Logistics service performance: Estimating its influence on market share

Stank, Theodore P

Logistics creates value by accommodating customers' delivery requirements in a cost effective manner. Logistics service performance, therefore, assesses a provider's ability to consistently deliver requested products within the requested delivery time frame at an acceptable cost (Bowersox, Closs, and Cooper 2002). Logistical services, a unique subset of industrial services that span the boundaries between suppliers and customers, have become increasingly important to successful supply chain operations. Logisticians understand that these activities constitute the very essence of their business. Communicating the importance of logistical activities to other functional activities, as well as to corporate officers, has been a difficult feat. Professor Donald Bowersox, speaking at the Council of Logistics Management Annual Conference in Toronto in 1999, described establishing the link between functional logistics performance and overall firm performance as our discipline's equivalent to finding a cure for cancer.

Definitive empirical results that link improvements in logistics performance to overall firm performance have been difficult to achieve. Some progress, however, has been made. Recently, Daugherty, Stank, and Ellinger (1998) conducted an in-depth assessment of the relationships among logistics service performance and customer satisfaction, loyalty, and market share in an industrial setting. Their findings indicated that high levels of logistics service are indirectly related to market share through satisfaction and loyalty.

The current research seeks to expand the knowledge of logistics service performance and its link to overall business performance in four significant ways. First, a more complex operationalization of logistics service performance consisting of three distinct constructs is used to capture logistical service in a comprehensive, yet parsimonious manner. Second, the model is applied in a unique industrial setting, the third party logistics (3PL) industry. Third, a secondary source of objective market share data (from Armstrong and Associates 1999) is used to anchor respondents' assessments of market share. Fourth, relationships among core dimensions of logistics service performance, satisfaction, loyalty, and market share are tested using structural equations modeling (SEM) in contrast to regression in the Daugherty, Stank, and Ellinger research to simultaneously assess the structural paths among the constructs.

The paper is organized as follows. First, logistics operational, cost, and relational performance are defined. A synthesis of the logistics, operations, and marketing literature forms the basis of a conceptual model relating the operational, cost, and relational determinants of logistics service performance to customer satisfaction, loyalty, and market share. The methods used to collect data and test the model are described next, followed by a discussion of results. Finally, the managerial implications of the findings and directions for future research are examined.

THEORETICAL BACKGROUND

Logistics service performance research can be broadly segmented into work oriented toward: 1) service dimensions and 2) service outcomes. This section reviews relevant work in these two areas, leading to a discussion of the research model and hypotheses.

Logistics Service Performance Dimensions

Logistics operational and relational performance construct definitions were drawn from the dimensions of overall service quality originally defined by Parasuraman, Zeithaml, and Berry (1985, 1988). Parasuraman, Zeithaml, and Berry's research identified five broad dimensions of service quality: 1) reliability (the ability to perform the promised service dependably and accurately); 2) responsiveness (the willingness to help customers and to provide prompt service); 3) assurance (the knowledge and courtesy of employees and the ability to convey trust and confidence); 4) empathy (the provision of caring, individualized attention to customers); and 5) tangibles (the appearance of physical facilities, equipment, personnel, and communications materials). The dimensions were operationalized in a measurement scale called SERVQUAL that assessed service quality as the gap between pre-transaction customer expectations of quality and their perceptions of service quality after consumption.

A substantial amount of research has documented the relative importance of service quality dimensions to end-use consumers, i.e., customers of retail services (Babakus and Mangold 1989; Berry 1995; Berry and Parasuraman 1991; Bojanic 1991; Carman 1990; Crompton and Mackay 1989; Johnson, Dotson, and Dunlop 1988; Parasuraman, Berry, and Zeithaml 1991; Zeithaml 2000; Zeithaml, Parasuraman, and Berry 1990).

Empirical evidence suggests that the proposed delineation of five components is not consistent when compared across different types of service industries (Babakus and Boller 1992; Carman 1990; Cronin and Taylor 1992; Finn and Lamb 1991). In particular, researchers have had difficulty replicating the SERVQUAL dimensions in industrial service contexts (Bienstock, Mentzer, and Bird 1997). One possible explanation for the disparate results is that the dimensions of service quality vary from one industry to the next. This is especially true for industrial services like logistics that focus on tangible things directed toward physical objects versus intangible actions directed toward thoughts and attitudes (Lovelock 1983). Accordingly, Brown, Churchill, and Peter (1993) advise researchers to carefully assess which issues are important to service quality in their particular situations and to modify the SERVQUAL scale accordingly. Another possible explanation is that a more generic conceptual scheme has yet to be identified.

Recently, Stank, Goldsby, and Vickery (1999) used the SERVQUAL dimensions as a starting point for producing a more generic conceptualization of logistics service performance, a unique example of industrial service. Their research identified two core dimensions of logistical service: 1) operational performance and 2) relational performance. Operational performance consisted of two key elements - reliability (which captured the dependability and accuracy of a service a la Parasuraman, Zeithaml, and Berry (1985) and related to the consistent quality aspect of operational performance) and price. Parasuraman, Zeithaml, and Berry's responsiveness, assurance, and empathy attributes were encompassed in relational performance, the second dimension of service performance in their study.

While Parasuraman, Zeithaml, and Berry's original research considered the price or cost of service to be part of communication, Stank, Goldsby, and Vickery included cost as a key aspect of logistics operational performance in their fast food industry research. In the current research, cost is conceptualized as a unique, third dimension of logistics service performance, separate and distinct from the operational and relational components of service. Literature in manufacturing and service operations provides substantial support for treating price (or cost) as a separate dimension of service performance (Cleveland, Schroeder, and Anderson 1989; Ferdows and De Meyer 1990; Hayes and Wheelwright 1984; Hill 1989; Krajewski and Ritzman 1987; Roth and Miller 1990; Roth and Van der Velde 1991; Wood, Ritzman, and Sharma 1990). Additional support for this approach is provided by Porter's generic strategies scheme in which cost leadership (in contrast with quality differentiation, for example) appears as a distinct, yet viable, path for attaining competitive advantage (Porter 1980, 1985).

Aside from theoretical considerations, there are two practical advantages of modeling cost (or price) separately from logistics operational performance or consistent quality. First, it allows us to determine the individual effects of these entities on the endogenous variables of the research model and, second, it enables us to examine the relationship of antecedent variables to cost.

Outcomes of Logistics Service Performance

Research frequently has sought to determine the relationship between service performance and perceptual factors such as customer satisfaction and repurchase intentions (Zeithaml 2000). Customer satisfaction may be considered a cumulative evaluation based on the total purchase and consumption experience with a good or service over time (Fornell 1992; Fournier and Mick 1999). The evaluation is based upon post purchase confirmation or disconfirmation of the buyer's preconceived expectations of product or service standards. High customer satisfaction has been linked to improvements in a firm's economic returns, including market share and profitability (Anderson, Fornell, and Lehmann 1994; Crosby, Evans, and Cowles 1990; Leuthesser and Kohli 1995; Reichheld and Sasser 1990).

A significant number of findings strongly support the notion that logistics service quality improvements can increase customer satisfaction (Daugherty, Stank, and Ellinger 1998; Innis and La Londe 1994; Leuthesser and Kohli 1995). Operational elements of logistics service related to product availability, product condition, delivery reliability, and delivery speed, as well as relational elements such as communications and responsiveness have been shown to have a positive relationship with customer satisfaction (Daugherty, Stank, and Ellinger 1998; Innis and La Londe 1994; Stank, Goldsby, and Vickery 1999).

The literature suggests a strong link between customer satisfaction and loyalty (Anderson and Sullivan 1993; Bearden and Teel 1983; Fornell 1992; Innis and La Londe 1994; Jones and Sasser 1995). Customer loyalty is conceptualized as having both behavioral and perceptual/psychological components (Anderson and Sullivan 1993; Jacoby and Kyner 1973). Repeat purchase behavior stemming from positive assessments of product and/or service offerings has often been used as a measure of customer loyalty (Leuthesser and Kohli 1995; Sharma and Lambert 1990). Repeat purchase behavior, however, may result from a number of underlying factors that are not related to a customer's favorable impression of a provider's product or service offering. Jacoby and Kyner assert that loyalty, in contrast to simple repurchase behavior, is the result of a psychological decision-making process that results in a nonrandom, behavioral purchase response with respect to one alternative out of a set of such alternatives expressed over time by a decision-making unit. Repeat industrial purchases representing increased resource expenditure between parties in industrial buyer-seller relationships are likely to result from an enduring desire on the part of the customer to maintain an important, valued relationship with a supplier (Cronin and Morris 1989; Morgan and Hunt 1994). Consistent with Morgan and Hunt we define loyalty as a long-term commitment to repurchase involving both a favorable cognitive attitude toward the selling firm and repeated patronage.

The literature hints at a strong link between customer loyalty and market share. Reichheld and Sasser claim that loyal customers improve market share by purchasing a greater volume and variety of product. In addition, loyal customers demonstrate greater resistance to counter persuasion and negative word-of-mouth (Rust and Zahorik 1993). Innis and La Londe confirmed the relationship in logistics, linking customer service, attitudes, satisfaction, and repurchase intentions. In their work, they refer to "repurchase intentions" as a "proxy for market share" (1994, p. 19).

Dick and Basu (1994) introduced a model that illustrated a linkage between consumer loyalty and specific consequences that may contribute to market share. Similarly, Daugherty, Stank, and Ellinger (1998) concluded that logistics operational elements related to product availability, product condition, delivery reliability, and delivery speed, as well as relational elements such as communications and responsiveness were found to positively influence customer satisfaction and purchasing patterns that contribute to market share growth. Their research in the personal products industry, however, concludes that loyalty predicts market share variance only when operating with satisfaction. The authors note that future research should attempt to further capture the interrelationships that exist among logistics service, customer satisfaction, loyalty, and market share given the lack of support for the hypothesis solely linking loyalty to market share.

Conceptual Model and Research Hypotheses

This research seeks to expand upon past research to further understand the relationships among logistics service performance, satisfaction, loyalty, and firm market share. Importantly, it does so in a model that provides a comprehensive picture of the relationships among key logistics performance elements. The research also explicitly examines the linkage between loyalty and a measure of market share that is anchored by actual outcomes.

The model is tested in the context of logistical services provided by 3PL providers. The 3PL industry was considered an excellent setting for examining the interrelationships of interest since the evaluation of the service provider (3PL) is independent of the manufacturer/shipper of the materials or goods. In essence, the provision and quality of the logistics service is distinct from the materials or goods themselves, reducing the potential for a confounding or "halo" effect. In other words, the nature of the relationship between the service provider and the customer is based solely on the nature of the service provision and not on the quality of the materials or goods. In addition, there are usually multiple logistics service alternatives available to shippers in the U.S. market. Therefore, customer loyalty is typically based upon the quality of logistical service rather than a lack of alternatives (i.e., "captive commitment"). For these reasons, the customers of third-party logistics service providers served as the target population for the research.

The conceptual model is presented in Figure 1. The theoretical foundations for the relationships depicted in this figure are summarized below based upon the prior review of relevant literature.

In Figure 1, customers' perceptions of logistics relational performance are portrayed as the antecedent of logistics operational and cost performance. Stank, Goldsby, and Vickery (1999) present empirical evidence to suggest that relational performance is an antecedent of operational performance (which included a cost component). Their research indicates that creating strong relationships with important customers allows select service firms to achieve sustainable advantage by tailoring logistical operational offerings to the needs of each customer. Such firms go beyond logistics basics by developing a meaningful understanding of customers' needs supported by flexible processes that enable customized solutions. The ability to customize logistics operational services with a high level of relational performance represents an "order winning" combination. It becomes significantly more difficult for rivals to intercede in supplier-customer relations once they reach this level of intimacy. In the current research, relational performance encompasses attributes such as responsiveness, assurance, and empathy for customer needs. Measures include assessments by customers' of whether the service supplier "knows your needs well," "cooperates with you to help do the job well," and "makes recommendations for continuous improvement on an ongoing basis." Conclusions regarding intimacy level, therefore, are based upon such assessments rather than on operational elements such as information system linkages.

In the Stank, Goldsby, and Vickery study, logistics cost performance was considered to be an element of operational performance. In the current research, however, cost performance was decoupled from operational performance since it is considered to be a third, unique dimension of service performance. While the literature provides little guidance on the relationship between relational performance and cost performance, it is reasonable to extrapolate a relationship similar to that found between relational performance and operational performance in Stank, Goldsby, and Vickery. It also seems reasonable to surmise that a firm's relationships with customers might yield benefits similar to those resulting from a firm's relationships with suppliers. Scannell, Vickery, and Droge (2000) found a positive and highly significant relationship between a firm's use of supplier partnerships (a relational item) and cost performance. Thus:

H1: Logistics relational performance has a positive effect on operational performance.

H2: Logistics relational performance has a positive effect on cost performance.

Empirical studies in operations and marketing provide considerable support for links between service performance and customer satisfaction (Cronin and Taylor 1992; Crosby, Evans, and Cowles 1990; Leuthesser and Kohli 1995; Youngdahl and Kellogg 1997). Relational performance provides service suppliers with enhanced insight regarding customer needs and wants. Upon learning of these needs and wants, the service provider can focus on operational means of meeting them at the lowest cost to the customer as possible. Previous research on logistics service performance provides some indication that these relationships hold in a logistical service context (Daugherty, Stank, and Ellinger 1998; Innis and La Londe 1994; Stank, Goldsby, and Vickery 1999). The following hypotheses are based upon the findings from this stream of logistics research:

H3: Logistics relational performance has a positive effect on customer satisfaction.

H4: Logistics operational performance has a positive effect on customer satisfaction.

H5: Logistics cost performance has a positive effect on customer satisfaction.

Customer satisfaction with service capabilities has been shown to have significant and positive impact on cognitive attitudes and repurchase intentions in multiple business settings (Anderson, Fornell, and Lehmann 1994; Cronin and Morris 1989; Youngdahl and Kellogg 1997). Oliva, Oliver, and MacMillan (1992), for example, found that when satisfaction increased above a critical level, repeat purchases increased rapidly. An equal decline in repeat purchases occurred when satisfaction fell below a critical threshold. In logistics, the relationship between customer satisfaction and loyalty has also been strongly supported (Daugherty, Stank, and Ellinger 1998; Innis and La Londe 1994; Stank, Goldsby, and Vickery 1999). These findings provide the theoretical basis for our next hypothesis:

H6: Customer satisfaction has a positive effect on customer loyalty.

The literature supports a strong link between customer loyalty and market share. Loyal customers improve market share by purchasing a greater volume and variety of products (Reichheld and Sasser 1990). Loyalty increases repurchase behavior because loyal customers demonstrate greater resistance to counter persuasion and negative word-of-mouth (Dick and Basu 1994; Rust and Zahorik 1993). They often are less price sensitive, decreasing risk of defection due to competitive price undercutting. It follows that firms that successfully prevent defection enjoy a lasting, stronger market presence than those that incur frequent defections. Hence, our last hypothesis:

H7: Customer loyalty has a positive effect on market share.

Note that the model postulates that logistics service performance is linked with market share through its direct relationship with customer satisfaction and its indirect relationship (via customer satisfaction) with loyalty.

RESEARCH METHOD

This section reviews the research design. Discussion of the sample design is followed by descriptions of the measurement scales. Results of the measurement model assessment then precede a presentation of hypotheses tests.

Sample Design

The research model calls for data requirements from both the service provider and customer in the supplier-buyer relationship. Data from the service provider were necessary for the market share assessment. The providers' customers then provided responses to measures for the remaining five constructs. The sampling method proceeded with a two-step "snowball" or hierarchical sampling approach, where the 3PL participants were first contacted and asked to identify prospective respondents for the customer sample. Snowball sampling uses a procedure in which initial respondents are selected and additional respondents are then obtained from referrals or by other information provided by the initial respondents (Green and Tull 1978). The snowball approach was deemed most appropriate given the inherent difficulty of identifying extensive users of outsourced logistics services. This basic sampling approach has been used to yield sample frames that may be difficult to otherwise garner participation (Daugherty, Stank, and Ellinger 1998). Statistics referenced in Tables 3 and 4 illustrate the diverse composition of the customer sample.

The population for the 3PL dataset included all providers listed in Armstrong's Guide To Third Party Logistics Services Providers that offered comprehensive logistics services across the domestic US (1999). Firms that provided comprehensive services (either on an asset or non-asset basis) were preferred because they serve as "one-stop shops" for logistical services and are ultimately responsible for the quality of customer service rendered. This requirement narrowed the population to 68 firms. Of these 68, two firms were acquired by other 3PLs during the data collection timeframe and were dropped from consideration, reducing the number of 3PLs to 66 firms.

The top executive (president or CEO) of each 3PL was contacted by telephone to determine his or her willingness to respond to the questionnaire and to ascertain the appropriate contact for the information being sought. Letters containing the Internet address of a web-based questionnaire were mailed or e-mailed (depending upon respondent preference) to the 3PL executive to further encourage participation in the research. The survey instrument requested information regarding the 3PL's basic service characteristics, the executive's perceptions of the firm's own service performance, and assessments of the firm's market performance for the previous fiscal year. In communications by telephone and letter, the executive was asked to direct the survey to individuals best suited to complete the survey's various sections. The CEOs indicated in many instances that their marketing or finance executives would be best suited to answer the financial performance aspects of the survey while sales or operations executives would address the remainder. The web-based survey instrument afforded respondent firms the opportunity to gather the input of multiple executives by simply accessing the survey on-line to complete designated sections. The survey instrument was developed with the assistance of a panel of industry experts, consisting of two 3PL CEOs, two senior logistics researchers (external to the research team), and one president of a major consulting firm. Of the 66 3PL firms contacted, 35 (53%) fully completed the survey instrument. Descriptive statistics for these 35 firms appear in Table 1.

The 3PL respondents were also asked to identify up to eight primary customers. A second questionnaire tailored to the customer perspective of 3PL service performance was then e-mailed to these customers. Customers were asked to evaluate the performance of the named 3PL and to determine their levels of satisfaction and loyalty. When 3PLs failed to identify customers, the research team referenced the Armstrong and Associates (1999) directory once again to select at least two customers among those listed for that 3PL. In total 129 customers were identified: 99 by 3PLs themselves and 30 from the Armstrong and Associates directory. A total of 113 responses were obtained (89 from the 3PL provided sample and 24 from those identified from the directory) for an overall customer response rate of 89%. Of the 113 surveys returned 111 were deemed complete for the measures of interest in this research. This represents an effective response rate of 86%.

The target respondent within each customer firm was the individual that works most closely with the 3PL and is believed to be in the best position to evaluate service performance as well as the firm's satisfaction and loyalty to the 3PL. In the vast majority of cases, this resulted in responses from those in functional management, director, or senior management positions. Tables 2, 3, and 4 provide demographic information for the 111 complete responses. The data were segmented by industry, firm size, and magnitude of business relationship, respectively, in order to assess the possibility of firm biases in the sample. Analysis of variance results indicated no presence of biases by industry, firm size, or annual transaction size.

The mean and standard deviation figures reported for the measurement items (Appendix) provide insight regarding differences between customers identified by providers and customers identified through the Armstrong and Associates listing. While customers generally responded very favorably to their 3PL provider (as illustrated by mean values ranging between 4.76 and 5.16 for satisfaction and 4.78 and 5.68 for loyalty), the standard deviation values suggest that the degree of favorable attitudes was not universal across the sample. The same may be said of the measures of service performance. Statistical analyses were conducted to determine if differences existed across the customer groups. Results from these analyses indicated little difference between respondent groups. Specifically, t-tests of differences in means of responses to all customer perception variables between the provider identified customer group and customers identified through the Armstrong listing revealed only 5 significant differences out of 15 variables at the p

Measurement Scales

A review of literature in logistics, marketing, and operations provided measures for the customer survey. Service performance measurements across the operational, relational, and cost dimensions are derived primarily from the work of Bienstock, Mentzer, and Bird (1997), Parasuraman, Zeithaml, and Berry (1985) and Stank, Goldsby, and Vickery (1999). Consistent with Stank, Goldsby, and Vickery, these measures were submitted to protocol analysis within the specific industry of interest for assessment of measurement item relevance and importance, per Brown, Churchill, and Peter (1993). The aforementioned panel of industry experts provided critical input toward the final selection of appropriate service performance measures.

Measurements of customer satisfaction and loyalty were originally drawn from the marketing literature and subsequently validated in Stank, Goldsby, and Vickery (1999). Satisfaction measures are derived from the work of Leuthesser and Kohli (1995). These measures are global in nature, providing an overall judgment of the extent to which service matches expectations. They do not assess satisfaction with any single element of the service received or specific interaction associated with the service relationship (Anderson and Sullivan 1993). Likewise, measures of loyalty determine the general relationship between the customer's relative attitude toward the service provider and the customer's subsequent patronage behavior (Dick and Basu 1994; Morgan and Hunt 1994). Cronin and Morris (1989) and Morgan and Hunt (1994) are the primary sources for these measures.

The service provider's market share measurement is a relative measure reflecting the 3PL managers' assessments of their share of the North American 3PL market on a 7-point scale where 1 = Worst in Industry, 4 = Average, 7 = Best in Industry. In addition to this relative assessment of share, each service provider was asked to supply a good-faith estimate of the firm's actual market share (percentage of total North American 3PL market). The "good-faith estimate" is a statement of the actual percentage share enjoyed by the firm (open-ended). Estimated market share values ranged from a low of .0125% to 10.0%.

Thirty-three firms provided a subjective assessment, 15 provided an estimate of the percentage of share enjoyed by the firm, and 33 reported gross revenues in the Armstrong guide. The subjective scale was used in this research since it provided a larger available sample size than that available with the reported objective measure of market share. Past research has also found that managerial assessments are consistent with objective internal performance (Dess and Robinson 1984) and with external secondary data (Venkatraman and Ramanujam 1986). It should be noted that while the subjective assessment is used in this research, there was a significant, positive correlation between the subjective measure and actual measure of market share (r = 0.634). The correlation between the subjective measure and gross revenue figure reported by Armstrong (r = 0.679) is based on 32 common observations.

Upon concluding data collection, the measurement items were chosen for this research based on a review of all items collected in the survey instrument. The review of measurement items sought to identify measures that provided parsimonious representation of the model constructs and complete data for the sake of analysis. The final selection of appropriate items was verified in a confirmatory factor analysis of the measurement model (presented next). The Appendix provides a list of all measurement items used in the research. Basic descriptive statistics (mean and standard deviation) are provided as well as a correlation matrix.

Measurement Model Test

The measurement and structural models were tested by performing latent variable structural equation modeling using Bender's (1998) EQS for Windows (version 5.7) software. Structural equation modeling (SEM) is a statistical approach that has the capacity to comprehensively and simultaneously test hypotheses among observed and latent variables (Hoyle 1995). Prior to testing the structural model, confirmatory factor analysis (CFA) was performed to further assess the construct validity of the latent variables. CFA provides a more rigorous method for assessing unidimensionality than Cronbach alpha, exploratory factor analysis, and item-total correlations (Anderson and Gerbing 1988).

Table 5 reports the results of the CFA. The primary outputs of the CFA are the assessments of measurement model fit. The traditional chi-square fit test indicates how well the model-implied covariance matrix matches the covariances among the measured variables in the sample data (Bollen 1989; Hayduk 1987; Marsh and Balla 1994). In a reversal of the typical testing assumption, support for the null hypothesis of equal covariances is sought to demonstrate sound model fit. In our case, the chi-square statistic of 185.324 (90 d.f.) results in a p-value below 0.001 - indicating rejection of the null hypothesis and poor model fit.

Chi-square, however, is not the sole measure of fit. Other fit statistics have been developed to provide further indication of goodness-of-fit. These include the Bentler-Bonett Normed Fit Index (BBNFI), Bentler-Bonett Non-normed Fit Index (BBNNFI), and Comparative Fit Index (CFI). Two of the three statistics reported for the current model have values greater than the 0.90 cutoff suggested by the literature to indicate reasonable fit (Bagozzi and Yi 1988; Fornell and Larcker 1981; Hu and Bentler 1995). Bender (1990) and Byrne (1994) claim that the CFI is the single most important index since it accounts for sample size - a common bias in index calculations. The CFI for the current model is 0.931, indicating sound fit. Likewise, the root mean square residual (RMSR) is acceptable at 0.040. A measure of parsimonious fit, the normed chi-square, is the ratio of the chi-square value to degrees of freedom. The normed chi-square for our CFA is approximately 2.06:1. While no consensus regarding a specific value for this normed chi-square statistic has been reported, ratios varying from 2:1 to 5:1 have been offered as upper thresholds for acceptable fit (Arbuckle 1997).

According to Anderson (1987), convergent validity is supported when factor loadings (lambdas) demonstrate that measurement items load significantly on their designated latent variables. The standardized lambda estimates in Table 5 present ample evidence for this form of construct validity. The lowest value among the estimates is 0.585 (item OP2). Further support of convergent validity is provided by the lack of significant, theory-driven modification indices (e.g., the Wald and Lagrange Multiplier (LM) Tests in EQS). Significant modification indices suggest that better model fit is possible by either dropping a "trouble" item (per the Wald Test) or respecifying a measurement item to a latent variable other than that proposed in the CFA (per the Lagrange Multiplier Test). While the Wald Test indicated that no parameters should be dropped in the analysis, the LM Test suggested four model respecifications (i.e., measurement items that cross-loaded with latent variables beyond those for which they serve as indicators). In each case, however, the research team deemed that the current model specification provides for stronger nomological validity than that suggested by the LM Test statistic. Therefore, the measurement model remained unchanged.

Discriminant validity assesses whether two or more constructs are the result of a single underlying construct (Dr6ge and Germain 2000). The most common method used to assess discriminant validity is the nested model approach, where comparisons are made between the original measurement model and successive models with correlations (phis) among latent variables fixed equal to one. As long as the alternative measurement models fail to demonstrate better fit (significantly lower chi-square goodness-of-fit values) than the original, support for discriminant validity among constructs exists (Bagozzi and Yi 1988). This was found to be true in the current research. Given the overall sound assessment of the measurement model, attention will turn to the structural model and testing of hypothesized relationships.

RESULTS AND DISCUSSION

The seven hypotheses illustrated in Figure 1 were tested simultaneously in a single structural equation model with EQS (Bollen 1989). Results of the full model analysis are provided in Table 6. The fit statistics for the full model are comparable to those of the measurement model (chisquare = 197.129; df = 98; p

The first two hypotheses examine the direct influence that relational performance exerts on the other two dimensions of service performance (i.e., operational performance and cost performance). Model results indicate strong support for hypothesis HI, demonstrating that relational performance is positively related to operational performance. Likewise, hypothesis H2 is strongly supported, suggesting that relational performance also is positively related to cost performance in a positive manner. Support for each hypothesis is illustrated in the significance of the standardized parameter estimates (gamma values) and the associated t-values in Table 6.

Hypotheses H3 through HS suggest that each of the three service performance dimensions are positively related to customer satisfaction. Results of the analysis provide support for only one of the three hypotheses. Relational performance demonstrates a positive, significant relationship with satisfaction (at the 0.01 level of significance), but interestingly, operational performance and cost performance are found to have no significant relationship with satisfaction. Therefore, support is found for H3, but not H4 and H5.

Hypothesis H6 suggests that customer satisfaction is positively influenced by loyalty. The analysis provides strong support for this contention. Hypothesis H7 subsequently identifies market share as positively related to customer loyalty. The data support this hypothesis as well (at the 0.05 level of significance). Figure 2 portrays the hypotheses test results.

Similar to earlier work conducted by Daugherty, Stank, and Ellinger, the research identifies service performance as a key antecedent of market share through its relationship with customer satisfaction and loyalty (1998). The operationalization utilized in the current research, based upon the two-dimensional performance portrayal of Stank, Goldsby, and Vickery, refined with a third dimension representing cost to customer and re-explored the basic premise of the model introduced by Daugherty, Stank, and Ellinger (1998). The new operationalization fostered a deeper understanding of the interrelationships among the performance dimensions as well as providing some indication as to which elements are key to improvements in satisfaction, loyalty, and market share.

The current operationalization gave rise to an interesting result. Previous findings revealed that both relational and operational performance were significant predictors of customer satisfaction. The current findings refuted this. Specifically, the current findings cast relational performance as having the only significant relationship with customer satisfaction, while the operational performancesatisfaction and cost performance-satisfaction relationships were not significant. The results support the belief that basic operational service and cost performance are recognized as order qualifiers and not differentiators in the eyes of customers. Relational performance elements are the key to differentiating excellent 3PL service from the ordinary. Service providers must not, however, look past the fundamental delivery of time and place utilities that meet customer expectations at a rate commensurate with delivered value.

The findings support the critical importance of establishing strong relationships with customers to gain the insights needed to tailor services to enhance both operational effectiveness and cost efficiency. Strong relational performance allows firms to proactively seek information on customer preferences and use it to be more responsive. In addition, service providers ensure that resources are invested only in areas that customers perceive to be important, improving asset and capital utilization and enabling cost efficiency. Ultimately, it becomes difficult for competitors to intercede in supplier-buyer relations once they reach this level of intimacy.

Conversations with practitioners following data analysis provided anecdotal support for the finding. A global supply chain manager for a large computer manufacturer, for example, discussed his firm's supplier scorecard with the authors. The manager noted that quality service on key operational elements was the central element of the supplier scorecard. This element, however, was not given the highest weighting in the supplier evaluation algorithm because it represented a basic qualifier used to determine whether a service provider could even remain a potential supplier. In other words, 3PLs were not even considered unless they could demonstrate sustainable levels of high performance on basic operational service elements. Rather, suppliers were rewarded based upon their ability to respond to the manufacturer's specific requests, and do so under increasingly stringent cost requirements.

Such findings are not new in logistics. Stank, Daugherty, and Ellinger (1998) interviewed restaurant managers in support of research conducted in the food service industry and noted that distributors that go beyond "core" operating capabilities to do "whatever it takes" to serve customers on the customers' terms earn the continuing business of fast food restaurants. Such distributors provide frequent communications and easily accommodate special requests. Poorly regarded distributors exhibit deficiencies in non-core distribution capabilities characterized by communications breakdowns and inflexibility in responding to special requests or service failures. They may have developed basic delivery capabilities but tend to miss the point regarding customer focus. One manager told of a story that exemplifies the finding. The delivery truck blocked the restaurant parking lot and the manager could see potential diners pulling away because they could not enter the lot. When told of the problem, the driver expressed his need to proceed with his delivery so that he could complete his route on time. The manager noted that the driver was ensuring that the food distribution firm performed well on logistical service, although it was the wrong service to satisfy the customer. The driver was working toward an inappropriate and myopic performance goal that totally ignored the supply chain bottom line (Stank, Daugherty, and Ellinger 1998, p. 78).

Partial explanation for the finding may be found in the industry context of the research. In a situation where an external supplier assumes ownership of all or part of a key business process, successfully forging a strong relationship between the customer and service provider may be the single most critical determinant of customer satisfaction. The influence of relational characteristics may dwarf all other considerations. Our results suggest that the development of an effective relationship is prerequisite to success on all other performance dimensions. Strong relationships can even ensure rapid service recovery when operational or cost activities falter.

Previous, more narrowly focused research has identified a direct link between customer satisfaction and overall firm performance (Anderson, Fornell, and Lehmann 1994; Ittner and Larcker 1996). Although we did not hypothesize a direct link between customer satisfaction and market share (the overall firm performance measure in our study), it should be noted that the SEM model did not suggest one (per the LM Test). Similar to earlier logistics research, our findings indicate that customer satisfaction does not directly affect overall firm performance, but rather affects overall performance indirectly through customer loyalty. Additional research is needed to examine relationships among customer satisfaction, loyalty, and other measures of overall firm performance (e.g., ROI, ROA) in a holistic, service performance context.

An important contribution of the research was that the data represented both sides of the supplier-buyer dyad. Customers of third-party logistics providers evaluated their perceptions of service performance as well as their overall satisfaction with and loyalty to the provider. Our research linked customer perceptions to a market share indicator provided by the 3PL suppliers themselves that correlated significantly with objective data of market share drawn from a secondary data source. Such efforts in data collection help ensure that provider biases regarding relationships among model antecedents and outcomes are removed.

CONCLUSION

This research examined the relationships of service performance, customer satisfaction, customer loyalty, and market share in the context of a holistic model that allowed for the simultaneous testing of these relationships. The model introduced a three-dimensional conceptualization of service performance for the 3PL industry using data from service providers' customers to assess key constructs. The results indicated that this conceptualization might provide a more generic approach for capturing service performance in an industrial services context than previous customizations of Parasuraman, Zeithaml, and Berry's five-dimensional measurement scheme (1985, 1988). Future research, however, might explore alternate operationalizations of the logistics service performance constructs to ensure generalizability of results. The results also support Stank, Daugherty, and Ellinger's finding that relational performance is antecedent to operational performance. This antecedent relationship extends to a discrete assessment of cost performance. Finally, the research establishes an empirical link between customer loyalty and a perceptual measure of market share that is anchored by objective data from secondary sources. Future research should seek to assess the direct relationship between perceptions of service performance and objective measures of firm performance. Future research should also investigate whether length of relationship has an impact on the interrelationships identified in the model.

NOTES

Anderson, James C. (1987), "An Approach for Confirmatory Measurement and Structural Equation Modeling of Organizational Properties," Management Science, Vol. 33, No, 4, pp. 525-541. Anderson, Eugene W., Claes Fornell, and Donald R. Lehmann (1994), "Customer Satisfaction, Market Share, and Profitability: Findings from Sweden," Journal of Marketing, Vol. 58, No. 3, pp. 53-66.

Anderson, James C. and David W. Gerbing (1988), "Structural Equation Modeling in Practice: A Review of the Two-Step Approach," Psychological Bulletin, Vol. 103, No. 3, pp. 411-423. Anderson, Eugene W. and Mary Sullivan (1993), "The Antecedents and Consequences of Customer-Satisfaction for Firms," Marketing Science, Vol. 12, No. 2, pp. 125-143.

Arbuckle, James L. (1997), Amos Users' Guide, Version 3.6, Chicago: SmallWaters Corporation. Armstrong and Associates (1999), Armstrong's Guide To Third Party Logistics Services Providers, Vol. 7, Stoughton, WI: Armstrong and Associates, Inc.

Babakus, Emin and Gregory W. Boller (1992), "An Empirical Assessment of the SERVQUAL Scale," Journal of Business Research, Vol. 24, No. 3, pp. 253-268.

Babakus, Emin and W. Glynn Mangold (1989), "Adopting the 'SERVQUAL' Scale to Health Care Environment: An Empirical Assessment," in AMA Educators' Proceedings, P. Bloom et al. eds., Chicago, IL: American Marketing Association, p. 195.

Bearden, William 0. and Jesse E. Teel (1983), "Selected Determinants of Consumer Satisfaction and Complaint Reports," Journal ofMarketing Research, Vol. 20, No. 1, pp. 21-28.

Bagozzi, Richard P. and Youjae Yi (1988), "On the Evaluation of Structural Equation Models," Journal of the Academy of Marketing Science, Vol. 16, No. 1, pp. 74-94.

Bender, Peter M. (1990), "Fit Indexes, Lagrange Multipliers, Constraint Changes, and Incomplete Data in Structural Models," Multivariate Behavioral Research, Vol. 25, No. 1, pp. 163-172. Bentler, Peter M. (1998), EQS for Windows, Version 5.7, Encino, CA: Multivariate Software. Berry, Leonard L. (1995), On Great Service, A Framework for Action, New York: The Free Press.

Berry, Leonard L. andA. Parasuraman (1991), Marketing Services, Competing Through Quality, New York: The Free Press.

Bienstock, Carol C., John T. Mentzer, and Monroe Murphy Bird (1997), "Measuring Physical Distribution Service Quality," Journal of the Academy ofMarketing Science, Vol. 25, No. 1, pp. 31-44. Bojanic, David C. (1991), "Quality Measurement in Professional Services Firms," Journal of Professional Services Marketing, Vol. 7, No. 2, pp. 27-36.

Bollen, Kenneth A. (1989), Structural Equations with Latent Variables, New York: Wiley. Bowersox, Donald J. (1999), "21" Century Logistics: Making Supply Chain Integration a Reality," Keynote address to the Council of Logistics Management Annual Conference, Toronto, Canada, October 17-20.

Bowersox, Donald J., David J. Closs, and M. Bixby Cooper (2002), Supply Chain Logistics Management, New York: McGraw-Hill/Irwin, p. 34.

Brown, Tom J., Gilbert A. Churchill, Jr., and J. Paul Peter (1993), "Improving the Measurement of Service Quality," Journal of Retailing, Vol. 69, No. 1, pp. 127-139.

Byrne, Barbara M. (1994), Structural Equation Modeling with EQS and EQS/Windows, Thousand Oaks, CA: Sage Publications.

Carman, James M. (1990), "Consumer Perceptions of Service Quality: An Assessment of the SERVQUAL Dimensions," Journal of Retailing, Vol. 66, No. 1, pp. 33-55.

Cleveland, Gary, Roger G. Schroeder, and John C. Anderson (1989), "A Theory of Production Competence," Decision Sciences, Vol. 20, No. 4, pp. 655-668.

Crompton, John L. and Kelly J. Mackay (1989), "Users' Perceptions of the Relative Importance of Service Quality Dimensions in Selected Public Recreation Programs," Leisure Sciences, Vol. 11, No. 4, pp. 367-375.

Cronin, J. Joseph Jr. and Michael H. Morris (1989), "Satisfying Customer Expectations: The Effect on Conflict and Repurchase Intentions in Industrial Marketing Channels," Journal of the Academy ofMarketing Science, Vol. 17, No. 1, pp. 41-49.

Cronin, J. Joseph Jr. and Steven A. Taylor (1992), "Measuring Service Quality: A Reexamination and Extension," Journal ofMarketing, Vol. 56, No. 3, pp. 55-68.

Crosby, Lawrence A., Kenneth R. Evans, and Deborah Cowles (1990), "Relationship Quality in Services Selling: An Interpersonal Influence Perspective," Journal of Marketing, Vol. 54, No. 3, pp. 68-81.

Daugherty, Patricia J., Theodore P. Stank, and Alexander E. Ellinger (1998), "Leveraging Logistics/Distribution Capabilities: The Impact of Logistics Service on Market Share," Journal of Business Logistics, Vol. 19, No. 2, pp. 35-51.

Dess, Gregory G. and Richard B. Robinson (1984), "Measuring Organizational Performance in the Absence of Objective Measures," Strategic Management Research, Vol. 5, No. 3, pp. 265-273. Dick, Alan S. and Kunal Basu (1994), "Customer Loyalty: Toward an Integrated Conceptual Framework," Journal of the Academy ofMarketing Science, Vol. 22, No. 2, pp. 99-113.

Dr/ge, Cornelia and Richard Germain (2000), "The Relationship of Electronic Data Interchange with Inventory and Financial Performance," Journal of Business Logistics, Vol. 21, No. 2, pp. 209-230. Ferdows, Kasra and Arnoud De Meyer (1990), "Lasting Improvements in Manufacturing Performance: In Search of New Theory," Journal of Operations Management, Vol. 9, No. 2, pp. 168-184. Finn, David W and Charles W. Lamb, Jr. (1991), "An Evaluation of the SERVQUAL Scales in a Retail Setting," in Advances in Consumer Research, R. H. Holman and M. R. Solomon eds., Vol. 18, Provo, Utah: Association for Consumer Research.

Fornell, Claes (1992), "A National Customer Satisfaction Barometer: The Swedish Experience," Journal of Marketing, Vol. 55, No. 1, pp. 1-21.

Fornell, Claes and David F. Larcker (1981), "Evaluating Structural Equation Models with Unobservable Variables and Measurement Error," Journal ofMarketing Research, Vol. 18, No. 1, pp. 39-50.

Fournier, Susan T. and David G. Mick (1999), "Rediscovering Satisfaction," Journal of Marketing, Vol. 63, No. 4, pp. 5-23.

Green, Paul E. and Donald S. Tull (1978), Research For Marketing Decisions, 4th Ed. Englewood Cliffs, NJ: Prentice Hall, Inc., pp. 210-211.

Hayduck, Leslie A. (1987), Structural Equation Modeling with LISREL: Essentials and Advances, Baltimore, MD: Johns Hopkins University Press.

Hayes, Robert H. and Steven C. Wheelwright (1984), Restoring Our Competitive Edge: Competing Through Manufacturing, New York: Wiley.

Hill, Terry (1989), Manufacturing Strategy Text and Cases, Homewood, IL: Irwin.

Hoyle, Rick H. (1995), "The Structural Equation Modeling Approach: Basic Concepts and Fundamental Issues," in Structural Equation Modeling: Concepts, Issues, and Applications, R. H. Hoyle ed., Thousand Oaks, CA: Sage Publications, pp. 1- 15.

Hu, Li-Tze and Peter M. Bentler (1995), "Evaluating Model Fit," in Structural Equation Modeling: Concepts, Issues, and Applications, R. H. Hoyle ed., Thousand Oaks, CA: Sage Publications, pp. 76-99.

Innis, David E. and Bernard J. La Londe (1994), "Customer Service: The Key to Customer Satisfaction, Customer Loyalty, and Market Share," Journal of Business Logistics, Vol. 15, No. 1, pp. 1-27.

Ittner, Christopher D. and David F. Larcker (1996), "Measuring the Impact of Quality Initiatives on Firm Financial Performance," in Advances in the Management of Organizational Quality, S. Ghosh and D. Fedor eds., Vol. 1, Greenwich, CT: JAI, pp. 1-37.

Jacoby, Jacob and David B. Kyner (1973), "Brand Loyalty vs. Repeat Purchasing Behavior," Journal ofMarketing Research, Vol. 10, No. 1, pp. 1-9.

Johnson, Linda L., Michael J. Dotson, and Brian J. Dunlop (1988), "Service Quality Determinants and Effectiveness in the Real Estate Brokerage Industry," The Journal of Real Estate Research, Vol. 3, No. 4, pp. 21-36.

Jones, Thomas 0. and W. Earl Sasser, Jr. (1995), "Why Satisfied Customers Defect," Harvard Business Review, Vol. 73, No. 6, pp. 88-99.

Krajewski, Lee J. and Larry P. Ritzman (1987), Operations Management, Strategy and Analysis, I't Ed. Reading, MA: Addison Wesley.

Leuthesser, Lance and Ajay K. Kohli (1995), "Relational Behavior in Business Markets," Journal of Business Research, Vol. 34, No. 1, pp. 221-233.

Lovelock, Christopher H. (1983), "Classifying Services to Gain Strategic Marketing Insights," Journal ofMarketing, Vol. 47, No. 3, pp. 9-20.

Marsh, Herbert W. and J. R. Balla (1994), "Goodness-of-Fit in Confirmatory Factor Analysis: The Effects of Sample Size and Model Parsimony," Quality and Quantity, Vol. 28, pp. 185-197. Morgan, Robert M. and Shelby D. Hunt (1994), "The Commitment-Trust Theory of Relationship Marketing," Journal ofMarketing, Vol. 58, No. 3, pp. 20-38.

Oliva, Terence A., Richard L. Oliver, and Ian C. MacMillan (1992), "A Catastrophe Model for Developing Service Satisfaction Strategies," Journal of Marketing, Vol. 56, No. 3, pp. 83-95. Parasuraman, A., Valerie A. Zeithaml, and Leonard L. Berry (1985), "A Conceptual Model of Service Quality and Its Implications for Future Research," Journal of Marketing, Vol. 49, No. 4, pp. 41-50.

Parasuraman, A., Valerie A. Zeithan-fl, and Leonard L. Berry (1988), "SERVQUAL: A Multiple-Item Scale for Measuring Consumer Perceptions of Service Quality," Journal of Retailing, Vol. 64, No. I pp. 12-40.

Parasuraman, A., Leonard L. Berry, and Valerie A. Zeithaml (1991), "Refinement and Reassessment of the SERVQUAL Scale," Journal of Retailing, Vol. 67, No. 4, pp. 420-450.

Porter, Michael E. (1980), Competitive Strategy, New York: The Free Press. Porter, Michael E. (1985), Competitive Advantage, New York: The Free Press.

Reichheld, Frederick R and W. Earl Sasser, Jr. (1990), "Zero Defections: Quality Comes to Services," Harvard Business Review, Vol. 68, No. 5, pp. 105-111.

Roth, Aleda V. and J. G. Miller (1990), "Manufacturing Strategy, Manufacturing Strength, Managerial Success, and Economic Outcomes," in Manufacturing Strategy, the Research Agenda for the Next Decade, Proceedings for the Joint Industry University Conference On Manufacturing Strategy, 1. E. Ettlie, M. C. Burstein, and A. Fiegenbaum eds., Ann Arbor, MI, pp. 85-96.

Roth, Aleda V. and Marjolijn van der Velde (1991), "Operations as Marketing: A Competitive Service Strategy," Journal of Operations Management, Vol. 10, No. 3, pp. 303-328.

Rust, Roland T. and Anthony J. Zahorik (1993), "Customer Satisfaction, Customer Retention, and Market Share," Journal of Retailing, Vol. 69, No. 2, pp. 193-215.

Scannell, Thomas V., Shawnee K. Vickery, and Cornelia L. Dr6ge (2000), "Upstream Supply Chain Management and Competitive Performance in the Automotive Supply Industry," Journal of Business Logistics, Vol. 21, No. 1, pp. 23-48.

Sharma, Arun and Douglas M. Lambert (1990), "Segmentation of Markets Based on Customer Service," International Journal of Physical Distribution and Logistics Management, Vol. 20, No. 7, pp. 19-27.

Stank, Theodore P., Patricia J. Daugherty, and Alexander E. Ellinger (1998), "Pulling Customers Closer Through Logistics Service," Business Horizons, Vol. 41, No. 5, pp. 74-80.

Stank, Theodore P., Thomas J. Goldsby, and Shawnee K. Vickery (1999), "Effect of Service Supplier Performance on Satisfaction and Loyalty of Store Managers in the Fast Food Industry," Journal of Operations Management, Vol. 17, No. 2, pp. 429-447.

Venkatraman, N. and V. Ramanujam (1986), "Measurement of Business Performance in the Absence of Objective Measures," Strategic Management Review, Vol. 11, No. 4, pp. 801-814.

Wood, Carolyn H., Larry P. Ritzman, and Davendra Sharma (1990), "Intended and Achieved Competitive Priorities: Measures, Frequencies, and Financial Impact," in Manufacturing Strategy, the Research Agenda for the Next Decade, Proceedings for the Joint Industry University Conference On Manufacturing Strategy, J. E. Ettlie, M. C. Burstein, and A. Fiegenbaum eds., Ann Arbor, MI, pp. 225-232.

Youngdahl, William E. and Deborah L. Kellogg (1997), "The Relationship between Service Customers' Quality Assurance Behaviors, Satisfaction, and Effort: A Cost of Quality Perspective," Journal of Operations Management, Vol. 15, No. 1, pp. 19-32.

Zeithaml, Valerie A. (2000), "Service Quality, Profitability, and the Economic Worth of Customers: What We Know and What We Need to Learn," Journal of the Academy ofMarketing Science, Vol. 28, No. 1, pp. 67-85.

Zeithan-il, Valerie A., A. Parasuraman, and Leonard L. Berry (1990), Delivering Service Quality, Balancing Customer Perceptions and Expectations, New York: The Free Press.

Theodore P. Stank

Michigan State University

Thomas J. Goldsby

The Ohio State University

Shawnee K. Vickery

Michigan State University

Katrina Savitskie

University of Memphis

ABOUT THE AUTHORS

Theodore P. Stank (Ph.D. in Marketing and Distribution from The University of Georgia) is Associate Professor of Logistics at Michigan State University. He is coauthor of 21st Century Logistics: Making Supply Chain Integration a Reality, and has published articles in the areas of logistics strategy, customer relevance, and integration in various journals including Business Horizons, Journal of Business Logistics, Journal of Operations Management, Supply Chain Management Review, and Transportation Journal.

Thomas J. Goldsby (Ph.D. in Marketing and Logistics from Michigan State University) is Assistant Professor of Logistics at Ohio State University. His research interests focus on logistics customer service and supply chain integration. He also has interest in the development and implementation of environmental or "green" business practices. He has published articles in academic and professional journals, such as the Journal of Business Logistics, International Journal of Logistics Management, Supply Chain Management Review, and Journal of Operations Management.

Shawnee K. Vickery (Ph.D. in Business Administration from the University of South Carolina) is Professor of Operations Management at Michigan State University. Her research has been published in Decision Sciences, Journal of Operations Management, The European Journal of Operational Research, The International Journal of Production Research, Journal of Product Innovation Management, Production and Inventory Management, The Journal of Supply Chain Management, International Journal of Physical Distribution and Logistics Management, International Journal of Operations and Production Management, and Journal of Business Logistics.

Katrina Savitskie is an Assistant Professor in Marketing and Supply Chain Management at the University of Memphis. She is completing a doctorate in Marketing, Logistics, and International Business from Michigan State University. Her main research interests include information technology and its role in the supply chain. She has been published in Decision Line and several conference proceedings.

Copyright Council of Logistics Management 2003
Provided by ProQuest Information and Learning Company. All rights Reserved