertical coordination in the agri-food value chain and has enhanced emphasis on GVCs (see the discussion on vertical integration in Box 1.4 in Part 1).

There is clear evidence of market concentration, especially in seeds,⁹⁷ fertilizers,⁹⁸ the international commodity trading sector,⁹⁹ and food processing and retailing. Other parts of the agri-food sector are characterized by a large number of suppliers.¹⁰⁰ Figure 2.13 illustrates how market concentration can vary widely across crops and regions by examining the market for seeds.

Figure 2.13

THE DEGREE OF SEED MARKET CONCENTRATION VARIES BY CROP AND REGION

Back

To some extent, market concentration and market power can be driven by the existence of a natural monopoly or oligopoly, especially when scarce natural resources are used, such as in the case of fertilizer production. Another driver is the research and development (R&D) intensity of the sector. For example, high R&D investments in the seeds and biotechnology industries could create entry barriers that could hamper competition.

Often in GVCs, market power can be linked to innovations that create a local and temporary monopoly generating excess profits. For example, packaged mixed vegetable salads, introduced through specific requirements by the innovator to upstream suppliers and processors, is seen as conferring some market power in the short term, at least until other competing firms enter the market.¹⁰¹ Therefore, value chain innovations and product differentiation often give rise to market power in particular locations, and temporarily until competitors arise.

Traditionally, market concentration in value chains has been linked to collusive behaviour and market power. This increases prices for consumers (due to oligopoly rents) and lowers it for farmers (due to oligopsony rents), reducing welfare for both, and transferring gains to the large food processing companies and food retailers.¹⁰² Nevertheless, market concentration does not necessarily lead to collusion or imperfect competition. Empirical evidence of market power abuse in agricultural and food markets remains scarce, despite high market concentration in parts of the value chains that are dominated by few firms which rely on vertical coordination.^103,104

To some extent, this may reflect the difficulty and complexity of identifying market power. Anecdotal evidence points out that some of the more powerful firms unilaterally impose contract conditions and may practice “unfair” business.¹⁰⁵ But “unfairness” is also difficult to identify and may include a refusal to have a written contract, excessive transfer of costs and risks between transaction parties, or frequent changes in prices. Regulation by competition authorities is hindered by difficulties to prove perceived unfair trading practices, but formalising transactions along the value chain through contracts may overcome some of these issues (see Part 3).

In general, the literature does not support the claim of systematic market power abuse.¹⁰⁶ For example, the penetration of food markets in emerging and developing economies by large food and retail companies from developed countries has boosted GVC participation rates, but there is no clear evidence that it has induced large-scale market power abuse. The effects of market power along the value chain could also be positive. For example, there is evidence that increasing concentration and market power of the downstream buyers could potentially help overcome local market and government failures in rural areas where upstream suppliers are located, as they may alleviate structural market barriers through reduced transaction costs and improved contract clarity.¹⁰⁷

PART 3 shifts the analysis from the global level to the farm household. Smallholder farmers face a number of constraints that determine their participation in markets and value chains. These constraints also affect their aspirations for better livelihoods. The discussion places the farm within the process of development to look at markets and market behaviour. Business models such as contract farming and value chains that integrate sustainability certification schemes are examined. These can help address the constraints farmers face, include them in markets and contribute to economic, environmental and social outcomes.

Markets lie at the heart of the development process, allocating activities and resources where they are most productive. In food and agriculture, well-functioning markets and trade are vital to improving the livelihoods of millions of people and can provide additional benefits, such as contributing to food security by ensuring that food moves from surplus to deficit areas.

The development process is characterized by the structural transformation of the economy. It is the path towards higher incomes and poverty eradication, but also the expansion of opportunities for better living standards and the ability of people to choose among different livelihoods.

Although earlier analyses viewed agriculture as a traditional low-productivity sector that had to provide labour and other resources to fast-growing modern sectors, no country has been able to achieve a transition out of poverty without a dynamic agricultural and food sector.¹ At the same time, agriculture’s growth depends crucially on how productivity and employment evolve in manufacturing and services. Structural transformation integrates the growth paths of all sectors of an economy, and this process depends on well-functioning markets (see Box 3.1 for a discussion on the role of markets in development).

BOX 3.1

HOW WELL-FUNCTIONING MARKETS CONTRIBUTE TO DEVELOPMENT

Broadly, structural transformation can be initiated by increases in the productivity of labour, followed by increases in income, which can stimulate demand, create employment and spur economic growth. Improved technologies, investment in education and skills, and well-functioning markets for labour, capital and products are key to this development process.

In agriculture, productivity increases mean that fewer people can produce more food. People start leaving agriculture in search of better economic opportunities in cities. With well-functioning labour markets, workers move from agriculture to other sectors of the economy, such as manufacturing and services, to find employment. Their savings provide capital fueling growth. Societies urbanize, and rural households diversify their income sources by obtaining better-paid work in the non-farm rural sector that also bridges agriculture with the rest of the economy.

Product markets also link agriculture with other economic sectors. Forward linkages are identified with agricultural output markets, as agriculture (1) provides low-cost food to workers, underpinning productivity growth in the broader economy; (2) supplies inputs to the food and manufacturing sectors; and (3) markets exports to earn foreign exchange that is necessary to import capital goods that add to economic growth. There are also important backward linkages through markets for inputs for agricultural production such as seed, labour, machinery, agrochemicals and, increasingly, services. Through these linkages, agriculture can be an engine of economic growth early in the development process and a major force for poverty reduction.

Along this path to structural transformation, agriculture’s relative importance in the economy declines as per capita incomes increase (Figure 3.1). As people become progressively richer, they consume more manufactures and services, while the demand for food rises at a lower rate. The final stage of structural transformation is an economy where GDP per capita is relatively high and agriculture contributes towards a small part of the GDP. For example, in 2017 in the United States of America, agriculture contributed around 0.9 percent of GDP, compared to an average of 21.4 percent in the least developed economies. At this final stage of transformation in the United States of America – which was probably achieved in the 1980s – agriculture became fully integrated with other sectors through well-functioning markets, with the productivity of labour across sectors becoming equal.

Figure 3.1

STRUCTURAL TRANSFORMATION ACROSS COUNTRIES: GDP SHARE OF AGRICULTURE AND GDP PER CAPITA, 2017

Back

Source: Timmer & Selvin. 2008.¹

Back

As markets form an important part of the structural transformation process, farmers’ participation in market-based exchange is key for sustainable development and poverty eradication. Ensuring that the poor are connected to well-functioning markets links them better to the development process.

From a narrower perspective, agricultural market-based exchange generates benefits through the production and sale of products in which farmers are specialized and have a comparative advantage. This creates income which can be used to buy other goods and services, including food that other producers may be able to provide at lower cost. In the long term, markets can also bring sustained effects. As ideas are also exchanged through market transactions, better technologies are diffused, productivity increases, and farmers and their families build their productive assets and invest in education, health and their livelihoods.³

Increasingly, procurement is characterized by a shift from traditional spot markets, where farmers and traders meet at the farm-gate, towards sophisticated, global vertically integrated value chains where contracts specify the timing, scale and quality parameters of the transaction (see also Parts 1 and 2 for a discussion on the evolution of agri-food trade and markets and the emergence of global value chains). In developing countries, both traditional and modern value chains are present, with the latter mostly meeting urban demand for food. International trade is also increasingly taking place through global value chains.

Within this progressively transforming market environment, international trade has increased significantly, and sophisticated value chains, both global and domestic, link farmers with consumers, giving rise to better opportunities. Yet, many farmers in developing countries, especially smallholders, remain marginalized and excluded from the development process, having access to traditional or informal markets that function poorly or only very locally.

Currently, in developing countries, a range of value chains link farmers to both formal and informal markets. Some value chains are being developed to meet the demand of higher-income urban consumers through supermarkets (see Part 1). At the same time, global value chains offer significant opportunities for farmers in developing countries to participate in the international market based on their comparative advantage, rather than having to only rely on domestic processing industries that may be still less competitive (see Part 2).

However, not everyone can link to global value chains, especially when enabling conditions for market access are not met. Often, in developing countries, smallholder farmers sell to small and local markets, informal vendors and lower-income population groups. Indeed, most farmers sell into markets, but, on average, their participation in terms of how much of their production is commercialized is not high.

This does not mean that households are entirely isolated from markets. Most farmers in developing countries participate in markets, both formal and informal, but their sale volumes are small. Few of them are net-sellers.

Household survey data showcase farmers in developing countries who sell only a part of their production, and this part can often be small. For example, in Ghana, farmers sell, on average, about 46 percent of their crop production (in value terms), retaining more than half for in-household consumption (Figure 3.2). In Malawi and Uganda, farmers participate in agricultural markets selling about 21 percent and 30 percent of their crop production, respectively. In the more commercialized agriculture of Viet Nam, the data suggest that, in 2008, the average proportion of household production sold in markets was 52 percent.

Figure 3.2

MARKET PARTICIPATION: AVERAGE SHARE OF HOUSEHOLD PRODUCTION SOLD IN MARKETS IN GHANA, MALAWI, UGANDA AND VIET NAM

Back

These thin market-based transactions do not add much to the household’s liquidity which is crucial to lifting smallholder farmers out of semi-subsistence (Figure 3.3). For many farm households, a large part of income is made up of the value of production retained in-house for consumption, wages often earned in informal labour off-farm markets, remittances and transfers. Farm households rely on these earnings to supplement their own food production, and many are net-buyers in the food markets.

Figure 3.3

AVERAGE HOUSEHOLD SALES REVENUE OVER TOTAL HOUSEHOLD INCOME IN GHANA, MALAWI, UGANDA AND VIET NAM, PERCENT

Back

Market participation and transaction costs

The high costs of market-based transactions largely explain the low rates of market participation in developing countries. For example, many farmers may have limited opportunities to participate in markets due to poor infrastructure and limited road accessibility that translate into high transport costs. These variable transaction costs add to the price farmers pay for inputs and lower the price they receive for their products. As farmers are geographically dispersed, and their supply is both small and inconsistent, private traders either do not source from them or require high margins. Distance and the quality of transport infrastructure give rise to a range of different commercialization rates across farmers, which is not reflected by looking at the average picture as in Figure 3.2. For example, farmers located near cities and towns are, in general, characterized by higher market participation relative to farms that are distant from markets.

In rural areas, information is also costly, and farmers may not be able to access details on buyers, markets and price levels. Search costs relating to finding a trader, bargaining, negotiating and agreeing on a transaction are high. These costs are fixed in the sense that once farmers acquire the necessary information, they can sell any quantity without adding much to these costs. Smallholders, who lack scale in production and sales, may be unable to overcome these fixed costs. Often it is the larger and well-endowed farmers that make up most of the supply in markets, especially for staple foods.⁴

Nevertheless, for commodities such as coffee and cocoa or vegetables, smallholders can have significant market and global value chain participation rates.

Household survey data suggests that there is a positive relationship between the proportion of household production that is sold in the markets and farm size (Figure 3.4). For example, in Ghana, small farms at the bottom of the farm size distribution with a farm size up to 0.4 hectares sell 35 percent of their production in markets. For larger farms, at the top 20 percent of the farm size distribution, with a size of more than 6.2 hectares, the rate of market participation is over 50 percent. In Viet Nam, market participation rates follow a similar pattern but are significantly higher for all farm sizes, indicating that farmers in the country face lower transaction costs compared with farm households in Malawi and Uganda.

Figure 3.4

SHARE OF HOUSEHOLD PRODUCTION SOLD IN MARKETS ACROSS THE FARM SIZE DISTRIBUTION IN GHANA, MALAWI, UGANDA AND VIET NAM, QUINTILES

Back

High transaction and search costs in developing countries result in thin product and input markets. They also result in systemic market failures – price instability and missing markets for credit and insurance. For example, in the context of a developing economy, farms face considerable difficulties in accessing credit, as banks are often reluctant to lend to them due to poor collateral and lack of information. Lack of access to insurance limits farmers’ ability to mitigate production risks and hinders on-farm investment. Such market failures can create poverty traps characterized by a cycle of low investment, low productivity and low incomes, especially for smallholder farmers.

There are also other constraints which can isolate farms from modern markets. Sales through modern value chains, such as supermarkets, require from farmers an ability to provide continuity of supply and to meet demanding food safety and quality requirements. Lack of information on quality standards, limited access to technology and low managerial and logistic skills make it difficult for many smallholder farmers in developing countries to supply modern value chains.⁵

For example, pesticide residues inspections on imports in the European Union in 2013 resulted in 10 percent of the beans and peas that arrived at its ports being rejected. In the same year, Kenya’s USD 930 million horticulture export industry experienced a 50 percent decline in total exports. Kenyan smallholder farmers, who reportedly produced about 80 percent of these exports, were significantly affected.⁶

There are also success stories. In Ethiopia, the value chain for teff – the most important staple in the country – is transforming. Lower processing costs and increasing demand for convenience and quality has resulted in increases in market participation and improved farm-level productivity in areas that are well linked with urban centres.⁷

Across farms, access to markets is far from uniform as transaction costs give rise to a broad range of market participation rates. For many farmers in developing countries, decisions on production and exchange strategies are constrained because markets are missing or do not function well.

The most important implication of these constraints for growth and development is that a whole set of decisions at the farm household are not separable from each other. For example, when markets do not function well, the decision of what and how much to produce cannot be separated from the decision of what and how much to consume. Missing markets mean that farmers may choose to diversify their production to mirror their diets, rather than pursuing efficiency-increasing specialization strategies and relying on markets for their consumption.⁸

Women farmers face even greater disadvantages than their male counterparts, as they have less access to assets and social capital, and gender adds to the factors that determine the wide range of market participation rates in developing countries. Households headed by women generate significantly less income than those headed by men (see Figure 3.5). In many countries, households headed by women participate in markets to a significantly lesser extent compared with households headed by men (see Figure 3.6).

Figure 3.5

AVERAGE TOTAL HOUSEHOLD INCOME BY GENDER OF HOUSEHOLD HEAD (USD, VALUED AT 2011 PRICES)

Back

Figure 3.6

AVERAGE HOUSEHOLD SALES REVENUE OVER TOTAL HOUSEHOLD INCOME BY GENDER OF HOUSEHOLD HEAD, PERCENT

Back

Decisions on how to allocate labour across on- and off-farm activities also depend on markets. Off-farm employment can complement farm income and provide an important risk management tool by diversifying income sources. Lack of well-paid employment – which can also be associated with low education levels – results in farmers assessing the cost of their labour as being very low. And as they are faced with such a low “shadow wage”, farmers tend to supply more family labour on-farm. The smaller the farm, the greater the labour intensity, giving rise to an inverse relationship between yield per hectare and farm size – which is often observed but not entirely understood. Small farms tend to achieve higher yields per hectare than larger farms but significantly lower productivity per worker and thus relatively lower incomes per capita.⁹

Technology adoption rates can also be related to missing markets. Farmers who are integrated into markets are more likely to adopt new technologies than those who are characterized by low market participation. Households with limited access to markets have weak incentives to adopt new technologies and to increase productivity, as they face only their own demand, which becomes quickly satisfied with small increases in production. Farms that are well integrated into markets face an aggregate demand for commodities, and their behaviour towards technology adoption is in line with the returns they expect from selling more produce.³

In the context of developing markets, a myriad of decisions, including on how to meet social objectives such as investments in education and health, are affected by poorly functioning markets. Lack of insurance and credit markets in the face of adverse weather conditions can significantly influence critical investments in the education of children. In Côte d’Ivoire, for example, rural households facing adverse rainfall shocks tend to reduce school enrollment rates by between 30 and 50 percent.¹⁰ In Honduras, children from rural households that have limited access to credit markets attain lower educational outcomes. Such negative impacts appeared to heighten with weather shocks identified with Hurricane Mitch.¹¹

Farm size, markets and structural transformation

Agriculture is one of the main economic occupations in the world. In essence, more than 600 million farms provide income and employment for billions of people, while producing food and raw materials for a growing and increasingly affluent population. About 90 percent of these farms are estimated to rely predominantly on family labour, occupying 70–80 percent of global farmland and producing about 80 percent of the world’s food in value terms. Most of these family labour farms are small – about 70 percent of the 600 million farms worldwide are equal to or smaller than one hectare and operate 7 percent of total agricultural land.¹²

Family labour farms remain the dominant mode of agricultural production, even in high-income countries. This is because family members – being self-motivated – can carry out location-specific tasks, such as planting, fertilizing and weeding, without supervision and therefore at a lower cost compared to hired labour.¹³ Thus, along the structural transformation process, as people exit agriculture, farm sizes evolve driven by changes in rural population growth, technology improvements and rural-to-urban migration.¹⁴

Evidence from agricultural censuses indicates that, between 1960 and 2000, the average farm size declined across low- and lower middle-income countries, while it exhibited increases in high-income countries.¹²

Currently, Africa remains mostly rural, with about 40 percent of its population living in urban areas. Its rural population increased more than threefold between 1950 and 2018, from 196 million to 740 million, and although urbanization will continue, it is projected that more than 1 billion people will continue to live in the rural areas of Africa by 2050.¹⁵

Figure 3.7

EVOLUTION OF AVERAGE FARM SIZE, HECTARES

Back

Declining farm sizes may have significant implications for market participation, especially in the context of modern food markets and global value chains, where access entails substantial fixed costs in terms of information. Assets, especially land, together with access to transport infrastructure, can be strongly associated with market participation.

Low market participation can hinder the development process and can also have negative implications for poverty eradication and food security for a large part of the population. Indeed, increasing the commercialization of small farms can increase income and reduce poverty. For example, in Kenya, a study of a sample of farm households which on average sell about 44 percent of their output, suggests that an increase of 10 percentage points in market participation results in a 17 percent increase in mean income per capita and reduces the prevalence of poverty among households by 16 percent.¹⁸

Positive effects of market participation were also identified in addressing deprivations in education, health and nutrition, and living standards, suggesting that market participation results in increases in human capital and capabilities that are key in including household members in the development process. At the same time, although markets can contribute to poverty reduction, increased market participation could lead to increased inequality, as gains in per capita income tend to be larger for relatively well-endowed and wealthier households than for poor ones.

Well-functioning markets can facilitate the process of structural transformation. However, the heterogeneity that characterizes agriculture and farms, but also the value chains and the agri-food enterprises within and across countries, will necessitate multifaceted actions (see Box 3.2 for a discussion on rural small and medium-sized enterprises).

Focusing on how to promote productivity per capita, integrate into markets and ignite growth in the sector is crucial for development, but, at the same time, the process of structural transformation entails different livelihood strategies for farm households within or even outside agriculture. Such strategies take into account several factors, including commercialization, which depends on farm size, transaction costs and the ability to meet the demand for food by urban consumers, but also on education, diversified skills and health that shape the ability of a household to transition entirely out of agriculture to other sectors of the economy.

In developing countries, policy prescriptions should address many challenges of integrating farmers into the process of economic growth. For example, social protection mechanisms reduce vulnerability, extreme poverty and deprivation but can also have positive impacts on market participation, as they enable households to manage risks better and engage in agricultural production.¹⁹ A range of interventions – both broad-brush and specific – will require investments to address market failures and many dimensions of development. Market-oriented farm households will directly benefit from improved transport and communication infrastructure that lower transaction costs and give rise to markets for products, inputs and financial services. Participating in these markets can promote investments, productivity and income growth but will also allow the households to diversify their income sources in the non-farm sectors.

Other households can leverage labour market linkages and public investments in education to build assets and upgrade their skills to transition out of agriculture and into different sectors of the economy. A significant challenge for policy-makers relates to semi-subsistence households that may be poor and geographically isolated, with unclear property rights and little productive assets. For these households, conventional markets may not work, and the options for improving livelihoods become limited.

The Green Revolution – driven by technology improvements targeting small farms in the 1960s and 1970s – witnessed governments undertaking a role in addressing transaction costs and coordinating markets (through, for example, marketing boards) at a considerable fiscal cost. Subsequent market liberalization policies tended to underplay market failures and the need for complementary actions to enhance the coordination of market participants. Since then, the transformation of food systems has shifted the focus on how private sector coordination mechanisms can promote access to modern value chains and include farmers in the process of economic growth.

Among such coordination systems, contract farming presents an institutional solution to address transaction costs and market failures across commodities, inputs, credit, insurance and information.³⁷ Contract farming arrangements are increasingly seen as a means to include smallholder farmers in remunerative markets for added-value foods that are shaped by urbanization and income growth. Contract farming can also integrate these farmers into markets for export commodities that are driven by the expansion of global agri-food value chains.

Contract farming can be defined as a forward agreement between farmers and processing and/or marketing firms for the production and supply of agricultural commodities, often at predetermined prices. The arrangements can also involve the purchasing firms’ pursuing a degree of vertical coordination through, for example, the supply of inputs and the provision of technical advice. Broadly, the contract binds the farmer to deliver a specific commodity in quantities and at quality standards determined by the purchaser and requires the firm to provide the farmer with either inputs or technical know-how and purchase the commodity.³⁸

Contracts can take a range of forms with varying terms, requirements and conditions. As an institution, contract farming can link farmers to consumers through sophisticated supply chains that add value to food by transport, grading, marketing and processing, ensuring that food meets specific quality and safety requirements.

In many developing countries, firms are not able to entirely bypass smallholder farmers, either because they dominate the agricultural sector or because firms need to secure a continuous supply of commodities to cover their fixed costs. Often, sourcing locally can be attractive relative to imports because it minimizes the impact of currency depreciation. As economic growth results in an expanding middle class, traceability and food safety issues become increasingly important and contract farming provides a measure of vertical coordination and control over farming methods, use of agricultural inputs, volumes supplied, and quality and safety standards. At the same time, contract farming arrangements can link developing countries’ farmers to global value chains and export markets, spurring growth (see Part 2).

For farmers, this form of vertical coordination – with contracts including the provision of inputs, such as seeds and fertilizers, technical assistance, credit and insurance, and a guaranteed price at harvest – addresses a number of constraints, such as price risk and the lack of access to markets, credit and information. Today, such new coordination systems may involve not only agribusinesses and individual farmers but combinations of government agencies, civil society, farmer groups, banks, digital technology and mobile telephone companies (see also the discussion in Part 4 on digital technology applications and market failures).

Empirical evidence on contract farming

The impact of contract farming on small farms in developing countries has been a topic of interest and analysis since the 1970s. Most studies utilize household-level data and focus on the average effects of contracting on outcomes, including crop yields, food security, assets, incomes and poverty status.

In practice, the results of such studies provide an indication of the average effects of contract farming. It is difficult to disentangle the specific outcomes of the various components of contracts that include predetermined minimum farm-gate prices, provision of inputs, technical support, credit and other services, which also may vary within and across household samples (see Figure 3.8 and Table 3.1).

Figure 3.8

CONTRACT FARMING INCENTIVES SETS

Back

TABLE 3.1

Contract farming studies reviewed in this report

Back

Madagascar case studies: various crops

In Madagascar, contract arrangements with 1 200 farm households covering different crops across six regions with different agro-ecological conditions, generated significant positive impacts on participants’ total household incomes.³⁹ On average, a 10 percent increase in the likelihood of participating in contract farming was estimated to bring a 6 percent increase in total household income (for more details on the Madagascar study, see Table 3.1).

A subsequent analysis of the same sample of households in Madagascar also suggests that participation in contract farming promotes food security by reducing the duration of a household’s hungry season – that is the period during which one or more household members have less than three meals per day. This hungry season, which lasts between 3.3 and 3.7 months, coincides with the period before harvest when households – contracted or not – receive cash from selling their produce. Extra income from contract farming reduces the hungry season by about eight days on average, and households that participated in contract farming schemes were found to be about 18 percent more likely to have a short hungry season.⁴⁰

Across these households, participation in contract farming was found to depend on several characteristics. For example, female-headed households were 45 percent less likely to secure contracts with purchasing firms, reflecting the constraints women face in entering markets.⁴¹ Farmers’ experience was also viewed as an important factor for contract farming participation with every additional year of agricultural experience relating to an increase of 1.2 percent in the likelihood of participation, suggesting that meeting the various contract requirements requires managerial and technical skills.

Farm size was estimated to be positively related to contract farming participation. On the one hand, the larger the farm, the more opportunities for diversifying production and thus participating in contract farming. On the other hand, as participation in contract farming is often found to be biased towards larger and relatively wealthier farms – that can better meet quality and quantity requirements – this indicates that contract farming may lead to greater inequality in rural areas. Indeed, it is likely that poorer farmers are not included in contract farming. In general, a systematic review of the literature on the effects of contract farming arrangements on income found that 61 percent of contract farmers had larger farms or more assets than their non-contract counterparts.⁴²

Senegal case study: cash crops

However, farm size and household wealth may not be the only significant driver of contract farming participation; the relationship may also depend on whether the production of contracted crops requires specific investments. For example, in Senegal, the participation of farmers in contract farming schemes for peanuts was shown not to depend on farm size. The cultivation of peanuts is a traditional cropping system that does not require any specific investments in either capital or knowledge as would be the case for an unfamiliar high-value crop.⁴³ Contract arrangements, based on local community information and reputation instead of farm assets, were found to increase farm income significantly, thus reducing poverty and inequality (see Table 3.1 for details on the contract components of the Senegal study).

Viet Nam, People’s Republic of China and Nicaragua case studies: fruit and vegetables

In Viet Nam, households with limited assets had the capacity to sell vegetables under contract to supermarkets, directly to consumers or to spot markets. Despite the small farm size, these households were found to be able to meet food quality and safety requirements and produce high-value products.⁴⁴

Food safety and quality requirements for fruit and vegetables by supermarkets and exporters give rise to increased vertical coordination and contract farming. In the People’s Republic of China, contract farming of apples and green onions was found to result in a 22 percent increase in the average income of apple growers and a 45 percent increase in that of green onion growers.⁴⁵ For labour-intensive products such as fruit and vegetables, participation was found to depend on the availability of family labour, rather than on farm size. For apple growers, income increases were due to higher yields as a result of technical advice and inputs provided through the contract. For green onion farmers, higher prices more than offset the input costs per unit, also resulting in increased incomes (see Table 3.1 for more details on the contract components). Higher income allowed these households to spend more on schooling, healthcare, food consumption and house improvements.

Contracts could have lasting effects on farmers’ livelihoods. In Nicaragua, participation in high-value fresh vegetable supermarket chains could result in a 16 percent increase in a household’s productive assets, such as tractors, plows and irrigation pumps over a period of 2.5 years.⁴⁶ Households located near a road and with access to irrigation water – factors that allow a steady supply of produce throughout the year – were found to be more likely to participate in these contract schemes. This increase in assets was the result of better access to credit and of the predetermined minimum prices that, as part of the contract, reduced farmers’ exposure to risk and promoted investments (see Table 3.1 for the contract details of the Nicaragua study). As assets determine productivity, contract arrangements that guarantee minimum prices could have long-term effects on household income and, thus, on poverty reduction.

Indeed, contract farming is often seen as addressing insurance market failures. In Madagascar, contracts where a guaranteed fixed price was offered to farmers were also associated with a decrease in average household income variability, as price risk was transferred to the purchasing firms.⁴⁷

Viet Nam and Senegal case studies: dairy

Milk is becoming an increasingly popular high-value food item, leading to high growth rates in the dairy sector in Viet Nam, and more generally, in Asia. In such value chains, product quality is an important factor in determining participation and farm-gate prices. When quality attributes are not observable, such as in milk, and individual testing is excessively costly, information asymmetries can lead to inferior market outcomes.⁴⁸ For example, purchasing firms may underreport quality levels to farmers to reduce the price that they have to pay. As a result, farmers may limit investment, thus negatively affecting farm productivity (see also Box 3.5 for an innovative solution to this information asymmetry problem).

A study assessed the impact of milk testing and quality verification by an independent third party agency on the behaviour of randomly selected smallholder dairy farmers in Viet Nam who were contracted by a dairy company. The addition of such a contract-enforcing institution to a rapidly grown market was shown to have a positive impact on input use (such as feed), quality and output levels (quantity of milk fat and total solids). This led to higher revenue and, on average, higher household welfare.

In Senegal, in another contract farming arrangement for milk, innovative contract design focused on building trust and social capital between the buyer and farmer. In order to secure adequate milk supplies from a large number of semi-nomadic farmers, a dairy firm provided a nutrition-based incentive that could improve children’s nutritional status as a component of the contract and in order to increase milk deliveries. This incentive – the daily provision of a micro-nutrient fortified yogurt for each young child in the household – compensated farmers for consistent milk deliveries. This innovative approach showed significant impacts on the frequency and amount of milk delivered, especially during the dry season, when it is difficult to meet delivery requirements compared to the rainy season. These effects were shown to be more significant when women managed contracts. The impact on total milk delivered in both the dry and rainy season is large for female-headed households, representing an increase of 64 percent in the dry season and 33 percent in the rainy season, thus underlining that the empowerment of women significantly improves nutrition and well-being for the entire household.⁴⁹

Benin case study: staple food value chains

Unlike fruit and vegetables, staples have limited possibilities for increasing value added and upgrading quality through contracts. Staple foods are not perishable like vegetables and can be stored and transported easily. However, increasing the efficiency of staples’ value chains through contract farming is important for developing countries. First, it could benefit a significant number of smallholder farmers and, second, it could contribute to strengthening access to food by a growing urban population thus promoting food security.

Although most contract farming analyses focus on high-value products, in Benin rice contract farming is found to have significant effects on household income, yields and farm-gate prices. Benin’s rice sector competes with imports but is characterized by low value added and low quality. Studies show that membership in an organized farmers’ group is important for participation in the rice contract scheme, together with the household size and the education level of the household head. Farm size and assets are found not to influence participation.⁵⁰

Improvements in quality due to better sorting that increased purity levels resulted in increases in farm-gate prices by 11 percent compared with average prices. Contract farming also brought improvements in yields through better access to inputs and an expansion of the area under rice, resulting in an average 60 percent increase in production. The effect of participation in contract farming on household income was estimated at an increase of 17 percent (see Table 3.1).

Nevertheless, contract farming for staple foods may be limited. The rice market in Benin is still small compared to staple markets across developing countries that are characterized by many farmers and traders. Rice production requires specific investments in levelling, flooding and draining the fields which, in conjunction with more possibilities for quality differentiation, can render contract farming possible.

Benefits of contract farming

In general, the evidence on the positive effects of contract farming on welfare is overwhelming, at least within the local contexts of the studies. However, impacts can be highly heterogeneous both across different contract schemes and between farmers participating in a particular scheme. Analysis of various high-value contract schemes in India revealed that, in some cases, participation resulted in significant increases in net profits per hectare while, in other cases, it had a negative effect on profitability per hectare (see Table 3.1 for more details in the various contract arrangements included in the India study).⁵¹

Indeed, in developing countries, the evidence suggests that participation in markets and contract farming is subject to reversals. Contract schemes collapse frequently, and there is a high rate of exit, as farmers move in and out of contracts. If markets are to contribute to development, sustained participation is necessary; the positive effects of contract farming on farmers will be larger if participation is continuous, as investments in productive assets, technologies and knowledge take time to generate benefits. This highlights the need to carefully analyse the contractual terms and arrangements against the effects on farmers’ welfare to understand better the dynamics of sustained participation in such markets.⁵²

A comprehensive review of the evidence on the income effects of contract farming suggests that participation increases farm income by an average of 63 percent. Out of the 26 contract farming schemes analysed, only two were found to have negative effects. This finding underlines the positive impact of contract farming on welfare but, at the same time, masks the heterogeneous effects contract farming might have.⁴² Similar conclusions are drawn from the analysis of empirical studies in this report.

Although all studies reviewed used statistical techniques that properly identified the causal impact of contract farming on welfare, these income effects may be overestimated. First, it is likely that non-significant income effects are not reported, as scientific articles are more likely to be published if they found a significant effect (this is called publication bias). Second, most of the studies may neglect contract farming schemes that failed as well as farmers’ exits from contracts (this is called the survivor bias). Both sources of bias may result in an overestimation of the income effects.

Participation in contract farming schemes is also subject to spillover and trade-off effects. For example, the higher labour requirements of contract farming can affect employment off-farm. A study, using data from Madagascar, suggests that contract farming is associated with a 79 percent decline in household income per capita derived from labour markets and a 47 percent decrease in the income generated by non-farm businesses. This is a consequence of increased specialization in the production that is necessary to meet contract requirements. It may also indicate that contract farming is more profitable relative to non-farm employment. This relationship between earnings from contract farming and earnings from the labour market could potentially explain why farmers move in and out of contracts frequently. At the same time, there are positive spillover effects, as knowledge and technology obtained by participation in a contract can affect non-contracted crops. Such technological spillovers could result in a 51 percent increase in agricultural income derived from non-contracted crops.⁵³

Despite its weaknesses, the analysis of contract farming participation can provide valuable insights into the effects that different forms of contracts and service provision may have in obviating market failures. Secure access to markets, the provision of inputs and credit, price premia that reward quality, predetermined farm-gate prices, extension services and technical advice form a complex service structure that addresses specific constraints and risks faced by farmers in developing countries.

Although more research is necessary, the evidence suggests that price premia, when combined with input provision and credit, have an important positive income effect in the context of annual crops. While predetermined prices address price risk for all contracted crops, the effect of price premia can be especially significant in the context of remunerative markets and global value chains for differentiated and certified products. Extension services and transport provision, when included in the contract, also have a strong impact on income, highlighting the importance of improved technologies and better transport infrastructure on market participation.⁴²

Various types of coordination mechanisms can simultaneously address different market failures faced by farmers in developing countries. Many innovative business models are designed to address multiple market failures simultaneously through “bundling” inputs and services together.

In development and poverty reduction programmes, where the objective is to promote self-employment, empirical evidence suggests that combined interventions may be needed to achieve a significant and persistent impact on a large part of the beneficiaries. Different actions simultaneously targeting the poor over a limited period, such as the transfer of a productive asset with consumption support, technical skills training, coaching, access to savings and health education, can complement each other in supporting households to improve their livelihoods.⁵⁴

Such a comprehensive approach could be effective in the context of multiple market failures that vary widely in terms of severity and across space. In agriculture, bundling inputs and services may perform better than standalone provision.

For example, linking the provision of modern inputs with insurance can lead to relatively higher productivity and income increases compared with facilitating the provision of technology and insurance separately. Investing in improved seeds is perceived as risky, as in the event of a drought, farmers may lose their investment. Using traditional low-quality inputs is preferable under uncertainty, especially for subsistence farmers for whom the additional cost of modern technology would make up for a significant share of their income. However, contract farming that links improved seeds to insurance could increase farmers’ demand for technology by reducing their exposure to risk. In Kenya, bundling crop insurance with improved seeds is found to increase investments on-farm, including on land and inputs such as fertilizers and machinery.⁵⁵

Innovative business models can also decrease the costs for buyers of contracting with smallholder farmers. Another set of innovations increases benefits to both parties through introducing product differentiation in terms of quality and other characteristics; these innovations can change the amount and nature of risks involved and can also provide access to niche and more remunerative markets.

Many attributes of these business models are not new, especially when they are considered on their own. Innovations are tailored in a way so that models address multiple market failures simultaneously to include smallholder farmers in value chains.

Bundling inputs and services with insurance to address production risk

There is a range of options for contract farming to include production insurance directly. Especially in developing countries, where farmers are characterized by low collateral, standalone insurance contracts can have minimal impact on the adoption of new technologies. In contrast, insurance interlinked with credit can be much more effective at promoting technological change.⁵⁶ Firms purchasing from farmers through contracts can better bundle credit and production insurance to farmers. This is because the contracting relationship itself and the related services provide additional means of enforcing the lending contract (see Box 3.3 for an example of bundling insurance).

The effects of bundling inputs and services through contracts on farm sales and income

Contract farming agreements that guarantee a minimum price can provide farmers with a measure of price insurance, thus creating powerful incentives for investment. Often, either traditional domestic markets for a contracted commodity are thin (as can be the case with rainfed horticulture), or the international price exhibits high volatility and extended periods in which prices are depressed (as in the coffee and cocoa markets). Therefore, contracts that include predetermined fixed prices can reduce farm income variability and promote investments.

For the purchasing firm, one important challenge associated with these contracts is the risk that farmers may opt to sell contracted production to buyers outside of the contract – a practice known as side-selling. For example, farmers may deviate and side-sell when market prices sufficiently exceed the contracted price, assessing that gains from such one-time defections exceed the longer-term benefits of adhering to the contract.⁵⁸

Contracts featuring protection from price volatility are likely to be most sustainable and prosperous, especially when farmers are risk-averse and value less exposure to price risk. For example, farmers in Nicaragua contracting with Walmart proved to be willing to accept a contract that featured an average contract price lower than the average price in the traditional market.⁶⁰

Moreover, price guarantees through contracts have been shown to induce investments in production. Bundling inputs and services together with a predetermined price can provide additional benefits, especially in terms of increased market participation. For example, researchers working with a rice processor in Benin (see Box 3.4) found that a contract guaranteeing a producer pre-determined price showed production impacts similar to contracts that also included the provision of extension services and input loans. Nevertheless, contracts that included only predetermined prices had a lower impact on the share of household production that was marketed compared with contracts that bundled predetermined prices with inputs and services.

Innovations in product quality differentiation

The large number of growers, intermediaries and traders involved in agricultural markets in developing countries makes it difficult for information on product quality to pass through the value chain. Potential product quality premiums are rare and, given the number of transactions and the scope of sourcing across many farmers and locations, it is difficult for quality signals and product differentiation based on brand or reputation to be transmitted through markets.

Quality heterogeneity can impede smallholder farmer market participation and also make household autarky more likely.⁶¹ Where contract farming does incorporate quality-based price premia (differential prices based on quality grades), this grading can provoke complaints from farmers about opportunistic product devaluation on the part of the purchasing firm to manipulate and reduce the contracted prices. This information asymmetry between buyers and sellers on quality grading can lead to chronic underinvestment in production by farmers, which in turn can adversely impact product quality and market participation.⁵⁸

Innovations in quality differentiation in contract farming can help “de-commodify” smallholder agriculture, i.e. move away from single-grade bulk production to graded-scale production. Coffee provides an example of an internationally traded commodity that is grown by millions of farmers in Africa, Latin America and Asia and is also characterized by low and volatile prices. At the retail level, coffee has become an increasingly differentiated product catering to a rising population of sophisticated consumers.

This quality differentiation creates opportunities for participants in the value chain to benefit from emerging price differences. Nevertheless, a quality-based model must provide additional returns and risk-mitigation to farmers through contracts that are long-term and that set fixed prices, quantity guarantees for multiple quality grades and transparent payment mechanisms (see Box 3.5).

Increases in farm productivity and commercialization can raise incomes and improve livelihoods but may also lead to undesirable outcomes in the context of the social and environmental dimensions of sustainable development. For example, modern value chains could exclude women farmers or those with small landholdings, resulting in a range of inequalities and lack of opportunities to integrate into the development process. Increasing pressure by markets to leverage economies of scale could further marginalize smallholder farmers, potentially creating social challenges.

There are also concerns that increased crop production for exports resulting from trade openness and globalization is the leading driver of deforestation (see Part 2). It is estimated that, in Latin America, commercial agriculture accounted for almost 70 percent of the deforestation in the period 2000–2010.⁶³ The loss of forests increases carbon emissions that contribute to climate change, as trees can store large amounts of carbon, but also reduces biodiversity, eradicating the natural habitat of fauna and flora.

Economists, in general, commend the market economy for creating incentives for people to deliver products and services, generating wealth and spurring economic growth. However, they recognize that in some instances markets may fail to reconcile the interests of individuals with those of society as a whole. Markets may result in negative environmental outcomes or may fail to address social objectives, such as reducing inequality.

Such environmental and social impacts are “external” to the market and not accounted for in the prices of agricultural products. To align markets with collective interest and social well-being, it is necessary to support the market economy with institutions. Governments commonly use direct regulation, as well as taxes and subsidies, so that markets account for costs that would otherwise not be included.

For example, some governments levy taxes on pesticides to “internalize” their environmental cost to society and to reduce their use or subsidize climate-smart agriculture practices. Across the world, social protection systems are established to address inequality. At the same time, institutional arrangements such as sustainability certification schemes can harness the market mechanism to produce public goods and sustainable outcomes.

Sustainability certification schemes and standards

Sustainability standards are gaining importance in global markets, especially for high-value products with established links to global value chains. They are often seen as strengthening the link between smallholder farmers in developing countries and affluent consumers in industrialized countries (see Part 2 for a discussion on the growing demand for sustainability certified products).⁶⁴ For farmers, higher and more stable prices for certified products and better market access provide the incentive to adopt sustainability standards, comply with standard-specific rules in production, and undergo regular inspections by independent certification agencies, such as FLOCERT for Fairtrade certification or the International Federation of Organic Agriculture Movements for organic certification. Often, higher prices compensate for the increased costs of production and farm management that are necessary to comply with the standards.

KENYA
An African woman collecting coffee berries from a coffee plant.

©iStock.com/Bartosz Hadyniak

Sustainability certification schemes have various objectives. For example, organic standards provide incentives to produce crops without synthetic fertilizers and pesticides; fair trade standards aim at improving market access and prices for smallholder farmers in developing countries. Other schemes include a range of requirements for environmentally friendly farm practices to promote agroecological management, such as agroforestry, the use of organic fertilizers and pesticides, and safer treatment and disposal of waste.

Some certification schemes include social rules aiming at improving the working and living conditions of farmers and workers in developing countries.⁶⁵ These rules relate to the safety and health of workers, social rights such as remuneration equal to or above the minimum wage, rights to education for children, and child labour policies. Other certification programmes include requirements for establishing effective producer or worker organizations in an attempt to strengthen the bargaining power of farmers (see Figure 3.9 for more information on the requirements of selected sustainability certification schemes).

Figure 3.9

SELECTED SUSTAINABILITY CERTIFICATION SCHEMES: STANDARDS AND POTENTIAL OUTCOMES

Back

Compliance with sustainability certification schemes often implies significant trade-offs. For example, organic or other environmental provisions tend to increase production costs, which farmers may not always be able to pass down to consumers. Sustainability certification can also exclude the most disadvantaged farmers when they cannot meet the requirements set by the standards.

Environmental outcomes of sustainability certification schemes

In general, sustainability certification schemes are found to improve environmental practices. For example, in Brazil, Colombia, Costa Rica, Guatemala and Mexico, standards set by a multinational corporation were seen to improve the environmental conduct of certified smallholder coffee producers compared with their non-certified counterparts.⁶⁶ This positive relationship between certification and environmental benefits was shown to be stronger if farmers were organized in cooperatives rather than if they were selling directly to private intermediaries such as traders and coffee roasters.

The institutional structure of the value chain plays an important role in how sustainability certification shapes economic, environmental and social outcomes, as different intermediaries may transmit different signals to farmers on the applied standards.⁶⁷ Often, farmers’ groups or cooperatives are seen as better placed to provide technical support and managerial advice to certified farmers.

In Costa Rica, organic standards contributed to reducing the use of fertilizers, pesticides and herbicides and to increasing the use of organic fertilizers among certified coffee farmers. Nevertheless, the analysis suggested that although standards can have significant environmental benefits, they are likely to entail high costs for farmers that should be compensated by increased price premia.⁶⁸

Shade-grown coffee can support multiple ecosystem services, such as climate-change adaptation, pest control by birds, and the production of food and other products of economic value by shade trees. In Ethiopia, shade-grown coffee Rainforest Alliance certification programmes effectively alleviated forest degradation.⁷² Adequate incentives – certified farmers received 15 to 20 percent higher prices for their coffee compared to the market – combined with a high standard of criteria for certification and monitoring increased the density of the certified forest coffee areas compared with those without the certification.

Economic outcomes of sustainability certification schemes

Improved welfare and incomes for smallholder farmers is one of the main objectives of many sustainability certification schemes. But the exclusion of the most disadvantaged smallholder farmers is a risk associated with many of these schemes.

Support to farmers to adopt and continue complying with stringent standards is important. In the case of vegetable producers in Thailand, the analysis suggests that for cooperative-led farmers, the costs of compliance with GlobalGAP standards make continued certification possible only for larger farmers, especially after the withdrawal of donors’ support. Receiving support from an exporter was shown to help overcome the initial costs of standards adoption and to increase the likelihood of being re-certified by 85 percent. Establishing farmers’ groups and long-term partnerships among actors along the value chains as well as with development agencies and non-governmental organizations is a crucial factor in including smallholder farmers in high-value certified product markets.

In Uganda, research indicates that the economic benefits of sustainability certification for coffee can only partly offset the compliance costs.⁷⁴ By establishing rural producer organizations, farmers can leverage technical support from non-governmental organizations to obtain group certification, as well as increase the volumes of certified coffee that is delivered.

In Côte d’Ivoire, cocoa farmers’ cooperatives are central to helping their members to comply with Fairtrade standards. Fairtrade aims at improving smallholder livelihoods and promotes collective action among farmers. Certification is awarded to cooperatives and offers minimum guaranteed prices for certified products, as well as a Fairtrade premium to provide cooperatives with technical advice and inputs.⁷⁵ The evidence suggests that Fairtrade certification increases the yield of certified farmers by an average of 13 percent compared with their non-certified counterparts, and a gain of 4 percent in the price received. On average, per capita consumption expenditure for certified farmers is 20 percent higher than that of uncertified producers.⁷⁶

The characteristics of cooperatives are also found to affect the likelihood of certification and to shape farmers’ productivity and income. In the case of cocoa in Côte d’Ivoire, while cooperatives with higher assets and better service provision were more likely to be certified, Fairtrade certification largely increased the income of cocoa farmers who were members of cooperatives that were less well-endowed. This suggests that the Fairtrade premium that targets support to cooperatives adds to their capacity to provide technical advice and inputs.

As economic growth, urbanization and rising living standards transform consumers’ preferences in developing countries, domestic certification schemes become increasingly popular, as they provide information on both the quality and safety of food to consumers. In Viet Nam, rapid supermarket penetration in domestic markets has promoted the use of domestic certification, such as VietGAP, that creates opportunities for smallholder farmers to enter high-value and differentiated products markets. Often, domestic standards may be less stringent than international ones – for example, VietGAP recommends using integrated pest management practices (IPM), while for GlobalGAP certification IPM is essential.

In Thai Nguyen province in the northeast of Viet Nam, VietGAP-complying green tea farmers, both individually and organized in cooperatives, were shown to increase their access to lucrative domestic value chains and to receive prices 11–20 percent higher than those paid for non-certified tea. At the same time, as certified farms used more labour to comply with the standards, their labour costs were found to be twice those of non-certified farms. Despite the increased costs of production, the net income of certified farms was estimated to be 30 percent higher compared to non-certified farms.⁷⁷

Generally, smallholder farmers’ integration in sustainability certified products value chains is found to generate economic benefits. However, recent reviews synthesizing the evidence suggest mixed results on the impact of sustainability certification on sales revenue, farm income and agricultural wages.^65,78 Such differences across studies can be attributed to context-specific factors that are often ignored or not fully captured by the analyses, but also on the range of requirements and service provision across various certification schemes.

Other studies suggest that smallholder farmers’ participation in sustainability certification schemes may improve welfare in the short term, but in the longer term, the evidence is mixed, and for some households integrating into the labour market provides a way out of poverty.⁸⁰ Although sustainability certification schemes are not the only pathway to sustainable growth, they are generally seen as providing a structured system to achieve and document improvements through clearly defined rules, indicators and mechanisms.

Social outcomes of sustainability certification schemes: education and gender

Many sustainability certification schemes include specific requirements that adhere to social principles. For example, Fairtrade requires that certified farmer organizations promote non-discrimination, ensure workers’ health and occupational safety, and ban child labour. Such schemes can encourage investments in the education of children. For example, data from smallholder coffee farmers in Uganda suggests that Fairtrade certified households spend 146 percent more on child education and keep children at school longer than non-certified households. For many certification schemes, income from cash crops is often earmarked for larger investments, such as education, thus directly contributing to child education.⁸¹

Investment in children’s education generally tends to increase with income, but the decisions of households on education can be complex and shaped by a range of factors. Many studies find mixed evidence, but, overall, a positive relationship between participation in certified product value chains and schooling can be ascertained.⁶⁵ For example, in Oaxaca and Chiapas in rural southern Mexico, households’ participation in Fairtrade–Organic certified cooperatives was found to increase schooling more for girls than boys. While for girls between 16 and 25 years of age, schooling was estimated to increase by 0.7 years; for boys, the impact was weaker, probably due to better rural labour market opportunities for males.⁸²

Certification schemes can also affect household members in different ways, depending on their roles in crop production, control over income and decision-making power. Often, the certified crops are traditional cash crops, over which men have more control. When certification increases the profitability of traditional cash crops, existing gender roles and inequalities might be reinforced or exacerbated.⁶⁵ In general, in farm households, commercialization may change gender roles, resulting in a lower share of the income being controlled by women.

Some certification schemes, such as Fairtrade and UTZ, involve specific gender and non-discrimination policies that might help promote women’s status and reduce prevalent gender disparities in access to information, inputs and services. For example, some standards require farmer organizations to encourage and document female participation in regular agricultural training, to organize workshops that raise awareness on gender issues, or to offer services that specifically target disadvantaged groups such as women.

An analysis of certified coffee-producing households in Uganda suggests that standards aiming to promote gender equity were successful in integrating women in the certified coffee value chain. The results indicate that in a certified household the probability that a man controls the revenues from coffee sales is reduced significantly, compared with a non-certified farm. This may be due to the certification scheme’s gender mainstreaming activities but also due to increases in labour by female household members. As quality standards increase the demand for labour and women work more, they increase women’s bargaining power and their influence on decision-making.⁸³

Sustainability certification can bring additional non-tangible social benefits. Fairtrade standards for wage labour include provisions for distributing the premium, facilitating freedom of expression, ensuring safe labour practices, and providing arrangements for collective bargaining on safe, decent and equitable working conditions. A study was conducted to survey the Fairtrade certification and well-being of wageworkers on banana plantations in the Dominican Republic, where banana production directly provides jobs to an estimated 32 000 workers. The banana is one of the most traded tropical commodities in the world, and only an estimated 5–8 percent is Fairtrade-certified or covered under another sustainability standard. Overall, the study found positive effects on the labour force, particularly by delivering in-kind benefits, offering a sense of job security, improving workers’ voice and enabling private savings.⁸⁴

PART 4 looks at how digital technology can make agricultural and food markets more efficient and more inclusive. The analysis investigates the digital divide in agriculture across and within countries and focuses on how digital technology can address market failures. A range of different applications are examined, from text messages relaying information on prices to complex e-commerce platforms that integrate farmers into markets and the use of blockchain on value chains. The discussion brings together the benefits of digital technology in contributing towards all dimensions of sustainable development, while addressing its risks and the need for policies and regulatory frameworks.

Digital technologies are rapidly transforming our economies and societies. Their adoption is driving down information and transaction costs, improving efficiency, creating new jobs, generating new income streams and saving resources. At the same time, they can be disruptive, modifying or displacing activities and products. Digital technologies can help agriculture meet the global challenges it faces. These include increasing the production of sufficient, safe and nutritious food for a growing population to ensure food security; generating jobs, improving incomes, reducing poverty and promoting rural economic growth; and sustainably managing natural resources.

Some digital technologies accelerate the evolution of agricultural and food value chains. Other technologies significantly affect the contribution of labour, capital and other inputs to the production, processing and marketing of food. Thus, the adoption of digital technologies can result in changes in relative prices, disrupting markets.

Sensors, satellites, robots and drones are examples of digital technologies that can revolutionize farming and value chains. Sensors and satellites provide information on soil conditions, weather and temperature, or crop growth. They enable farmers to achieve better yields by optimizing farm management, reducing the use of fertilizers, pesticides and water, and also contributing to better and more sustainable outcomes. The Internet of Things that connects robots, drones and vehicles to the internet can make labour-intensive tasks, such as monitoring plant health or sowing crops, more cost-effective.

These technologies also generate large amounts of data that can be combined with other information, stored and analysed to support decision-making. Such Big Data can contain high-variety information assets which can be processed by new methods of analysis, such as artificial intelligence, to assess possible outcomes based on a range of actions and conditions to help guide future interventions (see Box 4.1 for definitions of digital technologies and innovations).

Distributed ledger technology (DLT), such as blockchain, can offer many benefits downstream by providing a secure and decentralized way to perform transactions between untrusted parties along value chains. Combined with sensors that give information on the timing of delivery at each stage of the value chain as well as on the quality of the product, DLTs can disrupt vertical coordination activities, where numerous actors are involved from farm to fork.

These developments are taking place in the context of a broader evolution of global food systems; digital technology contributes to the pace of this evolution. Consumer preferences are changing – driven by economic growth, urbanization and modern lifestyles – which in turn affect markets. Consumers are demanding progressively higher-value foods, nutritional attributes and quality assurance.

Yet, a marked digital divide exists across countries, reflecting differences in access to information and technology. A digital divide is also present within countries, between rural and urban areas, between genders, and across sectors. And nowhere is the digital divide more evident than in agriculture. Commercial farms and businesses in developed countries and emerging economies already use technology intensively, while smallholder farmers in many developing countries continue to struggle to access information, markets and inputs.

Technological innovation is crucial for economic growth. Once an innovation takes place, it is common for improvements to follow and for the innovation to be also used differently than initially intended. Innovation can take time to reach the markets at scale. This is often due to the costs of technological adoption, but acceptance and familiarity also play a role, especially in the diffusion of more complex innovations.

Fixed-line telephones have been displaced by mobile phones as a mode of communication, and mobile broadband subscriptions have also significantly surpassed fixed broadband subscriptions (Figure 4.1). The speed of mobile phone technology uptake was made possible in part by the lower infrastructure costs it entailed. Communication improved significantly, and, on average, most of the world’s population now lives within range of a mobile-cellular signal, irrespective of whether or not they are subscribers or users.

Figure 4.1

Global Subscriptions to fixed and mobile telephones, and fixed and mobile broadband, 2005–2019 (per 100 people)

Back

Nevertheless, there are large discrepancies in network coverage and mobile phone ownership across countries, and these mostly reflect differences in average income per capita (Figure 4.2). Gaps between countries are lower in terms of network coverage when compared to the number of subscriptions that provide a better indication of mobile phone access. For instance, Thailand has nearly 180 subscriptions per 100 inhabitants – many people may own more than one SIM (Subscriber Identity Module) card or device, while some may not own a device at all. The latest data for Niger indicates only 40 mobile-cellular subscriptions per 100 inhabitants.⁶

Figure 4.2

MOBILE CELLULAR ACCESS IN SELECTED COUNTRIES, 2018

Back

Around 54 percent of the global population was estimated to use the internet in 2019.⁴ Internet access has spread rapidly, but gaps persist across countries and grow progressively as average income per capita decreases. Not only is access lower in least developed countries, the rate of adoption is also lower (Figure 4.3).

Figure 4.3

INDIVIDUALS USING THE INTERNET, PERCENT OF POPULATION

Back

Access to the internet in least developed countries is low, and about 19 percent of the population used the internet in 2019. In the same year in Africa only 18 percent of households had access to the internet at home. Active mobile-broadband subscriptions in Africa amounted to only 34 out of every 100 inhabitants in 2019.³

Important connectivity gaps remain between urban and rural areas, posing a challenge to farmers’ ability to adopt new technologies, innovate and participate in markets. On average, in rural Africa, only 10 percent of households have access to the internet, but these rates can be much lower for individual countries in the region.⁵ Gender imbalances also extend into the digital realm, with rural women having the least access to the internet. Worldwide, 48 percent of women have access to the internet, compared to 58 percent of men.³

Rural areas in developed countries are better connected to the internet. Denmark has the highest connectivity rate, with 97 percent of both rural men and women using the internet, and nearly no gap with respect to urban areas. In developing countries, there is a significant gap between urban and rural areas. In the Plurinational State of Bolivia, 15 percent of rural women reportedly use the internet, compared to nearly 53 percent of urban women. In Niger, only 0.6 percent of rural women use the internet (Figure 4.4).⁶

Figure 4.4

INDIVIDUALS USING THE INTERNET IN SELECTED COUNTRIES BY GENDER AND LOCATION, 2018 (PERCENT)

Back

Smartphones – mobile phones with touchscreen interface that perform a number of complex tasks like computers – were an important technological breakthrough. They make it possible for households to have access to the internet without a computer. Indeed, since 2014, more households have had access to the internet than have had a computer.³ Lowering the costs of smartphones can contribute to significantly reducing the digital divide.

Denmark and the Republic of Korea have subscription rates that surpass one smartphone per inhabitant. However, data and voice mobile-broadband subscriptions – that provide an indication of smartphone-based subscriptions and ownership – remain low in many countries (Figure 4.5).

Figure 4.5

RATIO OF DATA AND VOICE MOBILE BROADBAND SUBSCRIPTIONS OVER POPULATION FOR SELECTED COUNTRIES, 2018

Back

Access to the internet is indispensable to ensure equal access to information and services. Reducing the digital divide across countries, between urban and rural areas, and between men and women is a pressing need. Including the elderly and vulnerable groups is also needed, since they face additional constraints.

The role of governments will be significant in enabling adequate environments for innovations and further technological development.⁷ Long-established development pillars remain key to ensure that rural households will be able to take advantage of the digital revolution. Access to education and improved physical infrastructure will be indispensable to enable smallholder farmers to engage in the modern economy. A conducive environment for the digitalization of agriculture requires (i) expanding and improving infrastructure – both for ICT and otherwise; (ii) improving people’s ability to use the internet effectively so that they benefit from digitalization; and (iii) designing a regulatory framework that is both conducive to innovation and takes into account the specificities and risks digitalization entails.

The Taobao villages in China (see Box 4.3) make possible a new, innovative economic development model through e-commerce. Higher education levels, logistics and communication infrastructure were the preconditions for establishing digital business platforms that are inclusive of farmers. The villages’ novel business model sheds light on how to address challenges to regulation.

Innovative partnerships will be needed to increase digital inclusion. The successful digitalization of agri-food value chains – one which generates benefits across social, economic and environmental areas – will require public-private partnerships and multi-stakeholder cooperation.

The digitalization of food and agriculture has taken different paths in developed and developing countries. The internet made possible the creation of a plethora of technologies, some of whose potential and impacts are being witnessed at present, such as in the case of e-commerce platforms. Others, such as distributed ledger technology, have not yet been adopted at scale. A better understanding of their potential and limitations will be indispensable to ensure they make a positive contribution to the sustainable development of the sector and across economic, social and environmental objectives.

The digital divide is nowhere more evident than in agriculture. In developed countries and emerging economies, the use of technology in agriculture is advanced. The fast pace of innovation, which allows for digital technologies to collect, store and analyse data, is revolutionizing production systems and value chains. For example, precision agriculture is emerging as an innovation-driven solution in many regions and countries, such as central and northern Europe, North America, Argentina and Australia, where large farms provide economies of scale and higher returns to investments in technology.⁸

Precision agriculture methods rely on satellite positioning systems, remote sensing and the Internet of Things to manage crops and optimize the use of labour, fertilizers, pesticides and water. These methods improve efficiency but can also increase food safety, as well as reduce the negative environmental impacts of farming practices. Precision agriculture operations also generate data that can feed into Big Data and analytics, thus supporting decision-making. Such technological advances can have significant impacts on markets for agricultural labour, capital, and food and agricultural inputs.

However, in some developing countries, digital technology adoption rates are low. Often, applications are limited to text messages through mobile phones or simple offline farmer advisory digital videos that provide information to farmers in rural areas. However, a number of initiatives address specific challenges smallholders face and have produced multiple benefits (see Box 4.2).

On average, agricultural productivity is higher where countries adhere to good regulatory practices.⁹ Effective regulation can improve access to digital technologies, strengthen the coordination of actors along the food value chain, and promote productivity and income growth. Indeed, agriculture’s digital divide between developed and developing countries becomes more apparent when one focuses on the enabling environment.

A new dataset produced by the World Bank Enabling the Business of Agriculture (EBA) project allows benchmarking of regulations that promote an enabling environment for providing and using digital technology services, with a particular focus on rural areas. The EBA data covers information related to the licensing framework for mobile operators, spectrum management and infrastructure sharing.¹⁰

This EBA ICT score reveals the extent of the digital divide in agriculture across developed and developing countries (Figure 4.6). Sub-Saharan Africa and South and East Asia face significant constraints in promoting digital technologies in agriculture. None of the countries in these regions have introduced regulation that encourages competition among mobile telephony operators to enter telecommunication markets. In contrast, high-income OECD countries exhibit strong regulatory frameworks that provide incentives to the private sector to increase connectivity beyond urban centres.

Figure 4.6

ENABLING THE BUSINESS OF AGRICULTURE ICT SCORE

Back

To increase digital technology adoption rates in the rural areas of developing countries requires investments in supply-side and demand-side factors. On the supply side, rural network coverage and availability of digital applications are needed. Demand-side factors include digital skills and literacy, especially for smallholder farmers. Addressing these factors necessitates a range of public policy interventions and, most importantly, a regulatory environment that attracts private sector investments.¹¹ Effective regulation that also promotes market competition will be essential to expand broadband access and lower users’ costs across and within countries. The participation of governments in investments, through public-private partnerships, can ensure that gaps in infrastructure and access are bridged also in rural areas. Public-private partnerships will be important to provide incentives for private investment in poor developing countries.⁵

High information and transaction costs explain why agricultural markets in many developing countries are thin or missing. Improving infrastructure helps markets develop. Institutional arrangements, such as contract farming, aim to reduce costs related to searching for a trader to negotiate a deal, bargain, and reach and monitor an agreement (see Part 3 for a discussion on contract farming). Modern food value chains introduce additional costs that are often related to information about consumer preferences, especially on quality and food safety. Digital technologies can help reduce these costs and promote access to markets, addressing many of the constraints smallholders face to become part of the formal economy and value chains.⁷ For example, search costs are significantly lower in digital environments compared with the physical, analogue world, and this expands both the quality and scope of searches.

Lower search costs can significantly improve the match between buyers and sellers, such as in the context of a digital e-commerce platform, and can reduce bargaining costs while potentially adding to the bargaining power of the farmer.¹² Expanding the scope of matching through digital technologies means that buyers and sellers can agree on a contract that closely corresponds to their preferences. Facilitating exchange can affect prices, as well as price dispersion. For example, decreasing farmers’ costs of searching for traders offering the highest price could reduce price dispersion across farmers and markets. All these benefits help increase welfare.

Digital technologies also make it easier to ascertain the reputation and trustworthiness of both buyers and sellers. For example, DLTs can promote access to multiple dimensions of information on business history, including price levels, production methods, product quality and other attributes. This can facilitate both contracts and markets for differentiated and certified products that can attract price premia and generate environmental and social outcomes (see Part 3 for a discussion on sustainability certification schemes).

In addition, such low-cost exchanges can have an impact on the organization of firms and facilitate vertical integration and global value chains (see Part 2 for a discussion on global value chains). Evidence from the manufacturing industry suggests that low-cost information enables managers to understand better what is taking place at a distance, while also facilitating problem-solving by employees on the front line.¹³

Although the cost of transportation in the digital world is approximately zero – information can be replicated and disseminated easily – the role of physical distance in shaping trade costs remains significant. Digital technology allows producers and consumers everywhere in the world to access enhanced information on products; however, it is difficult to assess its impact on trade. There is little evidence of this, but some studies suggest that while trade flows decrease with distance, based on information available both online and offline, distance may matter less online.¹⁴

In sum, digital technologies hold the potential to address a range of information asymmetries in markets, improve farmers’ access and reshape value chain management.⁵ They also represent an important tool for achieving the 2030 Agenda for Sustainable Development and the Sustainable Development Goals as they can be used to promote a more productive, resilient, sustainable and transparent food system.⁷

Improving access to information

Information on prices is particularly important to farmers. Prices signal opportunities to producers, consumers and traders – such as when excess demand creates more profitable opportunities to sell. Prices reflect changing consumption preferences and contain information that enables farmers to decide what and how much to produce.

Today, mobile phones are the most widespread form of digital technology in use, and mobile phone applications that provide information on prices is the most frequent digital technology in agriculture. However, the evidence on the impact of price information is mixed.

A number of studies provide a range of estimates on the effects of disseminating price information on smallholders’ sale prices and profits. For example, in the central highlands of Peru, SMS-disseminated price information increased farmers’ sale prices by 13–14 percent, especially for crops that are perishable and for which market information is valuable.¹⁵ In Cambodia, where local rice markets can be characterized as oligopsonistic and in which farmers sell below the average wholesale price level, improved flows of information through mobile phones led to an increase of about 4–5 percent in the farm-gate price of rice.¹⁶ On the other hand, in West Bengal, in a market environment where transaction costs are high and where middlemen earn large margins, the bargaining power of potato farmers was found not to benefit from price information provided through various means including mobile phones.¹⁷

In general, most studies agree that mobile phone use reduces price volatility and improves market integration.¹⁸ In rural Niger, mobile phones helped decrease the dispersion of prices for cowpea, a perishable good, but not for millet and sorghum – both storable commodities. Although no increases in prices received by farmers were found, information reduced price volatility more in remote markets and during periods when markets were thin.¹⁹

Other studies suggest impacts of a different nature. In Colombia, information on prices shaped decisions depending on the size of the farm. Smaller farms responded by planting crops on which they received information through SMS, while larger farms used the information to search for new markets. For both small and large farms, price information did not result in higher farm-gate prices.²⁰

Similarly, in Niger, the use of mobile phones did not have any impact on quantities produced, market participation or the prices received for crops. Nevertheless, households with mobile phones were found to plant a more diverse basket of crops, particularly marginal cash crops grown by women.²⁶

Price information disseminated by mobile phones can successfully improve welfare when other market failures are not binding. This is the case, for example, when transport infrastructure is adequate to support arbitrage, output markets are competitive and related markets, such as for inputs and credit, are also functioning well.

Successful initiatives not only give price information via mobile phones, but they also combine a variety of digital technologies and tools to provide information on other market attributes, credit, farming practices and weather (Box 4.2).

Improving access to markets through e-commerce platforms

In agriculture, the use of e-commerce platforms is still in its infancy if compared to the online trade of consumer goods. The widespread use of such platforms could disrupt traditional agricultural value chains, reducing the need for the many intermediaries typically engaged across each stage of a chain or changing how these intermediaries work. Various digital e-commerce platforms have emerged to connect farmers with households or restaurants or to allow new wholesale intermediary modes for aggregating produce from many smallholder farmers and for reselling more efficiently.²⁷

In developing countries, e-commerce platforms can reduce search costs and promote efficient matching between farmers and consumers, leading to increased market access and better outcomes in terms of income and welfare. Shortening the value chain can also reduce overall transaction costs and improve price transparency, thus resolving a number of market failures. The exponential growth of the Taobao villages in the People’s Republic of China illustrates the potential of e-commerce to generate employment, income, and market participation growth. Increased participation of smallholder farmers in the digital economy through e-commerce is key for sustainable development, as it creates opportunities for marginalized groups to benefit from economic growth. About 3 000 Taobao villages have annual online sales of more than USD 1 million and also support a growing services sector (see Box 4.3).³²

At the retail level, recognizing an emerging demand for food online shopping, mainstream supermarket chains now also offer online shopping and delivery services (see Part 1 for a discussion on food retail e-commerce). During the COVID-19 pandemic, restrictions on movements to contain the spread of the virus generated a dramatic increase in the demand for online food shopping and home delivery services in some countries. Early market analysis forecasts the online grocery market to grow by 33 percent in 2020 in the United Kingdom of Great Britain and Northern Ireland, for instance.²⁸ In the People’s Republic of China estimates suggest that the share of the online market increased from 11 to 38 percent of total food retail purchases in February 2020.²⁹ As the importance of e-commerce increases worldwide, it is possible that negative effects arise such as environmental concerns related to overpackacing.

Improving access to financial services

Savings and credit facilitate on-farm investments and help farm households to accumulate assets that promote productivity, adding to food security and resilience. Low population density, poor infrastructure and lack of information on collateral increase the costs of financial services and result in missing credit and insurance markets. For a bank, the fixed costs to establish a branch in a remote and sparsely populated area are very high compared with the quantity of business it will conduct. Digital technologies lower costs and allow financial institutions to enter rural markets without establishing a costly physical presence, thereby leading to the inclusion of population groups that previously had no access to a bank.

Nevertheless, there is no clear consensus about the effects of the use of mobile banking by households. Some studies find that M-Pesa is mostly used for money transfers, especially remittances from cities to rural areas, rather than savings.³⁵ Other evidence shows that it is more common for the poor, the non-educated and women to not have an M-Pesa account, or to not save money in their accounts if they have them.³⁶

A study using data collected from 379 households in three Kenyan provinces found that M-Pesa money transfers increased market participation by 37 percent, resulting in higher household incomes.³⁷ There is also evidence that mobile phone-based transfers can increase resilience in times of distress through reduced transaction costs. For example, it was estimated that M-Pesa contributed to lifting 2 percent of Kenyans out of poverty, with households that used it being better able to mitigate negative shocks. Such impacts were found to be more pronounced in female-headed households.³⁸

Digital platforms that facilitate linkages between value chain actors can increase financial access (see Box 4.4). In Ghana, the AgroTech Smartex mobile application – designed and implemented by the Grameen Foundation – aims to strengthen linkages between farmers, extension workers, input suppliers and traders. It also facilitates access to credit through better record-keeping and monitoring. The application collects data including farmers’ profiles and farm-related information, such as crops grown, yields, inputs and past credit history. This data can serve to attract formal lenders (such as banks and microfinance institutions) and traders or to encourage suppliers to provide inputs on credit.³⁹

Improving access to insurance

Climate change is likely to increase the frequency and severity of extreme weather events, and the uncertainty that surrounds climate variability hinders investment in productive technologies, which can result in poverty traps.⁴¹ Agricultural insurance can promote on-farm investments in technologies and inputs; but it can also build resilience by facilitating the adoption of sustainable production approaches.

Innovative insurance schemes, such as weather index-based insurance, differ from traditional indemnity insurance. The latter involves high costs of administering contracts and determining crop or livestock losses with large numbers of dispersed farmers. Index-based insurance, on the other hand, provides coverage based on an index of weather conditions that are correlated with those losses; the conditions include wind speed, the temperature or rainfall during a certain period. For example, with weather index-based programmes, farmers are paid whenever rainfall or temperature is higher or lower than specific thresholds likely to cause a significant fall in crop yields.

Digital innovations in earth observation, satellite rainfall estimations and remote sensing, combined with in situ data, can support index-based insurance programmes at lower costs. Insurers do not need to make field assessments, as in the case of multi-peril crop insurance schemes, thereby reducing insurance premiums. Index-based insurance programmes can provide coverage to millions of smallholder farmers, many of whom were previously considered uninsurable.

The Agriculture and Climate Risk Enterprise (ACRE) in sub-Saharan Africa, is the largest weather index-based insurance programme in the developing world for which the farmers pay a market premium. It is also the first agricultural insurance programme worldwide to reach smallholders using mobile phone technologies (see Box 4.5).⁴²

Distributed ledger technology is a disruptive technology with potential impacts across many sectors. It currently lies at the centre of discussions on digital applications, including those related to food and agriculture. In essence, DLT is a decentralized, consensus-based record-keeping system, and its use in agri-food value chains can have a significant impact. These value chains comprise a large number of production stages within and across countries and involve many actors, including farmers, traders, processors, banks, retailers and consumers.

Currently blockchain – the best-known DLT – is used only marginally in agri-food value chains, although many pilot initiatives are underway to assess its potential (for examples, see Boxes 4.7 to 4.12). Blockchain’s impact on food and agriculture will be more evident in the years to come when its use reaches a critical scale. Box 4.6 elaborates on the origins of blockchain, its purposes, and how it functions.

BOX 4.6

UNDERSTANDING DISTRIBUTED LEDGER TECHNOLOGY

Distributed ledger technology emerged in 2008 as a supporting system for the Bitcoin cryptocurrency. It was intended as a peer-to-peer consensus-based mechanism to carry out financial transactions without the use of a bank.

DLT makes possible the creation and use of decentralized consensus-based record-keeping of any type of information. For example, blockchain functions as a ledger in which all transactions are recorded chronologically. This record exists simultaneously across all computers of transaction parties, as well as those of the record keepers (that are called nodes in the blockchain jargon) in the network.

In a blockchain, every new transaction (called a block) is linked to its predecessor (and subsequently to its successor) through a highly complex code generated automatically by an algorithm. In practice, once a transaction takes place, information is entered in the blockchain, verified by the record keepers and replicated throughout the entire network. Verification is triggered through a complex consensus mechanism, with the record keepers (nodes) assessing the new information and agreeing to this new entry (Figure 4.7).

Figure 4.7

Illustration of blockchain BASED agri-food value CHAINS

Back

Once the transaction is verified, is it difficult to change, unless the same consensus mechanism is triggered again. The immutability of blockchain is a key characteristic, without which users could easily choose alternative solutions. In addition, the verification process is decentralized with dispersed record keepers reaching consensus and is not dependent on an arbitrator or a third-party. Another important characteristic is that, often, users also take the role of record keepers.

DLTs may be permissioned, which means that one or more participants retain some control over who can join and what actions a participant can take. This can influence the function of the blockchain-enabled platform. For example, with fewer participants, there is less information to verify transactions, and, with fewer record keepers, the DLT platform moves towards a more centrally controlled mechanism that resembles other digital solutions, such as normal databases.

Permissionless platforms, however, operate so that anyone can join. By joining, users agree to the rules of the platform. This enables peer-to-peer interaction, more information on transactions and a more effective consensus. DLTs can also allow transfers of assets without the use of an intermediary.

The blockchain is pseudonymous. In the traditional banking system, the identities of transaction parties are recorded. In the blockchain, each user and record keeper has a pseudonym in the form of a unique alphanumeric address (or a public key), and the technology makes it very difficult to reveal the real identities of any given user.

To use blockchain, an individual or firm needs access to an internet connection, an internet-enabled device and blockchain software. Users can either develop their own blockchain software or join a platform that provides blockchain-based software for multiple purposes. Etherium is an example of such a platform.

The perceived advantages of blockchain are (i) peer-to-peer interactions that forego an intermediary; (ii) increased transparency, as records are available to all, at all times; (iii) enhanced traceability, since the history of transactions is recorded and immutable, and all can see it; and finally, (iv) a significant reduction in the risk of data-tampering. These attributes contribute to market efficiency by both lowering transaction costs and enhancing information. As lack of information on past transactions influences perceptions on the expected capacity of a supplier to meet expectations, DLTs can significantly facilitate entry into markets and thus increase competition.

SOURCES: Đuric’. 2019; Cong & He. 2018; Catalini & Gans. 2019.^7,43,44

Back

In agri-food value chains, blockchain technology can be of particular importance for applying “smart contracts” that are designed to self-execute once a number of predetermined conditions are met. In a smart contract, the clauses that rule the exchange of goods or services are embedded in coding, and actions (such as payment) are triggered automatically once conditions are met (such as the delivery of products). Smart contracts can significantly reduce transaction costs and increase the efficiency and transparency of transactions.

For example, exporting agricultural commodities, say grains, involves a complex web of intermediaries. These include farmers, wholesalers and buyers, but also a large number of providers of logistic services such as transport, storage, quality control, shipping, ports and customs, and trade financing, as well as contract and authentication services. At each stage of this value chain, the commodity should be stored, handled and transported in line with specific standards that set humidity, temperature and impurity thresholds.

This global value chain involves considerable transaction costs and paperwork, which can be drastically reduced by blockchain and smart contracts (see Figure 4.7). As transactions are completed at each stage of the value chain, information is sent to record keepers. This is done by the supplier, the buyer, other service providers or IoT devices, such as sensors, that can follow the merchandise and signal its position, temperature and other quality parameters. For each stage of the value chain, the record keepers verify this information. Once the transaction at each stage is completed and a consensus is achieved, a block is added to the blockchain and payments are made to suppliers and service providers through smart contracts.

Blockchain technology could fundamentally change trade practices and reduce, modify or entirely eliminate the need for a number of intermediary services along a value chain. It holds potential for both developing and developed countries. In developing countries, it is being used to tackle market failures and empower smallholder farmers (see for example Boxes 4.8, 4.9 and 4.10). In developed economies, users are pursuing increased value chain efficiency and transparency (see Boxes 4.7, 4.11 and 4.12).

The public and decentralized nature of the technology enables participants to see each other’s data entries in real time, improving information flows, efficiency and coordination. Box 4.7 presents an example of a blockchain initiative that aims at improving coordination, efficiency and transparency of the agricultural commodity trading industry.

Market access, financial inclusion and social outcomes through blockchain

Blockchain technology can be applied to address multiple market failures. Smallholder farmers often do not operate in the formal economy, meaning transactions take place in cash and are not recorded over time. Blockchain technology can help build a record of financial information, creating a history and digital identity. This record can help farmers establish a business reputation, improving their market access and also enhancing their eligibility to obtain credit by formal financial institutions (see Box 4.8).⁷

Blockchain-based solutions could reduce the number of intermediaries in the value chain, providing farmers with a more direct connection to markets and shortening the value chain. Within the blockchain, smart contracts can also help build trust and promote transparency. For example, many crops are characterized by seasonal labour needs, and informal seasonal labour markets are common in agriculture. Smart employment contracts, both immutable and public, could reduce costs and increase transparency, especially when foreign seasonal workers are concerned. In these blockchain applications, information could be made available to the employer, employee and legal authorities, such as immigration departments and welfare and social insurance programmes.⁷ Some firms are reportedly exploring the use of smart employment contracts to eradicate unfair practices in hiring workers in their value chains.⁴⁵

Smart contracts could also considerably reduce agricultural insurance costs (see Box 4.9). In the case of climate risks, for instance, weather index-based insurers could combine information from multiple sources (weather stations, satellites and sensors) with blockchain technology to both determine whether farmers should be paid and trigger the payment.

Blockchain, traceability, transparency and sustainable outcomes

Blockchain can facilitate the traceability of food throughout the value chain, allowing for the record-keeping of a product’s origin and trajectory through all stages of production, processing and distribution. Increased ability to trace products can be valuable for many purposes. First, blockchain technology can enable actors to know the stage of a product in real time, helping identify delays, irregularities and bottlenecks and improving coordination. Second, it can significantly facilitate corrective action should unsafe food items reach the market. Finally, it can respond to a growing demand by consumers for more information on the location where food is produced and on the production methods. The ability to share and guarantee this information is becoming an important factor for gaining consumers’ trust.

Food products can be accidentally contaminated throughout the value chain. For example, in 2006 in the United States of America, health officials took nearly two weeks to identify the source of an E. coli outbreak related to spinach. In another case, about three weeks were needed to identify the source of a salmonellosis outbreak connected to papayas in 2017.⁵⁰

Such periods are necessary due to the complexity of the value chain and the need for many stakeholders to verify multiple records and trace backwards each step of the chain. In both cases, the outbreaks were connected to a specific supplier; but the time involved in establishing the identity and location of the producer resulted in both loss of confidence in these products and consumers completely foregoing consumption for fear of purchasing unsafe food. Many farmers lost income despite the safety of their products. Food safety and improved traceability were the key motivations for some supermarket firms to run blockchain pilots on product value chains (see Box 4.10).

Blockchain technology also holds the potential to curb intentional adulteration of food products. High-value food items are more prone to malicious actors’ adding or substituting with a cheaper alternative. The increased transparency that accompanies blockchain technology would render it more difficult, for instance, to inflate weight or replace ingredients while remaining anonymous (see Box 4.11 for an example of blockchain applications on spices). The immutability of blockchain could also deter other intentional malpractices.

Increased traceability through blockchain would facilitate verifying the authenticity of products that are certified by sustainability certification schemes (see also Part 2 for a discussion on sustainability certification in GVCs and Part 3 for a discussion on farmers’ participation in such schemes). Sustainable standards and labelling provide information to consumers on environmental and social dimensions of production and can result in better management of natural resources and the inclusion of smallholder farmers in global markets. Better traceability can promote trust and enable consumers to modify their consumption patterns, which in turn changes incentives and allocations through markets and can foster sustainable outcomes for all. Promising blockchain solutions are also emerging to tackle biodiversity challenges (see Box 4.12).

Barriers to blockchain adoption

Despite its potential for agri-food value chains, blockchain technology has not yet been adopted at scale. Its slow diffusion and adoption should not be interpreted as a failure. Adopting blockchain technology can take many years in spite of potential productivity gains across many industries.⁴⁴ The complexity of the technology could be a deterrent to adoption; as well as the important requirements in terms of computer processing capacity, and costs related to high electricity needs. These are issues that are expected to prevent wider adoption of DLTs in the short term.

Blockchain technology is cumulative, meaning that transactions build on each other. The trustworthiness of the system depends on having many record keepers to build the consensus mechanism and verify transactions that take place. This requires significant storage and computer memory capacities. It also results in a relatively slow speed of recording transactions, since the blockchain needs to synchronize transactions across all nodes.⁶¹ The constraints on the size and amount of blocks that can be created in a given time limit the number of transactions that can take place per second on the blockchain.⁴⁴

Developing and implementing a novel blockchain solution can be costly. While the entry costs reflect investments that will produce benefits over time, in its current form the technology’s energy costs will likely continue to rise, generating negative environmental outcomes. Running costs are high due to electricity needed to validate an increasing number of transactions constantly.⁶¹

Using blockchain technology does not entail greater digital literacy than mobile phone applications, but developing a blockchain solution does require substantial technological know-how. Many blockchain pilots in the agri-food value chain are underway in developed countries. Developing countries lag behind despite the potential the technology holds for them. This is because blockchain applications require stable electricity supply, hardware power and memory, high-speed internet access, and a skilled labour force, all elements that may not be present across the developing world. Not all countries have a labour force with the skills needed to apply blockchain across the agri-food markets or other sectors of the economy. This could have implications for the digital divide across countries and between sectors.

It is expected that as the technology evolves, these barriers will be reduced. Both public and private sectors will play key roles in its evolution and application to food and agriculture.⁵² Traditional development areas, such as infrastructure and education, including digital literacy, will continue to be essential to allow for actors to benefit from the digitalization of the economy and to facilitate the uptake of blockchain technology.

At the moment, many blockchain pilots are being pursued in parallel, using various blockchain systems, mostly by the private sector. Blockchain solutions by the public sector lag far behind. This could reveal a missed opportunity to increase the efficiency of agricultural policies, such as payments for environmental services, or compliance with food safety requirements and SPS measures. To fully integrate DLTs into the agri-food value chains would require interoperability between the systems used by different agents (governments, producers and trade partners) and across countries.

While digital technology can bring significant gains, many questions remain unanswered. It is still difficult to foresee the full impact that digital applications can have on agricultural and food markets.

Digital technologies still face many constraints to adoption at scale, and they would be best used where they can provide benefits that other technologies cannot. This is, first and foremost, where market failures can be directly and effectively addressed; second, where there are significant efficiency gains to be realized for all. Third, and especially in the case of blockchain, where trust between parties is missing.⁶²

There are many questions and potential risks to address in the context of agricultural and food markets. These relate to the impacts that digital technologies could have on market participation, data issues and market power.

Risks for market participation

Digital technologies can empower all value chain actors – including smallholder farmers in developing countries – by reducing transaction costs and barriers to entry. At the same time, digital technologies can exclude from markets those smallholder farmers who cannot afford the initial costs to become part of the digital economy or who lack the skills to do so. Exclusion from the digital economy could add to the challenges smallholders already face and further undermine the smallholder farm sector and the livelihoods of millions of people in the rural areas of developing countries. The risk of exclusion from an increasingly digital economy is particularly high for illiterate smallholder farmers. While some technologies may help foster inclusion of illiterate farmers (see Box 4.2, for instance), it is indispensible to redouble efforts to reach full literacy and ensure that all have the skills to use the internet fully and effectively.

Exclusion from markets can be an unintended outcome of digital technologies. In agriculture, failure to respect contract requirements may occur for many reasons. For example, a farmer may fail to meet obligations to deliver specified quantities complying with certain quality standards due to extreme weather events, pests, diseases, or lack of credit. In this case, the immutable, public and perennial nature of blockchain may work against smallholder farmers who are more susceptible to such difficulties. This could generate a new information asymmetry, which could lead to farmers’ exclusion from markets, thus limiting livelihood opportunities. At present, it is unclear if and how agents using the blockchain could adjust to such potential problems and other specificities of smallholder farming.

The digitalization of the sector is expected to affect agricultural labour markets considerably. Automation may reduce or eliminate the need for some types of manual jobs on the farm and some intermediary services, adding to the impact of structural transformation on labour in rural areas. Emerging employment opportunities will be skewed towards the higher end of the skills base. Progressively higher skills will be needed to farm and engage effectively in agri-food value chains as technology advancements spread further. This will increase employment opportunities for high-skilled labour but risks further marginalizing low-skilled workers.

To engage in agri-food value chains effectively, both farmers and labourers will need to be capable of accessing digital technologies and to have the skills to use them. Promoting capacity building and digital literacy will be key to the workforce across all levels of agri-food value chains.

Data collection, privacy concerns and regulatory gaps

Data management is at the forefront of current concerns related to digital technology, and lack of trust surrounding data issues is a major obstacle to the digitalization of agriculture. Information in agriculture has been created, disseminated and used by farmers for centuries. Since the middle of the nineteenth century, agriculture has been driven by data – information that was collected, analysed and communicated. For example, the establishment of the United States Department of Agriculture in 1862 resulted in annual reports which, based on national surveys, disseminated information on yields, prices and new farming practices. In 1905, with the creation of the International Institute of Agriculture – FAO’s predecessor – information on worldwide production, trade and prices was made available.⁶³

Digital technology revolutionized data collection, a traditionally costly and lengthy process, but also resulted in real time-data being captured and collected by computers, smartphones, the internet and IoT devices. Everyone is generating large amounts of personal data that, under the proper legal frameworks, can be of value to the public and private sectors. All economic sectors, including food and agriculture, are becoming progressively more data-intensive.

Big Data is different from the “analogue” data that was previously collected and analysed, both in terms of volume and the potential for analysis. Analysing such data can shed light on hidden patterns, or unexpected relationships, that can support decision-making. For example, in agriculture, analysis of ten years of weather and crop data in Colombia revealed specific patterns of climatic variation impacts on rice yields. This analysis could support accurate site-specific forecasts and provide advice to farmers to change the sowing date and take advantage of optimum solar energy during the ripening stage.⁶⁴

Such climate-smart, site-specific information can deliver significant and sustainable benefits to farmers and society in general. In developed countries, the private sector, such as large suppliers of seeds and agrochemicals and agricultural machinery manufacturers, already engages in such innovative Big Data “smart farming” applications. These firms have made significant investments in digital technologies and services, leveraging economies of scale and their market shares. Through a multitude of digital technologies and devices, they collect information on farming practices and operations of their clients, as well as data on weather and soil conditions. They process and analyse these and relay the knowledge they produce back to their clients. The firms thus enhance production efficiency and, in many cases, provide wider benefits, such as preserving natural resources and minimizing the use of fertilizers or pesticides. The sale of innovative inputs and the provision of specific know-how to farmers through digital technologies also generate returns for these firms which often are protected by patents and copyright – otherwise businesses would have no incentive to engage in research and development of such technologies.

Risks associated with non-competitive behaviour

Competition is necessary to capture the benefits of markets and promote economic growth. Markets should be competitive to contribute towards resource allocations that, together with effective policies and regulation, can advance sustainable development. Digital technologies can affect competition in agricultural and food markets. More specifically, the way a blockchain is set up influences the information available to participants and can have a broad range of impacts on competitiveness.

For example, blockchain can allow access to transaction records that provide information on suppliers’ reputation and thus can facilitate competition. In addition, the use of blockchain’s decentralized consensus for verifying transactions avoids third-party control mechanisms that are labour-intensive and often have excessive market power. But blockchain, by its nature, can also be designed to ensure the secrecy of certain information.⁶⁷ This is witnessed clearly with cryptocurrencies, such as bitcoin, which can be used by people who do not want their identity to be revealed.

The decentralized nature of blockchain can improve competition through increased information but also raises new concerns for potential departures from competitive behaviour.⁴⁴ In principle, increased information, the ability to commit to price agreements through smart contracts, and a reduction in the cost of transactions and access to markets can significantly promote competition in markets. In general, this implies a lower risk of non-competitive collusive behaviour – for example, when firms agree among themselves to offer a certain price level to farmers to increase profits.

In the analogue world, information is difficult to obtain – there is information asymmetry. Firms cannot fully observe either the quantities purchased by their rivals or the prices paid. To collude, they need to communicate and agree to coordinate their behaviour in the market.⁶⁸ In the blockchain world, there is no information asymmetry, and this may facilitate collusive and other forms of non-competitive behaviour in many ways.

When applying the information available in the blockchain world to economic models of non-competitive behaviour, the analysis suggests that, in theory, blockchains can lead to tacit collusion. Tacit collusion, in this case, refers to unspoken actions by firms that hinder competitive behaviour and can affect prices or quantities, and therefore welfare. This type of collusion may resemble that of a cartel.⁴³

The information available through the blockchain makes it easier for firms to infer the behaviour of their rivals. Because firms can observe each other’s actions in real time, it is possible to follow transactions and identify when a firm deviates from competitive behaviour. This could open the possibility for a response to restore competitiveness or the opportunity to join in the non-competitive action to maximize profits – a tacit collusion.⁴³

For example, tacit collusion in the blockchain would hypothetically take place if one firm sees another offer contracts to farmers for specified quantities at a lower price and, instead of setting price levels in line with demand and supply, also proposes lower prices to farmers. The number of firms involved in the blockchain could influence such theoretical outcomes. For example, in permissionless blockchains, the number of participating firms may be large compared with permissioned ones. However, it is possible that, as the technology evolves, firms acquire the capability to process and analyse large quantities of data in the blockchain in near real time, facilitating tacit collusion.⁶⁷

Blockchain could also be deliberately programmed to facilitate collusive behaviour. The technology makes possible the creation of smart contracts (through self-executing code) that coordinate and regulate collusive behaviour of many actors, which increases colluders’ capability to accompany each other’s behaviour.⁶⁷ This could be made possible by introducing “sidechains” that store confidential information parallel to the main blockchain.

In addition, some analysts suggest that smart contracts can make such non-tacit collusive agreements more stable. Smart contracts between firms can contain clauses to automatically punish deviations from collusive behaviour, reinforcing the incentives for participants to adhere to such behaviour and furthering the stability of the collusive agreement.⁶⁷

Blockchain technology can also influence the nature and options for regulatory oversight. In the blockchain, users are pseudonymous, which makes it difficult to identify and investigate the participants. Transactions can be encoded and only visible to the parties involved.⁶⁷ On the other hand, anti-trust and competition regulators could have access to information in blockchains and thus could observe market behaviour in more detail – just like firms can infer deviation from competitive behaviour, regulators can observe behaviour that reflects tacit collusion. Legal action to combat tacit collusion in blockchains is, however, far from straightforward. It will take time to crystallize the extent to which regulators can successfully prevent or correct tacit collusion through blockchain.

Governments should focus on providing an enabling environment that encourages new entrants and facilitates the innovation and diffusion of digital technologies. In blockchain, separating the consensus-generating record keepers from the participating firms is among the options suggested to prevent collusion.⁴³ Currently, blockchain users can also assume the role of a record keeper, having access to all information available.

Auditing record keepers in the blockchain or adding regulatory record keepers can also maintain competitiveness. Some people also argue that it would be possible to programme blockchain applications to restrict information sharing, but this would come at the cost of reducing the quality of the consensus and preventing the use of smart contracts since (encrypted) data cannot be validated.⁴³ Resorting to encrypted data would also invalidate one of the key advantages for using blockchain technology, namely increased transparency.

More work is needed to understand the risks posed by digital technologies on market power, the possibility for collusive behaviour and, more broadly, the formation of digital monopolies. Governments will need to equip themselves to effectively regulate the digital economy. Building a deep understanding of the evolving technologies that will shape our future within antitrust and competition agencies is of pressing importance. This will entail investing to increase the technical competencies needed by regulatory and enforcement agents to understand blockchain technology and detect and deter collusion.

Legal frameworks will need to evolve before they become obsolete to address the risks listed above. At the same time, it is important to ensure that legislation does not undermine investment and technological innovation. Responding to these contrasting needs will be a challenge of the future.