Changing Governance Patterns in the Trade in Fresh Vegetables between Africa and the

United Kingdom

Catherine DolanSchool of Development Studies, University of East Anglia, Norwich UK

and

John HumphreyInstitute of Development Studies, Brighton, UK

Acknowledgements: The authors gratefully acknowledge financial support from

the Department for International Development of the United Kingdom

government and the co-operation of many firms in the horticulture trade in the

UK and in Kenya.

Abstract

Over the past 20 years the marketing of African fresh vegetables in the UK has

become dominated by large retailers that have adopted competitive strategies

based on quality, year-round supply and product differentiation. This has led to

a dramatic change in marketing channels, from wholesale markets to tightly-knit

supply chains. Global value chain analysis is used to explain why the various

stages of production and marketing have become much more closely integrated,

and to consider the likely outcome of a further round of restructuring occurring

at the present time. While the current trends may lead to a changing role for

importers, the tendency towards the concentration of production and processing

in Africa in the hands of a few large firms is likely to continue.

2

Introduction

The trade in fresh vegetables between Kenya and the United Kingdom has grown rapidly in

the past two decades. This growth was accompanied by drastic restructuring. Loose trading

relationships in wholesale markets were replaced by tightly-structured supply chains. Certain

aspects of this transformation, such as the development of year-round supply of fresh

produce, expansion of product ranges and development of sophisticated "cool chains" are

similar to those described by Friedland (1994). However, the Kenya-UK fresh vegetables

trade is not dominated by transnational companies of the type described by Friedland. Rather,

it is based on networks of Kenya-based producer-exporters, medium-sized UK importers and

large UK retailers. The best way to understand how these networks have developed and how

they continue to change is not to look to the literature on transnational corporations, but rather

to the literature on global value chains.

Global value chain analysis emerged initially out of a recognition of the role of global

buyers in creating global production and marketing networks. It has become an analytical

approach which explains why different types of global production and distribution networks

arise and how the activities of firms or production units are co-ordinated explicitly through

the integration of operational decisions about what is to be produced, how and when.1 This

paper analyses the Kenya-UK fresh vegetables trade from a global value chain perspective. It

presents the concept of governance in value chains and then explains why the horticulture

trade has changed and the consequences of this change for the structure of horticultural

production and processing in Africa.2 It then considers the possible outcomes of two current

trends in the UK horticulture industry: the shift from company to generic standards and the

introduction of category management by UK supermarkets.

3

Governance in global value chains

Global value chain analysis emerged initially out of a recognition of the role of global buyers

in creating global production and marketing networks. Gereffi (1994) emphasised the

importance of what he called "buyer-driven global commodity chains" in the garments

industry, observing that in some cases large retailers or brand-name companies appear to

determine how geographically dispersed production and distribution systems operate without

necessarily owning any manufacturing or distribution facilities. The UK-Kenya horticultural

trade exhibits similar characteristics. Supermarkets exercise a considerable degree of control

over what happens in the value chain,3 but only take ownership of the product when it arrives

in their distribution centres. However, other parts of the global food industry are organised in

very different ways, as is evidenced by Gibbon's (2001) work on food commodities.

Therefore, it is essential to understand why particular types of value chain governance arise.

In any production system, there are two key decisions:

1. What is to be produced. What characteristics will the product have?

2. How it is to be produced. This involves the definition of production processes, which

can include elements such as the technology to be used, quality systems, labour

standards and environmental standards.

One way of organising a value chain would be for these parameters to be determined by

each agent at each point in the chain. Thee co-ordination of the activities of the different

agents would be achieved through arm's length market relationships. Agents in the market

make decisions about what to produce, buy and sell. What is to be produced is determined by

the producer in accordance with their understanding of the requirements of potential

customers. Therefore, buyers are "design takers", purchasing ready-made products. This

requires that there is "sufficient stability of products and manufacturing practices so that both

sellers and buyers can plan their activities rationally and make rational decisions to sell and

4

buy at the prices at which the markets equilibrate" (Simon, 2000: 750). The producer also

decides how products are to be produced. The buyer is indifferent to this choice except

insofar as it results in product characteristics that affect the value of the product for the buyer.

Co-ordination through markets can be facilitated by intermediaries, who transmit information

about buyers' wants to producers and information about what producers are making to sellers

(Spulber, 1996).

This co-ordination breaks down when markets are fast-moving, or when buyers are

concerned with product characteristics which cannot easily be assessed at the time of

purchase. However, there are a variety of ways in which buyers can transmit their own

requirements to producers or obtain more information about both product and process

characteristics, while still maintaining an ability to switch easily between suppliers:

• Products can be customised to meet buyers' specifications. As long as this does not

require transaction-specific investments by either party, it can be managed through

market relationships. In this case, the buyer chooses from a supplier-defined set of

options. Buyer specification of product parameters can go further. For example, it

used to be common for firms in the US auto industry to provide detailed design

drawings for suppliers to make components (Helper, 1993). The products were made

using generic machinery, and so no transaction-specific assets were involved on either

side. This allowed the auto companies to draw upon a wide range of potential

suppliers, choosing between them predominantly on the basis of cost.

• Product grading allows the classification of products into different groups according to

the measurement of specified characteristics.4 Once again, this allows the buyer to

choose between different product categories, which might be defined by governments,

industry associations or leading firms.

5

• Product characteristics can be established through product labelling which specifies

that they meet certain characteristics. This is particularly important with regard to

safety, as would be the case for the labelling of children's toys or electrical equipment.

• Information about processes along the chain can be provided through certification.

Generic standards such as ISO 9000 (quality systems), ISO 14000 (environmental

protection procedures) and SA 8000 (social standards), and sectoral standards such as

Eurepgap5 give buyers some indication about suppliers' adherence to particular norms.

Process standards respond to two different requirements. Firstly, they can be a more

effective way of controlling product characteristics than inspection of the finished

product. The development of total quality management has driven a shift from control

over product to control over process as a means of achieving quality. The introduction

of process controls such as HACCP (Hazard Analysis and Critical Control Point)

analysis can be seen in the same light. Secondly, an increasing demand in global

markets for process controls arising from concerns about safety, labour and

environmental standards creates " credence goods", which in the context of agricultural

produce have been described as follows: "A credence good is a complex, new product

with quality and/or safety aspects that cannot be known to consumers through sensory

inspection or observation-in-consumption... The quality and safety characteristics that

constitute credence attributes include the following: (1) food safety; (2) healthier, more

nutritional foods (low-fat, low-salt, etc.); (3) authenticity; (4) production processes that

promote a safe environment and sustainable agriculture; (5) “fair trade” attributes (e.g.,

working conditions)" (Reardon et al., 2001).

These mechanisms are important for ordering markets. They communicate information

about product and process parameters between buyers and sellers and provide buyers with

assurances about what they are buying.

6

Nevertheless, it is important to recognise that in some situations buyers will choose to

specify product and process parameters through explicit co-ordination along the value chain.6

Buyer specification of product parameters arises in two circumstances. Firstly, it is required

when buyers require a high level of customised inputs. This is particularly likely to arise

when manufacturers make products with integral product architecture. As the design of the

product changes, the components also have to be redesigned. Secondly, it arises when the

buyer has a better understanding of the demands of the market than the supplier. The buyer

then interprets the needs of the market and informs the supplier of what is required.

Buyer specification of process parameters arises, firstly, when buyers may not find any

suitable standard available for regulating particular process parameters. Secondly, buyers

may not regard generic standards as sufficiently credible. Thirdly, buyers may deliberately

create their own standards as a form of product differentiation.7 If buyers create private

process standards, it follows that they must not only incur the costs of developing and

operationalising the standard, but that the value chain must incur the cost of arranging for

standards to be enforced through monitoring procedures. The benefits from increased product

characteristics or more reliable adherence with product or process standards must be enough

to outweigh these costs.

The specification of product and process parameters can also lead to increased

specification of logistics parameters along the chain. This usually occurs when there is a

degree of task complexity and/or time pressure that requires co-ordination of tasks across

firms. This is frequently an issue in the perishables sector, and the setting of product and

process parameters reinforces it, because the number of potential suppliers is restricted and

products cannot be bought from independent intermediaries.

Parameters setting can occur in many value chain relationships. Global value chain

analysis suggests that the introduction of developing country suppliers into value chains

7

creates particular needs for parameter specification. When developing country suppliers are

integrated into global value chains they are exposed to the demands of more sophisticated

markets. As Hobday has argued, the ‘latecomer’ firm to the global economy is “dislocated

from the mainstream international markets it wishes to supply” (Hobday, 1995: 34).

Frequently, the standards and regulations for these markets are quite distinct from those in the

domestic market, and increasing concern with consumer safety and ethical trade in

industrialised countries further increases the gap between developed and developing country

markets.

If specification of product, process and logistics parameters requires explicit co-ordination

of activities along the value chain, what forms might this take? Clearly, vertical integration

(hierarchy) is one possibility, but the importance of outsourcing based on closer relationships

with fewer suppliers is now widely recognised. While most attention has been focused on the

formation of supply networks within countries, Sturgeon (2001: 16-17) notes that many

formerly vertically integrated firms have outsourced many of their activities and have created

vertically disintegrated global production networks.8 Networks come in all shapes and sizes,

varying in scale, composition and in the level of co-ordination between actors. Humphrey

and Schmitz (2000) argue that two distinct types of governance should be distinguished

within networks when parameters setting is being considered. On the one hand, there are

networks that bring together firms with complementary competences. These jointly set

parameters. They refer to these as "networks", as the term network is frequently used to

denote some form of co-operation between "equals". On the other hand, there are networks

characterised by a marked asymmetry of competence and power between the lead firm and

subordinate firms within the chain. The lead firm often specifies what is to be produced, how

it is to be produced and how the performance of firms in the chain is to be monitored.

Humphrey and Schmitz refer to this form of governance as "quasi-hierarchy".

8

The development of closer relationships with fewer suppliers creates the problems of

transactional dependency and opportunism. While these problems may be ameliorated

through persistent relationships that involve repeat transactions and the development of trust

between the partners, it is important to recognise that tensions and conflicts exist even in

long-term relationships. Transactions cost analysis points to the ways in which these can be

managed. Four particular costs are associated with working with a small number of

suppliers:9

1. Co-ordination costs - the cost of transmitting information between agents in the chain

and the delays arising from such co-ordination.

2. Operations risk. One of the partners may under-perform, failing to meet all of its

commitments. As relationships become more complex, this becomes more difficult to

monitor.

3. Vulnerability to opportunism arising from transaction-specific investments. These

increase the cost of switching between suppliers or buyers, so that bargaining power is

reduced compared to the situation prior to making the investments.

4. Loss of resource control: "Resources may be generated or transferred as the result of a

relationship that is difficult to control after the fact" (Clemons et al., 1993: 16).

Technology or expertise may leak from a "partner" to suppliers or buyers.

These costs are not necessarily an obstacle to the development of closer relationships with

fewer suppliers. However, they are a factor in relationships. The firms in such relationships

have to find ways of managing them.

This section has defined what is meant by the value chain governance, specifying the

reasons why firms might wish to specify parameters along the chain, distinguishing between

different governance structures and considering the costs associated with such governance.

These concepts will now be applied to explain the evolution of value chains in the UK fresh

9

produce industry. The reasons why the value chain linking UK supermarkets and their

suppliers in Africa has evolved and why different forms of governance have emerged at

different stages in the chain will be discussed.

The horticulture value chain: 1960s – 1980s

In the late 1960s, wholesale markets traded 90% of fresh horticultural produce in the UK,

linking dispersed producers with small retailers, greengrocers and market stalls (Gray and

Kleih, 1997). When Kenya began selling "Asian" vegetables and off-season temperate

vegetables in the UK during the early 1970s, these, too, were sold through wholesale markets.

These sales channels created relatively few barriers to entry for overseas producers and

exporters. For example, Kenya's horticultural trade, which began with a small number of

Asian-owned family enterprises during the 1960s, rapidly expanded to over 100 exporters by

the mid-1980s (Jaffee, 1995: 353). While the top ten exporters accounted for the greater part

of the business, there were many smaller firms supplying French beans and Asian vegetables

to UK wholesale markets during the peak season of October to April. Some of the larger

exporters had contracts with medium-sized farms, but the majority of exporters purchased

vegetables through spot markets in rural areas (Harris, 1992; Dijkstra, 1997). The fact that

many of these businesses operated for a short period of time and with limited capital, buying

produce when margins were good and withdrawing when conditions were difficult, reinforces

the point that barriers to entry were low.

Barriers to entry for producers were also low, and by the early 1980s the participation of

smallholders in fresh vegetable production had increased markedly. Whereas the majority of

Kenya’s fresh produce exports during the mid-1970s came from 150-200 medium- or large-

scale farms, by the mid-1980s there were an estimated 15,000 smallholders involved in the

trade, growing french beans, Asian vegetables and fruit (Jaffee, 1995: 347 and 352-53).

10

Many smallholders grew crops for exporters under contract who provided them with inputs

and technical assistance, but a larger number sold through intermediaries such as brokers or

middlemen. Smallholder production of horticultural crops continued throughout the 1980s

with close to 75% of fruit and vegetables in Kenya still grown by smallholders in the early

1990s (Harris, 1992). This system served Kenya well. In 1989, Kenya accounted for over

30% of all EU imports of legumes and "other vegetables", and the trade was worth Ecu 31

million.10

When the UK supermarkets first entered the fresh vegetables trade, they too purchased

product from the wholesale market, employing wholesale agents working on a commission

basis. However, this placed certain constraints on the supermarkets. The mixing of produce

by exporters and importers meant that they had little or no information about precise product

origin and could not exercise any control over how products were produced. Further, the

supermarkets could only purchase standardised products, having no control over the type or

quality of product in wholesale markets. Finally, production could not be scheduled in

advance. Each retailer competed for the same pool of produce as it arrived in the UK.

According to one former supermarket buyer, as recently as the late 1980s supplies for product

promotions would be secured by a team of buyers arriving simultaneously at the main London

wholesale market, Covent Garden, and buying with cash as much produce as possible from

the various wholesalers. If they could not buy all they needed at once, their increased demand

would quickly raise prices.

The system of fragmented production and export, combined with the wholesale market

distribution channel provided flexibility, but it also meant that the UK supermarkets were

unable to specify product, process or logistics parameters along the chain. This began to

change as the UK supermarkets began to reorganise supply channels.

11

The transformation of the fresh vegetables value chain: 1990s

The fresh vegetables trade continued to grow in the 1990s. Between 1989 and 1997, exports

from sub-Saharan Africa to the EU grew by 151%. Kenya remained the dominant supplier,

accounting for 56% of all vegetable exports from sub-Saharan Africa.11 In this period,

however, the fresh vegetables value chain was totally transformed. This transformation

stemmed from several factors. Firstly, UK multiple stores (supermarkets and major retail

chains) greatly increased their share of total fresh fruit and vegetables sales, from 44% in

1992 to 76% in 1997, the highest level in the EU (Nagarajan et al., 1994). Secondly, the

supermarkets by-passed the wholesale markets and worked directly with UK importers,

delegating lower-profit functions such as quality control, monitoring, and distribution to their

suppliers (Marsden and Wrigley, 1996). Thirdly, there was a marked shift away from

standardised, loose product to greater product variety, product innovation and increased

packaging and processing. Fourthly, traceability was established along the chain, and

monitoring and audit regimes put in place.

What drove these changes? And what were the consequences of these changes on the

structure of the value chain, the distribution of functions within it, and the inclusion and

exclusion of different agents?

Transforming the outputs of the chain

The two main factors driving the restructuring of the fresh vegetables value chain and the

increasing role of supermarkets in explicit co-ordination of the chain were, firstly, the

competitive strategies of the supermarkets around product differentiation, and secondly, the

need to control risk in the face of a more complex regulatory and consumer environment.

For UK supermarkets, fresh produce was a key item in competition for market share in the

1990s. It was not only profitable (it had one of the highest returns per square metre of shelf

12

space, rivalled only by wine and chilled food), but it was also a “destination category” - one

of the few products that that influence a consumer’s choice of stores. In the 15 years up to the

late 1990s, supermarkets doubled the shelf area of fresh produce departments in an effort to

strengthen their image as suppliers of quality products (Burch and Goss, 1999).

They extended the range of imported produce on offer by introducing new vegetables,

more sophisticated packaging and increased levels of post-harvest processing such as

washing, trimming and chopping. A visit to any major supermarket chain in the UK reveals

innovation in product range (particularly in exotic and tropical fruit and vegetables), product

variety (vine-ripened tomatoes, for example, and increasing ranges of organic produce), food

preparation (pre-washed and chopped food), packaging (including joint packaging of

complementary foods, such as sweet corn and sugar snap peas ready for stir-fry dishes), and

the introduction of new produce to the mass market (for example, through new recipe ideas).

These products had to be supplied consistently over the year, even if this meant sourcing

produce from different countries around the world, in order to prevent shoppers from

switching to rival stores. In other words, the value chain was not only expected to supply a

physical product, but also range of services associated with it such as consistency, reliability

of delivery and the capacity to innovate.

These trends also reflected increasing consumer concerns with health and a greater

willingness to cook and eat "ethnic" foods. However, the supermarkets were not merely

following consumer preferences. They were interpreting these preferences and translating

them into particular products, as well as playing an active role in creating the supply chains

required to produce them. The supermarkets took the lead in deciding which new products

would be brought to market and how inputs and knowledge from various sources would be

mobilised for use in the value chain.12

13

UK supermarkets increasingly specified the process parameters to be followed along the

value chain. In part, this was a consequence of the drive to improve product quality. Quality,

for example, was obtained through greater emphasis on how products were grown and

harvested, and the conditions and speed under which they were transported and stored.

Product differentiation itself, and in particular the introduction of ready-to-eat vegetables,

required conformance with EU laws on food preparation, necessitating strict process controls

during processing and packing. Much of this work was carried out in Kenya.

Nevertheless, the critical driver for increased process control was the increasingly

demanding regulatory environment. In 1990 the UK government established comprehensive

standards for food hygiene and safety in the Food Safety Act.13 The Act required that

retailers demonstrate "due diligence" in the manufacture, transportation, storage and

preparation of food (Marsden and Wrigley, 1996), holding them accountable for lapses in

their suppliers’ performance. Similarly, in 1993 the EU introduced a programme to

harmonise maximum pesticide residue levels (MRLs) on food sold in the EU, authorising

governments to ‘name and shame’ retailers selling products with residues that exceeded the

legal standards (Chan and King, 2000).

These regulations were part of a broader trend towards the increasing salience of credence

factors among consumers who were not only concerned about quality and safety, but also

about the social and environmental conditions under which products were produced (Reardon

et al., 2001). Labour and environmental concerns increased in importance in the UK, not least

because of media exposés of high-profile UK companies. The fresh produce industry, in

particular, fell under the media spotlight with Tesco’s alleged exploitation of Zimbabwe farm

workers and Delmonte’s alleged violation of workers’ rights, the subject of press coverage in

Kenya (The Daily Nation, 2001). In response to consumer and NGO concerns, most leading

supermarkets developed company standards14 that extended beyond regulatory issues (e.g.

14

food safety) to include such factors as working conditions (both in the fields and the

packhouses), the use of child labour and the environmental aspects of agricultural production.

Logistics parameters were also specified. The UK supermarkets sought to organise the

flow of products through the chain so that they were transported efficiently from farm to

supermarket shelf. These specifications extended from systems for postharvest cooling and

storage of produce on farm, to the conditions at packhouses and airport handling facilities,

and finally to the processing and storage at the UK importers' own facilities. Further, the

supermarkets, together with the importers and exporters, developed systems for planning crop

production so that supplies of produce matched expected consumer demand.

Introducing and enforcing these parameters required much greater co-ordination of

activities and increased information flows along the value chain. Product parameters were

revised through discussions with importers and exporters, who had to translate product and

packaging ideas into viable production and processing systems. Co-ordination of production

schedules involved not only establishing annual supply programmes, but also involved

weekly or daily contact along the chain so that decisions could be made about what crops to

harvest, which product promotions to undertake, and what adjustments were required to meet

unforeseen events. Process parameters were enforced by regular inspection and audit and the

creation of quality systems that allowed for traceability and record-keeping and such

questions as pesticide application. In some cases, the implementation of these systems

became a competitive advantage when UK supermarkets introduced codes15 of practice to

cover the production processes of their supply chains (e.g. Nature’s Choice developed by

Tesco). By the mid-1990s all the major supermarkets had implemented company-specific

codes and introduced detailed procedures for monitoring them.

It should be noted that while UK retailers were able to make considerable progress towards

ensuring traceability and applying process controls, systems controls are sometimes breached.

15

There are some indications that UK supermarkets are willing to relax tight controls over the

production base in circumstances of product shortfall, purchasing from intermediaries and

other non-audited suppliers.

The changing structure and governance of the chain

The delivery of product innovation, quality, reliability and safety required a considerable

restructuring of the fresh vegetables value chain. The supermarkets looked for greatly

increased co-ordination and control, but had no intention of extending ownership. Therefore,

it was necessary to construct a chain which allowed co-ordination, mobilised competences

and provided an adequate incentive structure for those involved. This meant developing

tighter relationships between UK-based retailers and importers, and Kenyan-based exporters

and growers.

The relationships between actors in the chain became more complex and exclusive. The

supermarkets worked directly with a limited number of UK importers. In some respects the

relationships had the characteristics of obligational contracting (Sako, 1992). Relationships

were long-term, but rarely formalised in contracts. There were also some elements of risk

sharing. Supermarkets might agree to pay importers a fixed percentage of the final product

price, which meant that both parties would gain or lose as prices rose or fell. The importers

acquired new functions, moving beyond a “trading” role toward providing a broader range of

services and assuming a more active role in the management of the value chain. As well as

providing some UK-based processing and handling, they assumed responsibility for

developing new sources of supply, supporting developing country producers and monitoring

their performance. Their technical capabilities, in particular, became increasingly important,

with importers often working in concert with exporters to solve production problems or to

develop new products and presentation ideas.

16

However, uncertainties remained in the supermarket-importer relationship. Each

supermarket sourced from a range of importers. A typical supermarket might have six

different suppliers of imported fresh vegetables. While these would not necessarily supply

the same produce, inspections of supermarket shelves in 1999, 2000 and 2001 showed that it

was possible to find similar or identical produce from different suppliers in the same

supermarket at the same time of year. While this enabled supermarkets to spread their risks

and ensure supply continuity, it created uncertainty for the importers. Relationships between

importers and supermarkets were generally long-term, but they were sometimes terminated,

and the threat of termination could be used to drive down prices. Furthermore, the use of the

multiple importers led to a loss of resource control. According to one leading UK importer,

new product ideas had to be hidden from supermarket buyers, otherwise they would be passed

on to competitors.

Relationships between UK importers and African exporters also changed in the 1990s. In

order to develop and maintain product and process parameters, importers developed closer

relationships with exporters. Generally speaking, each exporter would sell to only one UK

importer, which would not source from more than one Kenyan exporter. This provided some

degree of mutual dependence and ensured that neither importer nor exporter were competing

with other firms for the same produce. However, the degree of transactional dependence

between the parties was not the same. While four of the five largest Kenyan exporters were

sending between 70 and 100 percent of their output to the UK, the importers were buying

from various countries in order to maintain year-round supply and spread risk.16

Nevertheless, there were long-term relationships between UK importers and Kenyan

exporters,17 and in some cases these were reinforced by equity and financial ties. The

importers wanted to ensure continuity of supply and to guarantee a return on their investments

in the capabilities of African exporters. The exporters were concerned about operations risk.

17

If the importer had various suppliers, would one supplier be preferred over another? This

prompted two of Africa’s largest exporters (one in Kenya) to establish their own UK

importers, thus providing them with dedicated distribution channels in the UK.

The product and process parameters of UK supermarkets changed the roles of exporters

and producers, forcing them to acquire a range of new capabilities to retain their UK business.

They could no longer act purely as growers or traders. Product and process innovations

depended upon sophisticated technical knowledge of production, as well as close ties with

researchers, seed companies, and importers. Increased processing, including production of

ready to eat products, entailed heavy investments in cold storage, packhouses and high-care

facilities, so that produce could be harvested, processed and transported to the UK in

hygienically and temperature-controlled conditions (Hopwood, 2001). At the same time, the

imperatives for rapid and reliable delivery placed pressure on African exporters to gain

greater control over logistics by stabilising handling and transport costs.18 These challenges

provided the impetus for two of Kenya’s largest exporters to secure joint ventures with freight

forwarders, enabling them to control more closely the post harvest process and the efficient

flow of produce through the chain.

By the end of the 1990s, the demands for capital and technical capacity had led to the

exclusion of many small exporters who were unable to meet supermarket requirements. This

exclusion was clearly evident in all the major African FFV exporting countries, but was

particularly significant in Kenya, where the top seven firms controlled over 75% of all exports

by the end of the 1990s. Those small and medium-size firms that remained in the trade were

largely dependent on arms-length marketing relationships, exporting bulk produce to

wholesale markets in Europe and the UK.

The restructuring of the chain also led to two changes in production. Firstly, production

moved away from smallholders to large farms, many of which were owned by the exporters.

18

By 1998, four of the largest exporters in Kenya sourced only 18 percent of their total produce

from smallholders. This partly stemmed from the perception of supermarkets that

smallholders could not meet process controls, such as food safety and pesticide regulations,

but also from exporters who were concerned about the costs entailed in monitoring large

numbers of small farmers (Dolan et al., 1999: 29). Those smallholders that remained in the

value chain were organised into grower schemes with a high degree of supervision by the

exporters.19 Secondly, several large Kenyan exporters began to acquire their own growing

capacity, with an increasing number centralising production on their own estates. According

to one leading UK importer, a key factor in losing a supermarket contract was the fact that its

main African supplier had no farms of its own (Dolan and Humphrey, 2000).

New tendencies in the horticultural value chain: 2000 and beyond

Thus, by the end of the 1990s, the UK supermarkets had restructured the fresh vegetable

value chain, moving away from their initial reliance on wholesale markets to tightly-knit

supply chains. However, value chains and governance structures are dynamic, and two

factors are further changing governance structures in the chain. Firstly, external standards and

independent certification procedures are becoming more important for European retailers.

Secondly, the supermarkets are seeking to a further rationalise the value chain and outsource

more of the management of the chain. To what extent will these tendencies reinforce or

undermine existing governance structures in the chain?

The shift to external process standards

By the end of the 1990s the majority of UK supermarkets had developed their own company

standards, not only defining the product and process parameters to be met along the value

chain, but also establishing systems to monitor and control compliance. Within the last few

years, however, the range of standards adopted by UK supermarkets has extended to include

19

several broad-based external standards that generally cover a range of firms and address

several different areas of risk within the supply chain.

These later generations of standards address two sources of risk for the supermarkets. The

first relates to consumer pressure. In contrast to company standards, external standards have

been formulated by a wider range of actors including trade associations, NGOs, companies

and the public sector. The multi-stakeholder nature of these standards provides supermarkets

with greater legitimacy and credibility in the eyes of consumers, and has been a key factor

behind their widespread adoption. The second source of risk is the costs involved in ensuring

compliance, which can reduce supermarket competitiveness in an industry characterised by

extremely tight margins. As Nadvi and Waltring (2001) note, if a buyer is exposed to process

risks in the supply chain, it is in their interest to ensure that the costs entailed in mitigating

those risks are also borne by its competitors. This is a strong motivation for supermarkets to

implement standards that are broadly recognised in the market, for it reduces the “spillover”

effects of their investments (Reardon et al., 2001:10). In both these cases externalising

auditing and verification through independent certification processes allows supermarkets to

relax some of their control over chain governance.

In the fresh vegetables sector, external standards have taken two main forms. The first are

sectoral codes developed by industry-wide organisations and/or trade associations. Some

sectoral codes have their origin in the North, and are being adopted by African suppliers

either voluntarily or as a requirement to supply certain buyers. The most significant standard

for suppliers of horticultural produce is the Eurepgap protocol, developed by a network of

European retailers to ensure best practice in the production and sourcing of fresh produce.

Other sectoral codes have been established through consortia of trade associations and

producers in Africa, who moved early to introduce their own benchmark standards as a way to

promote ethical production in the horticultural sector (Barrientos et al., 2001).20

20

The second are generic social codes developed through consortia of trade unions, NGOs,

companies and enterprise associations. In contrast to sectoral codes, however, these codes

provide a global social minimum standard that can be adopted across firms and countries

(Pearson and Seyfang, 2001). Examples of multi-firm social codes adopted in the

horticultural sector include the social management standard Social Accountability 8000 (SA

8000) and the Ethical Trading Initiative (ETI) Baseline code. The latter has been particularly

significant in the sourcing of African produce, as seven of the UK’s largest retailers21 have

agreed to apply the ETI Baseline code to all their own brands as well as to fresh produce,

suppliers (Barrientos et al., 2001).

In contrast to supermarket own standards, which are monitored and enforced by importers

and supermarkets themselves, sectoral and generic standards tend to be verified by agents

outside the chain. This means that supermarkets are becoming less involved in the close

supervision and auditing of exporters. To what extent might sectoral and generic standards

substitute for company standards and shift governance away from inter-firm relationships

within the chain?

While the implementation of sectoral and generic standards and their independent

verification may lower transaction costs and obviate the need for direct monitoring, they may

not lead to a change in the nature of governance in the chain. Firstly, supermarkets would

need to have confidence in the credibility of external standards and in the reliability of private

certification schemes. In the fresh produce sector, standards are relatively new, and to date

only a handful of Kenyan exporters have had an audit visit by leading international

certification bodies such as SGS or BVQI. Until there is sufficient experience of

implementation and verification, it is likely that supermarkets will regard the costs of non-

compliance high enough to merit continued involvement in monitoring. Secondly, external

standards and independent verification procedures may not cover all of the areas of food

21

safety and environmental protection over which supermarkets might wish to retain oversight.

For example, Reardon et al. (2001) have discussed how some agroindustrial firms in Latin

America have been reluctant to relinquish control over food safety despite the widespread

prevalence of generic standards. A similar trend is apparent in Kenya. While most exporters

are now implementing Eurepgap, and social codes such as SA 8000 and ETI, there are still

criteria in supermarket codes that entail further requirements. Supermarkets, or the importers,

must still monitor these activities directly.

Thirdly, supermarkets may continue with company standards and their associated

governance structures because branding and product differentiation are key to their

competitive strategy. As increasing numbers of consumers make choices on the basis of social

and environmental concerns, supermarkets recognise the benefits that codes of practice can

provide in differentiating their products. This is particularly the case for supermarket-own

codes (e.g. Tesco’s Nature’s Choice) which clearly distinguish them from their competitors.

Finally, logistics integration and product development remain important factors in value

chain governance. Even if process parameters could be controlled effectively via generic

standards, neither supermarkets nor importers would countenance a return to spot markets.

The demands for product innovation, product quality and reliability of delivery entail co-

operation between exporters and importers in resolving unforeseen contingencies and

providing solutions to technical problems. However, the way these are managed in the future

will be influenced by a second significant change in the fresh vegetables chain, the

introduction of category management.

Category management

The previous sections have described how the UK supermarkets have reduced the number of

their fresh produce suppliers and extended the range of activities performed by them. A

former employee of a large supermarket estimated that in 1987 it had 800 fresh produce

22

suppliers, compared to less than 80 by 2000. Category management takes this process a

significant step further. Although implementations of category management vary between

supermarkets, and each supermarket is still working out how to apply the system in different

areas of its product range, the basic idea is simple. Products are grouped into a number of

categories, and within each category the value chain is consolidated and a large part of its

management transferred from the supermarkets to the "category captain" or "category

manager".

The category captain is expected to play a much broader role in managing not only the

supply chain but also the marketing side of its category. Category management involves a

shift of functions from the supermarket to the category manager, taking over functions

previously exercised by the supermarket, in particular:22

• Organising the supply chain. The category captain co-ordinates the supply chain, working

with other suppliers to ensure consistent, year-round supply of produce. This involves

anticipating demand and arranging supply, sequencing production from different areas.

Previously, the supermarket would have co-ordinated the different suppliers.

• Integrating the management of the whole value chain. One of the characteristics of

category management is that the interface between the supermarket and the category

manager is broadened to include specialists in logistics, packaging, quality, finance and

product development.

• Developing the category. The category manager takes responsibility for developing the

category's profile and maximising returns. This means taking over marketing activities,

such as working with consumer focus groups, collecting data on sales trends, comparing

like-for-like sales across the retail sector and analysing returns per square metre shelf

space, and taking more responsibility for product innovation and managing the innovation

process along the value chain.

23

• Information exchange. Category managers are expected to provide extensive information

about costing margins along the chain, although this information does not flow equally in

both directions

The category management model is being combined with a consolidation of the supply

chain. Typically, the number of suppliers for each fresh produce category is being reduced

from six or more to just three: the category manager and two other suppliers. The process of

consolidation in the fresh produce sector is continuing, and a number of UK fresh produce

importers report that they are being asked to focus on supplying each major customer with a

small range of products in larger volumes.23 The fresh produce value chain will be

consolidated around a much smaller group of larger importers with considerable technical and

financial resources. Category management represents a clear shift from quasi-hierarchy to

network relationships between supermarkets and importers.

This system addresses some of the problems of opportunism evident in the 1990s. The

supermarkets will have a greater commitment to fewer suppliers. Prices should be fixed more

on the basis of examination of costs and profit margins rather than conflictual bargaining, and

the supermarkets are talking in terms of guaranteeing a viable rate of return on sales to their

suppliers. The broader interface between supermarket and category manager should improve

co-ordination along the chain. Nevertheless, it is far from clear how the new system will

work, particularly when pressures on margins in the industry remain strong.

Further along the chain, the impact of category management will be less direct.

Irrespective of the introduction of category management, producers and exporters will still be

needed to grow and process crops. However, changes will occur. Firstly, there are already

signs of increasing volatility in the relationships between exporters and importers. Exporters

do not want to be tied to the weaker category importers that may be excluded from the chain.

For example, in 2001 one of Kenya’s largest exporters switched importers in anticipation that

24

their importer would lose its supermarket business. Similarly, in 2000-01 two of Kenya’s top

seven export firms lost their main supermarket business when their importer lost its market

for legumes, including the firm referred to in the previous footnote. Exporters are also taking

steps to reduce their vulnerability in the face of these risks. Firstly, the exporters are

diversifying their markets. Data from the five leading Kenyan fresh produce exporters

showed that all but the largest of them had reduced the share of their produce going to their

main market between 1998 and 2001. Secondly, there are signs that the largest exporters are

attempting to guarantee their own position by not only strengthening their forward integration

but also diversifying their production bases, investing in, or managing, production capability

in other countries. In some cases, this enables them to provide a more diversified growing

capability for the UK importers, and hence a broader portfolio of products. In others, it is

combined with forward integration into importing businesses, which in order to survive in the

UK market must source a broad range of produce from various different countries. This

means creating integrated, transnational production, a processing and importing businesses

similar in some respects to firms such as Delmonte.

It is too early to say how category management will eventually change the fresh produce

value chain. Even the supermarkets are not sure how it will develop. The tensions between

co-ordination and economies of scale (favouring consolidation), on one hand, and maintaining

competition to generate innovation and cost reductions on the other, have yet to be played out.

While category management may create a more effective interface between the retailer and

the category manager, the relationship between the category manager and the rest of the

supply chain will still have to deal with the problems of opportunism and incentives.

At the present time, 2001, the shift to category management is making linkages within the

value chain more exclusive (fewer importers and fewer exporters supplying particular product

lines to each supermarket) and strengthening the linkages between UK importers and UK

25

supermarkets. However, in the longer term, more substantial changes in the value chain may

result from category management. Firstly, the category managers may become a new,

specialist link in value chain, managing importers on behalf of the supermarkets, but not

themselves playing the importer role. In this scenario, the provision of specialist services

around marketing and logistics is separated out from relationships with developing country

producers and exporters. This would provide the problem of how one category manager deals

with both its own supply chain and other UK importers. Secondly, the process of outsourcing

of supply chain management functions in the fresh vegetables sector by UK supermarkets

could be taken one step further. The category managers could take on even more

responsibility, becoming independent "stores within stores", acting as franchisees within the

supermarkets. This scenario is most likely if fresh produce ceases to be a "destination

category". The prospects for this happening are unclear. Although supermarkets are looking

increasingly to non-food products for growth and margins, fresh produce remains important

for attracting customers. Finally, category managers could become suppliers of packages of

services which can be customised to meet the needs of different supermarkets. In other

words, transaction-specific investments would be reduced, and supermarkets could switch

their business between different supply chains. Able to source a wide range of products to

desired to standards, the category managers would compete against each other for segments of

the supermarket business at regular intervals. The supermarket relationships with category

managers would become more arm's-length, with co-ordination limited to the customisation

of the package of services, but without supermarket monitoring of activities along the chain.

This last development would be reinforced by the introduction of credible certification

systems that could be relied upon to deliver the quality, labour and environmental standards

required in the value chain. Already, product attributes such as "fairtrade" and "organic" are

delivered by broad-based standards independent of any particular supermarket. External

26

standards such as Eurepgap and SA8000 will further extend this process. This would allow a

further loosening of linkages in the value chain, while maintaining confidence in the attributes

of products available in the UK supermarkets. The process of increasing monitoring and

control and the development of intensive interaction between agents along the chain seen in

the past 20 years is by no means irreversible.

Conclusions

The way in which supplies of fresh vegetables for the UK market are grown and processed in

Kenya and distributed to UK supermarkets has changed beyond all recognition in the past 20

years. In this paper, global value chain analysis has been used to explain the changes in the

industry, and why a distribution system originally based on spot markets has been replaced by

tightly-integrated supply chains. Value chain analysis allowed an exploration of the

consequences of these changes on the structure of the value chain, the distribution of

functions within it, and the inclusion and exclusion of different agents in the chain. The

concept of parameter setting highlighted not only how key decisions are made in the chain,

but also how some parameter setting and enforcement might be removed from inter-firm

relationships altogether by the development of sectoral and generic standards.

The driving force behind the restructuring of the value chain has been the rise to

dominance of a small number of supermarkets in UK food retailing. It has been shown that

UK supermarkets exercise a decisive influence over all stages of the value chain, from the

way crops are grown (and the processes of innovation that lead to the introduction of new

crops and varieties) to their processing and storage, despite the fact that they do not take

ownership of produce until it is delivered to their regional distribution centres. Supermarket

requirements for increased processing of products and product differentiation, combined with

27

increasing external pressure to meet food safety, environmental and labour standards has led

to a radical restructuring of the fresh vegetables business. However, the process of

restructuring will not end. Current tendencies indicate ways in which market relationships

might be reintroduced into the chain without sacrificing assurances about process controls.

The introduction of category management and the imposition of external standards may

reverse the current tendency for supermarkets to exercise ever-closer monitoring and control.

28

References

Barrientos, S., C. Dolan and A. Tallontire, 2001. The Gender Dilemma in Ethical Trade.

NRI Working Paper, Chatham: Natural Resources Institute.

Burch, D. and J. Goss, 1999. "Notes from Global Sourcing and Retail Chains: Shifting

Relationships of Production in Australian Agri-Foods." Rural Sociology, 64: 334-

350.

Chan, M.-K. and B. King, 2000. Review of the Implications of Changes in EU Pesticides

Legislation on the Production and Export of Fruits and Vegetables from Developing

Country Suppliers. Final Report Prepared for the Rural Livelihoods Department

of the Department for International Development, Chatham: Natural Resources

Institute.

Clemons, E., S. Reddi and M. Row, 1993. "The Impact of Information Technology on

the Organization of Economic Activity: The 'Move to the Middle' Hypothesis."

Journal of Management Information Systems, 10: 9-35.

Coulter, J., J. Millns and A. Tallontire, 1999. Increasing the Involvement of Smallholders

in High-Value Horticulture, Lessons from Zimbabwe. Chatham: Natural

Resources Institute.

Dijkstra, T., 1997. Trading the Fruits of the Land: Horticultural Marketing Channels in

Kenya. Aldershot: Ashgate.

Doel, C., 1996. "Marketing Development and Organisational Change: The Case of the

Food Industry." Pp. 48-67 in N. Wrigley and M. Lowe eds., Retailing,

Consumption and Capital: Towards the New Retail Geography. Harlow: Longman.

29

Dolan, C. and J. Humphrey, 2000. "Governance and Trade in Fresh Vegetables: The

Impact of UK Supermarkets on the African Horticulture Industry." Journal of

Development Studies, 37: 147-176.

Dolan, C., J. Humphrey and C. Harris-Pascal, 1999. Horticulture Commodity Chains:

The Impact of the UK Market on the African Fresh Vegetable Industry. Working

Paper, 96, Brighton: Institute of Development Studies.

Foss, K., 1996. A Transaction Cost Perspective on the Influence of Standards on Product

Development: Examples from the Fruit and Vegetable Market. DRUIDS Working

Paper, 96-9, Copenhagen: Copenhagen Business School.

Friedland, W.H., 1994. "The Global Fresh Fruit and Vegetable System: An Industrial

Organization Analysis." Pp. 173-189 in P. McMichael ed. The Global

Restructuring of Agro-Food Systems. Ithaca: Cornell University Press.

Gereffi, G., 1994. "The Organisation of Buyer-Driven Global Commodity Chains: How

U.S. Retailers Shape Overseas Production Networks." Pp. 95-122 in G. Gereffi

and M. Korzeniewicz eds., Commodity Chains and Global Capitalism. Westport:

Praeger.

Gibbon, P., 2001. "Upgrading Primary Products: A Global Value Chain Approach."

World Development, 29: 345-363.

Gray, A. and U. Kleih, 1997. The Potential for Selected Indian Horticultural Products on

the European Market. Marketing Series, 11, Chatham: Natural Resources

Institute.

Harris, S., 1992. Kenya Horticultural Subsector Survey. Nairobi: Kenya Export

Development Support Project.

30

Helper, S., 1993. "An Exit-Voice Analysis of Supplier Relations: The Case of the US

Automobile Industry." Pp. 141-160 in G. Grabher ed The Embedded Firm.

London: Routledge.

Hobday, M., 1995. Innovation in East Asia: The Challenge to Japan. Cheltenham:

Edward Elgar.

Hopwood, D., 2001, 'Cold Storage Warms to the Modern Age', Fresh Produce Journal,

January 5.

Hughes, A., 2000. "Retailers, Knowledges and Changing Commodity Networks: The

Case of the Cut Flower Trade." Geoforum, 31: 175-190.

Humphrey, J. and H. Schmitz, 2000. Governance and Upgrading: Linking Industrial

Cluster and Global Value Chain Research. IDS Working Paper, 120, Brighton:

Institute of Development Studies, University of Sussex.

Jaffee, S., 1995. "The Many Faces of Success: The Development of Kenyan Horticultural

Exports." Pp. 319-373 in S. Jaffee and J. Morton eds., Marketing Africa's High-

Value Foods. Washington D.C.: The World Bank.

Jones, C., W. Hesterly and S. Borgatti, 1997. "A General Theory of Network

Governance: Exchange Conditions and Social Mechanisms." Academy of

Management Review, 22: 911-945.

Marsden, T. and N. Wrigley, 1996. "Retailing, the Food System and the Regulatory

State." Pp. 33-47 in N. Wrigley and M. Lowe eds., Retailing, Consumption and

Capital: Towards the New Retail Geography. Harlow: Longman.

Nadvi, K. and F. Wältring, 2001. "Making Sense of Standards: An Overview of Global

Process Standards and Implications for a Research Agenda," Paper presented at

Workshop on Local Upgrading in Global Chains, Brighton, Institute of

Development Studies, February.

31

Nagarajan, G., R.L. Meyer, D.H. Graham and K. Ouattara, 1994. Exporting Fresh

Fruits and Vegetables to Europe: Potential and Constraints for Gambian

Producers. Report prepared for USAID/The Gambia, ESO 2166.

Pearson, R. and G. Seyfang, 2001. "New Hope or False Dawn? Voluntary Codes of

Conduct, Labour Regulation and Social Policy in a Globalising World." Global

Social Policy, 1: 49-78.

Reardon, T., J.-M. Codron, L. Busch, J. Bingen and C. Harris, 2001. "Global Change in

Agrifood Grades and Standards: Agribusiness Strategic Responses in Developing

Countries." International Food and Agribusiness Management Review, 2.

Reardon, T. and E. Farina, 2001. "The Rise of Private Good Quality and Safety

Standards: Illustrations from Brazil," Paper presented at International Food and

Agribusiness Association Symposium, Sydney, June.

Sako, M., 1992. Prices, Quality and Trust. Cambridge: Cambridge University Press.

Simon, H., 2000. "Public Administration in Today's World of Organisations and

Markets." Political Science and Politics, 33: 749-756.

Spulber, D., 1996. "Market Microstructure and Intermediation." Journal of Economic

Perspectives, 10: 135-152.

Sturgeon, T., 2001. "How Do We Define Value Chains and Production Networks?" IDS

Bulletin, 32: 9-18.

The Daily Nation, 2001, 'Worker's Rights Violated', The Daily Nation, November 12.

32

ENDNOTES

1 The term "explicit co-ordination" is taken from Clemons et al. (1993).

2 This paper is based on research carried out in the United Kingdom and the Kenya in two stages, in 1997-98

and in 2000-01. This involved interviews with UK supermarkets, UK importers and Kenya exporters.

3 A value chain refers to all the activities involved in making a product and taking it through to the consumer.

The term "supply chain" is used in this paper to refer to the firms involved in providing the material inputs

into a particular point in the value chain. This distinction is discussed in Sturgeon (2001).

4 The role of product grading in reducing the cost suppliers of acquiring information about products is

discussed by Foss (1996).

5 Eurep is the Euro-Retailer Produce Working Group, consisting of large European food and flower retailers,

and suppliers of fresh food. The term gap stands for good agricultural practice.

6 This explicit inter-firm co-ordination activity is termed "governance" in the global value chain literature.

7 For a discussion of company standards and how they are developed, see Nadvi and Wältring (2001).

8 For further discussion of the reasons why firms might prefer network to hierarchy even when complex co-

ordination of activities is required, see Jones et al. (1997).

9 This discussion of transactions cost is based on Clemons et al. (1993: 14-16).

10 Data from Eurostat, referring to product categories HS 0708 (peas and beans) and HS 0709 (other vegetables,

including artichokes, asparagus, mushrooms, sweet peppers, capsicum, etc.). Throughout the 1990s, Kenya

accounted for more than half of EU imports of these products from sub-Saharan Africa.

11 For source and definitions, see the previous note.

12 For example, Doel (1996) describes how Marks & Spencer created the chilled, ready meals sector in the UK,

creating not only a new product segment, but also a whole new supply industry. Similarly, Hughes (2000)

describes how UK supermarkets use designers and consumer magazines to shape consumer preferences for

flowers. Various knowledge networks are mobilised by supermarkets to create demand and satisfy it.

13 There were also various international agreements, established by institutions such as the CODEX

Alimentarius Commission, WTO Agreements on Technical Barriers to Trade and on the Application of

Sanitary and Phytosanitary Measures, the Montreal Biosafety Protocol, and the rules and regulations set

under the International Plant Protection Convention.

14 This term is taken from Reardon and Farina (2001).

15 Standards are typically expressed through instruments known as codes of practice.

33

16 One UK supermarket concentrated all of its mangetout sourcing at a particular time of year in one country

during the mid-1990s, only to find that supply dried up when that country's output was adversely affected by

the El Niño phenomenon. Further, one UK importer described how his major customer insisted that he

signed an Egyptian supplier, as the lower cost of airfreight from Egypt meant that prices for certain types of

products would be lower.

17 In 1998 most UK importers had been involved with the same Kenyan exporter for between five and ten years

(Dolan and Humphrey, 2000) .

18 Airfreight absorbs close to 50% of the total CIF export cost (Dolan and Humphrey, 2000).

19 For an analysis of such grower schemes, see Coulter et al. (1999).

20 These include the following codes: Kenya Flower Council, Fresh Produce Exporters Association of Kenya,

Zambia Export Growers Association, and Horticultural Promotion Council, Zimbabwe.

21 Supermarket members of the ETI are ASDA, The Co-Op, J Sainsbury, Marks & Spencer, Safeway,

Somerfield, and Tesco. It is not uncommon, however, for both importers and retailers to allege "off the

record" that their competitors engage in "disloyal" competition", flouting commitments to ethical standards in

the pursuit of low prices.

22 This information is taken from discussions with managers in the fresh produce sector in the UK.

23 For example, one UK importer lost the legume business of a leading supermarket, but was compensated with

increased business in other product categories. This was little compensation for its Kenyan legume supplier,

which lost 20% of its total export business overnight, forcing it to close a newly-constructed high care

facility (USD 700,000) and lay off 50% of its total packhouse employees, who were involved in value-adding

activities such as chopping and packing vegetables.