I NTERNATIONAL M ARKET S ELECTION
Testing the practical usefulness of the trade-off model A case example at Campina
by
Maarten Idink
University of Groningen
Faculty of Economics and Management and Organization
MSC International Business & Management
September 2007
Maarten Idink Meddoseweg 37A 7142 HA Groenlo M.Idink@live.nl
(06)53972972
Student number: s1483986
I NTERNATIONAL M ARKET S ELECTION
Testing the practical usefulness of the trade-off model A case example at Campina
This paper serves as a Master Thesis in the field International Business &
Management at the Rijksuniversiteit Groningen, the Netherlands, and was written during the period from June 1st to September 10th 2007, during an
internship at Campina.
Supervisor: I. Haxhi (Ilir)
Co-assessor: F.A.A. Becker-Ritterspach (Florian)
Maarten Idink, September 2007
The author is responsible for the content of this paper.
ABSTRACT
Literature on International Market Selection (IMS) contains many proposed models which make significant contributions but do not effectively address the IMS problem for individual companies. In this study we aim to (1) theoretically determine the relative market attractiveness of several target markets for Campina’s whey derived food ingredients and (2) test the complexity and practical usefulness of the trade-off model for individual companies, by mapping and demonstrating the use of this analytical approach based on readily available secondary data in a specific case example. The results of this study make clear that the trade- off model is a systematic, efficient and effective method for scanning a large number of markets, which is also able to account for a firm’s strategic orientation. For these reasons in particular, the trade-off model provides a useful basis for individual companies to determine the relative attractiveness of markets within a considered set, prior to the final in-depth analysis.
Key words: International Market Selection/ Relative market attractiveness/ International expansion
TABLE OF CONTENTS
1. INTRODUCTION ... 1
2. LITERATURE REVIEW ... 4
2.1. LINESOFRESEARCHONIMS... 4
2.2. CLASSIFICATIONOFIMSMETHODS... 5
2.3. NORMATIVEIMSMETHODS ... 7
2.4. UPDATINGEXISTINGFINDINGSONIMS ... 9
2.5. CONCEPTUALMODEL... 13
3. METHODOLOGY... 14
3.1. SAMPLEANDDATA... 14
3.2. CONSTRUCTSANDVARIABLES ... 15
3.3. METHOD... 17
4. RESULTS AND ANALYSIS ... 19
4.1. RELATIVEMARKETATTRACTIVENESSOFTARGETMARKETS... 19
4.2 COUNTRYSPECIFICCHARACTERISTICS... 22
5. DISCUSSION... 25
6. CONCLUSION... 30
REFERENCES ... 33
TABLES ... 38
FIGURES ... 41
APPENDICES ... 42
APPENDIXI ... 43
APPENDIXII... 44
APPENDIXIII ... 45
APPENDIXIV ... 46
APPENDIXV ... 47
APPENDIXVII... 49
1. INTRODUCTION
One of the very first concerns of firms that plan to expand their international activities is the choice of one or more countries as target markets (Papadopoulos & Denis, 1988). Several authors (Ayal & Zif, 1979; Welch & Wiedersheim-Paul, 1980; Papadopoulos
& Denis, 1988; Bradley, 1999; Philp & Chaiwun, 2003) showed that identifying the right market(s) for expansion is one of the most important steps in a firm’s export strategy.
However, faced with so many countries to evaluate, a manager can be overwhelmed with the diversity and complexity of alternative market opportunities. There are vast differences among countries in terms of size, income, language, infrastructure, market access, culture, and many other important dimensions. Yet, the differences and similarities among countries are fundamental in determining which markets are suitable for entry (Cavusgil, Kiyak &
Yeniyurt, 2004).
Cooper and Kleinschmidt (1985) showed that firms who select target markets from a total set of available countries, realize more rapid export growth than those who consider only a few (or no) alternatives. Screening a large set of countries however, asks for a systematic market selection approach. The need for such an approach is underscored by the complexity of current markets and the growing importance of global strategic positioning (Douglas & Craig, 1983; Albaum, 1998). Yet, the evidence is that markets are usually chosen without much systematic analysis (Cavusgil et al., 2004).
The International Market Selection (IMS) approach is an efficient and effective method for analysing and selecting a foreign target market(s) (Papadopoulos & Denis, 1988;
Douglas & Craig, 1992). Selection is used here to mean the choice among two or more alternatives; thus, the IMS process would normally follow the firm’s decision to start or expand international operations, and precede the final in-depth assessment of a specific market just prior to entry. Martín (2006), in a review of various IMS models, suggests that notwithstanding their many strengths, only one of the proposed IMS models to date offers a reasonably complete solution to the IMS problem: the trade-off model by Papadopoulos, Chen and Thomas (2002). Despite the fact that the model has been empirically tested and its predictive power points to practical benefits (Papadopoulos et al., 2002), the model is only been tested for exports from one country to another and not for exports from individual companies to potential target markets. In this study we aim to map and test the trade-off
model and its practical usefulness for individual companies for selecting foreign target markets.
To test the model in real case we looked for a company planning to expand its international activities. In this study we followed Campina, one of the largest dairy co- operatives of Europe, in its search for foreign markets in the whey derived food ingredients market. An examination of the distribution of whey production around the world leads to the conclusion that the global availability of whey is highly concentrated in the EU (40%) and the US (25%) (Rabobank, 2002). Several trends (e.g. rising rates of consumption, increasing per capita income and deepening distribution) however, lead to a growing importance of the whey derived food ingredients markets in developing countries, located in Asia, Latin America and the Middle East. Though consumption rates for dairy products are still relatively low in these countries, there is considerable scope for volume growth (Euromonitor, 2007). Because the whey derived food ingredients market is still subject to fierce price competition in both the EU and the US, many manufacturers are trying to expand their geographical coverage towards the abovementioned regions. To avoid the major competition in the EU and the US and to anticipate the future challenges developing countries offer, Campina is planning to expand its existing international activities.
Therefore, combining all these aspects, the main question addressed in this paper is:
Does the trade-off model provide a useful basis for individual companies to determine the relative attractiveness of markets within a considered set, prior to the final in-depth analysis? Within this question there are two primary elements which are reported on in this study. The first is to theoretically determine the relative market attractiveness of several target markets for Campina’s whey derived food ingredients; the second is to map the trade- off model approach and to test its complexity and practical usefulness for individual companies.
The methodology used, is in accordance with the study of Papadopoulos et al. (2002).
The trade-off model uses two key constructs, demand potential and trade barriers, as well as firm strategy as a contingency construct. Each key construct is measured by only four variables, resulting in simplicity and low cost. Eight countries were selected as the target markets’ set for testing the model in a six year reference, with 2000 as base year and 2005 as final year. Information was gathered from publicly available secondary data sources. Data for each variable was scaled using the procedure of Liander et al. (1967). Based upon scale intervals, a country score from 0 to 10 was assigned to each country, based on its absolute value. The scores of each country on each variable were then average to get a total score for
each of the potential and barrier constructs. The main results are presented by using two different approaches; the two-dimensional approach and the total score approach.
We expect not only to provide a list of countries that are candidates for market entry for Campina’s whey derived food ingredients. Rather, the objective was to test the complexity and practical usefulness of the trade-off model for individual companies by demonstrating the use of this analytical approach based on readily available secondary data in a specific case example. Next to testing the practical benefits of the trade-off model, we attempted to update existing findings on IMS to rekindle interest in IMS, which seems absolutely relevant since many firms look for opportunities for expanding their international activities. An update of existing findings on IMS can be found in the literature review. In here, we will discuss not only several normative IMS models, we will also shed light to the many proposed IMS taxonomies which emerged since the late 1980s.
After conducting the study, the results made clear that the trade-off model is a systematic, efficient and effective method for scanning a large number of markets, which is also able to account for a firm’s strategic orientation. For these reasons in particular, the trade-off model provides a useful basis for individual companies to determine the relative attractiveness of markets within a considered set, prior to the final in-depth analysis.
A literature review is carried out first, in which traditional lines of research and gaps in literature focusing on IMS are identified. Thereafter follows a presentation of the method used in this study. Third, the main results and analysis of the study are presented. In the discussion section our critical appreciation of the model is described. Finally, the last section addresses the main conclusions which can be drawn for the study.
2. LITERATURE REVIEW
2.1. LINES OF RESEARCH ON IMS
The essence of the international market selection (IMS) problem is developing an efficient and effective method for selecting a foreign target market(s) (Papadopoulos &
Denis, 1988; Douglas & Craig, 1992). Selection is used here to mean the choice among two or more alternatives. The IMS process would normally follow the firm’s decision to start or expand international operations and precede the final in-depth assessment of a specific market just prior to entry (Papadopoulos & Denis, 1988). According to Papadopoulos et al.
(2002), IMS is the first and most important step in export strategy. Identifying the right market(s) for expansion is important because (a) it can be a major determinant of success or failure, especially in the early stages of internationalisation; (b) target market decisions precede the development and thus influence the nature of foreign marketing programs; (c) the nature and geographic location of selected markets affects the firm’s ability to eventually coordinate its foreign operations; and (d) establishing bases at appropriate foreign markets can be a key ingredient in the firm’s global competitive positioning strategy (Ayal & Zif, 1979; Welch & Wiedersheim-Paul, 1980; Papadopoulos & Denis, 1988; Bradley, 1999;
Philp & Chaiwun, 2003).
In the light of its importance, IMS attracted significant research attention from the 1960s to the late 1980s. However, interest waned in later years, mainly because of the difficulty in developing IMS models that would be generalizeable to various industries (Douglas & Craig, 1992; Papadopoulos et al., 2002). In the abundance of earlier research, all covering some aspects of selecting the right markets, Leonidou, Katsikeas, & Samiee (2002) acknowledge IMS as one of the marketing strategy variables for which more research is needed (2002). In general, empirical studies addressing this important topic are scarce and, therefore, new investigations using empirical data are expected to emerge (Martín, 2006). In addition, when it comes to the IMS behaviour of the firm and its management, it has not received abundant coverage in the literature, either theoretically or empirically (Philp &
Chaiwun, 2003).
IMS research can be classified into two main groups: theoretical and empirical (Martín, 2006). Both groups, at the same time, can also be classified into two sub-groups, see
also Figure 1. The first subgroup in theoretical studies deals mainly with the classification of IMS methods (e.g. Papadopoulos & Denis, 1998). The second subgroup addresses the classical IMS normative methods and deals with the question of how foreign markets must be selected. Most IMS literature follows this approach (e.g. Douglas & Craig, 1982;
Papadopoulos, 1983; Cavusgil, 1985; Minifie, 1998).
The number of studies published, using the empirical approach is very limited. Two sub-groups however, can be distinguished. The first subgroup addresses IMS relationships as a determinant of other variables. Studies dealing with the impact of IMS on company performance (e.g. Christensen, da Rocha & Gertner, 1987; Brouthers & Nakos, 2005;
Martín, 2006) and literature addressing the determinants of IMS behaviour (e.g. Bradley, 1995) are examples of IMS relationship theories. The objective of the second sub-group of empirical studies is to describe or model the actual IMS procedure followed by firms. It usually identifies or describes the sequence followed by a sample of firms in their internationalization or expansion process in terms of markets entered (e.g. Dow, 2000;
Brewer, 2001).
--- Insert Figure 1 about here ---
2.2. CLASSIFICATION OF IMS METHODS
In this sub-section we will update the traditional inventory of IMS taxonomies. The following presentation of the most comprehensive IMS taxonomies may be seen as a contribution of this paper next to testing the trade-off model itself. In a relatively early work, Papadopoulos & Denis (1988) classified over 40 proposed models into qualitative and quantitative approaches. The former involves the rigorous and systematic gathering of analysis of qualitative information about one or a handful of potential country markets. The qualitative approach is however, open to the potential biased opinions of information sources and the subjective judgement of the decision maker (Papadopoulos & Denis, 1988). The latter is based on analysing large amounts of secondary statistical data about many (or all) foreign countries (Papadopoulos & Denis, 1988). The quantitative approach can be divided
into market grouping methods, which cluster countries on the basis of similarity and market estimation methods, which aim at differentiating markets on the basis of their potential.
Another comprehensive taxonomy of IMS models can be found by the work of Sheridan (1988). By taking into account the time required, the degree of formalization of the technique, the use of resources and the amount of information collected in the IMS decision- making process, Sheridan (1988) proposed the following IMS taxonomy: the intuitive approach, the unstructured data collection approach and the systematic method approach.
The intuitive approach and the unstructured data collection approach are comparable with the qualitative approach as proposed by Papadopoulos & Denis (1988). Sheridan’ systematic methods approach (1988) is classified into four broad groups: grouping methods; potential market methods; product-specific methods; and strategy-based methods and is comparable with Papadopoulos’ quantitative approach (1988).
Chen (1996) also presents an inventory and taxonomy of IMS approaches, which is an adaptation of the classification given by Papadopoulos and Denis (1988). Chen (1996) however, includes next to the market grouping and market estimation models, also decision making models to the quantitative approach. Decision-making models are concerned with establishing ordered sets of rules and procedures that describe how to go about the selection and make no use of statistical data-analysis techniques (Chen, 1996; Martín, 2006).
Albaum, Strandskov & Duerr (1998) distinguish between reactive and proactive market selection approaches. A reactive approach refers to a situation where the exporter acts passively in selecting a market by filling an unsolicited order or waiting at the initiatives of foreign buyers, importers or others who indirectly select the market for the firm. The domestic market is seen as starting point and expansion is based on market similarities.
Proactive behaviour is more market oriented. The exporter is more active in initiating the selection of foreign markets and the approach is more formal and systematic, using screening procedures or decision-making models.
Bradley (1999) described international market selection behaviour as being either opportunistic or systematic. Sometimes, a third market identification approach is added:
evolution from the opportunistic approach to the systematic approach. Opportunistic market selection occurs when a firm becomes aware of specific marketing opportunity through external stimuli, for example a foreign customer or an agent. In this situation, a foreign market opportunity is brought to the firm’s attention and the firm responds by entering that particular market. Systematic market selection occurs when a specific marketing opportunity comes as a result of systematically comparing potential markets. Systematic market selection
is a logical and more formal process, which usually starts with a preliminary screening of potential markets followed by a thorough investigation of both the industry and the firm’s sales potential in each market. Potential markets are classified by specific criteria and markets are prioritized for immediate study or for development at a later date. Finally, in evolution from opportunistic to systematic selection, a firm that is already aware of a specific marketing opportunity assesses a potential market and compares it with available opportunities in other foreign markets, before entering a particular country.
More recent classifications distinguish between systematic, semi-systematic and non- systematic approaches (Martín & Papadopoulos, 2005) according to two criteria: the degree of systematization of the search for information; and the degree of systematization of the information analysis. In non-systematic methods no formal approaches are used at any step of the process; in semi-systematic approaches only the information collection is approached systematically, and in systematic methods both the search for and analysis of information for IMS are done using formal collection and analytical techniques.
2.3. NORMATIVE IMS METHODS
Since the classification of Papadopoulos & Denis (1998) is followed in most of the IMS literature, we will make use of this taxonomy to describe some of the proposed IMS approaches and models. By reviewing existing IMS literature we conclude that quantitative IMS methods constitute the vast majority of IMS models. Despite the fact that qualitative approaches are systematic, they are open to the potentially biased opinions of those who provide the information (Papadopoulos & Denis, 1988). Several authors (Piper, 1971; Vogel, 1976; Johansson & Moinpour, 1977) found evidence that the outcomes of qualitative IMS methods can be largely inaccurate. Besides, qualitative approaches are limited by their nature of considering a limited number of countries (Papadopoulos & Denis, 1988). Quantitative approaches on the other hand, make use of a formalised statistical analyses based upon secondary data, making it possible to compare and screen a significant larger amount of countries (Papadopoulos & Denis, 1988; Papadopoulos et al., 2002). Papadopoulos & Denis (1988) divide the quantitative approach into two categories: market grouping methods; and market estimation methods.
Market Grouping Methods
Market grouping methods cluster countries on the basis of similarity. Clustering yields a group of countries with similar commercial, economic, political, and cultural dimensions (Cavusgil et al., 2004). These similarities not only help managers compare the countries, but also provide information on possible synergies among markets (Papadopoulos
& Denis, 1988; Cavusgil et al, 2004). The market grouping method clusters similar countries on the basis of their overall status, making no attempt at measuring demand levels.
Several shortcomings of the market grouping method have been identified. First of all, its exclusive reliance on aggregate, macro indicators at the neglect of specific- product/service market indicators (Cavusgil & Nevin, 1981; Douglas & Craig, 1983;
Papadopoulos & Denis, 1988; Cavusgil et al., 2004). Proponents of the market grouping approach state that product-specific variables should only be included during the later stages of the market opportunity analysis, when a reduced set of countries has been identified. In their opinion, such indicators are not readily available as secondary data, and require extensive and costly market research. Therefore, their inclusion is only appropriate during the final stage of research, prior to market entry (Cavusgil et al., 2004). Opponents of the market grouping method state that a preliminary market assessment based on aggregate data is still a necessary initial step (Papadopoulos & Denis, 1988). Another criticism on the market grouping method focuses on the assumption that countries are indivisible, homogeneous units and that within-country heterogeneity is totally ignored (Kale &
Sudharshan, 1987; Jain, 1996; and Cavusgil et al., 2004). Their assumption is that because similarities among groups of consumers within national boundaries are not considered, possible economies of scale in production, R&D, marketing, and advertising are lost. A third criticism on the market grouping approach is presented by Luqmani, Yavas & Quraeshi (1994) and is in line with other criticisms of approaches based on macro factors. They argue that international markets should be seen as a continuum rather than as entirely similar or dissimilar. In their opinion, the level of convenience demanded in products and services by consumers worldwide represents such a continuum. This perspective provides a rationale for constructing an index that places countries on a continuum rather than forcing them into distinct and mutually exclusive clusters.
Market estimation models
Methods in the market estimation category evaluate foreign markets on the basis of several criteria, and those with the highest scores are selected (Papadopoulos & Denis,
1988). Whereas the market grouping method groups countries on the basis of general similarity, this method differentiates countries on the basis of potential (Papadopoulos &
Denis, 1988; Papadopoulos et al., 2002). The criteria used in this approach vary across different methods and may include indicators of wealth, size, growth, competition and access. In comparison with the market grouping method, which is concerned with macro- indicators, the market estimation method is concerned with industry-specific indicators. The market estimation approach can be subdivided into two groups; those that evaluate the total potential of foreign markets (total demand methods) and those that focus on the import component of market potential (import demand methods).
The market estimation method also has its limitations. Due to the difficulty in determining which indicators are relevant to the product being reviewed, their main limitation is that there is no agreement on which variables to use (Papadopoulos & Denis, 1988; Papadopoulos et al., 2002). Besides, there is no agreement in assigning importance to the weights to the various variables (Papadopoulos & Denis, 1988; Papadopoulos et al., 2002). Some approaches limit the analysis to only one or two types of variables and therefore, do not capture the full set of influences (Day, Foc & Huszagh, 1988). As a remedy, other approaches have included large numbers of variables; these however, do little more then introducing redundancy (Nachum, 1994). Next to the general limitations of market estimation methods, the import demand models have the additional weakness that they cannot account for the market potential that is available to the exporter. In some cases, the exporter is able to compete against the domestic producers and/or can generate new customers, thereby stimulating primary demand (Papadopoulos & Denis, 1988; Root, 1994;
Papadopoulos et al., 2002). A final limitation of the market estimation method is that many models are not empirically validated. Many studies (e.g. Kumar, Stam & Joachimsthaler, 1994; Hoffman, 1997) have contributed to the approach by focusing on the analysis of using input indicators, they however, lacked to test and validate the indicators themselves (Papadopoulos et al., 2002).
2.4. UPDATING EXISTING FINDINGS ON IMS
Although literature on IMS contains a large number of useful theoretical and applied models, almost none of the models addresses the IMS problem effectively (Papadopoulos et al., 2002; Cavusgil et al., 2004). It goes beyond the scope of this paper however, to review
all proposed IMS models individually. Besides, such an inventory and assessment is already been provided by Papadopoulos & Denis (1988). While reviewing existent IMS literature, we found that none of the proposed models however, combines the characteristics of being industry specific, generalizeable, relatively simple to use, able to reflect the total demand available to the company and is empirically validated (Papadopoulos et al, 2002). Since the particular work of Papadopoulos & Denis (1988) only two new studies emerged, which both tried to overcome (some) of the limitations of previous work (Martín, 2006). Since updating the existing findings on normative IMS models itself may be seen as a contribution of this paper, both models are presented at some length.
Country clustering and ranking model
The method by Cavusgil et al. (2004) illustrates the application of two preliminary market assessment techniques and the synergy that arises by using them simultaneously:
country clustering and country ranking. In other words, the main limitations of country clustering are overcome by including a second technique to the model; country ranking.
The first step of the model is to group countries according to their similarity. The variables used were identified through a review of previous literature (e.g. Liander et al., 1967; Sethi, 1971; Douglas & Craig, 1995) and by incorporating several new variables. This resulted in a final set of five factors each representing one or several variables:
infrastructure; economics; standard of living; market size; and dynamism and future potential. After selecting the variables, a hierarchical clustering technique was selected to group the countries on the abovementioned variables.
While clustering identifies markets in terms of macro similarities, it does not indicate which countries may be more attractive. The objective of the second technique, ranking, is to order countries on the basis of aggregate demand level (Cavusgil et al, 2004). The ranking technique used here, is derived from an example of indexing offered by Cavusgil (1997a, 1997b). After filtering some variables out and incorporating new variables, Cavusgil et al.
(2004) make use of seven dimensions, each representing one or several variables: market size; market growth rate; market intensity; commercial infrastructure; market receptivity;
free market structure; and country risk. The dimensions are derived by standardizing the variables and then converting them to a scale of 1–100. Finally, the seven dimensions are combined into an overall market opportunity index by using corresponding weights (Cavusgil et al., 2004). Managers can choose to weight these factors to reflect the specific characteristics of their own business. The relative weights of the dimensions can also be
changed to specific industries and/or business, which are determined by a Delphi process of international business professionals and educators (Cavusgil, 1997a, 1997b).1
The abovementioned approaches of preliminary market assessment could be used independently; the paper of Cavusgil et al. (2004) clearly shows however, the synergy that arises by using country clustering and country ranking simultaneously. According to the authors; “clustering produces structurally similar groups, but does not reveal much about market potential… ranking identifies the most attractive markets, but does not help the manager understand similarities and differences among them… the methods in combination provide unique and highly valuable information for companies looking for potential target markets” (Cavusgil et al. (2004). Each country is clustered in a group with similar markets.
Afterwards, each country is ranked inside its cluster. Thus, the model provides an overview about market attractiveness within and among clusters.
Although the authors of the country clustering and ranking model claim to have overcome some of the limitations of previous work, it is clear that the model still hides some limitations itself. Its major shortcoming is its exclusive reliance on macro indicators at the neglect of specific product and/or service market indicators. This shortcoming is not solved with the incorporation of the ranking technique, since the model also ranks countries based upon macro-indicators. The authors state however, that managers can select additional dimensions and/or variables that more closely represent the desirable market characteristics for specific products or services. Besides, manager can also adjust the weights of the dimensions according to the requirements of their products or industry (Cavusgil et al., 2004). The problem is however, that the additional variables and weights have not been tested empirically before a manager decides to use them. Besides, the incorporation of the ranking technique leads to another limitation: the efficiency of the model. In total, the model consists of 35 variables. The overwhelming task for a manager to screen countries on 35 variables, will irrevocably lead to the exclusion of a number of potential target markets. Yet, as a preliminary market assessment method, the method should be able to screen many foreign markets. Therefore, the efficiency of the country ranking and clustering model can be seen as another limitation. In sum, the country clustering and ranking model still hides some limitations, the model is: (a) not industry specific; (b) not simple to use; (c) not able to reflect total demand available to the company; and (d) not empirically validated.
1 The index is available online through Michigan State University’s GlobalEDGE knowledge portal (www.globaledge.msu.edu)
Trade-off Model
The shift-share model of Green & Alleway (1985) has been the starting point for the trade-off model. Two key strengths of that particular model provide a strong base for further development: (a) the shift-share model’s core logic, which captures both market potential and import barriers; and (b) its underlying objective of developing a simple and flexible model which emphasizes industry-specific demand estimation (Papadopoulos et al., 2002).
The major steps undertaken in the development of the Trade-off Model consisted of: setting relevant criteria; specifying key constructs and their measures and allocating weights to them; setting rules for aggregating scale values and their weights; and using these to transform the data into an overall score. The trade-off model makes use of only those criteria, constructs and measures which have theoretical support (Papadopoulos et al., 2002).
The models key constructs are demand potential and trade barriers. While many researchers have identified trade barriers as important export determinants, it has hardly been accounted for in previous IMS models. The key constructs of the model are represented by the following variables: apparent consumption; import penetration; origin advantage; and market similarity, representing demand potential and tariff barriers; non-tariff barriers;
geographic distance; and exchange rate, representing trade barriers.
In addition to these market characteristics, the trade-off models also accounts for the firm’s strategic orientation. Despite the fact that this dimension has been largely neglected by previous research, the trade-off model has added strategic orientation as a contingency construct to guide the weighting of the key constructs and their variables. The rationale behind this is accentuated by the potential versus barrier specification of the model:
different firms with different needs would attach different degrees of importance to each side of the trade-off (Ekeledo & Sivakumar, 1998; Papadopoulos et al., 2002). According to Ayal
& Zif (1978) firms can adopt two main strategies: offensive and defensive. These strategies distinguish between firms that seek growth at their competitors’ expense and value opportunities more than being concerned about risks, versus firms that focus more on preventing competitors from making inroads on their market share. This is a dichotomous representation of a continuous variable, but for purposes of simplicity and given the needs of the model it was adopted as contingency variable in the study of Papadopoulos et al. (2002).
Data for each variable is scaled for assigning country scores on each construct. The results of the trade-off model are presented twofold. The first clusters countries by their values on each construct into even quadrants (low/low; high/low; low/high; high/high) based on a plotting technique. The second ranks countries by their overall value on both constructs.
2.5. CONCEPTUAL MODEL
Based on existing literature on IMS, an appropriate model for individual companies should meet the following criteria: (a) the model should use a multiple variable approach, since these produce more meaningful results (Baalbaki & Malhotra, 1993); (b) the model should use as few variables as possible to achieve simplicity and low cost (Papadopoulos &
Denis, 1988); (c) the model should be able to assess general environmental conditions, but also be able to assess the industry-specific conditions (Denis, 1978; Papadopoulos et al., 2002); (d) the model should be capable of accounting for the strategic dimension of the individual firm (Denis, 1978); and (e) the model should be grounded in theory (Papadopoulos & Denis, 1988; Papadopoulos et al., 2002). After reviewing various IMS models, we can state that, notwithstanding their many strengths, only one of the proposed IMS models to date offers a reasonably complete solution to the IMS problem: the trade-off model by Papadopoulos et al. (2002). To meet the abovementioned criteria of efficiency, only four variables were used for each of the two key constructs. The authors however, succeeded in selecting those variables in the light of the criteria of developing an industry- level screening model. Besides, the model has been empirically tested and the model is grounded in literature. Based on the abovementioned, the model was specified as shown in Figure 2 and acts as the conceptual framework for gathering the data of this study.
--- Insert Figure 2 about here
---
3. METHODOLOGY
3.1. SAMPLE AND DATA
The conceptual model, presented in the concluding section of the literature review, acted as a framework for gathering the data in a coherent fashion. Initially, the following countries were selected for the study: Argentina; Brazil; China; Malaysia; Taiwan; Thailand;
Iran; Saudi Arabia; and United Arab Emirates. Due to missing data, Taiwan had to be filtered out, leaving a final set of 8 target markets. According to Euromonitor’s report “The World Market for Dairy Products” (2007) emerging markets in Asia, Latin America and the Middle East are leading growth in dairy products. The research predicts that the abovementioned countries show the highest growth ratio compared to other countries worldwide. Several trends (e.g. rising rates of consumption, increasing per capita income and deepening distribution) have lead to the growing importance of the whey derived food ingredients in the markets concerned.
In accordance with Papadopoulos et al. (2002) we made use of a six year period. The year 2000 is selected as base year and 2005 as final year. According to Papadopoulos et al.
(2002) the exact time period is however, of no consequence for the model, since there is no clear guidance in the literature at this point. Yet, it is important to select a period long enough not to be influenced by lagged effects. On the other hand, the difficulty to forecast over a long period means that the effects of decisions drop off as the period gets to long (Papadopoulos et al., 2002). Some data however, was not available for the selected years. To determine market similarity of the potential target markets we made use of several World Bank World Development Indicators (worldbank.org). This report contains data however, which is not available for each selected year. In case of missing data we made use of the latest available data.
Data was collated from several statistical databases. Data concerning production, imports and exports was collected from the International Dairy Federation’s (IDF) national committees along with several national andinternational organizations, such as the Food and Agriculture Organization (FAO), the United StatesDepartment of Agriculture (USDA) and theStatistical Office of the European Communities(EUROSTAT). Besides,Zentrale Markt- und Preisberichtstelle (ZMP) alsoprovided part of the data concerned. Besides, we made use
of the databases of the World Bank and the World Trade Organization. Data concerning exchange rates was collected from x-rates.com. The distances between the world’s main seaports have been determined through Logisticsworld.com.
3.2. CONSTRUCTS AND VARIABLES
The trade-off model by Papadopoulos et al. (2002) acted as the framework for determining the model’s key constructs and their variables. The model’s key constructs are demand potential and trade barriers. Strategic Orientation was treated as the contingency construct to guide the weighting of the key constructs and their variables. The model’s weighting scheme is presented in the next sub-section.
Four variables were used for each of the two key constructs. The first variable of the demand potential construct was apparent consumption. Root (1994) recommends apparent consumption as the appropriate reflection of true market size in a given industry. Apparent consumption was measured by the country’s domestic whey production plus imports minus exports (App. Consumption = Domestic Prod. + Imports – Exports). The second variable was import penetration, which measures imports as percentage of apparent consumption [Import Pen. = (Imports / App. Consumption) * 100]. A high ratio means import market openness and low domestic producer competitiveness, signaling an attractive target market (UNCTAD, 1968). Origin advantage was the third variable of the demand potential construct. Origin advantage was measured by the exporting country’s share in a target market’s total imports [Origin Adv. = (Whey Exports NL. / Imports) * 100]. Companies from countries with a high overall share in the target country’s imports in a given industry enjoy advantages. Besides, strong trade relations between the exporting and importing countries, often lead to greater trade promotion effort and local representation (Alexandrides
& Moschis, 1977; Bilkey, 1987; Papadopoulos, 1999). Market similarity, the final variable of the demand potential construct, was measured through an overall score of four indicators (the indicators’ individual loadings are presented in brackets): Life Expectancy (0.77);
GNP/capita (0.87); Electricity Production (0.48); and Imports-to-GDP ratio (-0.48). Every individual indicator is scaled on a 0-10 scale first based on its absolute value, afterwards the indicators are multiplied with their loadings to get an overall score [Market Sim. = (Scale Value LE * 0,77) + (Scale Value GNP * 0,87) + (Scale Value EP * 0,48) + (Scale Value Imp. * -0,48)]. For a rational behind this variable see Sethi (1971) and Davidson (1983).
Tariff barriers was the first variable of the trade barriers construct. A weighted mean of annual tariff rates over the study period was used to measure this variable [Tariff Barrier = (Tariff rate ‘00 + Tariff rate ‘01 + …. + Tariff rate ‘05) / 6]. For a rationale behind this variable see Papadopoulos et al. (2002). Non-tariff barriers was the second variable of this construct. The study by Papadopoulos et al. (2002) made use of composite quantitative index of 20 barrier item to measure this variable. In this study however, we filtered out this variable since article 4.2 of the Agreement on Agriculture prohibits the use of agriculture- specific non-tariff measures. This means that border measures, other than normal customs duties, are no longer permitted (wto.org). In order not to disturb the plotting and weighting effects of the trade-off model, we choose to assign an average score of the other trade barriers construct variables as the score for non-tariff barriers. The third variable of the trade barriers construct was geographic distance. Geographic distance is directly related to transport costs and can act as a major barrier through its effect on export price. This variable was measured by determining the distance between the target market’s main seaports to Rotterdam, the exporting country’s main seaport (Alexandrides & Moschis, 1977; Aksoy &
Kaynak, 1994). The final variable of this construct was exchange rate. Given their volatility, currency exchange rates between the exporting and importing countries are a major risk element in exporting and can have a major impact on pricing and strategy. The exchange rate variable was measured by the percentage change in official exchange rate versus its previous year [Exchange rate = (Exchange rate ‘01 – Exchange rate ‘00) / Exchange rate ‘00] * 100].
For a rationale behind this variable see Papadopoulos et al. (2002).
The trade-off model treats strategic orientation as a contingency construct, which was used to guide the weighting of the potential and barriers constructs and their variables.
The rationale behind this is accentuated by the potential versus barrier specification of the model: different firms with different needs would attach different degrees of importance to each side of the trade-off (Ekeledo & Sivakumar, 1998; Papadopoulos et al., 2002). Firms can adopt two main strategies: offensive and defensive (Ayal & Zif, 1978). Firms with an offensive strategy favour the demand potential construct, because they would focus on market opportunities and put more effort into markets that may be difficult to penetrate but present strong potential. Firms with an defensive strategy on the other hand, favour the trade barrier construct, because they would focus more on markets that appear easier to penetrate and be more likely to be deterred from countries with high trade barriers. (Ayal & Zif, 1978;
Papadopoulos et al., 2002).
3.3. METHOD
The analysis of the data included several steps. The information gathered from statistical databases and other sources was transcribed into a matrix for each of the constructs, their variables, eight target markets and a six year reference. Data for each variable was scaled using the procedure of Liander et al. (1967), by subtracting the lowest country value from the highest and dividing the difference by 10, forming 10 equal scale intervals. Based on these scale intervals, a country score from 0 to 10 was assigned to each country based on its absolute value. The scores of each country for each variable were then averaged to get a total score for each of the potential and barriers constructs. High scores on the 0–10 scale correspond to high potential or low barriers.
The results of the trade-off model are presented by using two different approaches:
the two-dimensional approach; and the total score approach. The two-dimensional approach makes use of a plotting technique. Countries were plotted and grouped into four clusters (high/low potential and low/high barriers) in a 2x2 matrix. Target markets in the upper right (high potential/low barriers) offer the best export opportunities, those in the lower left (low potential/high barriers) are the least promising, and so on. The tests for the two-dimensional approach were carried out first and weights were not yet assigned to the constructs and their variables.
Next, weights were assigned to the constructs and their variables to develop total country scores that combine the scores on both constructs. Papadopoulos et al. (2002) designed hierarchical weighting schemes by assigning different weights to the potential and barrier constructs, with a sum equal to one, and then to each variable, also with a sum equal to one, see also Table 1.
--- Insert Table 1 about here
---
The scores of the two dimensions were summed to generate an overall composite score for each country. Three weighting schemes were used: Weq, weighting both constructs and all variables equally; Wdp, favouring demand potential; and Wtb, favouring trade barriers. The logic was to down weigh certain trade barrier variables in the Wdp scheme, and
vice-versa. Market similarity for example, was down weighted in Wdp, since more aggressive firms are also likely to be more adventurous and less driven by psychic distance.
In accordance to the study by Papadopoulos et al. (2002), target markets were ranked into four groups by their total score to reflect four levels of opportunity: high; medium-high;
medium-low; and low. Besides, countries are also ranked ordered by their total scores.
4. RESULTS AND ANALYSIS
4.1. RELATIVE MARKET ATTRACTIVENESS OF TARGET MARKETS
As foreshadowed, the first part of the research question sought to theoretically determine the relative market attractiveness of several target markets for Campina’s whey derived food ingredients. The results of the trade-off model are presented through two different approaches, the two-dimensional approach and the total score approach respectively.
Two-dimensional approach
The final set of 8 target markets was clustered by their scores on both constructs into even quadrants. The calculation of these scores can be found in the appendices. The results are shown in Table 2, target markets in the upper right (high potential/low barriers) offer the best export opportunities, those in the lower left (low potential/high barriers) are the least promising, and so on.
--- Insert Table 2 about here
---
The first cluster (low potential/high barriers) includes two countries, Iran and Thailand respectively. If we take a closer look at the two key constructs we can conclude that Iran scores low on demand potential mainly because the country has the lowest apparent consumption of all investigated target countries. Iran does not have any domestic production, imports are low and even a small amount of its imports is exported again. Another reason for Iran’s low score is its exchange rate volatility. Given its volatility, currency exchange rates between the exporting country, in this case the Netherlands and Iran are a (major) risk element in exporting and can have a (major) impact on pricing and strategy of the firm concerned. Thailand scores relatively low on apparent consumption and market similarity, resulting in an overall low score on the demand potential construct. Besides, Thailand also
has a low score on the trade barrier construct due to its enormous tariff rate of 30% for whey derived food ingredients.
The second cluster (low potential/low barriers) is composed by three countries which are relatively easy to enter but do not show high demand potential: Malaysia; Kingdom of Saudi Arabia; and the United Arab Emirates. The countries’ low overall score is mainly due to the fact that all three countries have a relatively small domestic market. Besides, also the exporting share of Dutch firms in the target market’s total imports is relatively low, which results in a low score on origin advantage.
The third cluster (high potential/high barriers) does not include any country. The fourth and final cluster (high potential/low barriers) is composed by three countries:
Argentina; China; and Brazil. All three countries show relatively high demand potential due to their high apparent consumption. Argentina and Brazil however, have a large domestic production which means that import penetration is low. China on the other hand, has no domestic production which means that the entire market depends on imports. If we take a look at the trade barriers construct, we can conclude that the tariff rate in Brazil is high compared to the tariff rates in Argentina and China. Transportation costs of bulky products to China on the other hand, are much higher due to its geographical distance. Finally, exports to Argentina include a risk element due to its volatile exchange rate.
To summarize, based upon the two dimensional approach three countries show the highest relative market attractiveness: Argentina; China; and Brazil. However, like we have seen in the abovementioned, countries in the same cluster may have different implications for the exporting firm depending on the individual variable values. Therefore, the individual variable values can help firms in developing more effective strategies for individual markets.
The country specific characteristics are addressed in the following sub-section.
Total score approach
Next to plotting the target countries on two separate dimensions, the trade-off model also makes it possible to rank countries on a single overall score. Ranking makes it possible to order countries in terms of their overall market attractiveness. The results of the total score approach are presented twofold. In accordance with the study of Papadopoulos et al. (2002), target markets are ranked into four groups by their total score to reflect four levels of opportunity: high; medium-high; medium-low; and low. Besides, the countries have been ranked ordered by their total score. If we take a look at the main results we can see that the
relative standings of the 8 target countries differ strongly depending of the relative weights, see also Table 3.
--- Insert Table 3 about here ---
The countries showing the highest overall market attractiveness with assigning weights equally (Weq) are, alphabetically: Brazil; China; Kingdom of Saudi Arabia; and The United Arab Emirates. The results for firms adopting an offensive strategy, thus favouring demand potential above trade barriers, show differences compared to the abovementioned results. Argentina, Brazil and China are the potential target markets showing the highest overall market attractiveness. For firms adopting a defensive strategy, thus favouring trade barriers above demand potential, China, Kingdom of Saudi Arabia and the United Arab Emirates show the highest overall market attractiveness. Table 4 is a simple, yet intuitive presentation of the rankings for illustrative purposes.
--- Insert Table 4 about here ---
To summarize, not surprisingly, the relative target market attractiveness differs depending on the weights assigned. Notwithstanding a firm’s strategy, China shows high overall market attractiveness and Iran shows low overall market attractiveness. It is striking however, that the overall highest scores can be found in the Kingdom of Saudi Arabia and the United Arab Emirates, which both did not show high market attractiveness based upon the two-dimensional approach. We also ranked countries ordered by their overall total scores representing their market attractiveness. This lead to the following results, for Weq: (1) Kingdom of Saudi Arabia; (2) United Arab Emirates; (3) China; (4) Brazil; (5) Malaysia; (6) Argentina; (7) Thailand; and (8) Iran. For Wdp: (1) Brazil; (2) Argentina; (3) China; (4) Thailand; (5) Malaysia; (6) Kingdom of Saudi Arabia; (7) United Arab Emirates; and (8) Iran. For Wtb: (1) Kingdom of Saudi Arabia; (2) United Arab Emirates; (3) China; (4) Argentina; (5) Brazil; (6) Malaysia; (7) Thailand; and (8) Iran.
4.2 COUNTRY SPECIFIC CHARACTERISTICS
Although the model aims to present its results through the two-dimensional and the total score approach, also the individual variables provide some meaningful additional insights for firms. Below the individual variable effects for each country are presented.
These can help Campina to gain more understanding of its target markets.
Argentina
The Argentinean whey market can be characterized by a very low import penetration ratio (0,24 percent). This means that the exporting firm faces very strong competition, not from other exporting firms, but from domestic whey producers in particular. Next to this, if we take a look at the ratios of total imports in 2005 versus 2000, these tell us that imports have grown significantly in most countries (e.g. China 1,49; Malaysia 1,58; Thailand 1,45;
Kingdom of Saudi Arabia 1,80), but declined in Argentina in the same period (0,80). It must be said though, that the origin advantage of Dutch export in the Argentinean whey market is high compared to the other target markets. Its exchange rate volatility however, seems to be a risk element.
Brazil
Also the Brazilian whey market, with an average import penetration of only 7,78 percent, seems to depend strongly on domestic production. Despite its low import penetration, its market size based on imports is still the third largest, right after China and Thailand. Besides, with an origin advantage of 10,62 percent Dutch whey exporting firms seem to enjoy some advantages compared to other exporting countries. The Brazilian tariff rate on whey derived products on the other hand, is one of the highest compared to other target countries and therefore has a negative effect on import prices.
China
Both approaches classified China as one of the most important target countries.
Imports have grown with a ratio of 1,49 in 2005 compared to imports in 2000. The import tariff rate is low and its exchange rate does not seem to hide any risks. Chinese imports from the Netherlands (6,17 percent) are still relatively limited however. Transportation costs to China are high due to the largest geographical distance to the main Chinese seaports compared to other potential target countries.
