SUPPLY CHAIN AND INTER- ORGANISATIONAL RELATIONSHIPS

Transcription

1 SUPPLY CHAIN AND INTER- ORGANISATIONAL RELATIONSHIPS

2 RIRL 2010 The 8th International Conference on Logistics and SCM Research BEM Bordeaux Management School September 29, 30 and October 1st 2010 Customer-specific adaptation by providers and their perception of 3PL-relationship-success Rudolf O. Large Department of Business Logistic, University of Stuttgart, Germany Nikolai Kramer Department of Business Logistics, University of Stuttgart, Germany Rahel Katharina Hartmann Department of Business Logistics, University of Stuttgart, Germany Abstract This paper strives to answer the question whether there is an impact of customer-specific adaptations by third-party logistics (3PL) providers on the success of 3PL-relationships from a provider s perspective. A document analysis is presented and hypotheses based on research in third-party logistics, relationship marketing and transaction cost theory are developed. Structural equation modeling and causal analysis with PLS (Partial Least Square) were used to test the hypotheses. This study provides evidence, that customer-specific adaptation by providers is an important prerequisite to 3PL-success. Managerial implications and implications for further research are provided. Key words: Third-party logistics (3PL), customer-specific adaptations, relationship performance, satisfaction, loyalty 1

3 1. Introduction Logistics service providers have encountered growing competitive pressures throughout the last decade. (Persson & Virum, 2001; Yeung et al, 2006). Especially in recent years the traditional transportation market has faced a dramatic slowdown (Klaus et al, 2009), marked by stagnating or shrinking volumes. This has led to overcapacity, and low margins in turn increasing problems with returns and financing (Klaus et al, 2009). Within this challenging environment logistics service providers tend to additionally suffer from a high degree of replaceability as their traditional services are simple and not customized. Therefore logistics service providers need to confront of a strategic reorientation towards differentiation and encourage service innovation in order to reach a higher degree of customer orientation and offer more complex and customer-specific services (Panayides, 2004; Ellinger et al, 2008). Transforming into a third-party logistics service provider that offers a bundle of customized logistics services (Rabinovich et al, 1999; Maltz & Ellram, 2000) provides opportunities to enter a growing market with higher margins, than obtainable in the traditional transportation market (Lai, 2004; Klaus et al, 2009). In comparison to traditional transport and warehousing services, third-party logistics are more complex, encompass a broader number of functions, and are characterized by longerterm, more mutually beneficial relationships (Africk & Calkins, 1994). Hertz and Alfredsson (2003) emphasize the importance of the ability to solve problems and the ability to undergo customer adaptations. Both characteristics are used to differentiate between third-party logistics providers and traditional logistics firms. Furthermore, Hertz and Alfredsson develop a typology of third-party logistics providers based on these characteristics. These business models of third-party logistics are essentially based on the creation of customer specific services and hence on adaptations by providers. Such a creation of specificity in logistical services leads to a deeper integration and decreases the providers replaceability. While there are various characteristics of third-party logistics service providers, Hertz and Alfredson (2003) point to specific adaptation as a key capability. Specific adaptations to customers systems and procedures as well as extensive monitoring and reporting responsibilities are natural in 3PL-relationships. Third-party logistics contracts can include detailed stipulations concerning a provider s responsibilities (van Hoek, 2000) and many third-party logistics providers complain about one-sided adaptation to customers systems and procedures (Lieb & Bentz, 2005). For example, the customer insists on a specific location, demands specific procedures, expects the usage of their equipment or requires the 2

4 provider to report a specific set of key performance indicators. Following these circumstances the research question of this paper emerges as follows: o Is there any influence of third-party logistics providers customer-specific adaptations on their perception of the success of a specific business relationship and if indicated, how strong is this relationship? Analysis draws upon literature on third-party logistics, transaction cost theory and relationship marketing to deduce constructs covering the wide-ranging concept of relationship success and to analyze the role of customer-specific adaptations in the 3PLservice industry. In addition, studies of tender documents have been conducted to identify the required degree of customer-specific adaptations. Both sources literature and documents were used to create a set of hypotheses, because hypotheses tend to be more stable and reliable due to their consideration of different categories of sources (Ellram 1996). A sample of third-party logistics providers was drawn to collect data on relationship success and customer-specific adaptations. Structural equation modeling (SEM) was applied to test the hypotheses. 2. Literature The following literature studies focus on third-party logistics, transaction cost theory and relationship marketing. In general, there is a lack of theoretical foundations in research of third-party logistics (Selviaridis & Spring, 2007). Most of the previous studies have focused on outsourcing and have, therefore, taken customers perspective on 3PL-relationships (Lieb & Kendrick, 2002). Relationship marketing was chosen, because general insights into the nature of supplier-customer relationships can be transferred to the topic of 3PL-relationships. Transaction cost theory deals with the effects of specific investments on the efficiency of business transactions. Therefore, a better understanding of the impact of adaptations on provider-specific assets can be expected. As a prerequisite, the definition of what constitutes success in 3PL-relationships should be discussed in detail. 2.1 Providers perception of 3PL-success In general, success could be understood as the degree of goal accomplishment in 3PLrelationships. As accepted in marketing science, success is conceptualized in a broader sense covering the performance of the relationship as well as the satisfaction and the loyalty of the business partners (Oliver, 1997). In this research, each of these constructs is configured from 3

5 a provider s point of view. Consequently, this research focuses equally on the relationship performance as perceived by the provider as well as on the provider s loyalty to and satisfaction with the 3PL-relationship. Most of the previous research focused on customers perceptions of 3PL-performance (Knemeyer & Murphy, 2005). From a customer perspective Knemeyer and Murphy (2004) define 3PL-performance as the perceived performance improvements that the logistics outsourcing relationship has provided the user. Performance improvements include, e.g. reduced logistics costs, reduced cycle times, more efficient handling of exceptions and improved system responsiveness (Knemeyer & Murphy, 2004; Sinkovics & Roath, 2004). These factors are inadequate to define performance from the providers perspective. At first glance, the logistics service providers performance could be measured in terms of growth in market share, growth in return on investment and growth in annual sales (Sum & Theo, 1999; Ellinger et al, 2008). On closer examination, these indicators are also not suitable, for this research paper as it focuses on specific 3PL-relationships and their performance rather than business performance of a logistics service provider in general. Therefore, a broader measurement of perceived relationship performance, covering the overall perception of the provider instead of general figures of business performance is necessary. Consequently in this research a reflective four-item-scale of Stank, Daugherty & Ellinger (1996) was modified and used to measure the outcomes of a specific business relationship as perceived by the provider: (1) My firms association with this customer has been a highly successful one (PERF1). (2) This customer leaves a lot to be desired from an overall standpoint (PERF2). (3) If I have to give this customer an appraisal for the past year, it would be outstanding (PERF3). (4) Overall, I would characterize the results of my firm's relationship with this customer as having exceeded our expectations (PERF4). Previous literature focuses on customer evaluation of satisfaction with logistics partnerships (Gibson et al, 2002). Accordingly, Stank et al (2003) for instance describe customer satisfaction in 3PL-relationships as a customer s contentedness with the overall relationship with their provider. In the context of logistical partnerships Gibson et al (2002) emphasize the importance of satisfaction of both partners in order to generate a sustainable relationship. In general, it can be assumed that providers satisfaction depends on their expectations versus the degree of actual goal achievement. Considering providers motivations for entering the 3PL-market, higher margins and customer retention rates are assumed to be important factors influencing provider satisfaction. Hofmann (2009) points to the importance of efficient 4

6 operational processes and punctual payments for providers to be satisfied. Additional factors indicating provider satisfaction with 3PL-relationship can be considered. Gibson et al (2002) for example prove that the level of trust, the total cost, and the flexibility of shipper-carrier partnerships are important indicators that characterize the degree of provider satisfaction. Therefore, to operationalize provider satisfaction, a formative scale that covers the totality of these indicators could be applied. In contrast, provider satisfaction much like customer satisfaction is a latent variable or in other words an intrinsic construct. It was concluded that direct measurement with formative scales is not appropriate and hence a modified combination of proven reflective satisfaction scales, common in relationship marketing and logistics was employed in this research (Cannon & Perreault, 1999; Daugherty et al, 1998): (1) Our firm regrets the decision to do business with this customer (SAT1, reverse). (2) Overall, we are very satisfied with this customer (SAT2). (3) We are very pleased with this customer s compensation (SAT3). (4) Our firm is not completely happy with this customer (SAT4, reverse). (5) If we had to do it all over again, we would still work for this customer (SAT5). (6) We are delighted with our overall business relationship with them (SAT6). (7) We wish more of our customers were like this one (SAT7). (8) It is a pleasure to deal with this customer (SAT8). (9) There is always one problem or another with this customer (SAT9, reverse). Loyalty is a valuable concept reflecting the long term success of a relationship (Daugherty et al, 1998). Since customer loyalty is one of the central constructs in consumer marketing, there are countless approaches to loyalty operationalization (Bandyopadhyay & Martell, 2007). Oliver (1997, p. 392) defines customer loyalty as a deeply held commitment to rebuy or repatronize a preferred product or service consistently in the future, despite situational influences and marketing efforts having the potential to cause switching behavior. In the 3PL-industry, customer loyalty stands for the commitment of the customer to maintain the relationship and if necessary to renew the contract. In a broader sense loyalty points out the mutual relatedness between the provider and the customer. Therefore, loyalty reflects the commitment of both partners. Consequently, provider s loyalty covers the commitment of the provider to continue and support the business relationship to the customer. Along the lines of common loyalty scales in the 3PL-business (e. g. Daugherty et al, 1998) a new reflective scale of providers loyalty was built for this research: (1) The relationship that my firm has with this third-party logistics provider is something we are very committed to (LOY1). 5

7 (2) The relationship that my firm has with this one is something we intend to maintain indefinitely (LOY2). (3) The relationship that my firm has with this provider deserves our maximum effort to maintain (LOY3). (4) Maintaining a long-term relationship with this provider is very important to my firm (LOY4). 2.2 Customer specific adaptations by the provider As shown above, third-party logistics consist of recurrent, complex services based on a longterm contract between a provider and a customer. For such settings, the transaction cost theory predicts the existence of specific investments by the provider (Williamson, 2008; Williamson, 1979). The transaction cost theory highlights specific investments as a decisive factor on the level of transaction costs and the business relationship contract. Therefore transaction cost theory is of vital importance to gain a better understanding of 3PLrelationships (Maloni & Carter, 2006). Asset specificity indicates a specialized investment that cannot be redeployed to alternative uses or by alternative users except at a loss of productive value (Williamson, 1996, p. 377). Transaction specific assets invested by the provider enable the usage of efficient processes and procedures to generate third-party services (Williamson, 1996). Furthermore, asset specificity is a precondition to meet the specific requirements of the customer and to efficiently support recurrent transactions (Williamson, 1984; Williamson, 1985). Williamson distinguishes at least four important types of asset specificity: site specificity, physical asset specificity, human asset specificity and dedicated asset specificity (Williamson, 1984; Williamson, 1991). Following Williamson (1985) third-party logistics contracts can be characterized by a recurrent frequency and a high level of asset specificity. Figure 1 displays the relationship among frequency of exchange, asset specificity and logistics contract characteristics. Detailed and long-term agreements (hybrid contracting) like third-party contracts are necessary to safeguard these specific investments and to reduce the risk of opportunism (Williamson, 2008). Additionally, if the frequency of service transactions is low it is difficult to recoup investments made in the 3PL-relationship. Therefore, third-party logistics contracts are not appropriate for transactions with an occasional frequency. Van Hoek (2000) proved that customer-specific third-party logistics arrangements including services such as final assembly, display building or special warehousing activities are positively related to the existence of detailed contracts. In conclusion, transaction cost theory predicts intensive 6

8 investments by third-party logistics providers as a prerequisite to meet the customer s requirements and to realize high relationship performance. Asset specificity No Medium High Frequency Recurrent occasional contract of carriage contract of carriage / warehousing contract forwarding contract forwarding contract / cooperation agreement forwarding contract / contract of employment third-party logistics contract / contract of employment Figure 1: Asset specificity and logistics contract characteristics. Generally, relationship marketing has emphasized the importance of behavioral adaptations by sellers to customers systems and procedures. Cannon and Perreault (1999) developed a typology of customer-supplier relationships from a variety of characteristics which can be regarded as relationship connectors. These relationship connectors are: information exchange, operational linkages, legal bonds, cooperative norms, adaptations by sellers, and adaptations by buyers. Based on relationship marketing, partner-specific adaptations can be regarded as important characteristics of close relationships. Two types of relationships with extensive adaptations were detected by Canon and Perreault (1999). The first one is the customer is king type which involves intensive adaptations only by the seller. The second type of relationship is mutually adaptive which requires adaptations by both the seller and the supplier. Ellinger et al (2008) generally emphasize the importance of customer orientation of logistics service provider. The business model of third-party logistics is essentially based on the creation of customer-specific services and hence on adaptations by the provider. Specific adaptations to the systems and procedures of the customer as well as extensive monitoring and reporting responsibilities are natural. Third-party logistics contracts can include detailed stipulations concerning a provider s responsibilities (van Hoek, 2000) and many third-party logistics providers complain about one-sided adaptations to customer s systems and procedures (Lieb & Bentz, 2005). In many cases, the customer insists on a specific location, demands specific procedures, expects the usage of his equipment or requires periodical reports of specific key performance indicators. Consequently, Hertz and Alfredsson (2003) 7

9 emphasize the ability of customer adaptations as a crucial characteristic of third-party logistics providers. In summary, the construct of providers adaptations covers both the investments in specific assets and the behavioral adaptations by the provider. Therefore, a reflecting scale covering both issues was used to measure the degree of providers adaptations. This scale is adapted based on items used by Knemeyer and Murphy (2004) and Sharland (1997): (1) We have gone out of our way to link us with the customer s business (PSPEZ1). (2) We have tailored our services and procedures to meet the specific needs of this customer (PSPEZ2). (3) We would find it difficult to recoup our investments in this customer if the relationship were to end (PSPEZ3). (4) We made considerable investments in tools and equipment in our relationship with this customer (PSPEZ4). (5) Gearing up to deal with the customer required highly specialized tools and equipment (PSPEZ5). 3. Documents Study Organizational documents are a source of insights into organizational relationships and are an expression of the interactions in terms of problems and behaviors between different parties (Forster, 1995). Especially since documents as written texts are capable, in a very truthful manner to reveal the reality of business activity (Hodder, 2000). The application of a qualitative research method allowed for a beneficial combination with the subsequent use of a quantitative approach, from which a broader perspective on the research phenomena emerged (Frankel et al, 2005). However, analyzing documents leads to issues with interpretation and subjectiveness (Forster, 1995; Hodder, 2000). Document studies of tender documents have been conducted to evaluate the required degree of customer-specific adaptations in the 3PL-business (Large, 2007). Altogether 15 tender documents (requests for quotation) have been analyzed. Two major European third-party logistics companies made these documents available to the author. The subject of 8 documents is customer-specific distribution and warehousing. Seven documents are requests for physical supply and logistics in manufacturing, e.g. sequencing activities and materials handling. Most of the customers belong to the automotive industry. Typically a request for quotation consists of a main text of more than 50 pages that describes the basic conditions and the specific customer requirements. Additionally, most of the 8

10 requests include an extensive appendix. Examples are warehouse layouts, annual demand figures and performance indicators of the existing equipment. Each document describes an individual case and has individual structure and style. In accordance, the qualitative method of explorative document analysis was applied (Frankel et al, 2005; Spens & Kovács, 2006). The analysis focused on the content of the described documents. Adopting qualitative content analysis as a research method (Ryan & Bernard, 2000), the available documents were analyzed systematically and theory-guided and processed with the required openness of the researcher (Marasco, 2008; Kohlbacher, 2006). The systematic frame of this document study was based on the findings obtained by discussing transaction cost theory, 3PL-literature and relationship marketing. In detail, this systematic frame consists of four content categories, as follows: required specificity (site specificity, physical asset specificity, and human asset specificity), the intended procedure of performance evaluation, the expected behavioral adaptation by the provider and the willingness of the customer to adapt to the provider. Analyzing the documents, the considerable amount of (required) site specificity is striking. Most customers insist on a specific location or at least stipulate that the warehouse must be located in proximity of their own manufacturing facilities. Most customers expect specific investments by the provider such as warehouses, warehousing equipment or computer systems. This highlights that physical asset specificity is a frequent characteristic of 3PLrelationships. For outsourced tasks, the provider is typically requested to use the customer s existing assets. Likewise, human asset specificity is a recurring requirement throughout. Typically, there is a need for additional personnel at the desired location, or at least a need for training in order to meet the specific requirements of the customer. As expected, most customers place specific demands on the service provider concerning performance measurement and reporting. The vast majority of the documents illustrate the amount onesided adaptations by the third-party logistics provider. 4. Hypotheses Provider capability of customer-adaptation is a crucial characteristic of 3PL-business (Hertz & Alfredsson, 2003) and thus a requirement of 3PL-business relationships. Customers of third-party logistics firms expect tailored logistical solutions (Sink et al, 1996). The document study shows that specific adaptations by the provider to the customer s systems and procedures seem to be a matter of course. A provider s capability and willingness to adapt are also prerequisites to transaction-specific investments made by that provider. Following transaction cost theory, specific assets improve the performance of 3PL-relationships, 9

11 because specific assets enhance the productivity of logistical activities in comparison to general purpose technology (Williamson, 1996; Williamson, 2008). Consequently, 3PLproviders should understand and recognize the necessity of specific adaptations to the customer. These ideas suggest the following hypothesis: H1: The provider s perception of 3PL-relationship performance is positively influenced by the degree of their own specific adaptations. Transaction cost theory indicates that intensive investments can be construed as means to stabilize business relationships, because transaction-specific assets can be redeployed to alternative uses and users only at a loss of productive value (Williamson, 2008, p. 8.) Providers expect ongoing relationships and a continuous use of their specific investments. The higher the degree of specific investments and adaptations the higher are the switching costs and the degree of the provider s loyalty towards the relationship. Asset specificity contributes to the commitment of both parties, resulting in a trustful relationship between the partners. Kwon and Suh (2004) proved that supply chain partners investments increase the level of trust between the partners, because these investments are perceived as a signal of commitment. Therefore, the effect of providers adaptations on the degree of loyalty is assumed as being positive: H2: The provider s loyalty is positively influenced by their own specific adaptations. The positive relationship between performance and customer satisfaction is a widely recognized phenomenon in consumer marketing as well as in business-to-business relationships (Patterson et al, 1997; Homburg et al, 2002). Similarly, the satisfaction of the provider with the business relationship is essential too. A significant level of provider satisfaction is a prerequisite to facilitating an enduring effort to provide outstanding logistical service by the provider. On the other hand, satisfaction is the result of an ongoing evaluation of the perceived outcome of the 3PL-relationship. Consequently, it can be suggested that a high level of perceived performance exerts a positive influence on the degree of customer satisfaction as well as on the degree of provider satisfaction. For that reason a positive relationship between the perceived performance and the satisfaction of the provider is assumed: H3: The provider s satisfaction is positively influenced by the perceived relationship performance. 10

12 In a broad sense loyalty indicates the reciprocal relatedness between the customer and the provider in a 3PL-relationship. Relationship commitment is fundamental to the concept of loyalty (Daugherty et al, 1998). Provider loyalty therefore stands for the commitment of the provider to support the ongoing 3PL-relationship. In marketing research customer satisfaction is recognized as a main influence on customer loyalty (Luo & Homburg, 2007; Seiders et al, 2005). In a figurative sense a high level of provider satisfaction is an important precondition of the willingness to maintain and continue the relationship by the provider. This analogy suggests the following hypothesis: H4: The provider s loyalty is positively influenced by the provider s perceived satisfaction. The influence of specific investments and adaptations on the satisfaction of partners in business relationships is assumed to be contradictory. On one side, 3PL-relationships based on specific investments and adaptations by the provider meet customers special requirements efficiently. Specific adaptations are thus at the core of the business and a crucial element of 3PL-relationships from the provider s point of view. Therefore, an indirect positive influence of the degree of adaptations on providers satisfaction mediated by the level of performance is reasonable (see H1 and H3). In contrast, many third-party logistics providers complain about one-sided adaptations to customers systems and procedures (Lieb & Bentz, 2005). The providers associate a higher degree of customer specific adaptations with additional costs and a growing dependence on their customers. Consequently, they feel uncomfortable with a high degree of own adaptations in addition to a substantial amount of asset specificity. To cover this phenomenon the suggested model includes a direct negative influence of adaptations on providers satisfaction. This leads to the last hypothesis: H5: A provider s satisfaction with the 3PL-relationship is negatively influenced by his own specific adaptations. 5. Methodology 5.1 Structural equation modeling with Partial Least Square (PLS) The five hypotheses derived in the previous section indicate a closely connected set of relationships between the theoretical constructs that form this research. The employed constructs are not directly observable or measurable. To account for each theoretical construct it is necessary to define the reflective multi-item scales involved (Hair et al, 2009). To meet these two requirements, the model is constructed using structural equation modeling 11

13 (SEM). The SEM approach combines a path model (relationship among the construct) and a measurement model (set of items for each construct) (Giménez et al, 2005; Hair et al, 2009). Figure 2 shows the structure of the path model. SmartPLS (Ringle et al, 2005) was executed for the analysis of the path model shown in Figure 2. This structural equation modeling software package is an application of the Partial Least Square Method (PLS) (Chin, 1998; Tenenhaus et al, 2005). In comparison to covariance based procedures the PLS algorithm is advantageous if the model is complex and the sample size is small (Chin, 1998). Covariance based SEM procedures such as LISREL or AMOS perform a simultaneous estimation of the totality of the model parameters. Therefore, these procedures require very large samples, especially if the models are complex. According to the recommendations of Bentler and Chou (1987) LISREL or AMOS would need more than 175 cases to analyze the path model shown in Figure 2. In contrast, the PLS estimation is based on a set of distinct multiple regressions. Following the recommendations of Chin and Newsted (1999) the sample size in PLS estimation should be at least ten times either the largest number of formative indicators or the largest number of independent variables influencing a dependent variable of the structural model. In this research the measurement model consists of reflective indicators exclusively. It follows that only the second criterion is relevant. The dependent variables with the largest number of predictor variables are loyalty and provider satisfaction. This adds up to two, thus requiring the number of usable cases to be at least 20 as per the recommendation by Chin and Newsted (1999). Accordingly, the sample meets the sample size requirements of PLS. The applied PLS approach is also more suitable for explorative studies where the level of theoretical knowledge and scale development is rather low (Chin, 1998) making PLS the best-fitting and most appropriate approach to analyzing the data of this study. 12

14 Performance of the Relationship Adaptations by the Provider -H5 Provider s Satisfaction Provider s Loyalty Figure 2: Hypothesized path model. 5.2 Sampling and data collection To examine the five hypotheses a two-part questionnaire was designed. The first part of the questionnaire consists of general questions about third-party logistics. The second part refers to a specific 3PL-relationship of the company. Reflective multi-item scales (Stank et al, 1996; Daugherty et al, 1998; Knemeyer & Murphy, 2004; Sharland, 1997) were adopted to measure the four constructs involved in this research. The questionnaire was sent by to 129 chief executives or sales managers of third-party logistics companies. The sample was drawn from a mailing list of the University. Additionally, the logistics newsletter of the German Association of Purchasing and Logistics (BME) was used to enlist additional participants. Altogether 45 responses were available for statistical analysis. 42 of the providers have already established at least one third-party relationship. Based on the number of questionnaires distributed the response rate is 27.9%. 5.3 Measurement assessment An important prerequisite of structural equation modeling is an accurate scale purification for each construct individually. This is particularly important in the case of new or adapted scales. In this study the path model consists of four latent variables. Thus a reflective measurement model was chosen. The questionnaire includes 22 indicators. SPSS was used in the first instance to perform reliability analysis and explorative factor analysis. The evaluation was undertaken using the criteria provided by Hair et al (2009). Some items were dropped because of low loadings or insufficient reliability of the scale (PERF4, LOY3, 13

15 SAT1, SAT5, SAT9, PSPEZ2 and PSPEZ3). After scale purification, measurement assessment showed satisfactory results. Only one loading (PSPEZ1) was still slightly below the benchmark value of 0.7. Since this item represents an important facet of behavioral adaptation it was not excluded. In total the calculations showed sufficient degrees of reliability and validity (Table 1). Table 1: Reliability and validity of the measurement model (calculations with SPSS). Construct Indicator Cronbach Loading Variance explained Alpha >0.7 >0.7 >50% Performance of the relationship PERF1 PERF PERF Loyalty LOY1 LOY LOY Provider s Satisfaction SAT SAT SAT SAT SAT SAT Adaptation by the provider PSPEZ1 PSPEZ PSPEZ Finally, SmartPLS was used to evaluate the scales of the model. Common criteria to evaluate reflective measures of PLS path models are the average variance extracted, the composite reliability and the communality (Stone-Geissers Q 2 ) (Chin, 1998). The results of these calculations are shown in Table 2. Each of the constructs meets the requirements. Table 2: Evaluation (calculation with SmartPLS). Average variance Extracted Composite reliability Stone-Geissers Q 2 (communality) Cronbach Alpha > 0.6 > 0.7 > 0 > 0.7 Specific adaptations by the provider (PSPEZ) Performance of the relationship (PERF) Provider s satisfaction (SAT) Provider s loyalty (LOY) The path relationships (standardized regression coefficients) of the hypothesized model were estimated performing SmartPLS 2.0 (Ringle et al, 2005). The bootstrap procedure (Efron, 1979) was used to obtain t-statistics in order to evaluate the significance of the parameters. 14

16 Each of the estimators is significant at the 5% level. The results are shown in table 3 and figure 3. Table 3: Parameter estimation (calculation with SmartPLS). PLS Path Bootstrap Standard t-value Significance coefficient sample mean error PSPEZ PERF (H1) PSPEZ LOY (H2) PERF SAT (H3) SAT LOY (H4) PSPEZ SAT (H5) Each of the hypotheses is supported by the analysis. In support of hypothesis 1, there is evidence that behavioral adaptations and specific investments by the third-party logistics providers exert positive influence on the performance of the relationship perceived by the provider. As hypothesis 2 predicted, the estimation indicates that third-party logistics provider s adaptations also exert positive influence on the degree of loyalty. Hypotheses 3 and 4 offer positive connections among relationship performance, provider satisfaction and loyalty. The data also strongly support these hypotheses. Performance of the Relationship R 2 = 0,16 Adaptations by the Provider H5: ** Provider s Satisfaction R 2 = 0,74 *** p < 0.01 ** p < 0.05 * p < 0.1 Provider s Loyalty R 2 = 0,49 Figure 3: Results of the PLS-estimation (standardized regression coefficients). Since one s own adaptations are recognized as an additional effort, there is a direct negative influence (-0.162) of specific adaptations by the providers on the level of satisfaction as predicted (hypothesis 5). However, there is a positive indirect effect of the specific adaptations by providers on their satisfaction (0.365) mediated by a very strong impact of 15

17 performance on satisfaction. Altogether, adaptations exert a positive total influence on the provider s satisfaction (0.203). The coefficients of determinations (R 2 ) for each dependent construct deliver insights as to whether the independent variables of the model exert substantial influence on this construct (Chin, 1998). The coefficient of determination of the performance of the relationship is rather small (R 2 =0.160), because there is only one exogenous variable. The R-squares of the two other endogenous constructs show sufficient (LOY: R 2 =0.494) and high values respectively (SAT: R 2 =0,742). 6. Discussion and implications This study delivers a better understanding of the nature of providers specific adaptations and the influence of these adaptations on the success of 3PL-relationships from a providers perspective. The findings have some consequences and helpful managerial implications. The first implication of this study relates to the importance of providers specific adaptations to the relationship performance. The impact of specific adaptations by the provider on the provider s perception of performance is positive. Sufficient behavioral adaptations and transaction-specific investments by providers are crucial for 3PL-performance. As proposed by Hertz and Alfredsson (2003) adaptations by the service provider are an essential characteristic of the 3PL-business. Therefore, own adaptations are accepted by the provider as a critical element of 3PL-relationships. Regarding the transaction cost theory the results confirm that asset specificity is a precondition to realize efficient 3PL-relationships. Furthermore, the results show that the amount of this influence is considerable. Consequently, we suggest that third-party logistics providers should adapt their systems and procedures to customers specific requirements to ensure high relationship performance. As suggested in relationship marketing, the results of this research provide additional evidence that 3PL-providers satisfaction is strongly influenced by the degree of perceived relationship performance. Since one s own behavioral adaptations and specific investments are sensed as an effort, there is a direct negative impact of providers adaptations on providers satisfaction. In other words, an increase in behavioral adaptation and higher specific investments by providers lower their level of satisfaction. This effect expresses the provider s displeasure over adaptations to customer s systems and procedures. However, there is an indirect effect of specific adaptations by providers on their level of satisfaction mediated by the very strong impact of performance on satisfaction. This positive indirect effect exceeds the direct negative impact leading to the total influence of providers 16

18 adaptations on providers satisfaction to be positive. This research illuminates and confirms the contradictory statements concerning the influencing factors on providers satisfaction. On one side, providers are dissatisfied with specific adaptations on the other they are aware of the necessity of such adaptations to enhance relationship performance. As expected by transaction cost theory, there is evidence that the direct effect of adaptations by 3PL-providers exerts positive influence on the degree of loyalty, because this theory construes specific investments as means to stabilize business relationships. This suggests that customers should promote providers adaptations. Furthermore, the degree of loyalty is positively driven by the satisfaction of the provider, because a satisfied logistics service provider, who perceives a relationship as positive, wants to continue this business. Besides the positive effects, providers adaptations also cause a low indirect negative effect on the degree of customer loyalty mediated by its negative impact on satisfaction. Nevertheless, the direct influence plus the positive effect mediated by performance and satisfaction outweigh this indirect negative influence. This result expresses the importance of providers adaptation to maintain 3PL-relationships. Satisfied customers should promote the providers adaptations, because these adaptations enhance the probability of contract renewal and reduce the probability of providers unexpected termination of the contract. Summing up, this study provides evidence that from a providers perspective the customerspecific adaptation of third-party logistics providers is an important precondition of thirdparty logistics relationship success. 7. Conclusion, limitations and further research This study has delivered first ideas concerning the providers adaptations and their impact on providers perspective of relationship success. The research is based on a sound theoretical foundation and suitable qualitative and quantitative methods for evaluation were employed. Firstly, customer-specific adaptations by third-party logistics providers have a considerable influence on the success of 3PL-relationships. This is indicated by the influence of customerspecific adaptations on relationship performance as well as on provider satisfaction and provider loyalty as the three examined aspects of relationship success. Secondly, evidence is provided, illustrating that third-party logistics providers are aware of the importance of their own partner-specific adaptations and accept these adaptations for the purpose of relationship success. Thirdly, light is shed on the contradictory statements regarding the relationship between the degree of customer-specific adaptations and provider satisfaction. Even though third-party 17

19 logistics service providers associate dissatisfaction with their customer-specific adaptations, ultimately they recognize a positive influence on satisfaction. This positive influence is mediated by relationship performance. Finally, 3PL-providers loyalty is exposed as being positively influenced by the degree of customer-specific adaptations as well as by the level of satisfaction. Nevertheless, there are some limitations that make further research necessary. Firstly, this research is based on adapted scales. Further evaluation and improvements of these scales are necessary. A more general problem is the small sample size. The reason for this small sample size is the comparatively small number of third-party logistics firms operating in Germany. Although PLS is a suitable method, larger samples would allow for the use of covariance based methods like AMOS or LISREL. The most important advantage of AMOS and LISREL is the availability of goodness-of-fit statistics to evaluate the overall quality of a structural equation model. An appropriate approach to solve the problem of small samples could be the broader collection of providers data to include providers from other countries. Above all, the growing market of third-party logistics providers, which already plays an important role for European countries economies, should generally be examined more closely using quantitative methods basing research on a strong theoretical foundation. Finally, this research is focused on providers perceptions of partner-specific adaptations and third-party logistics relationship success. It is feasible to deduce that customers will have divergent perceptions and points of view. Consequently further research should compare customers perceptions of partner-specific adaptations and third-party logistics relationship success with the results of this study. References o Africk J. M., Calkins C. S. (1994), Does asset ownership mean better service?, Transportation and Distribution, Vol. 35, N o 5, pp o Bandyopadhyay S., Martell M. (2007), Does attitudinal loyalty influence behavioral loyalty? A theoretical and empirical study, Journal of Retailing and Consumer Services, Vol. 14, N o 1, pp o Bentler P. M., Chou C.-P. (1987), Practical Issues in Structural Modeling, Sociological Methods & Research, Vol. 16, N o 1, pp o Cannon J. P., Perreault W. D. (1999), Buyer-Seller Relationships in Business Markets, Journal of Marketing Research, Vol. 36, N o 4,

20 o Chin W. (1998), The Partial Least Squares Approach to Structural Equation Modeling, in Marcoulides G. A., (Ed.), Modern Methods for Business Research, Erlbaum, Mahwah and London, pp o Chin W. W., Newsted P. R. (1999), Structural Equation Modeling Analysis with Small Samples Using Partial Least Squares, in Hoyle R. H. (Ed.), Statistical strategies for small sample research, Sage, Thousand Oaks, pp o Daugherty P., Stank T., Ellinger A. E. (1998), Leveraging Logistics / Distribution Capabilities: The Effect of Logistics Service on Market Share, Journal of Business Logistics, Vol. 19, N o 2, pp o Efron B. (1979), Bootstrap Methods: Another Look at the Jackknife, Annals of Statistics, Vol. 7, N o 1, pp o Ellinger A. E., Hult Jr. D. J., Elmadag G. T. M. (2008), Market orientation, employee development practices, and performance in logistics service provider firms, Industrial Marketing Management, Vol. 37, N o 4, pp o Ellram L. M. (1996), The Use of Case Study Method in Logistics Research, Journal of Business Logistics, Vol. 17, N o 2, pp o Forster N. (1995), The Analysis of Company Documentation, in Cassell, C., Symon, G. (Eds.), Qualitative Methods In Organizational Research, Sage, London, pp o Frankel R., Naslund D., Bolumole Y. (2005), The White Space of Logistics Research: A look at the Role of Methods Usage, Journal of Business Logistics, Vol. 26, N o 2, pp o Gibson B. J., Rutner S. M., Keller, S. B. (2002), Shipp-carrier partnership issues, rankings and satisfaction, International Journal of Physical Distribution & Logistics Management, Vol. 32, N o 8, pp o Giménez C., Large R., Ventura E. (2005), SCM Research Methodologies: Employing Structural Equation Modeling, in Kotzab H., Seuring S., Müller M., Reiner G. (Eds.), Research Methodologies in Supply Chain Management, Physica, Heidelberg, pp o Hair J. F., Black W. C., Babin B. J., Anderson, R. E. (2009), Multivariate Data Analysis, Pearson, Upper Saddle River (7 th edition). o Hertz S., Alfredsson M. (2003), Strategic Development of Third Party Logistics Providers, Industrial Marketing Management, Vol. 32, N o 2, pp o Hodder I. (2000), The Interpretation of Documents and Material Culture, in Denzin, N. K., Lincoln, Y. S. (Eds.), Handbook of Qualitative Research, Sage, Thousand Oaks (2nd edition), pp