VALUE CREATION IN THE AGRI-FOOD VALUE CHAIN

BY

MARIA EMILIA CUCAGNA

THESIS

Submitted in partial fulfillment of the requirements

for the degree of Master of Science in Agricultural and Applied Economics
in the Graduate College of the
University of Illinois at Urbana-Champaign, 2014

Urbana, Illinois
a

Adviser:

Professor Peter Goldsmith

 ABSTRACT

Global agricultural markets reflect the increasing complexity of modern consumer demand for

food safety and quality. This demand has triggered changes throughout the food industry, and

led to greater opportunities for product differentiation and the potential to add value to raw

commodities (Goldsmith et al., 2002; Humphrey and Memedovic, 2006).Greater

differentiation and value adding have in turn dramatically changed the price spread or

marketing bill between the farm value of products and the retail value during the past 18 years.

Thus a significantly greater percentage of the final price paid by consumers is garnered down

chain rather than up chain. This apparent shifting of value creation, as measured by the

marketing margin, has invigorated empirical questions as to where and how much value is

created along the agri-food value chain. Using a regression model with a ten-year panel data of

financial information taken from 454 agri-food companies worldwide, this paper empirically

analyzes value creation in the agri-food value chain. The agri-food value chain can be split into

four main stages: input, production, processing and retailer. The main purpose of this study is

to analyze the differences in stages within the agri-food value chain by identifying which firms

and nodes excel or underperform in the value creation process. For the purpose of this paper,

we estimate the Economic Value Added (EVA) metric for each firm. We validate our findings

by creating and employing three additional value creation measures—the modified economic

value added, the percentage of companies that create value and the persistent value creation—

which improve the robustness of our findings and allow this study to find measures that are not

biased by company size. The results indicate that agricultural producers—the most

commoditized sector—contribute the least amount of value to the chain, while further

processing and retailing contribute significantly higher levels of value.

ii
 TABLE OF CONTENTS

INTRODUCTION.................................................................................................................... 1

LITERATURE REVIEW ....................................................................................................... 4

DATA ...................................................................................................................................... 23

EMPIRICAL MODEL .......................................................................................................... 41

RESULTS ............................................................................................................................... 44

DISCUSSION ......................................................................................................................... 53

CONCLUSION ...................................................................................................................... 58

REFERENCES....................................................................................................................... 62

FIGURES AND TABLES ..................................................................................................... 70

iii
 INTRODUCTION

Global agricultural markets have become increasingly complex due to changes in

consumer demand, the development of complex food standards mainly related to food safety

and food quality, advances of technology, and changes in the industry structure along the value

chain (Goldsmith et al., 2002; Humphrey and Memedovic, 2006). These changes have led to

greater opportunities for product differentiation and the potential for adding value to raw

materials (Coltrain et al. 2000).

In recent years, value creation in agriculture and its management has emerged as a

business survival strategy (Kampen, 2011). In the past, firms may not necessarily have needed

to focus on creating value since companies could be profitable as a consequence of controlled

distribution channels, regulated markets, the acquisition of badly performing firms, or a

scarcity of resources to produce (Doyle, 2000; Lindgreen et al., 2012). Historically, the

agribusiness sector has been a commodity-oriented industry, with strong focus on maximum

efficiency, homogenous products and economies of scale (Grunert et al, 2005). Although agri-

food markets have been efficient at converting raw commodities into homogeneous products,

the evolution of consumer demand for safe, high quality, and convenient products,

technological advances and increased competition are changing the agribusiness sector. New

agribusiness markets are characterized by companies that need to focus on value added

activities that allow them to meet the expectations of consumers by offering high quality and

differentiated products (Streeter, 1991; Alexander and Goodhue, 1999; Coltrain et al., 2000).

In response, companies’ survival relies on their ability to satisfy the end consumer demand

with differentiated and value added products while achieving high profits that capture the

benefits of the value creation process.

1
 During the past 18 years, greater differentiation and value adding along the value chain

have in turn dramatically changed the price spread or marketing bill between the farm gate and

the retail sector (Fig.1). The average farm value of food has been 16.1 percent since 1993,

while the other 83.9 percent came from post-farm activities of the value chain stages. Thus, a

significantly greater percentage of the final price paid by consumers is garnered down chain as

opposed to up chain. This change in the farm and marketing bill share suggests a dramatic shift

over time in the loci of food and agribusiness value creation.

As a result, the main objective of business has been to continuously create value

(Conner, 1991; Sirmon et al., 2007). In order to achieve this goal, understanding the concept

of value creation and the way to add value has attracted the interest of researchers and managers

during the past decade (Wilson and Jantrania, 1994; Anderson, 1995; Parasuraman, 1997;

Bowman and Ambrosini, 2000; Mittal and Sheth, 2001; Payne and Holt, 2001; Walter et al.,

2001; Anderson et al., 2006; Möller, 2006; Lepak et al., 2007; Sirmon et al., 2007; Smith and

Colgate, 2007; Ambrosini and Bowman, 2009; O' Cass and Ngo, 2011). Therefore, the idea of

“value creation” is of predominant importance in business and marketing. Previous literature

indicates that value creation is a “dynamic and multi-stage process involving different users of

value” (O'Cass and Ngo, 2011), and many studies find a variety of benefits for firms,

stockholders and customers that can be traced to value creation (Anderson and Narus, 1998;

Lindgreen and Wynstra, 2005). These benefits include an increase in customer loyalty, an

upsurge in value to stockholders and rising profits (Srivastava et al., 1998; Zeithaml, 2000;

Priem, 2007; Sáenz and Aramburu, 2010).

Although value creation is an increasingly relevant concept in agribusiness, it is perhaps

surprising that managers and even academics “often do not know how to define value, or how

to measure it” (Anderson and Narus, 1998, Lindgreen et al., 2012). Furthermore, there is no

study on a worldwide basis that has analyzed or compared the role of members of the agri-food

2
value chain in the value creation process. How can value creation be achieved in agribusiness?

Which chain members are creating the most value? How can this be measured? Are any of the

stages of the agri-food value chain “more” responsible for the value creation process?

This paper aims to answer these questions and clarify this gap in the literature. The

current study focuses on analyzing the differences in stages within the agri-food value chain

by identifying which stage nodes excel or underperform in the value creation process. For the

purpose of this paper, we estimate the Economic Value Added (EVA) metric for each firm. We

validate our findings by creating and employing three additional value creation measures—the

modified economic value added, the create value metric and the persistent value creation —

which improve the robustness of the findings by using a measure that it is not biased by

company size. Additionally, the impact of the drivers of value creation is evaluated in terms of

firm size, product differentiation, intangible assets and research and development (R&D)

expenditures on our value creation metrics. We use a regression model with a ten-year panel

data of financial information for 454 agri-food companies worldwide. We conduct an empirical

analysis within the entire agri-food value chain by examining and comparing the contribution

of each value chain member to the value creation process. Finally, this study will contribute

additional understanding of value creation in the agri-food industry and will identify the factors

that influence this process.

The paper is organized as follows: In Section 2 we present the literature review.

Description of the data is presented in Section 3 before introducing the methodological

approach used in this study in Section 4. Section 5 follows with analysis of the results and

Section 6 present a discussion of the main findings. Concluding remarks and research

limitations are discussed in Section 7.

3
 LITERATURE REVIEW

The concept of value creation

The popularity of the concept of value creation has increased since the 1990s, and its

processes have been widely studied in the business discipline (Wilson and Jantrania, 1994;

Anderson, 1995; Parasuraman, 1997; Walter et al., 2001). In this context, Coltrain et al. (2000)

suggest that the concept of value creation is widely used when analyzing the potential

profitability of agriculture. However, the majority of value creation literature fails to provide a

deep and complete definition of value creation. As a result, the concept of value creation is still

poorly understood by managers and academics. Most of the literature defines value creation in

agribusiness as “to economically add value to a product by changing its current place, time,

and form characteristics to characteristics more preferred in the marketplace” (Coltrain et al.,

2000; Anderson and Hanselka, 2009). Coltrain et al. (2000) gives a more specific definition by

giving an example of value adding in agriculture: “to economically add value to an agricultural

product (such as wheat) by processing it into a product (such as flour) desired by customers

(such as bread bakers)”. The 2002 Farm Bill (Farm Security and Rural Investment Act of 2002)

also defines value creation in agriculture as a change in the physical state of any agricultural

commodity through a production method or handling process by which the agricultural

commodity or product is produced or segregated. The end result of changing value expands the

customer base for the product or makes available a greater portion of revenue derived from the

marketing, processing, or physical segregation of the agricultural commodity or product as

achieved by the producer (Farm Security and Rural Investment Act of 2002).

Amanor-Boadu (2003) attempts to clarify the concept of value creation by using the

USDA concept and by distinguishing between the economic metric of value-added and the

descriptive concept of value-added. There are two conditions that a value creation activity must

4
satisfy in order to qualify as value adding “(1) if one is rewarded for performing an activity

that traditionally has been performed at another stage farther down the value chain, or (2) if

one is rewarded for performing an activity that has never been performed in the value chain”

(Amanor-Boadu, 2003; Evans 2006), (Table 1). This definition focuses on the reward that a

company’s output/benefit obtains for a given value creation activity. The reward may be

“higher prices, increased market share and/or increased market access,” and this reward must

“be large enough to increase the total profitability of the performing organization.” “The size

of this reward is directly proportional to the level of customer satisfaction engendered by the

activity and not by the work or effort on the part of the performing organization.” “If the total

profitability of the performing organization is not increased by the value-added activity, then

the activity cannot be deemed to have contributed any value to the value chain or to customers,

and thus it fails to qualify as a value-added activity.” (Amanor-Boadu, 2003).

Therefore, in order to create value and to achieve a sustainable business, companies

must focus on their customers’ perception of value. Customer value reflects the relationship

between the benefits customers receive from a product, and the price they pay for it. The price

of a particular product may be high, but if the associated benefits are high as well, customers

perceive the product as valuable. The marketing discipline defines product as “a bundle of

benefits, and the more benefits there are, the more customers will perceive the product as

having value” (Anderson and Hanselka, 2009). The uniqueness of a product or service (the

value you add) is what ultimately attracts customers (Anderson and Hanselka, 2009).

Although these definitions are the closest terms to what we refer to value creation, most

practitioners still lack a deeply understood definition of value added. For instance it is still not

clear how one can measure customer benefits. Furthermore, the definitions of value creation

provided by previous research are only a weak description of what value creation really means

since these definitions focus solely on the process and product levels and do not take into

5
account two essential aspects of the value creation process: the firm structure capital needed to

perform a value creation activity, and the level of efficiency of companies in using their capital

(Table 1). This study aims to fill this gap in the literature by adding clarity to the term ‘value

creation’. This manuscript strives to provide a novel and complete definition of the term ‘value

creation’ and to consistently measure value creation across the agri-food value chain and

identify which factors affect firm value creation levels.

Value chain in the agri-food industry

The value chain framework emerges as a key aspect for value creation and has been

recognized as a crucial field for business success (Matopoulos et al., 2007). A value chain is

composed of a number of stages, namely, input seller, processor, distributor, and end consumer.

It is a system of firms that work together to positively impact one another’s performance

(Bigliardi and Bottani, 2010). For the purpose of this study, following Humphrey and

Memedovic (2006), agri-food value chain is split into four main stages: inputs, production,

processing and delivery to consumers. Agri-food firms are classified into these four stages

based on a firm’s main activity (Fig. 2). A description of each stage of these activities is

outlined below.

Stage 1- Inputs

Inputs are represented by biotechnological, agro-chemical and fertilizer, animal health,

animal breeding and farm equipment companies. Firms in this sector are suppliers of

agricultural products and services such as seeds, agricultural nutrients and farm machinery.

Firms on this stage focus on farmers, who are its primary customers.

6
Stage 2- Production

This stage includes all of the activities involved in the production of raw food materials.

In particular, crop production, beef cattle, poultry, and hog and milk production are included,

among others. The customers for these firms are food manufacturers, the next node of the chain.

Stage 3- Processing

This sector is composed of food manufacturing, soft drinks producers, brewers,

wineries and packaged food companies. Companies within this node are engaged in the

production of meat, dairy and food products. Specifically, soft drink firms produce, market and

distribute nonalcoholic beverages such as sodas, energy drinks, still and sparkling waters,

juices and juice drinks, sports drinks, and ready-to-drink teas. Brewer companies are involved

in the production, marketing, and distribution of beer. Winery firms make and sell branded

alcoholic beverages such as spirits, wines and various categories of beer. Finally, packaged

food companies process, distribute and market food products. Processing firms convert raw

materials into food products, as either brand or unbranded products, which are in turn sent to

the retail market (the last stage) for distribution and sale to consumers.

Stage 4- Delivery to consumer

The main activity of this stage is the distribution, sale and marketing of food products

to the end customer, the consumer. In the last stage of the agri-food value chain we include

firms concerned with food distribution, grocery chains, and restaurants. This category contains

firms that wholesale and/or retail food products, from convenience stores to supermarket

chains, as well as hypermarkets and international distributors.

Chikán (1997) and Martin and Jagadish (2006) describe the value chain as a “series of

value adding processes which flows across many companies and creates products and services

7
which are suitable to fulfill the needs of customers” (Vanyi, 2012). The aim of each value chain

actor is to meet customer desires while collaborating in the value creation process (Vanyi,

2012). Today, as members of a value chain, companies fulfill and satisfy costumers’ needs by

collaborating and working together in coordinated activities rather than individually competing

with one another in the market. According to Opara (2003), each step in the chain, from basic

inputs to final outputs, is defined as a link in the value chain. Within this context, the main

objective of firms within each stage is to offer maximum value to the customer and to improve

company performance while obtaining a competitive advantage from being part of the value

chain (Ketchen et al., 2008; Bourlakis et al., 2012). Consequently, value creation does not occur

in isolation but rather within a value chain framework where companies can operate most

efficiently as members of the chain. In this position, a company can more easily establish

collaborative relationships based on trust, commitment, and cooperation among the chain’s

members, which helps to improve all the companies’ performance and thus create value

(Lindgreen et al., 2012; Vanyi, 2012). That is why it is paramount to evaluate and measure

value creation in agri-business within a value chain context. Despite the remarkable number of

studies analyzing the need for collaboration among value chains and measuring the benefits of

collaboration among the value chain, little attention has been paid to measuring value creation

in the agri-food value chain context. In this study a value creation analysis is performed within

the context of the value chain to show which actors excel in the value creation process.

Marketing bill

Changes in consumer demand toward more high-quality, ready-to-eat, safe and healthy

food products have increased the price spread in the food marketing system. Price spreads are

based on food price differences along the agri-food value chain as measured at the farm,

wholesale, and retail nodes (Coltrain et al., 2000; Hahn, 2004; MacDonald, 2011). During the

8
past 18 years, the spread between the farm value of products and their retail value, often called

the marketing bill, has increased (Fig. 1). Food retail sales can be split into two components of

market value—the farm value and the marketing bill (USDA, 2013). Farm value is a measure

of the compensation received by farmers for the raw materials they produce, as a proportion of

the food purchase by end customers (Elitzak, 1999; Oswald, 2011). Farm value can be

estimated to be the retail price minus the spread, which implies that “higher price spreads cause

lower farm prices” (Hahn, 2004). Price spreads can be reduced if farm prices raise and/or retail

prices decline (Hahn, 2004).

On the other hand, the marketing share can be calculated as food dollar expenditures

minus farm share commodity sales (Canning, 2011). The farm share in 1993 was 18.4, while

in 2011 the farm share was 15.5 cents for each food dollar expenditure (“a food dollar

represents a $1 expenditure on domestically produced food by U.S. consumers” (USDA, 2013).

At the same time, the marketing share was 81.6 in 1993 and 84.5 cents in 2011 (Fig. 1). The

marketing bill represents a calculation of the value created by the value chain for farm products.

Rising marketing costs affect either farm or retail prices (Oswald, 2011). Changes in the farm

and marketing share over time suggest that not all stages in the value chain are capable of

creating value at the same levels. The price spreads show that value creation distribution in the

agri-food value chain is changing, giving a more important role to the last stages of the value

chain.

The number of value creation activities designed to meet customer needs is expected to

increase in the coming years (Coltrain, 2000). This may present an opportunity for farmers up-

chain to engage in value adding activities that directly benefit consumers and earn higher

economic benefits. The question that arises is: which member of the agri-food chain is

contributing more to the value produced along the value chain? Given that price spreads are

increasing, the level of value creation is thought to increase as we move downward along the

9
agri-food value chain. A higher spread and share by retail indicates that more value is being

created down chain. But price spreads do not connote value creation. Therefore, we explicitly

tackle this empirical problem by measuring the value created at various stages of the value

chain. . By analyzing and measuring value creation in a value chain framework, we expect to

find different levels of value creation among actors due to the characteristics and opportunities

to create value at each stage of the agri-food value chain.

Hypothesis I: Value creation levels differ across stages. Each stage contributes

differently to the value creation process.

Hypothesis II: Value creation levels will be higher as we move downward along the

agri-food value chain.

Drivers of value creation

There are two main drivers of value creation activities: innovation and coordination

(Coltrain et al., 2000; Amanor-Boadu, 2003). “An innovation is an idea, practice, or material

artefact perceived to be new by the relevant adoption unit” (Zaltman et al., 1973). This is

similar to Luecke and Katz (2003), who wrote ‘‘Innovation is generally understood as the

introduction of a new thing or method. Innovation is the embodiment, combination, or

synthesis of knowledge in original, relevant, valued new products, processes, or services’’

(Sawang and Unsworth, 2011). According to Coltrain et al. (2000), agri-food innovation is

commonly characterized by the introduction of new products, processes or markets. Innovation

refers to the ability and capacity to produce or create something unique, valuable and hard to

imitate by performing a series of activities such as new processes or modified old processes.

Within this framework, innovation associated with value creation activities refers to the

performance of activities that enhance “existing processes, procedures, products, and services

10
or creating new ones” using organizational structures (Amanor-Boadu, 2003). In the agri-food

industry there are a variety of agro-industrial innovative processes such as producing ethanol

from corn, biodiesel from soybeans, and particleboard from straw; adding value through the

production of enhanced grain crops such as, non-GMO, organic, Nutridense® maize, and low

linolenic soybean (Coltrain et al. 2000; Goldsmith and Silva, 2006).

Value creation can also be achieved through coordination activities. Peterson and

Wysocke (1997) explains that “coordination focuses on arrangements among those that

produce and market farm products. Horizontal coordination involves pooling or consolidation

among individuals or companies at the same node of the food value chain. Vertical coordination

includes contracting, strategic alliances, licensing agreements, and single ownership of

multiple market stages in different levels of the food chain” (Peterson and Wysocki, 1997;

Coltrain et al., 2000). The value chain framework provides numerous opportunities to obtain

significant rewards and create value along the chain by strengthening the coordination of

products, services, and information. Coordination value creation activities emphasize the

vertical and horizontal relationships among value chain actors (Barkema and Drabenstott,

1996; Coltrain et al., 2000). From a reward perspective, for a coordination or innovation

activity to be part of the value creation process, the activity should provide the performer with

a net positive benefit as reward for the effort of performing the specific activity (Amanor-

Boadu, 2003).

Firms employ a number of activities to create value along the agri-food value chain

(Table 2).

Stage 1- Inputs companies: Empirical evidence suggests that the increase in private

investment in agricultural research has been driven by the establishment and strengthening of

intellectual property (Moschini and Lapan, 1997; Goldsmith, 2001; Fulton and Giannakas,

11
2002). Effective intellectual property development in seed and equipment markets, for

example, allows suppliers to establish internationally recognized brands reflecting highly

differentiated products. Horizontal mergers through acquisition allows seed firms to achieve

significant research and development, marketing, and production economies, in part due to

enhanced coordination. This in turn has led to the development of new products such as

enhanced grains, insect and herbicide resistance crops, and higher yielding varieties and

hybrids.

As member of the agri-food value chain, the inputs stage is positioned in a relative weak

position, as it supplies products to the least profit-making member, the production stage.

Farmers as well as seed companies benefit from using GM technologies (Goldsmith, 2001;

Demont et al., 2007). As a general rule, within the context of GM development, inputs

companies obtain one third of the benefits, while two thirds of the rewards are occur down

chain. A value chain perspective is vital for becoming sustainable within this stage.

Consequently, success and value creation are conditioned to the ability to develop strategic

relationships with other members of the chain in order to provide high-quality products to the

end consumer (Vorley, 2001).

Hypothesis III a: Agricultural inputs is expected to be a high value creator node given

the evidence of branding and agricultural research as well as the evidence of horizontal

coordination among the companies of this stage.

Stage 2- This node can be described as the most commoditized sector of the value chain.

A commoditized industry is characterized by companies selling goods that are

indistinguishable from others, often produced in large quantities and sold inexpensively

(Carlson, 2004; Phillips et al., 2007). Thus, this stage is characterized by having low margins,

12
high price dependence and a low level of product differentiation. Companies often operate

within competitive markets and strive to compete on cost management and economies of scale.

In recent years, an increasing number of structural changes have taken place within this

stage, showing a decline in the number the firms but an increase in those firms’ overall size.

Large firms that have been able to obtain the benefits of large-scale business may be

responsible for making this stage profitable (Lobao and Meyer, 2001; Humphrey and

Memedovic, 2006).

Firms within this stage may focus on performing value added activities that permit them

to capture a greater proportion of the consumer dollar (Born and Bachmann, 2006; Kampen,

2011). Previous literature shows that vertical coordination, such as contractual arrangements,

or vertical integration, contribute to achieving efficiency and to creating more value (Hendrikse

and Bijman, 2002; Sporleder, 2006). Contracts, as a form of vertical integration, allow

agreements between buyers and sellers such as prices, quality and time to take place

(Humphrey and Memedovic, 2006). But spot transactions are still common in production

agriculture in the form of active commodity markets like the majority of U.S grain and oilseed

markets (Goldsmith and Bender, 2004; Goldsmith and Silva, 2006). Thus if producers and

processors are increasingly using contracts to operate (Collins, 2001; USDA, 2013b). One

would expect higher levels of value creation rather than lower, since producers are more closely

connected to customers and consumers.

The main opportunities for producers to innovate rely on their ability to enhance and

add value to their products by moving down chain and capturing part of the higher benefits of

post-producer stages (Coltrain et al., 2000). Producers that are able to create value by producing

what consumers desire rather than offering solely commodities will be able to obtain a larger

share of the food dollar (Coltrain et al., 2000). For example, value added grains are enhanced

13
quality crops with herbicide tolerance and resistance to particular insect pests or with changed

oil composition. These value-added crop products are distinguished from standard

commodities in that they can help improve contractual links among seed firms, farmers,

logistics companies and end customers. The shift from commodity grain to differentiated

grain/oilseed allows producers to create and capture added value (Hayenga and

Kalaitzandonakes, 1999; Goldsmith and Silva, 2006). “Value is created and premiums are paid

because the grain is doing more for someone along the value chain” (Goldsmith and Silva,

2006). This suggests that the value creation generated in one stage of the agri-food value chain

impacts the benefits garnered at other stages. Innovation creation and benefit sharing present

an agri-food value chain as a value creation system of complementarities where firms and nodes

create and share value.

Hypothesis III b: The lack of coordination and the commercialization of

undifferentiated products as well as the condition of price taker make the production

sector the most commoditized chain member. Consequently, Stage 2 is expected to

exhibit low levels of innovation and thus, low levels of value creation.

Stage 3- The process of turning raw agricultural outputs into food and beverage

products ‘adds value’ to raw commodities in an economic sense, but these activities may also

significantly alter the appearance, storage life, nutritional value, and content of the raw

materials (Gopinath et al., 1996). A processor’s main activity is to transform commodities into

food products, a process that adds economic value to the products. However, to qualify as a

value creation activity these actions should create products that satisfy consumer demand by

modifying products’ “appearance, storage life, nutritional value, and content of the raw

materials” (Gopinath et al., 1996)

14
 Increasingly, consumer demand for high quality and premium food products with

accompanying services along with the technological characteristics of the modern market are

making innovation crucial for the food processing node of the value chain. Innovation not only

enables firms to reduce cost but also to successfully adapt to the expectations of the end

customer (Capitanio et al., 2009).

Companies at this stage may focus on producing higher-quality products, as well as

manufacturing more branded and differentiated products, to achieve a competitive advantage

and thus create value (Omidvar, 2006, USDA). Oustapassidis and Vlachvei (1999) also

discusses advertising as one of the strategies to create value in the manufacturing sector. In this

context, product differentiation of new and unique products, in addition to advertising, can be

achieved through investment in R&D (USDA, 2013a).

Along the agri-food value chain, processors experience a competitive purchasing

advantage. Since processing is the most innovation-oriented node, food manufacturers acquire

low cost inputs from the production sector, giving them a greater opportunity to add value to

farm products.

Hypothesis III c: Food manufacturing is expected to exhibit a high level of value

creation due to strong product differentiation, and coordination with retailers.

Stage 4- The last stage of the agri-food value chain focuses on addressing the dynamics

of consumer expectations. One of the main drivers of innovation in this stage for food retailers

are differentiation through service and retail brands to better meet consumer demand (Burch

and Lawrence, 2005; Humphrey and Memedovic, 2006). Retailers offering food services such

as training their personnel to assist the consumer and designing a convenient shopping

experience, do so by focusing on increasing consumer satisfaction and customer value

(Rintamäki et al., 2007). Another source of value creation for food retailers is the development

15
of supermarkets’ own brands to increase their share of the food and beverage market (Burch

and Lawrence, 2005). This upstream integration into retail brands and private label products

changes the relationship between food manufacturing and retail in order to leverage “close

collaboration” in support of innovation (Burch and Lawrence, 2005).

There is evidence of increasing market power at the retail end of the agri-food value

chain as a consequence of increasing concentration and consolidation in the sector (Viaene et

al., 1995; Humphrey and Memedovic, 2006). The influence of retailers over processors,

manufacturing and also the consumer is allowing the retailer sector to obtain a competitive

advantage (Burch and Lawrence, 2005). This is because first, the growing market power of

downstream members may limit up-chain firms from moving to high-value-added activities

such as distribution, marketing and retailing (Farfan, 2005; Niemi and Xing, 2011). Greater

market power down chain allows for greater control over information flows and the resulting

innovation (Farfan, 2005; Humphrey and Memedovic, 2006). Second, the dominant position

of retailers along the value chain, because of their market power, affects the interlinkages

among the four nodes and allows retailers to depress farm prices and low input costs as well as

to establish the best possible terms and standards with suppliers (buyer power), thus extracting

additional value from agri-food commodities (Burch and Lawrence, 2005). Furthermore,

supermarkets, as owners of ‘critical value chain assets’ such as supermarket shelf space, are

able to create extra value by negotiating with suppliers.

Hypothesis III d: Delivery to consumer is expected to be a high value creator member

due to their strategic position along the value chain as well as the evidence of innovation

and coordination within this the node.

16
 R&D expenditures (Meisel and Lin, 1983; Cohen, 1996; Artz et al., 2010; Damanpour,

2010; Zona et al., 2013) and goodwill and intangible assets (Kramer et al., 2011) are key factors

in determining firm level of innovation. Product differentiation is also a key component of

value creation and firm success. The level of product differentiation can be measured as sales

over cost of goods sold (COGS) (Kotha and Nair, 1995; Goldsmith and Sporleder, 1998;

Berman et al, 1999; Nair and Filer, 2003; Balsam et al., 2011). Although several researchers

have identified different categories and activities that determine firm-level value creation (see

Kalafut and Low, 2001; Kale et al., 2001; Zéghal and Maaloul, 2010), there has been little

empirical research on how firms along the agri-food value chain successfully use these

activities to create value. This research aims to fill this gap in the literature by providing

evidence for how expenditures in R&D and the level of intangible assets affect value creation

in the context of the agri-food value chain. This study also measures the effect of the degree of

potential product differentiation on the level of value creation. As the body of literature

suggests, the expectation is that expenditures in R&D, intangible assets and product

differentiation positively affect value creation.

Hypothesis IV: Innovation is a critical factor that positively affects firms and industry

value creation. An increase in the level of innovation will lead to an increase of the

level of value creation.

Hypothesis IV a: Firms with higher levels of intangible capital will have higher levels

of value creation.

Hypothesis IV b: Firms with higher levels of research and development expenditures

will have higher levels of value creation.

17
 Hypothesis V: Firms with higher degrees of product differentiation will have higher

levels of value creation. The greater the ratio of sales over cost of goods sold, the lower

the value creation level.

Firm size

Previous studies suggest that firm size is a relevant characteristic that influences the

intensity of firm innovation (Damanpour, 2010; Hecker and Ganter, 2013; Zona et al., 2013).

Several studies provide evidence of a positive relationship between firm size and innovation.

Further, firm size —that is, large firms—is found to be related with the level of innovation

(Aiken and Hage, 1971; Moch et al., 1977; Kimberly and Evanisko, 1981; Ettlie et al. 1984;

Dewar and Dutton, 1986; Damanpour 1992; Sullivan and Jikyeong, 1999; Camisón-Zornoza,

2004; Damanpour, 2010; Zona et al., 2013). Large firms are more innovative because of the

availability of financial resources, technological possibilities, access to highly skilled labor, as

well as knowledge capability and economies of scope. Furthermore, large firms have a greater

capability to afford the cost of innovation as well as to manage the risk of unsuccessful efforts

(Nord and Tucker, 1987; Hitt, et al., 1990; Chandy and Tellis, 2000; Damanpour, 2010;;

Camison-Zornoza et al., 2004; Hecker and Ganter, 2013). Among the benefits related to size,

Vorley (2001) explains that large firms, in spite of having a cost advantage, are in a strong

position to invest in R&D and to access information that can benefit firm performance.

This theory suggests that firm size is one of the determinants of a firm’s level of

innovation and thus of value creation. As a result, large firms within the industry are capable

of creating cost advantages. Additionally, they have access to and control of the most valuable

intellectual property and can procure capital with which to innovate. It is important to account

18
for firm size when estimating value creation across the value chain to avoid possible

confounding effects of differing firm size across the four nodes of the value chain.

Hypothesis VI: There is a positive relationship between firm size and value creation.

Measuring Value Creation

A critical component of the agri-food value chain value creation question involves how

one measures value creation. While firms have become more focused on value creation, there

has been a growing concern about the use of value creation measures based on traditional

accounting information such as Return on Equity (ROE), Earning per Shares (EPS), Net

Operating Profit After Taxes (NOPAT) and Return on Investment (ROI) (Ibendahl and

Fleming, 2003). Although widely used, these measures fail both to capture a firm’s value

creation that results from management actions, and to account for the full cost of capital

(Sharma and Kumar, 2010). That is, accounting income is not a good estimator of true value

creation, and value creation efficiency (Sharma and Kumar, 2010). From an economic

perspective, to create value a company must produce higher benefits than its cost of capital. An

accurate measure of financial performance, economic value added (EVA), explains how well

a company produces operating benefits, given the amount of working capital invested

(Chmelikova, 2008). An increasing number of companies have begun to employ EVA to

measure the value created by each business unit, since this measure provides better information

than traditional measures (Geyser and Liebenberg, 2003; Ibendahl and Fleming, 2003).

The EVA of a company is defined as “a measure of the incremental return that the

investment earns over the market rate of return” (Sharma and Kumar, 2010). According to

Stewart (1991), this manner of calculating value creation comprises a unified financial

19
framework that includes “financial accounting, management accounting and business

valuation” (O'Hanlon and Peasnell, 1998). The formula for determining EVA is as follows:

AdjNOPAT
EVA = ( − Cost of Capital) ∗ NOA (1)
NOA

Where,

AdjNOPAT is the adjusted net operating profit after taxes. EVA measures the value created by

the operating activities of the company and thus involves an adjusted NOPAT that does not

include non-operating revenues and expenses. These would include such factors as training

costs or restructuring expenditures that might be better capitalized instead of counted as

expenses. Therefore, to accurately calculate economic profits, a variety of adjustments must be

made to the accounting measure of NOPAT (Ibendahl and Fleming, 2003; Chmelíková, 2008).

Net Operating Assets (NOA) represents the working capital invested or the total capital

employed by a firm via its main business activities. NOA involves an adjustment similar to that

of NOPAT and removes accounting items that are not used to generate an operating profit of

the core business, such as non-operating fixed assets. Other such factors might include

investment in R&D that may be activated, since this type of expenditure is considered a source

of value creation, and non-operating current assets that are not employed in operating activities

(Ibendahl and Fleming, 2003; Chmelíková, 2008).

The cost of capital is the weighted average of debt and equity cost. This study calculates

the cost of capital of each firm using the weighted average cost of capital (WACC).

The estimation of EVA measures the true economic profits which is different form

accounting profits in a variety of ways. The EVA takes into account the cost of capital that

20
managers must pay for the capital they employed. Under traditional accounting ratios, a firm

may seem to be profitable when in reality they are not, therefore, the cost of capital is one of

the most relevant component of the EVA estimation as it allow managers to determine the true

value generated by the capital employed (Anderson et al., 2004; Sharma et al., 2010).

According to Walbert (1994), businesses are no longer driven by making higher profits, but

focus on earning a return of capital over the cost of opportunity of the capital. Furthermore, the

EVA metric adjusts accounting information to combine operating efficiency and asset

management (Anderson et al., 2004).

Much of the previous research uses EVA as a comparative metric to stock price to test

market efficiency hypotheses rather than to explore levels and determinants of value creation,

which is our objective (Turvey et al., 2000; Keef and Roush, 2003; Sparling and Turvey, 2003).

Furthermore, previous studies measure EVA at the firm level, leaving open the opportunity for

us to study value creation across the value chain.

The EVA metric is a dollar scaled metric and thus is highly correlated with firm size.

Previous studies suggest that large companies could create more value than smaller companies

despite not using their assets as efficiently because the EVA metric is sensitive to size bias

(Ibendahl and Fleming 2003; Anderson et al., 2004). That is, larger firms, ceteris paribus, create

more nominal value. To understand how firm size affects the EVA metric, suppose that Firm

A has a zero cost of capital, net operating assets of 100 dollars and an AdjNOPAT of 10 dollars.

Calculating the EVA for this firm shows that Company A has an EVA value of 10 dollars.

Then, suppose that a smaller Firm B also has a cost of capital equal to zero and only 10 dollars

of net operating assets but has an adjNOPAT of 1 dollar. The EVA value for Firm B will be 1

dollar. Comparing these firms one in terms of EVA would incorrectly conclude that Firm A is

creating more value than Firm B. The rate at which both companies create value is exactly the

same, which means that both companies are using their capital equally efficiently. In other

21
words, the ratio adjNOPAT/NOA is equal to 0.1 for both companies, but since Firm A is

employing more capital, the EVA value for Firm A is higher than for Firm B. Our interest,

though, focuses more on real and relative value creation and those firms that are efficient or

inefficient users of capital when creating or destroying value. We develop and employ a second

value creation metric, the MEVA, which is scale neutral, is expressed as a percentage, and is

defined as follows:

𝐴𝑑𝑗𝑁𝑂𝑃𝐴𝑇
𝑀𝐸𝑉𝐴 = ( − 𝐶𝑜𝑠𝑡 𝑜𝑓 𝐶𝑎𝑝𝑖𝑡𝑎𝑙) ∗ 100 (2)
𝑁𝑂𝐴

A NOA weighted MEVA is estimated for each of the value chain nodes. Both the EVA

and MEVA may be positive, a net creator of value, or negative, a net destroyer of value. EVA

has no range limits while the MEVA ranges between -100% and 100%.

Interpretation of value creation metrics

A positive EVA or MEVA indicates that the firm or value chain node creates value

because the operating assets or working capital employed generate a return sufficient to cover

the capital costs of those assets. On the contrary, a negative EVA or MEVA indicates that a

company is destroying value because the returns from its net operating assets fall short of the

returns required to capitalize the company. EVA and MEVA, as value creation metrics, are not

a measure of company profitability. Consequently, companies with positive net operating

income may actually destroy value because the assets of the firm may have a higher and better

use. The measures show how efficiently the operating assets are being employed to generate a

determined level of operating income.

22
 DATA

An unbalanced panel of a ten-year time-series (2003-2012) of 454 agri-food companies

worldwide is analyzed. The data originate from income statement and balance sheets for each

of the firms collected from Morningstar Inc. Included are agri-food companies from 25

countries (64% are U.S. companies, 3.26% United Kingdom, 3.89% Japan, 3.36 % China, and

2.86% from Mexico and others). The data selected to conduct this study includes all the firms

that Morningstar classifies as belonging to the agricultural inputs, chemicals, beverages,

packaged food, farm construction, restaurants, food distribution, farm-products and grocery

industries. The agri-food value chain is divided into four stages, inputs, production, processing,

and delivery to consumers, and the companies are then reclassified into these four stages

according to the main activity of each. Table 3 provides a brief description of this database,

including a description of the main activities of each stage, as well as the number of companies

at each stage. This includes 97 companies in the input stage, 65 firms belonging to the

production stage, 156 companies within the processing stage, and 136 companies in the

retailer’s stage. The data represent firms that conduct activities that are part of the agri-food

value chain. Firms may conduct activities involving more than one sage along the value chain.

In such cases we follow the classification made by Morningstar Inc. identifying the main

activity of the each company. This main activity then tags the firm to a particular value chain

node. A number of quantitative measures and statistical tests are used to verify the differences

in the level of value creation at each stage.

The Morningstar data reported is expressed in the local currency of each firm’s country.

All the observations are converted into U.S dollar units based on the World Bank’s official

exchange rate. The official exchange rate refers to the exchange rate determined by national

23
authorities, or the rate determined in the legally sanctioned exchange market. It is calculated as

an annual average based on monthly averages (local currency units relative to U. S dollars).

Value creation metrics

Using this financial data, the EVA and MEVA metrics are calculated for each firm for

each year. The following formulas are used to compute the value creation measures:

𝐴𝑑𝑗𝑁𝑂𝑃𝐴𝑇 (3)
𝐸𝑉𝐴 = ( − 𝐶𝑜𝑠𝑡 𝑜𝑓 𝐶𝑎𝑝𝑖𝑡𝑎𝑙) ∗ NOA
𝑁𝑂𝐴

𝐴𝑑𝑗𝑁𝑂𝑃𝐴𝑇
𝑀𝐸𝑉𝐴 = ( − 𝐶𝑜𝑠𝑡 𝑜𝑓 𝐶𝑎𝑝𝑖𝑡𝑎𝑙) ∗ 100 (4)
𝑁𝑂𝐴

The adjustments made to NOPAT to obtain the calculation of AdjNOPAT (interest

expenses, R&D expenditures, restructuring, and merger and acquisition costs) are summarized

in Table 4. Table 5 presents the adjustments made to total assets. This table provides a brief

description of the adjustments made to total assets to calculate the NOA variable.

The cost of capital is the weighted average of debt and equity cost. To calculate the cost

of capital of each firm at each time period the weighted average cost of capital (WACC) is

used. This method considers the opportunity cost of each source of capital (debt and equity).

Formally,

𝐷 𝐸
𝑟𝑘 = 𝑟𝑑 (1 − 𝑡) ( ) + 𝑟𝑒 ( ) (3)
𝐷+𝐸 𝐷+𝐸

Where, 𝑟𝑘 is cost of capital, 𝑟𝑑 is cost of debt, 𝑟𝑒 is cost of equity, 𝐷 is the debt, 𝐸 is

total capital and 𝑡 is corporate tax rate

24
 Cost of debt is the average cost of the debt for each firm (Equation 4). It is estimated

using,

𝐼𝑛𝑡𝑒𝑟𝑒𝑠𝑡 𝐸𝑥𝑝𝑒𝑛𝑑𝑖𝑡𝑢𝑟𝑒𝑠
𝑟𝑑 = (4)
𝑇𝑜𝑡𝑎𝑙 𝑙𝑖𝑎𝑏𝑖𝑙𝑖𝑡𝑖𝑒𝑠

The variable tax (t) is used as the official corporate tax rate of each country. Weights

are calculated based on book value mix of debt and equity.

To estimate the cost of equity, the capital-asset pricing model (CAPM) is used

(Equation 5).

𝑟𝑒 = 𝑅𝑓 + 𝛽(𝑅𝑚 − 𝑅𝑓 ) (5)

where,

𝑅𝑓 = Risk-free bond

𝑅𝑚 = Expected market rate of return.

𝛽 = It is the relative risk of the particular asset

The estimation of the cost of equity of a company,𝑟𝑒 , is composed of three elements:

risk-free bonds (𝑅𝑓 ), the stock’s equity beta (𝛽 = 1.0 is average risk) and the market risk

premium (𝑅𝑚 − 𝑅𝑓 ) needed to attract investors to hold onto the market portfolio of risky assets

(Bruner et al., 1998; Sharma and Kumar, 2010). To calculate the risk-free rate of return a 10-

year Treasury bond return obtained from the Federal Reserve database is used. The long-term

bond yield carefully accounts for the default-free holding period return available and

consequently most closely represent companies’ investments (Bruner et al., 1998). The market

risk premium is calculated from the average of the last 25 years as the difference between the

return of the Standard & Poor’s 500 and the risk-free return. Since this database includes

25
companies from developing countries, the country risk premium to the cost of capital is

included. The country risk premium is the increased return that investors have to pay to invest

in a country that is deemed risky for investment. Risk premium estimates are based on the

country ratings assigned by Moodys. This premium measures a given country’s default risk.

Finally, the beta of each firm is estimated using the unleveraged beta by industry. The

unleveraged betas are estimated using the average market debt/equity ratio by industrial sector.

The betas are then leveraged using each company’s debt/equity ratio

This study also calculates a dummy variable called ‘create value’ (CEVA) which takes

value 1 when MEVA>0 and, 0 otherwise. This variable accounts for the proportion of

companies that have positive MEVA within a particular stage. Furthermore, a ‘persistent value

creation’ (PEVA) variable is calculated to identify the companies that have, in each node,

positive value creation levels for a period of at least five years. This variable is a dummy

variable that takes value equal to 1 if the firm uses its capital efficiently for at least 5 years, and

0, otherwise. This variable identifies and evaluates the characteristics of firms that are

sustainable in the efficient use of their capital. The objective of this analysis is to evaluate the

performance of the companies across time and to identify certain characteristics and patterns

of companies that use their capital efficiently and those that do not.

Firm Financial Characteristics

The analysis of the data also includes firms’ financial characteristics, such as firm size,

leverage, cost of goods sold (COGS) and innovation.

Firm size: Previous literature uses different proxies for firm size. Generally, firm size

is measured by using the total sales (Arundel and Kabla, 1998), total assets (Gopalakrishnan

26
and Damanpour, 2000; Zehri et al., 2012) or number of employees (Ettlie et al., 1987; Sawang

and Unsworth, 2011; Zona et al., 2013). For the purpose of this study, the term ‘total assets’ is

used as a proxy for firm size (Goldsmith and Sporleder, 1998; Gopalakrishnan and Damanpour,

2000; Camisón-Zornoza et al., 2004; Damanpour, 2010). From total assets the amount of

intangibles and goodwill is subtracted to avoid counting them twice and to allow them to be

included in the variable that measures innovation.

Leverage: In the process of value creation, capital structure is significant. One crucial

aspect is the proper usage of debt or leverage (Houle, 2008). Positive effects are associated

with increasing the leverage of a company in that it saves taxes, cures the risk of unproductively

reinvesting surplus cash flow, creates an urgency to perform well, and forces the sale of

underperforming or unrelated businesses or assets (Houle, 2008). However, prior evidence

from Finance and Accounting literature generally suggests that financial leverage is a risk

factor (Ely, 1995). This implies that firms with higher levels of financial leverage are riskier

than firms with lower levels of debt (Hua and Templeton, 2010). The interpretation of the

leverage variable is as follows: “the higher the ratio, the more debt a company uses in its capital

structure” (Morningstar). Defined as total assets divided by total shareholders' equity, financial

leverage is included in our analysis to measure the implications of capital structure on value

creation by using EVA, MEVA, CEVA and PEVA value metrics.

COGS: Cost of goods sold is the cost a company incurs to process, create, and/or sell a

goods or services including the price of the inputs as well as the costs of transforming it into a

final food product (Morningstar). Cost of goods sold is expressed as a percentage of revenue.

Since a high level of COGS will decrease firms’ revenues, it may negatively impacts the value

creation levels.

27
 Innovation: Following previous studies, this variable is composed of R&D

expenditures (Zona et al., 2013), goodwill (Degryse et al., 2012) and intangible assets (Kramer

2011). This study measures innovation in three variables. ‘Innovation 1’ is composed of the

three components described above; it is the sum of R&D expenditures, goodwill and intangible

assets. The study proceeds to separately evaluate the different sources of innovation to

determine whether there are differences in the source of value creation among the four nodes.

‘Innovation 2’ is composed of capital assets (intangible assets and goodwill) and ‘Innovation

3’ is composed of R&D expenditures. These variables are included to measure the impact of

innovation on value creation.

Summary statistics

This section includes the main characteristics of the data while presenting the definition,

means and standard deviation of the main variables (Table 6a) used in this study for each stage

of the agri-food value chain. Mean tests are calculated to determine whether the value outcomes

for each stage are statistically different from each other. Furthermore, a mean test is performed

to determine whether the outcomes of various stages are altogether different from the average

value outcome of the whole food value chain. In this way, this research provides insights

regarding the statistical differences in mean value outcomes among the various stages, and

between individual stages and the mean value outcomes of the whole value chain. Table 6b

illustrates the results from the mean test analysis and provides the p-values of those tests.

28
Value creation analysis

The EVA metric measures, in dollar terms, how well the firms are using their operating

capital to produce operating profits. In other words, the EVA value shows the net value created

by a firm after accounting for the cost of capital. The EVA metric of each stage is calculated

by taking the average EVA of the firms comprising each stage. Analysis shows that in terms

of EVA, agricultural inputs is the node that creates the most value, having, on average, an

added economic value of 264 million dollars per firm, which is 40% higher than the average

EVA of the whole value chain. Stage 3 has an average mean EVA value of 257 million dollars

per firm, which, in dollar terms, makes it the second highest contributor to the agri-food value

creation process. Stage 4 has a mean EVA value of 107 million dollars, having on average a

value creation level 43% below the value creation level of the whole agri-food value chain.

The production stage creates, on average, 43 million dollars, which represents 23% of the

average mean EVA value for the whole chain. The mean tests conducted to show whether

stages are statistically different in terms of the EVA variable do not allow a conclusion that

Stage 1 and Stage 3 are statistically different from each other. Similarly, the null hypothesis

for significant differences between Stage 2 and Stage 4 cannot be rejected.

The EVA value for chain members is determined by the level of net operating assets

and the adjusted NOPAT, as well as the cost of capital. Net operating assets differ across value

chain members. In Stage 1, the mean value of NOA is 5,979 million dollars per firm, which is

only 1% lower than average net operating assets per firm in Stage 3, in which the average NOA

is equal to 6,053 million dollars. Stage 1 and Stage 3 have, on average, net operating asset

values 34% and 36% higher than the mean operating assets of the value chain. The mean tests

conducted to measure differences among stages show that Stage 1 and Stage 3 are not

statistically different from each other in terms of NOA. Similarly, for Stage 2 and Stage 4 the

null hypothesis cannot be rejected for significant differences among them; thus, these two

29
stages are not shown to be different in terms of NOA. The average net operating assets for

Stage 4 total 2,200 million dollars, 12% lower than for Stage 2.

Likewise, this study does not find significant differences between Stage 1 and Stage 3

in terms of Adjusted NOPAT. Stage 1 produces on average an AdjNOPAT of 680 million

dollars, and Stage 3 generates net operating profits of 639 million dollars, on average, per firm.

Similarly, this study cannot state that Stage 2 and Stage 4 are different in terms of AdjNOPAT.

Stage 2 has an average AdjNOPAT of 210 million dollars, which is 14% lower than the average

net operating profits for Stage 4, which produces 245 million dollars. The rejection of the null

hypotheses of the mean tests for the variables described above, for Stages 1 and 3, and Stages

2 and 4, can be attributed to the differences in size among firms within those respective stages.

Firm size

The various nodes differ significantly in size across the value chain (Fig. 3). To measure

firm size, this study used adjusted total assets. Firm total assets are adjusted by intangible assets

and goodwill to avoid double accounting the intangible capital in the regression model. Stage

1 is significantly larger than Stage 3, Stage 2 and Stage 4, being composed of firms 1.5 times

larger than the average value chain. This fact gives an insight into the need for companies

within this stage to obtain benefits related to large-scale business to create value. This stage

has a coefficient of variability of 1.79 in size, an average 15% less variability than the overall

value chain, implying that in terms of size, Stage 1 is the most homogeneous node. Companies

in Stage 3 have, on average, total assets valuing 4,207 million dollars per firm—the second

largest stage within the agri-food value chain. Interestingly, Stage 3, which is smaller than

Stage 1, generates the same amount of EVA, suggesting that firms in Stage 3 are using their

capital more efficiently than those in Stage 1. In Stage 2, the average firm has 3,067 million

30
dollars in adjusted total assets, and Stage 4 firms have, on average, 3,444 million dollars per

firm and the highest coefficient of variation, making Stage 4 the most heterogeneous node in

terms of adjusted total assets. Stage 2 is a chain member composed of smaller firms within the

overall value chain (mean size of 3,067 million dollars). According to the mean test, this study

cannot affirm that, in terms of size, Stage 2 and Stage 4 are on average statistically different

from each other.

The size bias

Although the EVA metric is widely used by managers and academics because it has

better capabilities for measuring the firm value creation level in comparison with traditional

accounting measures, for the purpose of this study the EVA is a weak metric for comparing

value creation across the agri-food value chain. The statistical analysis of the mean EVA value

by stages supports this; the EVA metric is not an accurate measure for comparing value creation

levels among firms or stages because it is affected by the size of the companies. For this reason

it is not appropriate to conduct an analysis and evaluation of the differences in value creation

levels among stages using the EVA metric.

Across the agri-food value chain, each stage is composed of firms of different sizes.

This study analyzes the relationship between a company’s EVA and a company’s NOA to

understand bias toward the EVA by firm size. A company’s NOA variable is measured by the

average NOA per firm over a period of years. Similarly, a company’s EVA variable is created

to measure the average EVA per firm over a period of years. There is a positive relationship

between the size of the company and its level of value creation (Fig. 4). Meanwhile, there is a

positive correlation (0.77) between a firm’s size and the level of value creation measured by

EVA. This provides evidence that small firms may produce low levels of EVA, and large firms

31
may generate high levels of EVA. The differences in the EVA value can be due to differences

in firm size and not necessarily because of differences in the efficient use of firms’ operating

capital.

MEVA

For evidence of value creation, this study aims to evaluate how efficiently companies

among the agri-food nodes are using their capital. An accurate understanding of a firm’s value

creation requires that company size be controlled to avoid firm size bias. The best metric of

value creation to compare and correctly analyze the value creation of a stage composed of firms

of different sizes is the MEVA metric. This metric removes the distortion caused by firm size

and measures a comparable value creation level in terms of the rate at which each company is

creating value. The MEVA variable measures the standardized/normalized value creation.

Technically, the modified EVA is the EVA metric divided (normalized) by the capital

employed. It is expressed as a percentage of the rate of value creation in terms of NOA.

Down-chain is where the most value is created in terms of MEVA. On average, Stage

3 and Stage 4 are creating additional operating capital at the rate of 0.83% and 3.43 % per year

respectively, while upstream the average rate of value creation is negative. Down-chain there

is also less variability in terms of MEVA, suggesting that Stage 3 and Stage 4 are more

homogeneous in the rate at which companies create value than is the case with up-chain

members. Delivery to consumer is the node that uses its capital most efficiently, having a mean

value MEVA of 196% higher than the mean MEVA value of the overall chain. The average

firm in Stage 4 creates value at a rate 3.43% of net operating assets, in excess of the cost of

capital. This stage is composed of 136 companies, of which 71 have persistent positive levels

of value creation (52% of the firms have persistent positive MEVA value). This stage is

32
characterized by having the smallest percentage of negative MEVA observations and the

largest level of value creation rate (Fig. 5). The distribution of MEVA in Stage 4 is farther to

the right in comparison with other stages. In comparison with other nodes, Stage 4 is composed

of small firms with high a level of value creation. Taking into account the average NOA and

multiplying it by the average MEVA, Stage 4 is creating, on average, 7,555 million dollars of

net operating assets. Stage 3 has a mean MEVA value of 0.833%, being the second-highest

chain node contributor to the value creation process. In dollar terms, the average company in

this stage generates net operating total assets of 5,042 million dollars (if we take into account

the average NOA of the stage). In this stage 58% of companies create value across time and

represent the greatest proportion of persistent value creation companies. Firms that persistently

use capital efficiently have, on average, a value creation rate of 4.6% in Stage 3, while in Stage

4 persistent value creation companies create value at the average rate of 7.48% of net operating

assets.

On the other hand, up-chain is, on average, destroying value. Stage 1 is destroying value

at an average rate of 0.195% while Stage 2 is destroying value at an average rate of 0.806%.

On average, companies within these stages do not generate enough operating profit to cover

the cost of opportunity for their capital. Stage 2 is destroying value at the highest average rate

in the agri-food value chain. The production sector has 68% of its firms persistently destroying

value. It is the chain node with the highest number of firms having negative value creation

levels across time. This implies that, on average, the business nature of this stage is making

firms to create low value. This stage is the most commoditized sector of the value chain,

wherein a low level of product differentiation takes place, and the strategic advantage over

competitors is limited to achieving low cost and economies of scale and scope. In particular,

low-scale companies have a significantly lower value of MEVA in comparison with large-scale

companies within this stage (Fig 5). Persistent value-destroying companies are characterized

33
by having net operating assets 3 times smaller and an adjusted NOPAT 5 times smaller than

persistent value creation companies. Firms in the production sector are not persistent in value

creation. The main differences between observations with positive MEVA and negative MEVA

are that the positive ones rely on the level of net operating profits. Positive MEVA observations

have, on average, 21.6% more net operating assets than negative MEVA observations, but 3

times higher AdjNOPAT. Companies need to focus on obtaining higher levels of net operating

profits to create value. Furthermore, positive MEVA observations have COGS 12% lower than

those that destroy value. Overall, this stage is characterized by small firms that create low rates

of value.

Agricultural inputs companies destroy value at the average rate of 0.194%. This Stage

1 is the most heterogeneous stage in terms of MEVA, having a coefficient of variation of 1,

which is 507% higher than the coefficient of variation of the entire value chain. This means

that companies within Stage 1 create value at significantly different rates. The size of the firms

in this stage is crucial for creating value. Small firms have extremely low values of MEVA,

which increases the variability of the average MEVA for this stage (Fig 5). An analysis of the

data shows evidence of a strong relationship between size and the rate at which firms create

value (Fig. 4). In particular, 25% of agricultural inputs companies with the lowest MEVA

within the stage have net operating assets of 2,125 million dollars on average, and the range of

the MEVA value for these companies goes from -83% to -3%. The other 75% of companies

included in this stage have a positive value of MEVA (ranging from 2% to 34%) and a mean

NOA of 7,251 million dollars, which is 3.4 times higher than the mean NOA for the other 25%.

Therefore, Stage 1 is characterized by two main features. First, there is evidence of high

heterogeneity in the rate at which companies create value in terms of MEVA. Second, this

study finds that firms with a low value of MEVA have, on average, low values of net operating

assets.

34
 The distribution of MEVA differs across the four stages (Fig. 5). The mean test shows

statistically significant differences among the MEVA value for all chain members. In all stages,

small firms have lower rates of value creation in comparison with large ones, which implies

that value creation is, in part, a function of size. Despite the differences in value creation rates

among stages, large firms tend to have a positive value of MEVA. In other words, the benefits

associated with large business units are helping companies to create value. The nature of the

business of each stage seems to be playing an important role in determining the level of value

creation. Furthermore, the variability within stages in terms of MEVA decreases down-chain,

implying that down-chain companies are using capital more efficiently, but the down-chain

nodes are also more homogeneous in comparison with up-chain members.

Created value variable

The objective of this variable is to measure the number of observations in each stage

that create value. Technically, it is a dummy variable that takes value equal to 1 if the

observation has a positive value creation level and zero otherwise. So, this variable measures

whether a firm’s operating assets are being efficiently used in a particular year, but it does not

account for how well the capital is used. This metric is calculated as a manner of validating

results. The value member containing the highest number of observations with a positive value

creation level is Stage 4 with 67% of its observations demonstrating efficient use of operating

capital. With 8 % fewer number of observations creating value, Stage 3 is the second highest

chain member in this category, composed of 62% observations with positive value creation

levels, followed by Stage 1 with 60% of the firms creating value. Agricultural inputs has only

52% of its observations creating value, and therefore is the chain member with the least

proportion of observations creating value, having 16% fewer created value observations than

35
the overall value chain. In accordance with the MEVA analysis, the down-chain stages show a

higher proportion of observations creating value.

The mean test performed to reflect differences among the stages in terms of MEVA

shows statistical differences among all the nodes except for Stage 1 and Stage 3. For these two

stages the null hypothesis cannot be rejected. This variable only identifies whether a firm

capital is being used efficiently or not (without accounting for how much value is created, either

in percentage or in dollar terms). The ‘CEVA’ variable analysis shows that except for Stage 2,

more than 60% of the observations have positive value creation levels. However, the MEVA

calculation shows that from those observations that created value, the rate at which companies

add value varies across stages. Overall, this analysis of value creation definitely gives an insight

that down-chain is where most of value is being created. Furthermore, the three measures of

value creation evaluated in this section support the conclusion that agricultural producers are

the chain actors that least efficiently use operating capital.

Firm Characteristics Analysis

Innovation

Innovation is considered one of the drivers of value creation. The innovation variable

is composed of the sum of R&D, goodwill and intangible assets. Since firm innovation level is

related to firm size, it is more accurate to analyze the mean value as a percentage of total assets.

Stage 3 is the most innovative chain member, investing 66% of the average adjusted total assets

in innovative activities, and Stage 3 is the second stage that where the most value is produced,

in terms of EVA, MEVA and CEVA. Nevertheless, Stage 3 has the greater coefficient of

variation, suggesting heterogeneity in terms of innovation capital among firms. Stage 4 is the

36
second most innovative value chain actor, investing 21% of adjusted total assets in innovation

activities, which is one-third the level of Stage 4. Stage 1 invests in innovation 4.4 times less

than Stage 3, assigning 15% of the adjusted total assets to innovative activities. As the most

commoditized node, it is not surprising that Stage 2 is the least innovative chain member,

investing on average 9% of total assets in innovative activities. According to the mean test, the

mean value of innovation is significantly different among the chain actors in, except for Stage

1 and Stage 4, as well as Stage 2 and Stage 4, since the null hypothesis for the mean test

conducted to test differences in innovation cannot be rejected for these particular stages.

Down-chain is where most of the value chain value is being created, but also where

expenditures in innovation are more intense, suggesting a positive relationship between

innovation and value creation level. The distribution of the innovation variable varies across

stages (Fig. 6). Stage 3 has a higher level of investment in innovation capital but also greater

dispersion, suggesting that within this stage the level of innovation among firms differs.

Not only the level but also the source of innovation varies across the four nodes. In the

following paragraph, this study separately analyzes each component of the innovation variable:

‘Innovation 2’ variable is composed by intangible capital (intangible assets plus goodwill), and

‘Innovation 3’ is composed by R&D expenditures. This study also reports the proportion of

innovation that is represented by intangible capital and R&D expenditures.

The analysis of ‘Innovation 2’ shows that the main source of innovation in all the

stages—except for agricultural inputs—is intangible capital. For Stages 2, 3 and 4, the

innovation level is a function of intangible capital and not of R&D expenditures, as 99% of

innovation investments involve goodwill and intangible assets for the three nodes. In Stage 1,

intangible capital represents 86% of the total amount of innovation, and provides the main

source of innovation for this node. However, Stage 1 is where the greater investment in R&D

takes place. From the total investment in innovation, 16% is assigned to R&D expenditures,

37
which represents, on average, 141 million dollars (34% of the average net income). This is

eight times more than spent in Stage 3, the next highest member in R&D investment, Stage 3

companies invest on average 17 million dollars in R&D expenditures, representing 3% of the

average net income for this stage. The significantly higher level of investment in R&D in Stage

1 can be explained by the progressively higher need for product differentiation at earlier

production stages, and by the business nature of this member, which needs to be engaged in

investigative activities such as biotechnology, seed innovation and new technologies.

The analysis of innovation variables shows that down-chain is where the highest

proportion of total assets is assigned to intangible capital and also where the most value is

created. Consistent with the previous evaluation, Stage 2 having the lowest level of innovation

is the node that creates the least value. This suggests a positive relationship between innovation

and value creation.

COGS

The analysis of COGS focuses on identifying key relationships between buyers and

sellers to determine if there exists any type of complementarities in terms of costs and prices

among stages. Furthermore, this study aims to identify the implications of cost in the value

creation process. Cost of goods sold includes the inputs cost paid to the up-chain member plus

the cost of value-added activities performed to make the product ready for the next stage—and

not including indirect costs such as selling, general and administrative costs.

The distribution of COGS varies across chain members (Fig. 7). Consistent with the

MEVA analysis, down-chain is where the gross margin is higher than that of up-chain

members. Stage 4 is the only node that has COGS below 100%; the other stages have

companies with COGS above 100%. In particular, Stage 2 has the greatest proportion of

companies with COGS above 100% (3.5%), which can be traced to the nature of its productive

38
cycle. Production companies experience high uncertainty with selling prices. This risk is

increased by the timing between receptions of input to the moment when the product is ready

to sell to the following chain member.

On average, Stage 2 pays the highest cost in terms of revenues. Agricultural producers

spend, on average, 78% of their revenues on inputs costs. This can also be related to the fact

that product differentiation allows Stage 1 to force higher prices for products since companies

are offering unique products such as patented, high performance seeds. Stage 2 may offer low-

cost products to processors because they sell undifferentiated materials such as raw

commodities. This stage pays high inputs prices, having a gross margin of only 22% of its

revenues, and then sells to processors at low prices. Stage 3 is the chain member that pays the

lowest input prices (in comparison with all chain members), but it is the actor that adds the

most value to the products in gross terms. Specifically, 35% of its revenues represent the cost

of the value-added activities, and it pays 65% of its revenues in cost, having on average the

lowest rate of COGS in terms of revenue. In the previous analysis, this study finds that Stage

3 is the most intangible capital-oriented one, and clearly, Stage 3 is the chain member with

most opportunities for differentiating products and adding value to products through branding,

developing new products and new markets, and offering high-quality and ready-to-eat

products. The last stage of the agri-food value chain spends 71% of its revenues in cost of

goods sold, having one of the highest margin of the chain members. That can be attributed to

the fact that deliver to consumer sells its products to the consumer markets giving a key and

strategic position in the chain.

Buying and selling activities are related among members of the agri-food value chain,

and this study finds a negative relationship between COGS and MEVA outcomes (Fig. 8).

Consistent with previous analysis, down-chain is where companies create more value and also

pay the lowest input prices. In particular, processors can be seen as being in the strongest

39
position in terms of COSG, since companies within this stage buy at the lowest prices from the

least value-adding node and perform value-adding activities which allow them to obtain higher

margins. Furthermore, this analysis gives the insight that market power in down-chain stages

is causing downward pressure on the costs of farm products.

Financial leverage

The financial leverage variable is a measure of a firm’s capital structure. In particular,

this variable measures the proportion of equity and debt the company uses to finance the firm’s

assets. This analysis shows that there is no difference among stages in the level of financial

leverage. The mean test indicates that the mean leverage levels for each stage are not different

from one another. Additional investigation of the implications of financial leverage in the value

creation process will be discussed in the next section.

In order to deeply understand the distribution of the variables, the present study

evaluates the standard deviations of our main variables to identify whether the source of the

variance is due to differences across time or across firms. Each stage of the agri-food value

chain is composed of heterogeneous groups of firms. In other words, the variability among

firms within each stage is what explains the standard deviation. Firms across years seem not to

be experiencing significant variations regarding the main variables analyzed in this study.

40
 EMPIRICAL MODEL

To test our hypotheses and to identify which actors in the value chain are the highest

contributors to the value adding process we use an econometric model. This model essentially

consists of measuring the effect of inputs, production, processors and delivery to consumer

stages in the value creation process. It next determines the impact of firm size, firm capital

structure, and expenditures in innovation and COGS on stage value creation. The dependent

variables of the models are the value creation measures described in the previous sections:

EVA, MEVA, CEVA, PEVA variables. Formally, the models are as follows:

𝐸𝑉𝐴𝑖𝑡 = 𝛼 + 𝜌𝑆𝑡𝑎𝑔𝑒1 + 𝛽𝑆𝑡𝑎𝑔𝑒3 + 𝛾𝑆𝑡𝑎𝑔𝑒4 + 𝛿𝑙𝑜𝑔𝑆𝑖𝑧𝑒𝑖𝑡

+ 𝜃𝑙𝑒𝑣𝑒𝑟𝑎𝑔𝑒 𝑖𝑡 + 𝜗𝐼𝑛𝑛2𝑖𝑡 + 𝜏𝐼𝑛𝑛3𝑖𝑡 + 𝜋𝐶𝑂𝐺𝑆𝑖𝑡 (6)

+ 𝐶𝑜𝑢𝑛𝑡𝑟𝑦𝐹𝐸 + 𝑌𝑒𝑎𝑟𝐹𝐸 + 𝑢𝑖𝑡

𝑀𝐸𝑉𝐴𝑖𝑡 = 𝛼 + 𝜌𝑆𝑡𝑎𝑔𝑒1 + 𝛽𝑆𝑡𝑎𝑔𝑒3 + 𝛾𝑆𝑡𝑎𝑔𝑒4 + 𝛿𝑙𝑜𝑔𝑆𝑖𝑧𝑒𝑖𝑡

+ 𝜃𝑙𝑒𝑣𝑒𝑟𝑎𝑔𝑒 𝑖𝑡 + 𝜗𝐼𝑛𝑛2𝑖𝑡 + 𝜏𝐼𝑛𝑛3𝑖𝑡 + 𝜋𝐶𝑂𝐺𝑆𝑖𝑡 (7)

+ 𝐶𝑜𝑢𝑛𝑡𝑟𝑦𝐹𝐸 + 𝑌𝑒𝑎𝑟𝐹𝐸 + 𝑢𝑖𝑡

𝐶𝐸𝑉𝐴𝑖𝑡 = 𝛼 + 𝜌𝑆𝑡𝑎𝑔𝑒1 + 𝛽𝑆𝑡𝑎𝑔𝑒3 + 𝛾𝑆𝑡𝑎𝑔𝑒4 + 𝛿𝑙𝑜𝑔𝑆𝑖𝑧𝑒𝑖𝑡

+ 𝜃𝑙𝑒𝑣𝑒𝑟𝑎𝑔𝑒 𝑖𝑡 + 𝜗𝐼𝑛𝑛2𝑖𝑡 + 𝜏𝐼𝑛𝑛3𝑖𝑡 + 𝜋𝐶𝑂𝐺𝑆𝑖𝑡 (8)

+ 𝐶𝑜𝑢𝑛𝑡𝑟𝑦𝐹𝐸 + 𝑌𝑒𝑎𝑟𝐹𝐸 + 𝑢𝑖𝑡

𝑃𝐸𝑉𝐴𝑖𝑡 = 𝛼 + 𝜌𝑆𝑡𝑎𝑔𝑒1 + 𝛽𝑆𝑡𝑎𝑔𝑒3 + 𝛾𝑆𝑡𝑎𝑔𝑒4 + 𝛿𝑙𝑜𝑔𝑆𝑖𝑧𝑒𝑖𝑡

+ 𝜃𝑙𝑒𝑣𝑒𝑟𝑎𝑔𝑒 𝑖𝑡 + 𝜗𝐼𝑛𝑛2𝑖𝑡 + 𝜏𝐼𝑛𝑛3𝑖𝑡 + 𝜋𝐶𝑂𝐺𝑆𝑖𝑡 (9)

+ 𝐶𝑜𝑢𝑛𝑡𝑟𝑦𝐹𝐸 + 𝑌𝑒𝑎𝑟𝐹𝐸 + 𝑢𝑖𝑡

Explicative Variables

41
 Stage 1 is a dummy variable that takes value equal to one if the firm belongs to Stage

1, and zero otherwise. Stage 2 is a dummy variable that takes value equal to one if the firm

belongs to Stage 2 and zero otherwise, and Stage 3 is a dummy variable that takes value equal

to one if the firm belong to Stage 3 and zero otherwise. To avoid the dummy variable trap we

drop the Stage 2 variable. This theory suggests that Stage 2 (agricultural production) is the

more commoditized stage; therefore we drop the dummy variable for the second stage to

compare all the results against a stage that has a lesser probability of creating value. These

dummy variables are included to measure the contribution of each stage in the value creation

process and to test the stated hypotheses.

There are three main sources of bias in the present study. These are factors related to

each particular country, year or company. In order to avoid the endogeneity problem due to

unobservables in the error term, country-fixed effects (CountryFE) and year-fixed effects

(yearFE) are included. All country-fixed effects control for the unobservables in each country

that may affect the value creation process. In the same way, the year-fixed effects control for

shocks that affect all companies’ value creation process for each particular year. Since the

variables of interest are the stage dummy variables that changes at company levels, company-

fixed effects cannot be included. For this reason company characteristics are included instead

of company fixed effects such as size, leverage, goodwill and intangible capital, R&D

investments and COGS.

Firm characteristic variables are included in the regression model. The first is firm size,

measured as the logarithm of adjusted total assets to quantify the impact of firm size on the

value creation process. The second is firm leverage, measured as total current liabilities over

total equity, which is included in the model to measure the effect of the financial capabilities

of each firm with regard to the create value variables. This study includes two measures of

innovation: ‘Inn2’ that reports the value of intangible capital (intangible and goodwill assets),

42
and ‘Inn3’ that reports the value of R&D expenditures. Furthermore, the variable COGS is

included in the model to measure the potential product differentiation.

To validate the results and provide reliable findings, a t-test is performed to determine

whether stages are statistically different from one another after controlling for value creation

drivers. The primary interest relies on verification that the regression results of EVA, MEVA,

CEVA and PEVA metrics are statistically different among stages.

43
 RESULTS

This study estimates the effect of being part of the different stages of the value chain in

EVA, MEVA, CEVA and PEVA variables. The four models are designed to test Hypothesis I

that states that each stage contributes differently to the value creation process and to estimate

the contribution of each stage in the value adding process. In this section, the results of our

regression models are presented (Table 7). Only three dummy variables, Stage 1, Stage 3 and

Stage 4 are included in order to avoid the dummy variable trap, making Stage 2 the baseline

against which to compare. The test for significant differences across stages is also reported.

The four value creation metrics are calculated using each firm’s net operating profits

after taxes (NOPAT). The data used in the present study is comprised by firm from different

countries. Therefore, one interesting question that may arise is how local tax rates and policies

affect the NOPAT variable and the EVA estimation. For validation purposes we estimate the

value creation metrics using the net operating profits before taxes. The results are robust

showing no change in the EVA coefficient estimates. Additionally, the four regression models

control for the country fixed effects, which in part address differences in tax policies.

EVA as a measure of value creation

The EVA metric calculates the value creation level, weighted by the capital invested in

productive activity. In terms of EVA, Stage 3 is the largest contributor to the agri-food value

creation process. On average, this stage creates 40 more millions of dollars per year than Stage

2, ceteris paribus. The coefficient estimate is significant at the 0.1 level. Processors are

followed by Stage 4, delivery to consumers, which creates 39 million more dollars per year in

terms of EVA than Stage 2—independent of the country in which the company was located,

44
the year of each observation, and the individual characteristics of each firm. The coefficient

estimates are significant at the 0.05 level. Stage 1, agricultural inputs, destroys 15 million more

dollars per year in terms of EVA than Stage 2, independent of the country in which the company

was located, the year of each observation, and the individual characteristics of each firm;

however, the coefficient estimates are not statistically significant. Stage 2 (the constant) has an

estimated coefficient of -8 and it is not statistically significant, meaning that it cannot be stated

that the EVA level for this stage is statistically different from zero. The superiority of the value

creation levels of down-chain nodes in comparison with Stage 2 is consistent with

contemporary agribusiness literature that suggests that Stage 2 is the most commoditized stage,

for which possibilities for product differentiation are low and products offered by these

business units are characterized by being undifferentiated, homogeneous, and are a low level

of value-added goods.

Comparing the statistical differences for the dummy variables of each stage, results

show that the effect of being in Stage 3 cannot be statistically distinguished from the effect of

being in Stage 4. Additionally, the effect of being in either Stage 3 or Stage 4 is statistically

better than the effect of being in Stage 1 (t-test). To conclude, in terms of EVA, Stage 3 is the

sector that creates the most value in dollar terms, but in terms of EVA metric we cannot

statistically affirm that it is different from Stage 4, which follows Stage 3 in the chain. These

two nodes have a higher value creation level (measured as EVA) than Stage 2, which is not

statistically different from zero and Stage 1, showing that capital is being used more efficiently

by down-chain members.

45
MEVA as measure of value creation

In order to validate our results with a different measure of value creation, this research

estimates the MEVA metric that accounts for the standardized value creation level without

taking into account the capital invested in the productive process (see column 2 of Table 7).

In terms of MEVA, the value creation of the average company in Stage 3 or in Stage 4

increases to 1.59% and 4.71% per year respectively, in comparison with Stage 2, independent

of where the company is located, the year in which each observation took place, and the

individual characteristics of each firm. The coefficient estimates for Stage 1 is 0.60%, but not

statistically significant. For Stage 2, the coefficient estimate is negative but not statistically

significant. This is in accordance with the EVA results from Stage 2 that show that the

production sector is a poor contributor to the value creation process. The p-values for the t-

tests were performed to show whether stages are statistically different in term of MEVA.

Results show that there are statistically significant differences among the four nodes. Thus,

delivery to consumer is the value chain member that creates value at the highest rate in terms

of MEVA, followed by processors as the stage that contributes second most to the value

creation process. As hypothesized, the results show differences in the level of value creation

across the nodes and point out that down-chain nodes are the chain members that most

efficiently use capital.

CEVA variable as a measure of value creation

CEVA is s a dummy variable that takes value equal to 1 if the observation has a positive

value creation level, and zero, otherwise. This variable was analyzed to validate the results and

to examine a different dimension of the problem (Column 3 of Table 7).

46
 In particular, retailers have a highly significant coefficient estimate of 14.7%,

representing the chain member with the highest probability of creating value in comparison

with Stage 2. Stage 3 has a coefficient estimate of 8.8%, which is statistically significant at the

0.01 level, meaning that there is an 8.8% higher probability of having a positive value creation

level in Stage 3 than in Stage 2, ceteris paribus. The coefficient estimate for agricultural inputs

is 5.5% and is statistically significant at 0.01 level, which means that the probability of creating

value in Stage 1 is 5.5% higher when compared to Stage 2, which has a significant coefficient

value of 58.9%. In all cases, the CEVA variable is estimated independently of the country in

which company is located, the year of each observation, and the individual characteristics of

each firm.

The p-values for the t-test were calculated to show whether stages are statistically

different in terms of the CEVA variable. The results show statistical differences among the

four nodes except for Stage 1 and Stage 3, for which the null hypothesis for differences among

these stages cannot be rejected. In accordance with previous findings, Stage 2 has the least

probability of creating value in comparison with the other chain nodes. This result can be linked

with the nature of the business. Stage 4 is consistently one of the most efficient users of its

capital and captures the highest share of value-added. This can be linked to the fact that retailers

are controlling the flow of technology and knowledge along the chain—a finding that supports

previous research.

PEVA variable as a measure of value creation

This study generates a measure of persistent creation of value to examine how

sustainable each stage may be in the generation of value. PEVA is a dummy variable that that

takes value equal to 1 if a company uses its capital efficiently for at least five years (half of the

47
period covered by the database), and zero otherwise. This variable identifies the proportion of

companies that persistently create value within each stage. Stage 4 has a coefficient estimate

of 2.18%, meaning that the probability of a persistent create value in Stage 4 is 2.18% higher

than in in Stage 2. The coefficient estimate for Stage 3 is 2.16% and for Stage 1, 1.58%. The

probability of creating persistent value for Stage 2 is negative, showing the inability of the most

commoditized stage to create value in a sustainable way. Finally, all the coefficient estimates

for each stage dummy are statistically different at the 0.1 level, again showing that the down-

chain nodes excel in value creation.

The p-values for the t-test were calculated to show whether stages are statistically

different in terms of the PEVA variable. The results show statistical differences among the four

nodes except for Stage 3 and Stage 4, for which the null hypothesis for differences among these

stages cannot be rejected.

The four regression models were designed to test Hypothesis III a, Hypothesis III b,

Hypothesis III c, Hypothesis III d that predict the level of value creation at each stage.

Hypothesis III argues that agricultural inputs is a high value creation node. Although this study

finds statistically significant results for the CEVA and PEVA models, there is no significant

results for EVA and MEVA variables. Thus, this research cannot affirm that the conditional

mean of the value created by companies in this stage is different from zero. Therefore, there is

mixed evidence and thus no strong support for Hypothesis III a. According to Hypothesis III

b, Stage 2 is expected to exhibit low levels of value creation. In all the regression results, this

study finds that the production sector is the least contributor in the value adding process.

Therefore, the estimation results validate Hypothesis III b. Although the differences in the

conditional mean for EVA and MEVA metrics between Stages 1 and 2 are not statistically

significant, this study shows that the dimensions that distinguish companies in Stage 1 and

Stage 2 in terms of EVA are firm size, financial health, innovation levels, COGS, year fixed

48
and country fixed effects. The present analysis also finds no statistically significant differences

between Stage 1 and Stage 2 regarding the conditional and unconditional mean MEVA. One

explanation for this is that both stages have average low rates of value creation. For instance,

the conditional mean is -0.194 and -0.806 for Stage 1 and Stage 2 respectively, which are rates

very close to zero if one takes into account that the MEVA range is -100 to 100. Companies

within these stages show this type of pattern over the years of analysis, making the conditional

mean of EVA and the MEVA undistinguishable between these two stages.

Hypothesis III c and Hypothesis III d states that Stage 3 and Stage 4 are high value

creation members. The empirical analysis conducted shows that Stage 4 is the chain node that

most efficiently uses its capital, being the greater contributor to the value creation process. The

four regression models validate these results. Processors is found to be the second most

contributor chain member in the value adding process. Therefore, these results provide strong

support for Hypothesis III c and Hypothesis III d as well as for Hypothesis II which states that

the levels of value creation will be higher as one move down-chain.

The role of the Size, Leverage and Innovation, COGS.

In the four regression models this study includes four control variables, as well as

variables at the company level that can determine the value creation of each company such as

firm size, leverage, goodwill and intangible capital, R&D expenditures and COGS.

The logarithm of adjusted total assets as a measure of firm size has a positive and

statistically significant coefficient estimate (at the 0.01 level) in all the regression models,

meaning that size is a significant determinant of value creation. An increase of one percent in

size increases the EVA level by 40 million dollars. According to the MEVA variable, an

49
increase of one percent in the size of a firm increases the rate of value creation by 2.5 percentage

points. The probability of creating value (MEVA) is increased by 0.059% for the average

company when the firm size is increased by one percent. Finally, the probability of persistently

created value increases by 0.088% when firm size is increased by one percent. These results

are consistent with previous literature which states that larger companies are more innovative

because of their better availability of resources, economies of scale and scope to spread the

risk—all in contrast to small companies. Furthermore, large firms are associated with higher

possibilities for successfully commercializing innovative products, absorbing the costs of

innovation, having higher R&D intensity, and taking advantage of economies of scale. These

results confirm the Hypothesis VI and provides evidence of a positive relationship between size

and value adding supporting previous studies.

Leverage is measured as total liabilities over total equity. This variable measures the

source of capital within a company. For all these models the coefficient estimates are positive;

however only the EVA model (at the 0.1 level) is statistically significant. For the EVA model,

the coefficient estimate is 3.645, meaning that increasing firm leverage by one unit increases

the level of value creation by 3.6 million dollars.

All the regression models include two measures of innovation. The two measures of

innovation are included to test Hypothesis IV which states that innovation positive affects value

creation. Specifically, firms that have higher level intangible capital (Hypothesis IV a) or higher

level of R&D expenditures (Hypothesis IV b) may have higher levels of value creation.

Therefore, Intangible and goodwill assets (measured by ‘Inn 2’ variable) and R&D

expenditures (measured by ‘Inn 3’ variable) have been hypothesized as being key drivers of

innovation, and thus of value creation. The results of investment in R&D, differ depending on

the dependent variable. The coefficient estimate for the EVA model is 1.464 and is significant

at the 0.01 level, meaning that an increase of one million dollars in expenditures in R&D

50
increases a firm’s EVA by 1.4 million dollars. Furthermore, the coefficient estimate for the

CEVA model is 0.001 and is significant at the 0.01 level, meaning that the effect of increasing

expenditures in R&D by one million dollars increases, on average, the probability of creating

value by 0.1%, ceteris paribus. The coefficient estimates for R&D expenditures for the MEVA

and PEVA models are both 0.001; surprisingly, these are not statistically significant at 0.1

level. However, the coefficient estimates for the MEVA and PEVA models are significant at

0.2 level. Overall, these results are in accordance with Hypothesis V b and with previous

literature that highlights the role of innovation in value creation.

The second innovation variable is intangible capital, measured as the sum of goodwill

and intangible assets. As hypothesized, the regression results shows that intangible capital

positively affects value creation. For the first regression, the coefficient estimate is 0.045 and

is statistically significant at the 0.001 level, meaning that an increase in intangible capital of

one million dollars generates an increase of $45,000 dollars in a firm’s EVA. According to the

regression results for the MEVA model, an increase of one million dollars of intangible capital

increases the rate of value creation by 0.001. The coefficient estimate for intangible capital in

the CEVA model is 0.001, meaning that an increase of one million dollars in intangible capital

increases the probability of create value by 0.1 %. For the persistently create value model the

coefficient estimate is positive but not statistically significant. Therefore, the regression results

provide evidence of a positive relationship between innovation and value creation supporting

Hypothesis IV a.

Cost of goods sold is the ratio between cost and sales. The larger this ratio, the smaller

is the gross margin for the producer. Since this variable is the inverse of the gross margin of

the company, it is also an indirect measure of the level of product differentiation. This variable,

included in the four models to test Hypothesis V predicts a negative relationship between the

COGS ratio and the level of value creation. Increasing the ratio of COGS by one unit reduces

51
the firm’s EVA by $27.6 million dollars. The coefficient estimate is significant at the 0.05

level. Similarly, the probability of creating value decreases by 0.036 when COGS is increased

by one unit. The coefficient estimate is significant at the 0.01 level. For the MEVA and PEVA

models, the regression results are negative but not statistically significant. Since the variable

COGS can be interpreted as an inverse variable of product differentiation, the regression results

for the four models show a positive relationship between product differentiation and value

creation. In particular, the results show that firms producing more “commodity-like” products

have lower levels of value creation for the EVA and CEVA models providing support for

Hypothesis VI. Similarly, firms offering greater differentiation have a higher probability of

creating value. However, the results for the MEVA and PEVA does not provide strong support

for Hypothesis VI. Therefore, the results show a mixed evidence regarding the relationship

between COGS and value creation.

52
 DISCUSSION

This paper offers insights and implications for managers, shareholders and investors.

The results serve as a benchmark for designing and implementing business strategies aiming

to accomplish an efficient use of capital and thus, obtain higher value creation outcomes.

This study shows significant differences in the value creation levels of four stages in

the agri-food value chain. For example, up-chain members contribute least to the value adding

process. Stage 1 is the homogeneous in terms of size in comparison with other chain nodes, but

has significant variability in MEVA values. Agricultural inputs is an inefficient user of capital.

Stage 1 is the chain node with the highest level of investment in R&D expenditures (on average,

34% of net income is invested in R&D) but has relatively low returns in terms of MEVA.

Although Stage 1 is the chain node with the highest EVA mean value, it is characterized by

low rates of value creation (62% of the observations have an MEVA value below the mean

MEVA of the stage) but high amounts of capital invested, since it is composed, on average, of

the largest firms in the overall value chain. Stage 1 represents low value creation because of

the nature of the business, which provides agricultural inputs such as farm machinery,

genetically modified seeds and fertilizer, among others. Naturally, these activities require high

capital intensity and installed capacity to gain scale and thus reduce costs.

The analysis of previous sections shows that 53% of the agricultural input companies

have positive PEVA outcomes, being a persistent but low value creation member. However,

these companies have experienced a high variability in value creation outcomes over the ten

years of study. For example, although Alamo Group Inc. (AGL), that manufactures and sells

agricultural equipment and infrastructure maintenance equipment, has persistent value

creation, it has also experienced negative value creation levels in other years (Fig. 9). On

average, AGL destroyed 4.5 million dollars at the average rate of -1.12% (average MEVA).

53
Effective value creation appears unsustainable with respect to Alamo Group Inc. as even

though there are six non negative years of value creation, there are four years of highly negative

of significant capital destruction (Fig. 9). Like in this company, firms in this stage are

destroying value (on average, 40%), but they have periods of successful performance. This

example also shows the relevance of analyzing value creation across time. Managers should

take into account the importance of having persistent value creation business units, but they

should also be aware that new practices are needed to identify negative shocks and to mitigate

the effect of these shocks on value outcomes.

However, Stage 1 is comprised also by high value creating firms such as Monsanto.

Monsanto is a company that provides agricultural products for farmers such us seeds,

biotechnology trait products, and herbicides. Monsanto products are designed to improve

farmer’s productivity, reduce the costs of farming and produce better foods. Monsanto belongs

to Stage 1 in the agri-food value chain. The firm is a significant value creator over the 10 years

under analysis, having on average an EVA of $1,165 million dollars and the average annual

rate of value creation (MEVA) of 7.6%. Monsanto is a highly innovative company investing

10% of sales in R&D expenditures and maintaining 30% of the adjusted total assets in goodwill

and intangible assets. The firm is also a persistent value creator having positive MEVA value

outcomes in all the years under analysis. Furthermore, Monsanto attains an average COGS of

49.06% which is 24% lower than the average firm in the input stage. This firm as a high value

creator demonstrates that there are companies that are efficient users of capital and significant

value creators even in low value creation stages of the food and agribusiness value chain.

The production stage is characterized by being a poor value creator because it has the

lowest outcome in the four metrics of value creation analyzed in this study. The main distinctive

characteristic of this node is the inability to produce higher operating returns on capital. In

other words, companies intensely use capital but yet produce low returns. On average the

54
production stage annually produces 5 cents per each dollar invested in NOA, the lowest rate in

comparison with the other chain members. However, not all the companies in this stage are

underperforming in the value creation process. Successful firms are characterized by key

features. Analysis of the data shows that large companies have positive and persistent value

creation rates. Clearly, size plays a key role for production companies to create value. In other

words, companies that take advantage of the economies of scale in this stage perform better

than those that do not take advantage of size features. Furthermore, one of the main

characteristics of big firms is that they generate greater revenue per dollar invested in NOA (12

cents per dollar invested in NOA). In addition, these companies are more likely to invest in

intangible and goodwill assets. Small companies face more difficulties creating value,

especially when measuring the persistence of value creation outcomes. The small firms that

perform better are involved in value creation activities. For example, China Green Ltd, is a

vertically integrated, relatively small company (size: 82% less than the average size of the

stage) engaged in the cultivation, production and logistical activities of food products. The

firm uses its capital efficiently and produces 19 cents per each dollar invested in NOA (280%

higher than the average operating revenue of the stage). The company has an average MEVA

of 1.24% and an average EVA of 33 million dollars per years over the ten years under analysis.

Its average COGS is 49%, which is 36% lower than the average COGS for the stage, suggesting

that the company is adding more value than the average company in the sector. The company

is involved in differentiation activities through the production of branded products.

As this example clearly shows, there are opportunities in the production sector to obtain

higher returns and create more value. There are companies in this stage that are efficient users

of capital. However, most of the companies are unsustainable value creators. Furthermore, the

regression results show that the probability of creating persistent value for companies within

this stage is negative. Managers should be aware of that and need to focus on managing firms

55
assets more efficiently. The main weakness of this stage could be the product firms are offering.

Commodity production, which dominates Stage 2, is a low margin business, where price setting

is limited or nonexistent and cost management is essential. Producing non-commodity

products may allow companies some pricing power and the ability to obtain higher margins

that lead to higher operating profits and value creation outcomes. Overall, the efficient use of

capital is being poorly achieved within this sector, a fact related to the commoditization of

products offered. Thus, the companies in this stage can only compete in price reduction with

few possibilities for product differentiation.

As this study has already noted, up-chain nodes are the least contributor in the value

adding process along the agri-food value chain. To improve performance, a low value creator

might change the focus of its business strategies and aim to imitate the practices of down-chain,

high value-creating members. The implementation of new business strategies may be needed

to capture part of the value that is being produced down-chain. Undoubtedly, the products being

offered play a significant role. Although there may be other possible drivers of value creation,

results suggest that companies in the production sector need to move from the production of

undifferentiated products to differentiated ones in order to change their price orientation to one

more focused on value orientation. According to the results of the present study, investment in

innovation could dramatically help to improve firm performance and the creation of value.

Importantly, our findings also demonstrate that capturing some of the benefits related to size

could allow firms to create higher value outcomes. This finding may guide managers to

understand the relevance of forming strategic alliances, joint ventures or cooperatives to

increasing operating returns and improving the efficient use of capital.

This study finds that down-chain is where the most value is created. Food

manufacturing and delivery-to consumer-nodes use operating capital efficiently and make a

strong contribution to the value adding process of the entire value chain. Both stages have

56
positive and high value outcomes in the four measures of value creation used in this research.

Furthermore, Stage 3 and Stage 4 are associated with a strong focus on innovation and product

differentiation. This finding also contributes to the literature by providing managerial

implications about the relationship between innovation and value creation. Supporting previous

studies, this research shows that investment in innovation positively affects value creation. For

example, processors that excel in value creation outcomes invest on average 66% of the

adjusted total assets in goodwill and intangible assets.

Managers need to first understand the meaning of value creation in business. In this

sense, the present study contributes by providing a clear definition of value creation.

Additionally, this paper uses the well-known EVA metric but also introduces new financial

tools such as MEVA, CEVA and PEVA metrics to measure and analyze value creation.

Although these metrics are used at the stage level from a value chain perspective, they can also

be used to measure value creation at the business unit level. These metrics successfully measure

the value generated by a business unit after accounting for the cost of opportunity. Furthermore,

PEVA is a metric that allows measurement of value creation across time and gives an insight

into business sustainability.

Our findings may also help managers to understand how each stage excels or

underperforms in the value creation process along the agri-food value chain. This information

may be useful for managers and practitioners, since knowing about the performance of the

stage to which a particular firm belongs and how other nodes succeed or fail in the value

creation process could help managers to design more productive business strategies.

57
 CONCLUSION

This study evaluated the value creation process among the key stages of the agri-food

value chain. The main purpose of this manuscript was to analyze the differences in stages

within the agri-food value chain by identifying which nodes excel or underperform in the value

creation process. Furthermore, this research measured the effect of the drivers of value creation

in the contexts of value chain and agribusiness. An estimation of four value creation metrics

was conducted for each firm and each stage: the Economic Value Added (EVA) metric, the

Modified Economic Value Added (MEVA), the create value variable (CEVA) and the

persistent value creation (PEVA) variable. These four metrics were employed to improve the

robustness of the findings.

The analysis of value creation among the stages provides evidence that capital is not

being used equally efficiently across the four nodes. As hypothesized, each stage contributes

differently to the value creation process, according to the statistically significant regression

results and the mean tests performed.

The findings provide evidence that the production stage—the most commoditized

node— contributes the least amount of value to the chain, while additional processing and

retailing contribute significantly greater levels of value. Measuring value creation as MEVA

or CEVA, Stage 4 is the chain node that most efficiently uses its capital, being the greater

contributor to the value creation process. However, when value creation is measured using the

EVA metric, and accounting for the working capital invested, processors excel in creating

value. In all cases, down-chain is where most of the value is being created. The evidence shows

that the down-chain nodes create value more persistently than that found in the up-chain nodes.

The analysis of the drivers of value creation focused on innovation. As hypothesized,

regression results provide evidence of a positive relationship between innovation and value

58
creation. Analysis of the data also reveals significantly greater levels of innovation down-chain,

where capital is being used more efficiently, implying that innovative firms have higher levels

of value added. Two types of innovation were measured in this study: intangible capital and

R&D spending. The source of innovation was found to vary across stages. The main source of

innovation for all the stages is intangible capital, but the levels of investment in intangible

capital varies across nodes. However, agricultural input companies are more focused on R&D

investment than is the case for other stages.

This analysis also provides evidence that size is a significant determinant of value

creation, showing that firm size has a positive effect on the four value creation variables. In all

the regression models, firm size was found to be positive and significant, thereby affecting the

three value creation metrics as expected. Product differentiation was also found to positively

affect value creation, suggesting that companies offering homogeneous or commoditized

products provide a lesser level of value creation. A firm’s degree of product differentiation also

affects the probability of creating value.

One of the objectives of this paper was to clarify the term “value creation.” The present

study of value creation in agri-business, using the EVA concept as the base measure of firm

value added, allowed the researcher to conclude that to create value, firms must perform a

value-added activity that could consist of an innovation (which refers to the introduction of a

new method or the modification of an existing one), or they must perform an activity that was

previously performed by another member of the chain. Furthermore, business units could focus

on coordination as a method for creating value. In this case, investing in horizontal or vertical

coordination, as well as seeking collaboration among other firms, might be another effective

way to increase company value. But in either of these cases, to qualify as a value-added activity,

such an action must be capable of providing the company with a positive net benefit after

accounting for the cost of performing the activity.

59
The regression results show that there are still differences in the value creation levels among

the stages, even when controlling for widely considered drivers of value creation. This provides

the insight that there may be other critical factors at play, or, it may be the nature of the business

that explains differences in value creation levels among the four nodes. This study is restricted

insofar as it only uses financial data. Among the drivers of value creation enumerated in

previous sections, financial data only permit a measurement of innovation as driver of value

creation. Further information is needed to analyze and determine whether coordination, as a

source of innovation, can account for the different outcomes obtained for each stage. So, an

area for further research would be an analysis of the impact of coordination (horizontal and

vertical) on value creation. Coordination among value chain actors implies complementarities

among the activities of different chain members. According to Hendrikse et al. (2001), vertical

coordination of production, distribution, and marketing among firms in a value chain may have

an impact on the investment decisions of each firm individually. Investments by a firm of the

chain must be coordinated with investments by other firms to obtain optimal chain

performance. Furthermore, the lack of coordination among members in a chain is associate

with “ higher inventory costs, longer delivery times, higher transportation costs, higher levels

of loss and damage, and lowered customer service” (Lee et al., 1997; Simatupang et al., 2002).

Lee (2004) explains that a value chain to attain a competitive advantage not only need to be

cost effective and high speed but also need to develop three specific talents which are to be

adaptable, aligned and agile. Among the literature, it is widely agree that activities and

decisions making in one stage of the agri-food value chain impacts and benefits other stages,

implying that there exist intra-stages linkages and business relationships that make the agri-

food value chain a value creation system. Additional information would be needed to measure

coordination and the degree of collaboration among agri-food members and how interlink ages

60
among firms are contributing to the value creation process. To quantify this effect, a survey

and additional information data may be needed.

61
REFERENCES
Aiken, M., & Hage, J. (1971). The organic organization and innovation. Sociology, 5(1), 63-
82.
Alexander, C., & Goodhue, R. E. (1999). Production system competition and the pricing of
innovations: An application to biotechnology and seed corn. American Agricultural
Economics Association.
Amanor-Boadu, V. (2003). A conversation about value-added agriculture. Value-Added
Business Development Program. Department of Agricultural Economics. Kansas State
University.
Ambrosini, V., & Bowman, C. (2009). What are dynamic capabilities and are they a useful
construct in strategic management? International Journal of Management
Reviews, 11(1), 29-49.
Anderson, A. M., Bey, R. P., & Weaver, S. C. (2005). Economic value added adjustments:
much to do about nothing. WP-Lehigh University.
Anderson, D. P., & Hanselka, D. (2009). Adding Value to Agricultural Products. Retrieved
from from http://hdl. handle. net/1969, 1, 86940.
Anderson, J. C. (1995). Relationships in business markets: exchange episodes, value creation,
and their empirical assessment. Journal of the Academy of Marketing Science, 23(4), 346-
350.
Anderson, J. C., & Narus, J. A. (1998). Business marketing: understand what customers
value. Harvard business review, 76, 53-67.
Anderson, J. C., Narus, J. A., & Van Rossum, W. (2006). Customer value propositions in
business markets. Harvard business review, 84(3), 90.
Arundel, A., & Kabla, I. (1998). What percentage of innovations are patented? Empirical
estimates for European firms. Research policy, 27(2), 127-141.
Balsam, S., Fernando, G. D., & Tripathy, A. (2011). The impact of firm strategy on
performance measures used in executive compensation. Journal of Business
Research, 64(2), 187-193.
Barkema, A., & Drabenstott, M. (1996). Consolidation and change in heartland
agriculture. Federal Reserve Bank of Kansas City, Economic Forces Shaping the Rural
Heartland.

Berman, S. L., Wicks, A. C., Kotha, S., & Jones, T. M. (1999). Does stakeholder orientation
matter? The relationship between stakeholder management models and firm financial
performance. Academy of Management journal, 42(5), 488-506.
Bigliardi, B., & Bottani, E. (2010). Performance measurement in the food supply chain: a
balanced scorecard approach. Facilities, 28(5/6), 249-260.
Born, H., & Bachmann, J. (2006). Adding value to farm products: an overview. National
Center For Appropriate Technology.

62
Bourlakis, M., Maglaras, G., & Fotopoulos, C. (2012). Creating a “best value supply chain”?
Empirical evidence from the Greek food chain. International Journal of Logistics
Management, The, 23(3), 360-382. 3.
Bowman, C., & Ambrosini, V. (2000). Value creation versus value capture: towards a coherent
definition of value in strategy. British Journal of Management, 11(1), 1-15.
Bruner, R. F., Eades, K. M., Harris, R. S., & Higgins, R. C. (1998). Best practices in estimating
the cost of capital: survey and synthesis. Financial Practice and Education, 8, 13-28.
Burch, D., & Lawrence, G. (2005). Supermarket own brands, supply chains and the
transformation of the agri-food system. International Journal of Sociology of Agriculture
and Food, 13(1), 1-18.
Camisón-Zornoza, C., Lapiedra-Alcamí, R., Segarra-Ciprés, M., & Boronat-Navarro, M.
(2004). A meta-analysis of innovation and organizational size. Organization
Studies, 25(3), 331-361.
Canning, P. N. (2011). A revised and expanded food dollar series: a better understanding of
our food costs. US Department of Agriculture, Economic Research Service.
Capitanio, F., Coppola, A., & Pascucci, S. (2009). Indications for drivers of innovation in the
food sector. British Food Journal, 111(8), 820-838.
Carlson, W. (2004). Don’t produce a commodity. Greenhouse Grower, 22(12), 20-23.
Chandy, R. K., & Tellis, G. J. (2000). The incumbent's curse? Incumbency, size, and radical
product innovation. The Journal of Marketing, 1-17.
Chikan, A. (1997): Vállalatgazdaságtan, Aula Kiadó Kft
Chmelikova, G. (2008). Economic Value Added versus Traditional Performance Metrics in the
Czech Food-Processing Sector. Editorial Staff, 11(4), 49.
Collins, K. (2001). Statement of Keith Collins, Chief Economist, US Department of
Agriculture before the US Senate Committee on Appropriations, Subcommittee on
Agriculture, Rural Development, and Related Agencies: Economic Issues Related to
Biofuels. Unknown www. usda. gov/documents/Farmbill07energy. doc, 1-8.
Coltrain, D., Barton, D., & Boland, M. (2000). Value Added: Opportunities and Strategies.
Retrieved October 20, 2003.
Congress, U. S. (2002, May). Farm security and rural investment act of 2002. In107th
Congress, 2nd Session. Public Law (No. 107-171).

Conner, K. R. (1991). A historical comparison of resource-based theory and five schools of

thought within industrial organization economics: do we have a new theory of the firm?
Journal of management, 17(1), 121-154.
Damanpour, F. (1992). Organizational size and innovation. Organization studies, 13(3), 375-
402.

63
Damanpour, F. (2010). An integration of research findings of effects of firm size and market
competition on product and process innovations. British Journal of Management, 21(4),
996-1010.
Degryse, H., de Goeij, P., & Kappert, P. (2012). The impact of firm and industry characteristics
on small firms’ capital structure. Small Business Economics, 38(4), 431-447.
Demont, M., Dillen, K., Mathijs, E., & Tollens, E. (2007). GM Crops in Europe: How Much
Value and for Whom? Les cultures génétiquement modifiées en Europe: quels avantages
et pour qui? Genetisch veränderte Feldfrüchte in Europa: Welcher Wert und für
wen?. EuroChoices, 6(3), 46-53.
Dewar, R. D., & Dutton, J. E. (1986). The adoption of radical and incremental innovations: an
empirical analysis. Management science, 32(11), 1422-1433.
Doyle, P. (2000). Value-based marketing. Journal of Strategic Marketing, 8(4), 299-311.
Elitzak, H. (1999). Food Cost Review, 1950-97 (No. 34053). United States Department of
Agriculture, Economic Research Service.
Ely, K. M. (1995). Operating lease accounting and the market's assessment of equity
risk. Journal of Accounting Research, 397-415.
Ettlie, J. E., Bridges, W. P., & O'keefe, R. D. (1984). Organization strategy and structural
differences for radical versus incremental innovation. Management science, 30(6), 682-
695.
Evans, E. (2006). Value Added Agriculture: Is It Right for Me? EDIS document FE638, Florida
Cooperative Extension Service, Institute of Food and Agricultural Sciences, University
of Florida, Gainesville, FL. Published.
Farfan, O. (2005). Understanding and escaping commodity-dependency: a global value chain
perspective. documento preparado para la Corporación Financiera Internacional,
Washington, DC, Banco Mundial.
Fulton, M., & Giannakas, K. (2002). Agricultural biotechnology and industry structure.
Geyser, M., & Liebenberg, I. E. (2003). Creating a new valuation tool for South African
agricultural co-operatives. Agrekon, 42(2), 106-115.
Goldsmith, P. D. (2001). Innovation, Supply Chain Control, and the Welfare of Farmers: The
Economics of Genetically Modified Seeds. American Behavioral Scientist, 44(8), 1302-
1326.
Goldsmith, P. D., & Silva, C. (2006). Specialty Soybean and Corn Survey. National Soybean
Reseach Laboratory.
Goldsmith, P. D., & Sporleder, T. L. (1998). Analyzing Foreign Direct Investment Decisions
by Food and beverage Firms: An Empirical Model of Transaction Theory. Canadian
Journal of Agricultural Economics/Revue canadienne d'agroeconomie, 46(3), 329-346.
Goldsmith, P., & Bender, K. (2004). Ten conversations about identity preservation. Journal on
Chain and Network Science, 4(2), 111-123.

64
Goldsmith, P., Salvador, A., Knipe, D., & Kendall, E. (2002). Structural change or logical
incrementalism? Turbulence in the global meat system. Journal on Chain and Network
Science, 2(2), 101-115.
Gopalakrishnan, S., & Damanpour, F. (2000). The impact of organizational context on
innovation adoption in commercial banks. Engineering Management, IEEE Transactions
on, 47(1), 14-25.
Gopinath, M., Roe, T. L., & Shane, M. D. (1996). Competitiveness of US food processing:
benefits from primary agriculture. American Journal of Agricultural Economics, 78(4),
1044-1055.
Grunert, K. G., Jeppesen, L. F., Jespersen, K. R., Sonne, A. M., Hansen, K., Trondsen, T., &
Young, J. A. (2005). Market orientation of value chains: A conceptual framework based
on four case studies from the food industry. European Journal of Marketing, 39(5/6),
428-455.
Hahn, W. F. (2004). Beef and pork values and price spreads explained. USDA Economic
Research Service.
Hayenga, M., & Kalaitzandonakes, N. (1999, May). Structure and Coordination System
Changes in the US Biotech Seed and Valueadded Grain Market. In Prepared for
presentation at the 1999 World Food and Agribusiness Congress. Florence, Italy:
International Food and Agribusiness Management Association.
Hecker, A., & Ganter, A. (2013). The Influence of Product Market Competition on
Technological and Management Innovation: Firm‐Level Evidence from a Large‐Scale
Survey. European Management Review, 10(1), 17-33.
Hendrikse, G., & Bijman, J. (2002). Ownership structure in agrifood chains: the marketing
cooperative. American Journal of Agricultural Economics, 84(1), 104-119.
Hitt, M. A., Hoskisson, R. E., & Ireland, R. D. (1990). Mergers and acquisitions and managerial
commitment to innovation in M-form firms. Strategic Management Journal, 11(4), 29-
48.
Houle, M. D. (2008). Economic Value Added. Senior Honors Papers, 38.
Hua, N., & Templeton, A. (2010). Forces driving the growth of the restaurant industry in the
USA. International Journal of Contemporary Hospitality Management, 22(1), 56-68
Humphrey, J., & Memedovic, O. (2006). Global value chains in the agrifood sector (Working
paper). Retrieved from United Nations Industrial Development Organization website:
http://www.unido.org/fileadmin/user_media/Publications/Pub_free/Global_value_chain
s_in_the_agrifood_sector.pdf
Ibendahl, G. A., & Fleming, R. A. (2003). Using Economic Value Added (EVA) to Examine
Farm Businesses. In Annual Meeting of the Southern Agricultural Economics
Association. Mobile, Alabama. February.

Kalafut, P. C., & Low, J. (2001). The value creation index: quantifying intangible
value. Strategy & Leadership, 29(5), 9-15.

65
Kale, P., Dyer, J., & Singh, H. (2001). Value creation and success in strategic alliances:
alliancing skills and the role of alliance structure and systems. European Management
Journal, 19(5), 463-471.
Kampen, K. (2011). Financial analysis of three value-added dairy enterprises in Vermont,
Wisconsin, and New York (Doctoral dissertation, California Polytechnic State
University).
Keef, S. P., & Roush, M. L. (2003). The relationship between economic value added and stock
market performance: A theoretical analysis. Agribusiness,19(2), 245-253.
Ketchen Jr, D. J., Rebarick, W., Hult, G. T. M., & Meyer, D. (2008). Best value supply chains:
A key competitive weapon for the 21st century. Business Horizons, 51(3), 235-243.
Kimberly, J. R., & Evanisko, M. J. (1981). Organizational innovation: The influence of
individual, organizational, and contextual factors on hospital adoption of technological
and administrative innovations. Academy of management journal, 24(4), 689-713.
Kotha, S., & Nair, A. (1995). Strategy and environment as determinants of performance:
evidence from the Japanese machine tool industry. Strategic management journal, 16(7),
497-518.
Kramer, J. P., Marinelli, E., Iammarino, S., & Diez, J. R. (2011). Intangible assets as drivers
of innovation: Empirical evidence on multinational enterprises in German and UK
regional systems of innovation. Technovation, 31(9), 447-458.
Lepak, D. P., Smith, K. G., & Taylor, M. S. (2007). Value creation and value capture: a
multilevel perspective. Academy of management review, 32(1), 180-194.
Lindgreen, A., & Wynstra, F. (2005). Value in business markets: what do we know? Where
are we going?. Industrial Marketing Management, 34(7), 732-748.
Lindgreen, A., Hingley, M. K., Grant, D. B., & Morgan, R. E. (2012). Value in business and
industrial marketing: Past, present, and future. Industrial Marketing Management, 41(1),
207-214.
Lobao, L., & Meyer, K. (2001). The great agricultural transition: crisis, change, and social
consequences of twentieth century US farming. Annual review of sociology, 103-124.
Luecke, R., & Katz, R. (2003). Harvard business essentials: managing creativity and
innovation. Harvard Business School Press.
MacDonald, J. M. (2011). Agricultural contracting update: Contracts in 2008. DIANE
Publishing.
Martin, S. K., & Jagadish, A. (2006). Agricultural marketing and agribusiness supply chain
issues in developing economies: the case of fresh produce in Papua New Guinea.
Matopoulos, A., Vlachopoulou, M., Manthou, V., & Manos, B. (2007). A conceptual
framework for supply chain collaboration: empirical evidence from the agri-food
industry. Supply Chain Management: an international journal,12(3), 177-186.
Mittal, B. and J. N. Sheth (2001). Value Space: Winning the Battle for Market Leadership.
New York: McGraw-Hill.

66
Moch, M. K., & Morse, E. V. (1977). Size, centralization and organizational adoption of
innovations. American Sociological Review, 716-725.
Möller, K. (2006). Role of competences in creating customer value: A value-creation logic
approach. Industrial Marketing Management, 35(8), 913-924.
Morningstar (2014). Morningstar Inc. Retrived from http://www.morningstar.com/
Moschini, G., & Lapan, H. (1997). Intellectual property rights and the welfare effects of
agricultural R&D. American Journal of Agricultural Economics, 79(4), 1229-1242.

Nair, A., & Filer, L. (2003). Cointegration of firm strategies within groups: a long‐run analysis
of firm behavior in the Japanese steel industry. Strategic Management Journal, 24(2),
145-159.
Niemi, J., & Xing, L. (2011). Market power in the retail food industry: Evidence from Finland.
In The 21st Annual IFAMA World Symposium.
Nord, W. R., & Tucker, S. (1987). Implementing routine and radical innovations. Lexington,
MA: Lexington Books.
O'Cass, A., & Ngo, L. V. (2011). Examining the firm's value creation process: a managerial
perspective of the firm's value offering strategy and performance. British Journal of
Management, 22(4), 646-671.
O'Hanlon, J., & Peasnell, K. (1998). Wall Street's contribution to management accounting: the
Stern Stewart EVA< sup>®</sup> financial management system. Management
Accounting Research, 9(4), 421-444.
Omidvar, V., Brewin, D. G., & Carlberg, J. G. (2006, February). Meat Processing in North
America: Successes, Failures and Opportunities. In Southern Agricultural Economics
Association Annual Meeting. Orlando, Florida.
Opara, L. U. (2003). Traceability in agriculture and food supply chain: a review of basic
concepts, technological implications, and future prospects. Journal of Food Agriculture
and Environment, 1, 101-106.
Oswald, D. J. (2011). Economic analysis of the US Department of Agriculture’s value-added
producer grants program (Doctoral dissertation, Kansas State University).
Oustapassidis, K., & Vlachvei, A. (1999). Profitability and product differentiation in Greek
food industries. Applied economics, 31(10), 1293-1298.
Parasuraman, A. (1997). Reflections on gaining competitive advantage through customer
value. Journal of the Academy of marketing Science, 25(2), 154-161.
Payne, A., & Holt, S. (2001). Diagnosing customer value: integrating the value process and
relationship marketing. British Journal of management, 12(2), 159-182.
Peterson, H. C., & Wysocki, A. F. (1997). The vertical coordination continuum and the
determinants of firm-level coordination strategy (No. 11817). Michigan State University,
Department of Agricultural, Food, and Resource Economics.

67
Phillips, J., Holcomb, E. J., & Kelley, K. (2007). Determining Interest in Value-added Planters:
Consumer Preference and Current Grower and Retailer
Perceptions. HortTechnology, 17(2), 238-246.
Priem, R. L. (2007). A consumer perspective on value creation. Academy of Management
Review, 32(1), 219-235.
Rintamäki, T., Kuusela, H., & Mitronen, L. (2007). Identifying competitive customer value
propositions in retailing. Managing service quality, 17(6), 621-634.
Sáenz, J., & Aramburu, N. (2010). Towards a New Approach for Measuring Innovation: The
Innovation-Value Path. Identifying, Measuring, and Valuing Knowledge-based
Intangible Assets: New Perspectives, 87.
Sawang, S., & Unsworth, K. L. (2011). Why adopt now? Multiple case studies and survey
studies comparing small, medium and large firms. Technovation,31(10), 554-559.
Sharma, A. K., & Kumar, S. (2010). Economic Value Added (EVA)-Literature Review and
Relevant Issues. International Journal of Economics & Finance,2(2).
Sirmon, D. G., Hitt, M. A., & Ireland, R. D. (2007). Managing firm resources in dynamic
environments to create value: Looking inside the black box. Academy of management
review, 32(1), 273-292.
Smith, J. B., & Colgate, M. (2007). Customer value creation: a practical framework. The
journal of marketing theory and practice, 15(1), 7-23.
Sparling, D., & Turvey, C. G. (2003). Further thoughts on the relationship between economic
value added and stock market performance. Agribusiness,19(2), 255-267.
Sporleder, T. (2006). Strategic alliances and networks in supply chains: knowledge
management, learning and performance measurement. Quantifying the Agri-Food Supply
Chain, 161-171.
Srivastava, R. K., Shervani, T. A., & Fahey, L. (1998). Market-based assets and shareholder
value: a framework for analysis. The Journal of Marketing, 2-18.
Stewart, G., 199 1. The Quest for Value, New York, Harper Business.
Streeter, D. H., Sonka, S. T., & Hudson, M. A. (1991). Information technology, coordination,
and competitiveness in the food and agribusiness sector. American Journal of
Agricultural Economics, 73(5), 1465-1471.
Sullivan, P. & Jikyeong K. (1999). Quick response adoption in the apparel manufacturing
industry :Competitive advantage of innovation. Journal of Small Business Management
37: 1–13.
Turvey, C. G., Lake, L., Van Duren, E., & Sparling, D. (2000). The relationship between
economic value added and the stock market performance of agribusiness
firms. Agribusiness, 16(4), 399-416.
U.S Department of Agriculture, Economic Research Service. (2013, March).Food Dollar
Series.. Retrieved from http://www.ers.usda.gov/data-products/food-dollar-
series/documentation.aspx#.U45uMvldWSo

68
USDA (2013)a. Economic Research Service. [ONLINE] Available
at: http://www.ers.usda.gov/. [Last Accessed 1/11/2014].
USDA (2013)b. Farm Structure and Organization. [ONLINE] Available at:
http://www.ers.usda.gov/topics/farm-economy/farm-structure-and-
organization.aspx#.UtGYrvRDuSo. [Last Accessed 1/11/2014].
Vanyi, N. (2012). Members Of A Supply Chain And Their Relationships. APSTRACT: Applied
Studies in Agribusiness and Commerce, 6(5).
Viaene, J., & Gellynck, X. (1995). Structure, conduct and performance of the European food
sector. European Review of Agricultural Economics, 22(3), 282-295.
Vorley, B. (2001). The chains of agriculture: sustainability and the restructuring of agri-food
markets.
Walbert, L. (1994). The Stern Stewart Performance 1000: Using Eva™ To Build Market
Value. Journal of Applied Corporate Finance, 6(4), 109-112.
Walter, A., Ritter, T., & Gemünden, H. G. (2001). Value creation in buyer–seller relationships:
Theoretical considerations and empirical results from a supplier's perspective. Industrial
Marketing Management, 30(4), 365-377.
Wilson, D. T., & Jantrania, S. (1994). Understanding the value of a relationship.Asia-Australia
Marketing Journal, 2(1), 55-66.
Zaltman, G., Duncan, R., & Holbek, J. (1973). Innovations and organizations.
Zéghal, D., & Maaloul, A. (2010). Analysing value added as an indicator of intellectual capital
and its consequences on company performance. Journal of Intellectual capital, 11(1), 39-
60.
Zehri, C., AbdelBaki, A., & BouAbdellah, N. (2012). How Intellectual Capital Affects a Firm's
Performance. Australian Journal of Business and Management Research, 2(08), 24-31.
Zeithaml, V. A. (2000). Service quality, profitability, and the economic worth of customers:
what we know and what we need to learn. Journal of the Academy of Marketing
Science, 28(1), 67-85.
Zona, F., Zattoni, A., & Minichilli, A. (2012). A Contingency Model of Boards of Directors
and Firm Innovation: The Moderating Role of Firm Size. British Journal of Management.

69
 FIGURES AND TABLES

Table 1- The concept of Value Creation

Definition Focus Skills Source

Value added is to economically add value to a
product by changing its current place, time, and -Enhance firm
-Innovation
form characteristics to characteristics more income
preferred in the marketplace. Coltrain et al. (2000); USDA;
-Coordination
-Cost minimization Anderson and Hanselka (2009)
Value added means adding value to a raw
product by taking it to at least the next stage -Cooperative alliances
-Focus on prices
of production
-Uniqueness of -High Quality
Value-adding if it satisfies either of two products and products
Value creation

conditions: (1) if one is rewarded for performing services

an activity that traditionally has been performed -Minimization of the
Amanor-Boadu (2003);
at another stage farther down the supply chain, -Obtain a rewards, level of risk.
Born and Bachmann (2006);
or (2) if one is rewarded for performing an higher returns
Evans (2006)
activity that has never been performed in the -Introducing new and
value chain -Uniqueness of hard to imitate
process products

Value creation firms are those engaged in value -Efficiency: Increase -Optimal capital use
add activities that uses capital efficiently. return with lowest (debt and equity)
In other words, value creation companies are level of capital
Our definition
those that produce goods or services efficiently employed -Maintaining
obtaining a benefit higher than the cost of efficiency
producing them, including the cost of capital -Sustainability over the time

70
 Table 2 – Value creation activities

Stage Value-adding activity Examples

Investment agricultural
biotechnology Development of new crops
Inputs
Vertical and horizontal Consolidation, mergers and
integration acquisitions

Agricultural Economies of scale Large-scale enterprises

Production Vertical integration Contracts

Product differentiation High-quality, new products

Processors
Investment in R&D Branded products

Services and supermarkets’ own

Product differentiation brands
Delivery to Collaborative relationship with
Consumer Upstream integration processors

71
 Table 3- Agri-food value chain description

Number
Stages Sectors Description of
companies
Chemicals Biotechnology firms
Supplier of fertilizer,
1 Inputs Agricultural Inputs 97
insecticides and farm
Farm- Construction equipment

2 Production Farm Products Agricultural Production 65

Soft drinks
Produces, markets, and
Brewers
3 Processing distributes beverages and 156
Wineries food products.
Packaged food
Food Distribution
Delivery to Distribution and sale of food
4 Grocery 136
Consumer products to end consumer
Restaurant

72
 Table 4 -Adjustment made to NOPAT

Component Effect Explanation

From Net Income from continuing operations
Interest expenses should be added back to
NOPAT
Interest Expenses + because the cost of capital is subtracted from
NOPAT
when computing the WACC

Restructuring, Merger and These items are added back since we

Acquisition + activated them as long-term assets
R&D

Source: Chmelikova (2008)

73
 Table 5- Adjustment Made to Total Assets to Calculate NOA

Component Effect Examples Explanation

Long-term Assets
Other properties This type of investment may be singled out since
Non-operating fixed assets - Other long-term assets it is not associated with the main operations
Construction in process of the firm.

Research and We activate expenditures in research and

development development
Intangible Assets + Restructuring, Merger because they constitute one of the sources of
and value creation, so R&D is considered an
Acquisition investment by and for the firm.
Short-term Assets
Short-term Investments These items should be singled out since this type
Non-operating current of
-
assets short-term activity is not used to generate value
Other Current Assets during the firm’s normal activities.

Source: Chmelikova (2008)

74
 Table 6a- Summary statistics

Variable Definition Stage1 Stage2 Stage3 Stage4 All

Mean 5,979 2,494 6,053 2,200 4,467
Net operating assets (see
NOA SE (10,832) (4,927) (13,993) (5,188) (10,534)
Table 5)
CV 1.81 1.98 2.31 2.36 2.36
Mean 680 210 639 245 480
Adjusted NOPAT (see table
AdjNOPAT SE (1,466) (456) (1,994) (599) (1,444)
4)
CV 2.15 2.17 3.12 2.44 3.01
Mean 264 43 257 107 188
EVA Economic Value Added SE (829) (302) (1,361) (425) (944)
CV 3.15 6.96 5.30 3.97 5.03
Mean -0.194% -0.806% 0.833% 3.434% 1.162%
MEVA Modified EVA SE (17.652) (19.650) (15.714) (18.421) (17.544)
CV 91 24 19 5 15
Mean 60.78% 52% 62% 67% 62%
1= MEVA>0
CEVA SE (0.489) (0.500) (0.487) (0.472) (0.487)
0= MEVA<0
CV 0.80 0.97 0.79 0.71 0.79
Mean 51 21 90 71 233
1= MEVA>0 at least 5 years
PEVA % 53% 32% 58% 52% 51%
0= MEVA<0 at least 5 years
Mean 6,637 3,067 4,207 3,444 4,382
Size Adjusted Total Assets SE (11,896) (6,192.531) (8,815) (8,491) (9,322)
CV 1.79 2.02 2.10 2.47 2.13
Mean 1.449 1.122 1.099 1.561 1.315
Leverage Total Liabilities/ Total equity SE (4.893) (2.603) (20.719) (11.458) (14.123)
CV 3.38 2.32 18.85 7.34 10.74
Mean 1,021 283 2,790 721 1,480
R&D, Intangibles and SE (2,596) (691) (8,359) (2,087) (5,390)
Innovation 1
Goodwill CV 2.54 2.44 3.00 2.90 3.64
%AdjTA 15% 9% 66% 21% 34%
Mean 880 283 2,773 721 1,442
SE (2,272) (691) (8,312) (2,087) (5,332)
Innovation 2 Intangible +Goodwill CV 2.58 2.44 3.00 2.90 3.70
%AdjTA 13% 9% 66% 21% 33%
Inn2/Inn1 86% 99% 99% 99% 97%
Mean 141 0.238 17 0.0375 38
SE (392) (1) (139) (0) (210)
CV 2.78 5.50 8.10 10.69 5.60
Innovation 3 R&D expenditures
% Net
34% 1% 3% 1% 11%
income
Inn3/Inn1 16% 1% 1% 1% 3%
Mean 73% 78% 65% 71% 70%
COGS Cost of goods sold SE (16) (27) (17) (23) (21)
CV 21.91 34.58 26.48 32.76 29.51
Companies 97 65 156 136 454
Observations 770 431 1,251 1,014 3466

Notes: Values are expressed in USD in million dollars. Standard deviation is in parentheses (SE). CV is the coefficient of
variation. % AdjTA is the variable expressed as a percentage of Adjusted Total Assets. % Net income is the variable expressed
as a percentage of Net Income. Inn2/Inn1 is Innovation 2 divided by Innovation 1. Inn3/Inn1 is Innovation 3 divided by
Innovation 1. MEVA is a variable measure in percentage. CEVA is a binary variable that takes value equal to 1 if the MEVA
is greater to one and zero, otherwise. PEVA is a binary variable that takes value equal to 1 if MEVA>0 for at least 5 years and
zero, otherwise. COGS is a percentage variable that measure the cost as a percentage of the revenues.

75
 Table 6b- Mean tests

Null Hypothesis
All
S1 = S2 S1 = S3 S1 = S4 S2 = S3 S2 = S4 S3 = S4
Variable equal
NOA 0.00 0.88 0.00 0.00 0.62 0.00 0.00

AdjNOPAT 0.00 0.53 0.00 0.00 0.67 0.00 0.00

EVA 0.00 0.87 0.00 0.00 0.24 0.00 0.00

MEVA 0.56 0.20 0.00 0.09 0.00 0.00 0.00

CEVA 0.00 0.78 0.01 0.00 0.00 0.01 0.00

Size 0.00 0.00 0.00 0.03 0.48 0.05 0.00

Leverage 0.70 0.59 0.87 0.98 0.59 0.44 0.86

Innovation 1 0.02 0.00 0.24 0.00 2.07 0.00 0.00

Innovation 2 0.06 0.00 0.52 0.00 0.15 0.00 0.00

Innovation 3 0.00 0.00 0.00 0.14 0.99 0.05 0.00

COGS 0.00 0.00 0.03 0.00 0.00 0.00 0.00

Notes: The table shows the p-values of the mean tests calculated to determine whether the value
outcomes for each stage are statistically different from each other for each variable of interest.
‘S1’ means Stage 1, ‘S2’ means Stage 2, ‘S3’ means Stage 3 and ‘S4’ means Stage 4. The null
hypothesis is that the mean of particular variable are equal between two stages. Rejecting the
null hypothesis means that there are statistically significant differences between two particular
stages (Rejecting the null are shown in italics)

76
 Table 7 – Regression results

(1) (2) (3) (4)

VARIABLES EVA MEVA CV PCV

Stage 1 -15.928 -0.601 0.055** 0.158***

(22.532) (1.017) (0.028) (0.028)
Stage 3 40.634* 1.586* 0.088*** 0.216***
(24.768) (0.931) (0.025) (0.024)
Stage 4 39.273** 4.716*** 0.146*** 0.218***
(19.805) (0.958) (0.026) (0.026)
LN(Size) 40.346*** 2.544*** 0.059*** 0.088***
(7.133) (0.126) (0.005) (0.007)
Leverage 3.645* 0.029 0.001 0.001
(2.133) (0.019) (0.001) (0.001)
Innovation 2 0.045*** 0.001*** 0.001*** 0.001
(0.011) (0.000) (0.000) (0.001)
Innovation 3 1.464*** 0.001 0.001*** 0.001
(0.136) (0.002) (0.000) (0.001)
COGS -27.550** -0.113 -0.036*** -0.009
(11.316) (0.327) (0.009) (0.009)
Constant -8.626 -1.744 0.417*** -0.116*
(55.598) (1.232) (0.054) (0.064)

Country Fixed Effects Yes Yes Yes Yes

Year Fixed Effects Yes Yes Yes Yes

P-value Stage 1 = Stage 3 0.027 0.005 0.120 0.007

P-value Stage 1 = Stage 4 0.010 0.000 0.000 0.008
P-value Stage 3 = Stage 4 0.952 0.000 0.002 0.893
R-squared 0.505 0.186 0.264 0.310
Observations 3,466 3,466 3,466 3,466

Robust standard errors in parentheses

*** p<0.01, ** p<0.05, * p<0.1
Note: In column 1 and 2, the appropriate econometric model is to use a probit model.
Both the probit and the OLS econometric techniques were used: The two methods give
essentially identical results, with insignificant differences in the variables. For
simplicity, only the results of the OLS regression are shown.

77
Figure 1- Marketing bill and farm share of the U.S. real food dollar, 1993 to 2011
Avg. 1993 to 2011

Farm Share 16.1

Marketing
83.9
Share

Source: USDA (2013)-Authors calculation

78
Figure 2- Simplified Agribusiness Value Chain Diagram

79
 Figure 3- Firm Size per Stage

80,000
Firm Size (USD million dollars)

60,000

40,000

20,000

0
1 2 3 4
Stage

Notes: The graph shows the distribution of firm size per stage. Firm size is measured as
adjusted Total Assets. It is expressed in million dollars.

80
 Figure 4-Relationship between Firm EVA and Firm NOA

10,000

Firm EVA (USD million dollars) 8,000

6,000

4,000

2,000

0
0 20,000 40,000 60,000 80,000
Firm NOA (USD million dollars)

Linear (Fitted Values)

Notes: The graph shows the positive relationship between firm EVA and Net Operating
Assets (NOA). Firm EVA is calculated as the average EVA for each firm across the
years. Firm NOA is calculated as the average NOA for each firm across the years. The
variables are expressed in million dollars.

81
 Figure 5- MEVA Distribution by Stages

f f

ccccc f

Notes: The graphs show the distribution estimation of MEVA for each stage. Small Firms are those firms that have
a MEVA value below the mean MEVA of the stage. Large Firms are those firms that have a MEVA value above the
mean MEVA of the Stage.

82
 Figure 6- Innovation Level per Stage

80,000
Innovation 1 (USD million dollars)

60,000

40,000

20,000

0
1 2 3 4
Stage

Notes: The graph shows the level of innovation for each observation in each stage.
‘Innovation’ 1 is composed of goodwill, intangible assets and R&D expenditures. It is
measured in million dollars

83
 Figure 7- COGS Distribution per Stage

0.04

0.03
Density (%)

0.02

0.01

0
0 50 100 150
COGS (%)

Stage 1 Stage 2 Stage 3 Stage 4

Notes: The graph shows the distribution of Cost of Goods Sold (COGS) for each
stage. COGS is expressed as a percentage of sales.

84
 Figure 8- Relationship Between COGS and MEVA

150

100
COGS (%)

50

0
-100 -50 0 50 100
MEVA (%)

Linear (Fitted Values)

Notes: The graph shows the positive relationship between Cost of Goods Sold (COGS)
and MEVA variables. Fitted values show the correlation between the variables. The
variable COGS is expressed as a percentage of revenues. MEVA is expressed in
percentage.

85
Figure 9- Alamo Value Creation levels 2003-2012

86