RELATIONAL INVESTMENTS AND RESOURCE ALLOCATION IN THE BUYER SUPPLIER RELATION: THE MODERATING EFFECT OF SUPPLIER INTELLIGENCE

RELATIONAL INVESTMENTS AND RESOURCE ALLOCATION IN THE BUYER SUPPLIER RELATION: THE MODERATING EFFECT OF SUPPLIER

INTELLIGENCE

Thesis MSc Supply Chain Management

University of Groningen, Faculty of Economics and Business January 29, 2018 JESSICA SPEELMAN Student number 2158590 E-mail: j.j.speelman@student.rug.nl Supervisor/ University: dr. ir. N.J. Pulles

University of Groningen, Faculty of Economics and Business Co-assessor/ University:

dr. J. Veldman

2 ABSTRACT

The aim of this paper is exploring strategies that improve the resource allocation in a buyer-supplier relation, by investigating the influence of relational investment on the resource allocation. A detailed picture of the impact of supplier intelligence on this relation is provided. This study uses survey data collected by three firms in the food industry and 68 of their suppliers. First, no relation between social investments and the preferential allocation of innovation resources exists. Furthermore, it shows that social investments positively influence the resource allocation of physical goods, which becomes stronger for buyers with low supplier intelligence. This outcome shows that it is highly valuable for managers to make social investments. Since these investments have more influence to gain the preferential allocation of physical resources compared to economic investments. Furthermore, the relation between economic investments and preferential resource allocation becomes negative when buyers have low supplier intelligence.

Keywords: Supplier intelligence, preferential resource allocation, relational investments,

3

INDEX

3.2 The effects of economic and social investments with high/low supplier intelligence ... 9

4. METHODOLOGY ... 12

4.1 Research design ... 12

4.2 Sample and data collection ... 13

4.3 Measures ... 15

4.4 Analysis ... 18

5. RESULTS ... 21

5.1 Multigroup analysis – supplier intelligence ... 22

5.2 Multigroup analysis - manufacturer or retailer ... 24

5.3 Control variable ... 25

6. DISCUSSION ... 26

7. CONCLUSION ... 29

7.1 Managerial implications ... 29

7.2 Limitations and future research ... 30

8. REFERENCES ... 32

9. APPENDIX ... 36

9.1 Supplier - survey ... 36

9.2 Buyer - survey ... 38

9.3 Tables ... 39

9.4 Management summary (English version) ... 41

4

1. INTRODUCTION

Differences in performance within and between firms is still rather unexplained despite its frequent occurrence. Research shows that network resources affect a firm’s performance (Dyer & Hatch, 2006). Nollet, Rebolledo, & Popel (2012) state that 75% of suppliers offer products to their customers which are rare and not easy to replace. The offering of ‘rare products’ leads to high rivalry within the market to obtain the required resources. Especially if buying firms competitors source from the same suppliers (Pulles, Veldman, Schiele, & Sierksma, 2014). Therefore, companies should search for better resources compared to their competitors in the market to gain competitive advantage (Dyer & Singh, 1998; Ellis, Henke, & Kull, 2012; Pulles, 2014), and win the competition of the resource allocation which finally results in becoming a preferred customer (Pulles et al., 2014).

This study focuses on relational investments as a mechanism to improve the relation, and acquire better resources than competitors. In addition, the study is based on the social exchange theory. Previous research of Dyer & Singh (1998) suggests that giving and taking is the basis in a relation. Additionally, as stated by Moon & Bonney (2007), efforts can be made through social or economic investments. Research shows that it is essential to differentiate between social (intrinsic) and economic (extrinsic) investments, since the preferences of businesses differ per company (Dorsch, Carlson, Raymond, & Ranson, 2001).

The performance of a company highly depends on the suppliers in the value chain and their performance, and the knowledge of the buyer about the supplier (Choy, Lee, Lau, & Choy, 2005; Tanskanen, 2015). Supplier intelligence, which is the knowledge the buyer has about the supplier, can work as a source of innovation (Tanev & Bailetti, 2008). Furthermore, it affect the ratios in a relation (Tanskanen, 2015). However, ambiguity remains in supplier intelligence influencing the link between relational investments and the resource allocation of the supplier. Therefore, the following research question is developed:

How does the influence of economic or social investments on the preferential resource allocation of physical resources or innovation resources differ for high versus low supplier

intelligence?

5 resource allocation and if it differs depending on the different dimensions (social and economic) of investment. Besides, it is unclear whether relational investments will increase the change of becoming a preferred customer (Nollet et al., 2012). The influence of supplier intelligence is a contribution to the research field, since supplier intelligence is not yet included in a comparable model (Moon & Bonney, 2007; Nollet et al., 2012; Tanskanen & Aminoff, 2015).

This study will contribute to the Supply Chain Management (SCM) literature that focuses on the mechanisms that strengthen the relation between buyer and supplier by examining the link between relational investments and preferential resource allocation (Nollet et al., 2012; Pulles, Schiele, Veldman, & Hüttinger, 2016; Pulles et al., 2014). Furthermore, this study will investigate the influence of supplier intelligence on the relation between relational investment and resource allocation. An empirical study will be carried out within buying firms and their suppliers, using a survey carried out at two industrial and one retail company. The purchasing director of one of the companies said: "It is important to know how we can further improve our relations and strategies to gain the desired resources and stay ahead of our competitors." For purchasing professionals, it is important to gain further evidence on their contributions in strategic issues are critical to the firm (Carr & Pearson, 1999; Preston, Chen, Swink, & Meade, 2017). This study will help managers by providing insight in relational investments and their effects on the resource allocation of suppliers. Additionally, it will contribute to the knowledge on how to use their relational investment tactics to gain the desired and required resources and how to gain an advantage with their supplier intelligence. The key contribution is that this research identifies supplier intelligence as a mechanism to improve the effects of investments, except for social investments, on the preferential allocation of physical resources.

6

2. BACKGROUND

Buying firms are always trying to gain the best resources regarding quality and price, to enhance their competitive position (Ellis et al., 2012). Resources are defined as “tangible (physical) or intangible (innovation), human, financial organizational, intellectual, and physical entities which are available to the firm to increase their competitive advantage” (Skjott-Larsen, Schary, Mikkola, & Kotzab, 2007: 68). Physical resources are defined as tangible resources such as money, physical goods, love, and friendship (Cropanzano & Mitchell, 2005; Surroca, Tribó, & Waddock, 2010). Examples of innovation resources are enhanced knowledge and new patents which can be an advantage for other collaborations and can be leveraged towards a competitive advantage (Silver Coley, Lindemann, & Wagner, 2012).

A company’s competitive advantage derives from the resource and capabilities that the company can control. These resources are the basis for a company. The resource-based view suggests that firms gain and sustain competitive advantage by obtaining resources and capabilities that are rare in supply (Xiao, 2015). Since the resources are limited, other companies are searching for the same resources. A competitive advantage will only be achieved when the buying firm obtains better resources than its competitors (Capron & Chatain, 2008; Markman, Gianiodis, & Buchholtz, 2009; Schwieterman & Miller, 2016). Resources that contribute to a competitive advantage and are therefore mentioned as ‘better resources’ are valuable, rare, inimitable, and non-substitutable (Lin & Wu, 2014).

Suppliers do not treat all their customers equally, therefore customers have to find strategies to obtain the desired resource allocation (Nollet et al., 2012). Based on the literature, high relational investment leads to a higher commitment between buyer and supplier. Griffith, Harvey, & Lusch (2006) enhance this statement by suggesting that ongoing relations among supply chain members and investing in this relation, increase efficiency and effectiveness. Both Dyer & Singh (1998), and Griffith et al. (2006) suggest that putting the effort in a relation will lead to a more fruitful relation. However, this does not show if relational investments benefit to the supplier resource allocation.

7

3. HYPOTHESES DEVELOPMENT

3.1 Relational investment

Investments are the expenses in time, money, and life chances that are exchanged in structuring and maintaining relationships (Harris, O’Malley, & Patterson, 2003). Moon & Bonney (2007) developed the investment model in which relational investment is defined as investments which are of less value for other companies and specifically tailored to the relationship. Relational investments strengthen the relation. However, the social exchange theory assumes that actors are dependent on each other and learn from experiences. The comparison level explains the effect of previous experiences and the expectations of one actor against the other. If alternatives exist and those alternatives exceed the expectations and experiences of the current actor in the relation, an actor can decide to engage in a relation with the other actor and leave the current relation (Tanskanen, 2015). Some researchers have investigated the influence of investment on the commitment to buyer-supplier relations. Anderson & Weitz (1992) conclude that transaction specific investments (TSIs) lead to more committed suppliers. However, this research of Anderson & Weitz (1992) focuses on relations with an equal level of dependence. Research on the buyer-supplier relation with an asymmetrical level of dependence shows that bilateral TSIs can help to strengthen the relational commitment (Jap & Ganesan, 2000). Matevž & Maja (2013) state that both physical and investments in human resources are important in a relation to gain competitive advantage. Research suggests that relational-specific investments take on either a social or economic form (Dorsch et al., 2001).

3.1.1 Social investments

Direct investments into a relation, such as emotional effort, time, self-disclosure, love, status, and information can be described as a social investments (Dorsch & Carlson, 1996; Dorsch et al., 2001; Rusbult & Farrell, 1983). Silver Coley et al., (2012) extent his definition by adding employees' knowledge, such as specific insights about the product or process and specific know-how as social investments.

8 (Harris et al., 2003). In addition, the most important driver of being attractive was good communication between the buyer and supplier (Tanskanen & Aminoff, 2015). According to Kim (2000), social investments will stimulate the cooperative aspect of a relation and it is a useful instrument to control opportunism. Even though previous research state that social investments increase the commitment and cooperation between buyer and suppliers, and the attractiveness of a buyer, no research is conducted regarding the influence of social investments on the preferential resource allocation. Research of Morgan & Hunt (1994) suggests that commitment and trust are necessary for effective relation marketing, which entails activities directed toward establishing, developing, and maintaining successful relational exchanges. Since social investments lead to more commitment between buyers and suppliers, it might also positively relate to supplier resource allocations. Therefore, this research will investigate the following hypotheses:

Hypothesis 1a: Social investments of the buyer are positively related to supplier allocation of physical resources

Hypothesis 1b: Social investments of the buyer are positively related to supplier allocation of innovation resources

3.1.2 Economic investments

9 improvement of the buyer’s competitive advantage. The competitive advantage increases if the collaboration and communication between the buyer and the supplier are high. Economic investments could provide similar beneficial characteristics as observed by vertical integration (i.e., lower trading costs, better inter-firm communication, coordination, and quality). However, Carr & Pearson (1999) state that these investments may cause a loss of focus and diseconomies of scale and do not necessarily lead to competitive advantage (Carr & Pearson, 1999). Complementary to Whipple, Lynch, & Nyaga (2010), economic investments will guide the buyer-supplier exchanges. Next to that, it will reduce the uncertainty or conflict (Kotabe, Martin, & Domoto, 2003). Both social as economic investments will affirm the trust in each other and create a barrier to exit from the relation (Wilson & Jantrania, 1994). Furthermore, empirical results suggest that when suppliers do relation-specific investments, the supplier will be encouraged to suggest ideas for process innovations (Whipple et al., 2010). To conclude, previous research suggest that supplier development will positively influence the preferential resource allocation. Investing in supplier development can be seen as an economic investment since supplier development include direct capital investment, and investments in personnel. These supplier development programs, positively influence the reciprocity of the supplier. Therefore, this research investigates the following hypotheses:

Hypothesis 2a: Economic investments are positively related to supplier allocation of physical resources

Hypothesis 2b: Economic investments are positively related to supplier allocation of innovation resources

3.2 The effects of economic and social investments with high/low supplier intelligence

10 of knowledge on the performance of a supply chain. Research suggests that it can be used as a strategic resource to improve the performance of a supply chain (Hult, Ketchen, Cavusgil, & Calantone, 2006). However, Hult et al. (2006) cannot explain why companies which use knowledge as a strategic asset outperform other companies.

Yang, Yu, Liu, & Rui (2014) suggest that sharing knowledge will positively influence the relation between the buyer and the supplier. Furthermore, information sharing will positively influence the trustworthiness of a firm, especially for goodwill trust. In other words, goodwill trust will increase if buyer and supplier share a lot of information (Dyer & Chu, 2003). Goodwill trust will positively influence supplier resource allocations for buyers that contribute a large share for the supplier (Pulles et al., 2014).

Paulraj et al. (2008) show that communication between firms leads to an increase in behavioral transparency, which ultimately causes lower transaction costs and an enhancement of the transaction value. Not only transaction costs will decrease according to Choy et al. (2005), sharing knowledge within a supply chain is also crucial for the effectiveness of a manufacturer. The positive influence of sharing knowledge especially yields when the environment is dynamic and uncertain. Nonetheless, this research focuses on knowledge sharing, which is not the same as supplier intelligence, despite sharing knowledge within a buyer-supplier relation increases the supplier intelligence (Lau et al., 2010). Furthermore, this research does not explain whether it enhances the relation between relational investments and resource allocation. Knowledge sharing facilitates innovation and performance in organizations (Wang & Wang, 2012). People will not only increase their performance by means of their own experiences but also from the experiences of the supplier. In relations where direct interactions are common, the degree of knowledge transfer is higher, and the relation is stronger, which lead to better outcomes (Khan, 2014; Modi & Mabert, 2007).

11 investment is more effective compared to an investment without knowledge beforehand. A gap of the supplier can be a shortage in money, knowledge, or the wish to have more contact with the buyer. This can be related to the social exchange theory. If a buyer has the knowledge about the supplier, the buyer knows how interdependent the supplier and the buyer are on each other. Therefore, the buyer can exceed the expectations of the supplier which causes reciprocity (Tanskanen & Aminoff, 2015). Reciprocity will be enhanced if the supplier gives feedback and direction (Krause, Scannell, & Calantone, 2000). Furthermore, It is proven that a high degree of knowledge about the market gives a company more proactivity and flexibility to gain competitive advantage (Khan, 2014). According to this, it is hypothesized that supplier intelligence positively influence the relation between relational investments and resource allocation. The following hypotheses are drawn:

Hypothesis 3a: The effect of social investments on supplier resource allocation of physical resources will be stronger for buyers which have high supplier intelligence

Hypothesis 3b: The effect of social investments on supplier resource allocation of innovation resources will be stronger for buyers which have high supplier intelligence

Hypothesis 4a: The effect of economic investments on supplier resource allocation of physical resources will be stronger for buyers which have high supplier intelligence

Hypothesis 4b: The effect of economic investments on supplier resource allocation of innovation resources will be stronger for buyers which have high supplier intelligence

Having discussed all hypotheses, the following figure (1) shows the developed conceptual model which corresponds to the hypotheses.

12

4. METHODOLOGY

4.1 Research design

This study investigates the influence of relational investments on preferential resource allocation. To research this relation, both the supplier and the buyer perspective are necessary. The supplier perspective is needed to measure the level of resource allocation and the difference in physical and innovation resource allocations. On the other hand, the buyer perspective is needed to measure the degree of relational investments of this particular relation. Therefore, this study will investigate the dyadic data of the buyer and the supplier, which consider both perspectives. The buyer and supplier are treated as an independent source since the data for the variables "relational investment" are collected at the buyer and the variables “resource allocation” are collected at the supplier (figure 2). Therefore, the unit of analyses is the buyer-supplier relation.

Figure 2 Variables for each dyad

Since this study tests for hypothesized relations, that are exploratory, survey research should be considered (Karlsson, 2010). Therefore, a survey method is used to gather the necessary data. The survey consists of two different questionnaires, one directed at the supplier, and one directed at the buyer to collect data from both the buyer and the supplier since dyadic data is necessary to gain insights in both evaluations of the relation. The buyer fills in a survey for every supplier that provides data for this research. The data is gathered by using Qualtrics, an online survey tool.

13

4.2 Sample and data collection

To find a possible firm that was able to deliver data, a selection of companies is retrieved from the site of the Kamer van Koophandel (KvK), and by using own contacts. The companies are approached by telephone with the question whether are not they were willing to participate in this project. In most cases, the buyers ask to send a mail with more information. Eight companies were contacted in the first instance. From those companies, two directly said ‘no’ due to time constraints, or they did not want to share their supplier data. Finally, the companies who participate in this research come from personal network.

The primary data used for this research is collected in collaboration with three different companies. The first one (company 1), a large Dutch manufacturer of bread delivers their products to large Dutch supermarkets. This company was willing to support our research project. The second firm (company 2) that delivered a list of their suppliers and was able to assess the relation with their supplier is a nutrition company. The manufacturing industry is particularly interesting since the market pressures require an increase in product variety and complexity. Thereby, an increase in quality and reliability of the product is desired along with declining production costs. This demonstrates that a manufacturing organization cannot do it all by itself (Humphreys, Shiu, & Chan, 2001). Lastly, a retailer company (company 3) for nutritional products was able to cooperate. A retailer is attractive to include to see whether it can result in different outcomes for a retailer or manufacturer. As Pulles et al. (2016) suggest, more industries can increase the generalizability of the research, since industries can differ in preferential resource allocation.

14 During the data collection, 118 suppliers are asked to fill in the survey. After deleting partial responses, 68 surveys are filled in correctly and completely. This results in a supplier response rate of 57.63% (table 1). The buyers are able to fill in all the buyer surveys in a proper way, and therefore the 68 responses of the suppliers could be matched to the responses of the buyers.

A high number of non-respondents cause a sample not being representative regarding the population. It is highly debated what the limit is for non-respondents. The non-response effects could be resolved in two ways, namely increasing the response rate or identifying the non-respondents to control whether they differ from the non-respondents (Karlsson, 2010). In this research, the first method is chosen, to increase the response rate. This is done by sending a reminder after one week to kindly ask for a response. This increased the response rate from 46.7% to 79.5% for company 1. Since this already resulted in a high response rate, only the reminder mail was sufficient. However, this was not sufficient for the suppliers of company 2 and company 3. Both the suppliers of company 2 and company 3 are contacted by phone to ask if the mail was sent to the right person and if they are willing to participate. The reasons for the non-respondents are mainly a lack of time and not willing to cooperate in research like this. To test if there is a difference in the respondents who directly responded to the survey and the late respondents, a test for significant differences is done. The results of the test yields no significant differences amongst the two groups.

Suppliers contacted Suppliers responded Partially response Response rate

Company 1 44 35 7 79.55 %

Company 2 32 13 1 40.63 %

Company 3 42 20 10 47.62

Total 118 68 18 55.63% (Average)

Table 1 Response rate

15 The average length of the relation between the buyers and the suppliers is 13 years with a minimum of one and a maximum of 50 years. Next to that, the average length an employee was involved in the relation at the supplier's side was seven years. The average annual turnover for the suppliers is approximately € 120 million per year, based on the 46 suppliers who responded to that particular question. Most of the suppliers are located in the Netherlands which did not cause divergent results in the responses. Table 2 provides an overview of the profile of the sample.

Profile of the sample

Frequency Frequency

Length respondent involved in relation Length of relation

0-5 years 30 0-5 years 16 5-10 years 17 5-10 years 12 10-15 years 9 10-15 years 14 15-20 years 3 15-20 years 5 >20 years 7 >20 years 18 Unknown 2 Unknown 3

Country (supplier) Frequency Annual turnover (€) Frequency

The Netherlands 54 0-50 million 22

The United Kingdom 9 50-100 million 11

Germany 1 100-200 million 6

Belgium 1 >200 million 7

Ireland 1 Unknown 22

France 1

Table 2 Profile of the sample

4.3 Measures

The measurement and scales used in the survey are adopted mostly from previous quality research. Therefore, the variable validity can be ensured. In this study, we used multi-item scales to operationalize the key variables. A 5-point Likert scale with end points of “no, strongly disagree”/“very likely” and “yes, strongly agree”/“very unlikely" are used to measure the items. Except for the supplier operational performance which is measured by a 7-point Likert scale with end points of "poor performance" and "good performance." Likert scales are more readily analyzed and interpreted (Karlsson, 2010). The intelligence variables and the control variables are measured by using one or two items per variable and in different ways. A more extensive overview of the study’s measure is listed in table 3.

16 investments measure the degree of money put into the relation, and to which extent the buyer invest in human resources to benefit the relation with the supplier.

The dependent variables which are used in this study, supplier allocation of physical resources and innovation resources are measured based on scales of (Pulles, 2014). Supplier allocation of physical resources measures the extent to which physical resources are allocated to the buyer relatively to other buyers of the supplier. The variable items of supplier allocation of innovation resources measure the degree of information shared by the supplier compared to the degree of information sharing with other buyers.

The variable of the moderator "supplier intelligence" is measured using a question regarding share in turnover which are asked to both the supplier as the buyer. To measure the supplier intelligence, the differences between the answers from the buyer and the supplier are calculated. For example, the buyer and the supplier both give approximately the same answer regarding the share in turnover; the supplier intelligence is high. If the difference is high, supplier intelligence will be defined as low. The intelligence variable is measured based on one question. The share in turnover in which the supplier and buyer have to indicate the contribution of the share in turnover (in %). This question is based on Pulles et al. (2014).

Since the dependent variables in this research are explained by exogenous factors not captured by the scope of the model, additional a control variable is incorporated. This control variable is the length of the relation and is measured by using one item. The length of the relation is simply measured by asking the supplier how long the firm supplies the customer.

Table 3 Measurement items

Variable Measurement item (1 = strongly disagree; 5 = strongly agree) Item

loading

Social investments

Liu et al. (2012) We have invested a great deal in building up the relation with this partner. If we stopped working with this partner, we would waste a lot of resources and knowledge specifically tailored to this relation and suiting their needs.

0.949

Cronbach’s alpha 0.866 0.928

Composite reliability 0.937 Average variance extracted 0.882

Economic investments

Liu (2012) If we switch to another partner, we would lose a lot of investments we have made in this relation.

We have made a substantial investment in personnel development dedicated to this partner.

0.553 Cronbach’s alpha 0.553

Composite reliability 0.761 0.981

Average variance extracted 0.632

Preferential resource allocation (physical)

Pulles et al. (2014) We grant this customer better utilization of our production facilities.

We would choose to give this customer priority in the allocation of our products in the case of extreme events (e.g., natural disasters).

We allocate our scarce materials to this customer in case of capacity bottlenecks.

0.852

Cronbach’s alpha 0.814 0.843

Composite reliability 0.889

Average variance extracted 0.727 0.862

Preferential resource allocation (innovation)

Pulles et al. (2014) We are more willing to share key technological information with this customer. We share our best ideas with this customer first.

We dedicate more innovation resources to the relation with this customer.

0.912

Cronbach’s alpha 0.871 0.896

Composite reliability 0.902 0.830

Average variance extracted 0.754

Supplier intelligence

4.4 Analysis

To test the hypotheses, the partial least squares (PLS) approach to structural equation modeling (SEM) is followed with SmartPLS 3.0 as the most appropriate software. There are two approaches to SEM. The traditional covariance-based approach to SEM (CB-SEM) and the partial least squares approach to SEM (PLS-SEM). The choice to go for the (PLS) approach is based on the theoretical positioning of this paper, which is in nature explanatory. PLS is ideally for models with latent variables (Birkinshaw, Morrison, & Hulland, 1995). The research objective is one assumption for considering SEM. Next to the research objective, PLS-SEM is generally more powerful if the sample size is small, which is the case (n = 68). Since the sample size is lower than 250, the PLS-SEM approach is recommended (Hair, Hult, Ringle, & Sarstedt, 2014; Reinartz, Haenlein, & Henseler, 2009). The chance of not getting any results is possible by using CB-SEM with a small sample size. Furthermore, CB-SEM and PLS-SEM both have minor biases in their parameter estimates, and none of the variables follow a normal distribution, the PLS approach is chosen.

To ensure the reliability and validity of the measurement model, a number of tests was conducted. The key criteria are included in table 3. The Cronbach’s alpha for economic investments does not achieve the threshold of 0.7. However, the item responsible for the low Cronbach’s alpha is not deleted since the composite reliability is sufficient for this variable. The indicators outer loadings are all higher than 0.708, except the first item of economic investments. Again, this one is not deleted since it will decrease the measurement's content validity. Convergent validity is some sort of robustness check which indicates the correlation with an alternative measure of a similar variable. The average variance extracted (AVE) is a frequently used measure in establishing the convergent validity. The threshold is 0.5, which indicates that the majority of the variance in the indicators is explained by the variable (table 3).

19 Mean, Standard deviation, Squared AVE and Correlation table

Mean St. dev 1 2 3 4

1 Economic investment 2.72 1.287 0,795

2 Preferential resource allocation (innovation)

3.86 0.836 0.219 0.868

3 Preferential resource allocation (physical) 3.82 0.865 0.295 0.641 0.852

4 Social investment 3.38 1.015 0.566 0.096 0.339 0.939

Table 4 Means, Standard deviations, Squared AVE and Correlations table

Note: St. dev = standard deviation. Mean = sample mean. Numbers on the diagonal in bold show the square root of AVE. Numbers below the diagonal that are not bold show the correlations.

The second approach in assessing discriminant validity is testing the cross-loadings of the indicators. The cross-loading is an indicator's correlation with other variables in the model. The cross-loadings of all latent variables should be lower than the factor loading of its assigned latent variable (table 5). The cross-loading is for one item (economic investments), higher than its loading. However, this item is not deleted or changed, since the composite reliability is high enough and it meets the Fornell-Larcker criterion. Despite the lower loading of the indicator than its cross loading, previous research has proven that these measure are suitable (Liu et al., 2012; Pulles et al., 2014). Social investments Economic investments Preferential resource allocation (physical) Preferential resource allocation (innovation) RelaInvSoc1 0.949 0.505 0.356 0.049 RelaInvSoc2 0.928 0.563 0.274 0.139 RelaInvEco1 0.845 0.547 0.119 -0.016 RelaInvEco2 0.436 0.982 0.300 0.245 PrefResourceAlloPhys1 0.344 0.279 0.854 0.501 PrefResourceAlloPhys2 0.267 0.242 0.842 0.604 PrefResourceAlloPhys3 0.239 0.225 0.861 0.542 PrefResourceAlloInno1 0.079 0.226 0.519 0.931 PrefResourceAlloInno2 0.097 0.173 0.659 0.878 PrefResourceAlloInno3 0.079 0.025 0.608 0.790

Table 5 Factor loadings and cross-loadings

To assess the common method bias, the collinearity VIF is used in SmartPLS. The common method bias occurs if variations in responses are a result of an instrument rather than the actual effect carried by the respondent which the instrument attempt to expose. Mostly, this does not happen with formative factors. If the inner VIF values are lower than 3.3, no common method bias exists. The test is done for all the other latent variables to check if a common method bias exists. All results are lower than 3.3, and therefore no common method bias exists.

20 inclusion in the path model is definite and the model is shown to be sufficient in continuing the analysis (Hair et al., 2014).

Since the measures are considered reliable and validate, the assessment of the structural model is the next step. The effects of relational investments dimensions on supplier resource allocation (H1a – H2b) are tested by using the whole sample. Since the relations between the variables are unknown, there is a need to estimate the unknown parameters, which is done by a path analysis in SmartPLS. The coefficients of determination (R2), path coefficients, and the effect size are determined. To check whether the relations were significant or not, a bootstrapping is done with 5000 subsamples.

It has been tested if the effects of relational investments differed for groups of buyers with high or low supplier intelligence (H3a- H4b). To test the differences amongst the degree of supplier intelligence, pre-defined groups are made. To make the groups, the respondents were split based on their level of supplier intelligence. The median of supplier intelligence was taken as the measure for the groups. The descriptive statistics of the groups based on supplier intelligence are mentioned in the results section. The multi-group analysis is conducted to test if pre-defined groups have significant differences in their group-specific parameter estimates. Several parameters have to be specified, namely two group-specific numbers of observations, path coefficients and standard errors. These are used as input to calculate if there is a significant difference across the group.

21

5. RESULTS

After confirming the reliability and validity of the measurement model, the structural model is analyzed to obtain the results of this study. First, we tested the effects of social and economic investments on the resource allocation (i.e., H1a-H2b). Then, we tested how the influence of the relational investment dimensions on the resource allocation differed for groups with high and low supplier intelligence.

As figure 3 shows, social investments are positively related to supplier allocation of physical resources (b=.253, t=1.611, non-significant (n.s.)), but negatively related to supplier allocation of innovation resources (b=-0.041, t=0.211, n.s.). The first effect is as hypothesized. However, H1a and H1b are not supported because these effects are non-significant. The effect of social investments on supplier allocation of innovation resources is negative instead of the hypothesized positive relation. Both H2a and H2b are not supported since the effects are non-significant, although the relations are as hypothesized since economic investments have a positive effect on supplier allocation of physical resources (b=0.152, t=0.708, n.s.) and a positive effect on supplier allocation of innovation resources (b=0.242, t=0.833, n.s.).

In addition, the R2 of the dependent variables are 0.131 and 0.049 for respectively supplier allocation of physical resources and supplier allocation of innovation resources. Hair, Ringle, & Sarstedt (2011) describe these values as weak coefficients of determination. The variance observed in the dependent variable is not that well explained by the independent variables. These low values indicate that 13.1 percent of the variance in the preferential allocation of physical resources is explained by the independent variables and 4.9 percent of the variance in the preferential allocation of innovation resources is explained by the independent variables. This shows that a major part of the dependent variable is still unexplained.

22 Figure 3 Results of Structural Equation Model

Note: p < 0.05, n.s. = not-significant, t-value between parentheses

5.1 Multigroup analysis – supplier intelligence

To test the effect of supplier intelligence, the sample is divided into two subsamples with the median of supplier intelligence as the critical boundary. This results in two groups, one with 36 respondents and one with 32 respondents. The first group ‘high supplier intelligence’, in which the supplier and buyer responded with a score of less than 6 points difference on the question regarding share in turnover (“Please indicate the share (in %) that this customer has in your firm’s turnover.” (supplier) / “Please indicate your firm’s share (in %) in this supplier’s turnover.” (buyer)). The group ‘high supplier intelligence’ has an average supplier intelligence score of 3.03. The minimum score on the supplier intelligence variable was 0, and the maximum is 29. The ‘low supplier intelligence group’, has an average of 13.78. For a multigroup analysis, the samples can be considered relatively small. The number of observations in each group, when conducting PLS-MGA, also has a requirement regarding the minimum sample size. As the maximum number of arrows pointing at a latent variable is two, 20 observations per group are needed, according to the ten times rule (Barclay, Higgins, & Hompson, 1995). To test for variance between the two groups, a test which assures that possible differences are not caused by differences in the measurement model is conducted. Second, the results of the path coefficients were examined of both the groups to check whether different results yielded for the two groups, high supplier intelligence and low supplier intelligence. Third, the significance level is determined by conducting a bootstrapping test with a subsample of 5000.

23 measurement invariance test in SmartPLS, the outer loadings are compared. This measure showed that there was no significant difference between the outer loadings. Therefore, measurement invariance can be excluded (see Appendix table A1).

Hypothesis H3a examines the degree to which supplier intelligence moderates the effect of social investments on the preferential allocation of physical resources. The path coefficients were used to compare the results between the group with high supplier intelligence and the group with low supplier intelligence. The path coefficient of these relations are respectively 0.114, and 0.398 which yields a difference of 0.284 (t=0.893, and p=0.375). This indicates no significant difference in the effect of social investments and preferential allocation of physical resources between buyers with a high supplier intelligence and low supplier intelligence. Therefore, hypothesis H3a is not supported.

As shown in figure 4, the degree of supplier intelligence affects the relation between social investments and the preferential allocation of innovation resources in a different way. The results show that the relation between social investments and preferential allocation of innovation resources is negative for both groups (respectively -0.064, and -0.045). The difference (0.019) between the groups is not significant, and therefore H3b cannot be accept (t=0.051, and p=0.959).

Hypothesis H4a investigates if for buyers with high supplier intelligence the impact of economic investments on the preferential allocation of physical resources is stronger than for buyers with low supplier intelligence. In the association between economic investments and preferential allocation of physical resources, the corresponding coefficients are 0.299 for the group with high supplier intelligence and -0.025 for the group with low supplier intelligence. The resulting t-value is 0.727 which yields a p-value of 0.470, thus indicating there is no significant difference in the effect of economic investments on the preferential allocation of physical resources for the group with high and low supplier intelligence.

24 The model of the group with high supplier intelligence accounted for 14.6 percent of the variance in innovation resources and 14.4 percent of the variance in physical resources. The model of the low supplier intelligence group accounted for 5.2 percent of the variance in the innovation resources and 14.7 percent in the physical resources.

5.2 Multigroup analysis - manufacturer or retailer

To test whether the result differs for manufacturing companies or retailers, we divided the sample into two subsamples. This resulted in two groups, one with 48 respondents and one with 20 respondents. The first group was the group of the manufacturers and the second group consisted of the retailers. For a multigroup analysis, the samples can be considered relatively small, which was also the case in the multigroup analysis for supplier intelligence. To test for variance between the two groups, a test to assure that potential differences are not caused by differences in the measurement model is conducted. Second, the results of the path coefficients (beta) were examined of both the groups to check whether different results yielded for the two groups. Third, the significance level is determined by conducting a bootstrapping test with a subsample of 5000.

The first step tests for a significant difference between the responses of the two groups in the acquired responses. Differences between the two groups can also be caused by differences in the scale. Therefore, measurement invariance has to be excluded. By conducting the measurement invariance test in SmartPLS, the outer loadings are compared. This measure showed that there was no significant difference between the outer loadings. Therefore, measurement invariance can be excluded (see appendix table A3).

The results do not show a significant difference between the two groups, except for the relation between social investments and preferential allocation of physical resources for the Figure 4 PLS-MGA high/low supplier intelligence

25 retailer group. This relation is significantly positive (p= 0.014). Furthermore, the results show a negative relation between economic investments and preferential allocation of innovation resources (p=0.052, significant). Furthermore, the model of the group with manufacturers accounted for 13.5 percent in the innovation resources and 8.9 percent of the variance in the physical resources. The model of the group with retailers accounted for 29.9 percent of the variance in innovation resources and 28.8 percent of the variance in physical resources.

5.3 Control variable

To check whether the control variable ‘length of the relationship’ influences the relations in the model, a multigroup analysis is conducted. The sample is divided in two groups based on the median of the variable giving a group with suppliers who supplied the buyer for 10 years or shorter (n=35) and a group with suppliers who supplied the buyer for more than 10 years (n=30). The median of the variable is 10, the minimum length of a relationship in the sample is 1 year and the maximum is 75 years. Unfortunately, three suppliers did not answer the questions regarding the length of the relationship. However, the control variable of the length of the relationship did not show any significant different results (see Appendix table A5 and A6). Figure 5 PLS-MGA retailer and manufacturer

26

6. DISCUSSION

This study hypothesizes that social investments positively relates to supplier resource allocation since the social exchange theory emphasizes the positive influence of doing social investments. A positive effect of economic investments is also expected since economic investments increase the trust in each other and therefore causes a barrier to exit from the relation. Besides, supplier development can be seen as an economic investment. The supplier development programs positively influence the reciprocity of the supplier. Furthermore, high supplier intelligence would positively influence the relation between investments and preferential resource allocation. The mixed results, which are contradictory to the expectations of this study, will be discussed in this section.

First, the results indicate no relation between social investments and preferential allocation of innovation resources. As discussed by Harris et al. (2003), Kim (2000), and Tanskanen (2015) this is contradictory to the expectations, since previous research suggested that social investments would enlarge the commitment and trust, and therefore increase the chance of receiving a good treatment from the supplier. As suggested in this study, the presence of high commitment and high trust in a relation is necessary to achieve preferential allocation of innovation resources. However, this research did not include commitment and trust; therefore, it cannot be included in the model to test if it would have a mediating effect. Morgan & Hunt (1994: 34) state: "Relation commitment and trust develop when firms attend to relations by providing resources, opportunities, and benefits that are superior to the offerings of alternative partners." Commitment and trust are necessary when firms are in a relations and communicating valuable information. This valuable information can include expectations, market intelligence, and innovations.

27 decisive in this relation. The supplier also requires knowledge about the needs of the buyer. If the supplier knows the innovation is of value for the buyer, the supplier becomes more likely to share the innovation with the buyer.

Thirdly, the relation between social investments and the preferential allocation of physical resources reacts differently on supplier intelligence compared to the other relations. The relation between economic investments and preferential resource allocation is negative, where the relation between social investments and the preferential allocation of physical resources remains positive. Contrary, it can be concluded that knowledge becomes more important if money is involved in the investment. Economic investments are more efficient when supplier intelligence is high. However, social investments do not necessarily require high supplier intelligence. A possible explanation for this could be the lack of experiences and expectations in the relation. If the supplier does not have expectations, a social investment can be absorbed positively and more surprisingly. Such an investment can have a bigger impact and therefore be rewarded with preferential allocation of physical resources. Berg & Clark (1986) indicate that the interdependency in a relation is most important. Tanskanen (2015) explains dependency as one of the important factors of the social exchange theory. If the supplier is dependent on the buyer, it does not matter that the buyer does not have high or low supplier intelligence. The supplier does not want to lose the buyer. In other words, if the supplier intelligence is low, but the supplier is highly dependent on the buyer, the social investments will automatically lead to a preferential treatment. Previous research emphasize this by stating that collaborative relations will lead to more advantages than a transactional relation (Whipple et al., 2010). Transactional relations are relations where the level of interdependence is low.

Another explanation for this finding comes from Blau (1964) who states that parties remain in a relation, unless if there are other possibilities since it will be rewarding to do so. Continuing the relation with each other could already be a big enough ‘offer’ for a supplier to reward the buyer with preferential resource allocation and investments are therefore not necessary to improve the resource allocation.

29

7. CONCLUSION

This study provided insights into the effects of economic and social investments on preferential resource allocation and the influence of supplier intelligence on this relation. This research contributes to the existing literature in multiple ways.

First of all, the previous article of Humphreys et al. (2001), Lau et al. (2010), and Moon & Bonney (2007) only researched the influence of relational investments in general, without making a distinction between social and economic investments. Furthermore, the influence of relational investments on the preferential resource allocation, in which this research also emphasizes the different dimensions of relation investments and the different effects of these dimensions on the preferential resource allocation, has not been investigated yet. By differentiating between the social and economic dimension of relational investments, it highlights the importance of investing in either both or one of the dimensions depending on the degree of supplier intelligence. Contrary to the expectations, a social investment has a contradictory effect on the preferential resource allocation of innovation resources.

Second, this research contributes to the existing literature by taking a dyadic perspective on the relation between preferential resource allocation of the supplier and the relational investments of the buyer. Unless the fact that the relations are not significant, positive relations exist as discussed in the discussion section.

The use of supplier intelligence is new in this field of research and therefore a contribution to the existing literature. Despite the absence of significant differences between the two groups, this research shows that supplier intelligence is a relevant moderator to include in models in the field of supply chain management since unexpected results emerged.

7.1 Managerial implications

The previous review of literature show that social investments increase the commitment and cooperation between buyer and suppliers. Furthermore, the attractiveness of a buyer increases if the buyer does social investments. The aim of this study is measuring the impact of social investments on preferential resource allocation. Moreover, economic investments were investigated. The findings of previous research showed that economic investments do have a positive influence on the relation, however, it will not necessarily lead to competitive advantages. This research adopted and explored a cross-disciplinary theoretical perspective, which combined the buyer perspective on the relation with the supplier perspective.

30 be most beneficial for gaining preferential allocation of physical resources, which suggest that relational investments are most beneficial to achieve better physical resources compared to rivals in the market. Using the findings of this research, it is shown that social investments are useful for either buyers with high- and low supplier intelligence. If a buyer is not sure about the degree of supplier intelligence it possesses, it is recommended to do social investments instead of economic investments since economic investments possibly lead to negative effects if the supplier intelligence is low. Furthermore, buyers with high supplier intelligence which do economic investments are more likely to gain preferential allocation of innovation resources compared to buyers with low supplier intelligence. Therefore, gaining more supplier intelligence in combination with economic investments would be a useful strategy. When money is involved in the investment, the supplier intelligence becomes more important. Supplier intelligence can be achieved by using a reliable channel for all parties in which tacit knowledge can easily be exchanged. When the relation between a buyer and a supplier is relatively new, the chance that a buyer possess low supplier intelligence is high. Therefore, it is recommended for new relations to start with social investments instead of economic investments and first increase the supplier intelligence. By doing social investments, the trust and commitment increases and valuable information is shared more easily.

7.2 Limitations and future research

31 Lastly, due to the small sample size, potential bias in the parameter estimated produced by SmartPLS is likely to occur.

Despite these limitations, we believe our study offers opportunities for future research. First of all, future research could include more relations. For example, the suppliers of the suppliers or the buyers of the buyers to check if there is a significant difference between these two groups. This research is only conducted in the food industry. It could be interesting to include other industries and increase the sample size to improve the generalizability. Another recommendation for future research will be a longitudinal study instead of data gathering at a single moment. Due to for example a busy period, a relation can be assessed in a different way than when it is assessed in a quieter period. Since relations are constantly changing, it would be interesting to see if the outcomes of this research would also change over time. Next to that, longitudinal studies can also include observations, which are helpful in increasing the objectivity of the results. In future research, it would be of great value to include other variables in the survey. Namely, trust and commitment seem to have a mediating effect in the proposed model of this research. Due to time constraints, it could not be included in the survey for this research.

32

8. REFERENCES

Anderson, E., & Weitz, B. (1992). The use of pledges to build and sustain commitment in distribution channels. Journal of Marketing Research, 29 (1), 18–34.

Bai, C., & Sarkis, J. (2016). Supplier development investment strategies: A game theoretic evaluation. Annals of Operations Research, 240 (2), 583–615.

Barclay, D., Higgins, C., & Thompson, R. (1995). The partial least squares (PLS) approach to causal modelling: Personal computer adoption and use as an illustration. Technology Studies, Special Issue on Research Methodology, 2 (2), 285–309.

Berg, J. H., & Clark, M. S. (1986). Differences in social exchange between intimate and other relationships: Gradually evolving or quickly apparent? Friendship and Social Interaction, 101–128.

Birkinshaw, J., Morrison, A., & Hulland, J. (1995). Structural and competitive determinants of a global integration strategy. Strategic Management Journal, 16 (8), 637–655.

Blau, P. M. (1964). Justice in social exchange. Sociological Inquiry, 34 (2), 193–206.

Capron, L., & Chatain, O. (2008). Competitors’ resource-oriented strategies: Acting on competitors’ resources through interventions in factor markets and poutical markets. Academy of Management Review, 33 (1), 97–121.

Carr, A. S., & Pearson, J. N. (1999). Strategically managed buyer-supplier relationships and performance outcomes. Journal of Operations Management, 17 (5), 497–519.

Choy, K. L., Lee, W. B., Lau, H. C. W., & Choy, L. C. (2005). A knowledge-based supplier intelligence retrieval system for outsource manufacturing. Knowledge-Based Systems, 18 (1), 1–17.

Cropanzano, R., & Mitchell, M. S. (2005). Social exchange theory: An interdisciplinary review. Journal of Management, 31 (6), 874–900.

De Clercq, D., & Dimov, D. (2008). Internal knowledge development and external knowledge access in venture capital investment performance. Journal of Management Studies, 45 (3), 585–612.

Dorsch, M. J., & Carlson, L. (1996). A transaction-approach to understanding and managing customer equity. Journal of Business Research, 35 (3), 253–264.

Dorsch, M. J., Carlson, L., Raymond, M. A., & Ranson, R. (2001). Customer equity management and strategic choices for sales managers. Journal of Personal Selling and Sales Management, 21 (2), 157–166.

Dyer, J. H., & Chu, W. (2003). The role of trustworthiness in reducing transaction costs and improving performance: Empirical evidence from the United States, Japan, and Korea. Organization Science, 14 (1), 57–68.

Dyer, J. H., & Hatch, N. W. (2006). Relation-specific capabilities and barriers to knowledge transfers: Creating advantage through network relationships. Strategic Management Journal, 27 (8), 701–719.

Dyer, J. H., & Singh, H. (1998). The relational view: Cooperative strategy and sources of interorganizational comparative advantage. Academy of Management Review, 23 (1), 660– 679.

Ellis, S. C., Henke, J. W., & Kull, T. J. (2012). The effect of buyer behaviors on preferred customer status and access to supplier technological innovation: An empirical study of supplier perceptions. Industrial Marketing Management, 41 (8), 1259–1269.

33 Haenlein, M., & Kaplan, A. M. (2011). The influence of observed heterogeneity on path coefficient significance: Technology acceptance within the marketing discipline. The Journal of Marketing Theory and Practice, 19 (2), 153–168.

Hair, J. F. J., Hult, G. T. M., Ringle, C., & Sarstedt, M. (2014). A primer on partial least squares structural equation modeling (PLS-SEM). London: Sage Publications.

Hair, J. F., Ringle, C. M., & Sarstedt, M. (2011). PLS-SEM: Indeed a silver bullet. The Journal of Marketing Theory and Practice, 19 (2), 139–152.

Harris, L. C., O’Malley, L., & Patterson, M. (2003). Professional interaction: Exploring the concept of attraction. Marketing Theory, 3 (1), 9–36.

Hottenrott, H., & Peters, B. (2012). Innovative capability and financing constraints for innovation: More money, more innovation? Review of Economics and Statistics, 94 (4), 1126–1142.

Hult, G. T. M., Ketchen, D. J., Cavusgil, S. T., & Calantone, R. J. (2006). Knowledge as a strategic resource in supply chains. Journal of Operations Management, 24 (5), 458–475. Humphreys, P. K., Li, W. L., & Chan, L. Y. (2004). The impact of supplier development on

buyer–supplier performance. Omega, 32 (2), 131–143.

Humphreys, P. K., Shiu, W. K., & Chan, F. T. S. (2001). Collaborative relationships in Hong Kong manufacturing firms. Supply Chain Management: An International Journal, 6 (4), 152–162.

Jap, S. D., & Ganesan, S. (2000). Control mechanisms and the relationship life cycle: Implications for safeguarding specific investments and developing commitment. Journal of Marketing Research, 37 (2), 227–245.

Karlsson, C. (2010). Researching Operations Management. Abingdon: Routledge.

Khan, R. A. (2014). Sustainable competitive advantage through knowledge management. International Journal of Advanced Research in Computer & Technology, 3 (4), 1079– 1082.

Kim, K. (2000). On interfirm power, channel climate, and solidarity in industrial distributor-supplier dyads. Journal of the Academy of Marketing Science, 28 (3), 388–405.

Kotabe, M., Martin, X., & Domoto, H. (2003). Gaining from vertical partnerships: Knowledge transfer, relationship duration, and supplier performance improvement in the U.S. and Japanese automotive industries. Strategic Management Journal, 24 (4), 293–316.

Krause, D. R., & Ellram, L. M. (1997). Critical elements of supplier development: The buying-firm perspective. European Journal of Purchasing & Supply Management, 3 (1), 21–31. Krause, D. R., Scannell, T. V, & Calantone, R. J. (2000). A structural analysis of the

effectiveness of buying firms’ strategies to improve supplier performance. Decision Sciences, 31 (1), 33–55.

Lau, A. K. W., Tang, E., & Yam, R. C. M. (2010). Effects of supplier and customer integration on product innovation and performance: Empirical evidence in Hong Kong manufacturers. Journal of Product Innovation Management, 27 (5), 761–777.

Li, W., Humphreys, P. K., Yeung, A. C. L., & Cheng, T. C. E. (2012). The impact of supplier development on buyer competitive advantage: A path analytic model. International Journal of Production Economics, 135 (1), 353–366.

Lin, Y., & Wu, L. Y. (2014). Exploring the role of dynamic capabilities in firm performance under the resource-based view framework. Journal of Business Research, 67 (3), 407– 413.

Liu, Y., Huang, Y., Luo, Y., & Zhao, Y. (2012). How does justice matter in achieving buyer-supplier relationship performance? Journal of Operations Management, 30 (5), 355–367.

Markman, G. D., Gianiodis, P. T., & Buchholtz, A. K. (2009). Factor-market rivalry. Academy of Management Review, 34 (3), 423–441.

34 perspective: An illustrative example of relational and transactional drivers of competitiveness. Journal of Competitiveness, 5 (1), 16–38.

Modi, S. B., & Mabert, V. A. (2007). Supplier development: Improving supplier performance through knowledge transfer. Journal of Operations Management, 25 (1), 42–64.

Moon, M., & Bonney, L. (2007). An application of the investment model to buyer-seller relationships: A dyadic perspective. The Journal of Marketing Theory and Practice, 15 (4), 335–347.

Morgan, R. M., & Hunt, S. D. (1994). The commitment-trust theory of relationship marketing. Journal of Marketing, 58 (3), 20–38.

Morrison, P. D., Roberts, J. H., & von Hippel, E. (2000). Determinants of user innovation and innovation sharing in a local market. Management Science, 46 (12), 1513–1527.

Nollet, J., Rebolledo, C., & Popel, V. (2012). Becoming a preferred customer one step at a time. Industrial Marketing Management, 41 (8), 1186–1193.

Paulraj, A., Lado, A. A., & Chen, I. J. (2008). Inter-organizational communication as a relational competency: Antecedents and performance outcomes in collaborative buyer-supplier relationships. Journal of Operations Management, 26 (1), 45–64.

Preston, D. S., Chen, D. Q., Swink, M., & Meade, L. (2017). Generating supplier benefits through buyer-enabled knowledge enrichment: A social capital perspective. Decision Sciences, 48 (2), 248–287.

Pulles, N. J. (2014). The competition for supplier resources. Enschede: Universiteit Twente. Pulles, N. J., Schiele, H., Veldman, J., & Hüttinger, L. (2016). The impact of customer

attractiveness and supplier satisfaction on becoming a preferred customer. Industrial Marketing Management, 54, 129–140.

Pulles, N. J., Veldman, J., Schiele, H., & Sierksma, H. (2014). Pressure or pamper? The effects of power and trust dimensions on supplier resource allocation. Journal of Supply Chain Management, 50 (3), 16–36.

Reinartz, W., Haenlein, M., & Henseler, J. (2009). An empirical comparison of the efficacy of covariance-based and variance-based SEM. International Journal of Research in Marketing, 26 (4), 332–344.

Rosell, D. T., Lakemond, N., Melander, L., Rosell, D. T., Lakemond, N., & Melander, L. (2017). Integrating supplier knowledge in new product development projects: Decoupled and coupled approaches. Journal of Knowledge Management, 21 (5), 1035–1052.

Rusbult, C., & Farrell, D. (1983). A longitudinal test of the investment model: The impact on job satisfaction, job commitment, and turnover of variations in rewards, costs, alternatives and investments. Journal of Applied Psychology, 68 (3), 429–438.

Schwieterman, M., & Miller, J. (2016). Factor market rivalry: Toward an integrated understanding of firm action. Transportation Journal, 55 (2), 97–123.

Silver Coley, L., Lindemann, E., & Wagner, S. M. (2012). Tangible and intangible resource inequity in customer‐supplier relationships. Journal of Business & Industrial Marketing, 27 (8), 611–622.

Skjott-Larsen, T., Schary, P. B., Mikkola, J. H., & Kotzab, H. (2007). Managing the Global Supply Chain. Copenhagen Business School Press.

Surroca, J., Tribó, J. A., & Waddock, S. (2010). Corporate responsibility and financial performance: The role of intangible resources. Strategic Management Journal, 31 (5), 463–490.

Tanev, S., & Bailetti, T. (2008). Competitive intelligence information and innovation in small Canadian firms. European Journal of Marketing, 42 (7/8), 786–803.

35 Tanskanen, K., & Aminoff, A. (2015). Buyer and supplier attractiveness in a strategic relationship - a dyadic multiple-case study. Industrial Marketing Management, 50, 128– 141.

Tseng, S.-M. (2014). The impact of knowledge management capabilities and supplier relationship management on corporate performance. International Journal of Production Economics, 154, 39–47.

Tseng, S. M. (2009). A study on customer, supplier, and competitor knowledge using the knowledge chain model. International Journal of Information Management, 29 (6), 488– 496.

Wang, Z., & Wang, N. (2012). Knowledge sharing, innovation and firm performance. Expert Systems with Applications, 39 (10), 8899–8908.

Whipple, J. M., Lynch, D. F., & Nyaga, G. N. (2010). A buyer’s perspective on collaborative versus transactional relationships. Industrial Marketing Management, 39 (3), 507–518. Wilson, D. T., & Jantrania, S. (1994). Understanding the value of a relationship. Asia-Australia

Marketing Journal. 2 (1), 55–66.

Xiao, Y. (2015). Flexibility measure analysis of supply chain. International Journal of Production Research, 53 (10), 3161–3174.

36

9. APPENDIX

9.1 Supplier - survey

Thank you very much for participating in this research. The questions asked, are mostly related to providing your opinion on the relationship between your company as a supplier, and (COMPANY) as your customer. When a question is not completely applicable to the relationship with your customer, please provide the best suitable answer. There are no 'good' or 'bad' answers, please tick the answer that fits your firm's situation best. The information will be treated confidentially and your customer, (COMPANY), cannot gain access to your answers. The results of the data analysis will only appear in summary, assuring anonymity. In case of any doubt or confusion regarding the questions, please feel free to contact me via e-mail (EMAIL) or by phone (PHONE)

Statements about the preferential resource allocation (1= “no, strongly disagree; 5= “yes, strongly agree”) - Pulles et al., (2014)

Compared to our other customers...

(PrefResourceAlloPhys1) - We grant this customer better utilization of our production facilities. (PrefResourceAlloPhys2) - We would choose to give this customer priority in the allocation of our products in the case of extreme events (e.g., natural disasters).

(PrefResourceAlloPhys3) - We allocate our scarce materials to this customer in case of capacity bottlenecks.

Statements about the preferential resource allocation of innovation resources (1=”No, strongly disagree; 5 =”Yes, strongly agree”) - Pulles et al., (2014)

Compared to our other customers...

(PrefResourceAlloInno1) - We are more willing to share key technological information with this customer.

(PrefResourceAlloInno2) - We share our best ideas with this customer.

(PrefResourceAlloInno3) - We dedicate more innovation resources to the relationship with this customer.

Question about the share in turnover (0%-100%)

37 Final Questions

Please share the following general information about your company. If your company belongs to a group of companies please share the information and data of your site.

Questions about the length of the relationship (in years)