The effect of shipping fee structure on consumers' online evaluations ...

J. of the Acad. Mark. Sci. (2012) 40:759–770 DOI 10.1007/s11747-011-0281-2

ORIGINAL EMPIRICAL RESEARCH

The effect of shipping fee structure on consumers’ online evaluations and choice Nevena T. Koukova & Joydeep Srivastava & Martina Steul-Fischer

Received: 22 February 2011 / Accepted: 4 August 2011 / Published online: 21 August 2011 # Academy of Marketing Science 2011

Abstract This research examines how consumers respond to different shipping fee structures. Focusing on two of the most common shipping fee structures, flat rate shipping and threshold-based free shipping, we first demonstrate that offer evaluations are less (more) favorable with thresholdbased free shipping when order value is below (above) the free shipping threshold compared to flat rate shipping. However, when an alternative more important referent is present, the effect of shipping fee structure is attenuated. Second, we show that although perceptions of shipping fees as a profit generator are higher (lower) under thresholdbased free shipping relative to flat rate shipping for order values below (above) the free shipping threshold, providing a justification for the shipping fee by explicitly linking it to

Listed alphabetically, the three authors contributed equally to the article. This research was partially funded by the Alison and Norman Axelrod’74 endowed summer research fellowship awarded to the first author by Lehigh University and by the summer research grant awarded to the second author by the Robert H. Smith School’s Center for Excellence in Service and the Research Committee. N. T. Koukova (*) College of Business and Economics, Lehigh University, Bethlehem, PA 18015, USA e-mail: [email protected] J. Srivastava Robert H. Smith School of Business, University of Maryland, College Park, MD 20742, USA e-mail: [email protected] M. Steul-Fischer Institute of Marketing, School of Business and Economics, University of Erlangen-Nuremberg, 90403 Nuremberg, Germany e-mail: [email protected]

delivery costs encourages consumers to view the shipping fee as a cost of doing business rather than as a profit generator, thus raising offer evaluations. Keywords Shipping fee . Flat rate shipping . Thresholdbased free shipping . Free shipping . Progressive rate shipping . Online shopping

Direct-to-consumer business models, where the physical good is spatially separated from the consumer (Lewis 2006), necessitate firms’ incurring order assembly and delivery costs (Rosen and Howard 2000). Firms typically recover these costs by passing them on to consumers in the form of shipping fees (Barsh et al. 2000). The growth in e-commerce (online spending in the U.S. reached $142.5 billion in 2010, an increase of about 10.5% over 2009) has accentuated the importance of shipping fees. Designing a shipping policy is one of the biggest quandaries firms face as it requires balancing the need to recover delivery costs and the need to attract and retain consumers. Despite the topic’s importance, only a handful of studies have examined shipping fees (Lewis 2006; Lewis et al. 2006; Schindler et al. 2005). Further, the existing studies restrict their examination to comparing promotional offers of free shipping to threshold-based free shipping (Lewis 2006; Lewis et al. 2006). This paper attempts to contribute to the sparse literature by examining the effects of two common shipping fee structures—flat rate and threshold-based free shipping. With flat rate shipping, retailers charge a fixed fee regardless of order value. With threshold-based free shipping, retailers waive the shipping fee for orders above a specific amount, but charge a fixed fee otherwise. Estimates suggest that 49% of online retailers use flat rate

760

and 43% use threshold-based free shipping (ComCult Research 2003). The experimentation occurring online and the variety of fees indicate that retailers are grappling with the effect of shipping fees on consumer demand. For example, Amazon decreased its free shipping threshold from $99 to $25, while Ann Taylor Loft changed its policy from threshold-based free shipping to flat rate to progressive rate shipping. Thus, from a substantive as well as a practical perspective, there is a need to explore how consumers respond to different shipping fee structures. The primary purpose of this research is to experimentally examine how consumers evaluate and respond to flat rate and threshold-based free shipping. We focus on these two shipping fee structures because these are the most prevalent in the marketplace and most other structures are derived from them (e.g., flat rate shipping fee that increases stepwise based on order value). Importantly, we examine factors that may moderate the effect of shipping fee structure on consumer evaluations and thus provide a better understanding of the conditions under which different shipping fee structures affect consumer response. Since the shipping fee depends on order value for threshold-based free shipping, we first examine whether the effect of flat rate and threshold-based free shipping on online offer evaluations depends on order value. The results reveal that offer evaluations are less (more) favorable with threshold-based free shipping when order value is below (above) the threshold compared to flat rate shipping. The theoretical argument is that, for threshold-based free shipping, the possibility of free shipping is the natural referent against which any shipping fee is evaluated. While for order values below the threshold the shipping fee is likely to be coded as a loss relative to the referent and lead to lower evaluations, for order values above the threshold the shipping fee is likely to be coded as a gain and result in higher evaluations. However, if another referent that is more important than the shipping fee is present, such as product price discount, the effect of the shipping fee is attenuated regardless of order value. Given that the price of the focal product (and related discount) is more important than shipping fees (Hamilton and Srivastava 2008), such an alternative referent is likely to overshadow the effect of shipping or draw attention away from the shipping fee, particularly from the natural referent of free shipping associated with threshold-based free shipping. The presence of an alternative important referent thus moderates the interaction effect of shipping fee structure and order value on consumer evaluations of online offers. Second, we propose that the shipping fee structure itself may prompt consumers to think of retailer profit motives for employing this structure (Schindler et al. 2005). We examine consumers’ inferences about the extent to which shipping fees are used to generate profits and whether

J. of the Acad. Mark. Sci. (2012) 40:759–770

providing a justification for the shipping fee alters the effect of shipping fee structure on inferences and evaluations. Our results suggest that for order values below (above) the threshold for free shipping, consumers are more (less) likely to infer that shipping fees are used to increase profits when the retailer employs threshold-based free shipping than flat rate shipping. However, retailers may be able to channel such inferences in a way to improve offer evaluations by justifying the shipping fee structure employed. Specifically, we show that linking the shipping fee to the actual cost of delivery alters consumers’ inferred motive regarding shipping fees and improves evaluations. The provision of justification for the shipping fee thus moderates the interaction effect of shipping fee structure and order value on consumer evaluations. This research contributes to the literature by systematically examining the effect of two common shipping fee structures on consumers’ inferences and evaluations, and testing boundary conditions for the effects. Specifically, adding to previous research (e.g., Lewis 2006; Lewis et al. 2006), this paper investigates the effect of flat rate shipping and compares it with threshold-based free shipping. We demonstrate that a specific shipping charge may differentially affect offer evaluations depending on whether it is framed as a flat rate or a threshold-based shipping fee. We offer a theoretical explanation for the observed effect and provide recommendations to retailers on how to frame the shipping fee in order to increase consumers’ evaluations. Further, given that data constraints prevented an examination of moderators in previous research (e.g., Lewis 2006), the use of experimental methodology allows us to identify boundary conditions thus adding to our understanding of the effects of shipping fee structure. In particular, we demonstrate that the presence of an alternative important referent and the provision of justification for the shipping fee can systematically alter (moderate) the interaction effect of shipping fee structure and order value on offer evaluations. The next section provides the conceptual background and develops specific predictions for the effects of shipping fees and potential moderators. We then report three experiments that test the hypotheses and explore boundary conditions. The paper concludes by discussing the theoretical and managerial implications of the findings including directions for future research.

Conceptual background The explicit separation of shipping fee from the price of a product is an example of partitioned pricing where the total price is divided into two or more mandatory components such as a base price and a surcharge (e.g., Chakravarti et al.


2002; Cheema 2008; Hamilton and Srivastava 2008; Hamilton et al. 2010; Morwitz et al. 1998; Xia and Monroe 2004). However, partitioning the total price into product price and shipping fee has only been used as a context in which to study the effect of price partitioning rather than the differential effects of various shipping fee structures on consumer response. Specifically, while previous studies have compared partitioned (i.e., shipping charge listed separately) and non-partitioned (shipping included in the total price) offers, our focus is on the effect of different shipping fee structures such as flat rate and threshold-based free shipping. More relevant to the current investigation, in an analysis of online book purchase data, Smith and Brynjolfsson (2001) find that consumers are almost twice as sensitive to changes in the shipping fee than to changes in the price of the book. Hamilton and Srivastava (2008) also argue that consumers are more sensitive to the price of partitioned components that are less important and provide relatively low consumption benefits (e.g., shipping) than to the price of components that are more important and provide relatively high consumption benefits (e.g., the product). Lewis et al.’s (2006) analysis reveals that promotional free shipping increases order incidence but reduces order values, whereas threshold-based free shipping encourages larger orders but has little impact on order incidence. Although free shipping raises order incidence, the increased revenues are not enough to offset the cost of shipping. In sum, a shipping fee is an important component that is likely to influence consumers’ online purchasing behavior and decision making. While previous research has explored the effects of promotional free shipping and thresholdbased free shipping (Lewis et al. 2006), this paper is the first to examine the effect of flat rate shipping and compare it to the effect of threshold-based free shipping on consumer evaluations and inferences. Further, we study the role of order value on the effect of shipping fee structure on consumers’ inferences and evaluations, as well as the moderating role of an alternative important referent and shipping fee justification on the interaction effect of shipping fee structure and order value on consumer evaluations. Order value as a moderator of the effect of shipping fee structure on offer evaluations Proponents of both flat rate and threshold-based free shipping structures speculate on identical reasons for their success. Both are believed to be “clean” and “simple” for consumers and can be easily communicated such that it is clear to shoppers how much they are paying for shipping. Both shipping fee structures are believed to increase order value. While flat rate shipping presumably increases order

761

value because “it does not cost any more,” threshold-based free shipping encourages people to increase order value and reach the free shipping threshold to avoid the shipping fee (Lewis et al. 2006). Order value thus assumes particular significance for threshold-based free shipping. We argue that a critical difference in evaluating offers with flat rate and threshold-based free shipping lies in the referent used to evaluate the shipping fee. The notion that evaluations are reference dependent and based on a comparison with a standard or a referent is well acknowledged (Kahneman and Tversky 1979). Evaluations depend on both the absolute and relative value compared to some referent. For example, when a potential buyer assesses the value of a house, she is likely to anchor on some reference (e.g., comparable houses in the area, last house sold, average house) and make a judgment by comparing the current house with the referent (Hsee and Leclerc 1998). However, if the buyer is shown two houses, she is less likely to think about the referent used in the earlier evaluation and is more likely to make a judgment by comparing the two houses. The reason for the shift in reference from the single house to the two houses is that people rely more on vivid, easily available information than on pallid background information that requires more effortful processing to access (Hsee and Leclerc 1998). In general, the preference reversals between separate and joint evaluations can be explained by the “evaluability” hypothesis—attributes that are hard to evaluate independently loom larger in joint evaluations, while attributes that are easy to evaluate independently loom larger in separate evaluations (Hsee et al. 1999). We conceptualize a similar “shift in reference” mechanism underlying the effects of shipping fee structures on consumer evaluations. First, consider an online offer with flat rate shipping. In evaluating the fee, consumers are likely to use referents such as the shipping fee last paid, the fee charged by an online retailer used frequently, or some abstract distributional information such as “average fee” based on prior experience or knowledge. Now consider an online offer with threshold-based free shipping. For such offers consumers are likely to use the potential free shipping as the natural referent. Flat rate and thresholdbased free shipping are thus likely to differ in the frame of reference that consumers use in their evaluations. Specifically, while flat rate shipping is likely to invoke a memorybased referent based on prior knowledge or experience, in the case of threshold-based free shipping the referent shifts to stimulus-based free shipping. In other words, flat rate shipping is typically less evaluable as it is relatively difficult to assess given the lack of a precise referent. In contrast, threshold-based free shipping is more evaluable as it is relatively easy to assess, given the more vivid and precise referent of free shipping (Hsee et al. 1999). Unlike

762

Hsee et al. (1999), the evaluability of shipping fees is inherent to the shipping fee structure and does not depend on joint versus separate evaluation (which is not considered in the current research). Our focus is on how consumers evaluate a product offer with a specific shipping fee structure (separate evaluation), and we do not consider situations in which two offers with different shipping fee structures are evaluated simultaneously (joint evaluation). The implication is that because threshold-based free shipping provides a precise referent against which to evaluate the shipping fee, order value assumes differential significance. When the order value is below the threshold, the shipping fee is likely to be coded as a loss or penalty relative to the possibility of free shipping. For equivalent total prices, consumer evaluations are likely to be less favorable relative to an offer with flat rate shipping. Conversely, when the order value is above the threshold, the lack of shipping fee is likely to be coded as a gain or a bonus and offer evaluations are likely to be more favorable relative to an offer with flat rate shipping. The differential effect of the two shipping fee structures on offer evaluations for order values below or above the threshold is likely to be driven, in part, by perceptions of shipping policy fairness. Previous research highlights the importance of perceived price fairness in influencing consumers’ reactions to prices (e.g., Campbell 1999; Kahneman et al. 1986). Moreover, fairness perceptions mediate the effect of price discounts or price premiums relative to other consumers on purchase satisfaction (e.g., Darke and Dahl 2003). Thus, when the order value is below the threshold for threshold-based free shipping, the loss associated with incurring the shipping fee (compared to the referent of free shipping) will be perceived as less fair than a flat rate that every consumer must pay, leading to less positive offer evaluations. On the other hand, for offers above the threshold, the gain associated with free shipping will be perceived as fairer than flat rate shipping, leading to more positive offer evaluations. Alternative important referent as a moderator of the effect of shipping fee structure and order value on offer evaluations Our conceptualization of the effects of shipping fee structure is based on different referents being used in the evaluations. It follows that providing an alternative referent that is more important than the shipping charge may draw consumer attention away from the referents used in evaluating the shipping fee. The price of the focal product has been shown to be more important than the shipping fee (e.g., Hamilton and Srivastava 2008). Further, in general, consumers are bargain driven such that the subjective importance of a bargain is often greater than what would


be expected based on the economic value of the bargain (Darke and Dahl 2003). As such, an online offer that emphasizes a discount off the regular product price is likely to draw consumers’ attention away from the shipping fee, thereby attenuating the effect of shipping fee structure on consumer evaluations. Specifically, consumers are likely to shift their focus to the more important alternative referent when evaluating the online offer (Kahneman and Miller 1986). We thus predict that regardless of order value, consumer evaluations of online offers will not differ across flat rate and threshold-based free shipping when an alternative important referent such as a price discount is provided. Study 1a examines the effect of shipping fee structure, order value, and the presence of another important referent on offer evaluations. H1a: Compared to an online offer with flat rate shipping, a threshold-based free shipping offer will be evaluated less (more) positively when the order value is below (above) the threshold for free shipping in the absence of an alternative important referent. H1b: When an alternative important referent is present, consumer evaluations will not differ across flat rate and threshold-based free shipping structures, regardless of order value. In examining how threshold-based free shipping may affect offer evaluations, it is important to consider the level of the free shipping threshold itself. It is possible that the referent of free shipping may loom larger for thresholdbased free shipping when the order value is just below (versus well below) the free shipping threshold. Just missing out on reaching the threshold and incurring a “loss” may have a more pronounced effect on consumer evaluations when the threshold is easy to reach rather than difficult to reach (e.g., Kahneman and Miller 1986). For example, Lassar et al. (1997) found that just missing out on a warranty created a less favorable reaction relative to having the problem within the warranty period or long after its expiration. Study 1b examines the robustness of the results across different free shipping thresholds for an order value below the threshold and rules out the level of closeness of the threshold as an alternative explanation for the lower evaluations when threshold-based free shipping is employed (relative to flat rate shipping). Inferences of profit motive and the moderating role of shipping fee justification on the effect of shipping fee structure and order value on offer evaluations Given that consumers are skeptical of shipping charges and may view them as a way for firms to make additional profits (Schindler et al. 2005), we examine whether consumer perceptions of the extent to which a shipping


fee is used to generate profits rather than as a means to recover delivery costs is affected by whether the shipping fee structure is flat rate or threshold-based free shipping. According to attribution theory, consumers will seek to meaningfully explain firms’ shipping fee structure in terms of underlying reasons such as profits and/or cost (Fiske and Taylor 1991). Attribution research indicates that the inclination to search for causal explanations for others’ behavior is particularly high following an unexpected and/ or negative outcome (Folkes 1984; Kelley 1967). For example, the underlying motives of some sellers for a price increase may be perceived as positive or benevolent, whereas other sellers’ motives may be perceived to be negative, driven by the extent to which the seller is taking advantage of or exploiting consumers (Campbell 1999). Campbell (1999) demonstrates that a seller’s inferred motive is an important causal construct that underlies perceptions of price (un)fairness. As such, we argue that the extent to which consumers infer a profit generating motive for a retailer’s shipping fee is likely to affect how the retailer’s online offer is evaluated. Attribution theory highlights locus (whether the cause resides within or outside of the actor) and controllability (whether the cause is subject to volitional alteration or not) as two central characteristics of causality (Weiner 2000). They both trigger judgments about responsibility for the obtained outcome, affecting consumers’ evaluations and satisfaction (e.g., Tsiros et al. 2004). Given the salience of free shipping for a threshold-based free shipping structure, consumers paying a shipping fee for order values below the threshold are likely to seek an explanation for why the retailer charges a fee for orders below the threshold. They may believe that since profits are relatively low on small order values, retailers compensate by charging a shipping fee. Alternatively, they may believe that retailers attempt to make a profit on shipping fees from consumers with low order values in order to subsidize the high order value consumers. Consumers facing a reasonable flat rate fee will be less likely to search for a causal explanation for the fee because all consumers pay the same shipping fee, regardless of order value. Supporting this notion, Haws and Bearden (2006) demonstrate that differences between consumers result in the greatest perception of unfairness and the lowest overall satisfaction (relative to seller, time, and price setter differences). Thus, for order values below the free shipping threshold, consumers facing thresholdbased free shipping will be more likely to infer a profit motive for the shipping charge (e.g., retailer is trying to make money on shipping) relative to consumers presented with flat rate shipping that every consumer is paying. In contrast, consumers with order values above the threshold are less likely to seek an explanation and less likely to

763

regard the shipping fee as a way for retailers to generate profits. Thus, consumer perceptions of shipping fees as a profit generator are likely to be higher (lower) when the order value is below (above) the threshold for thresholdbased free shipping relative to flat rate shipping. Further, we examine whether providing a (cost) justification for the shipping fee moderates the effect of shipping fee structure and order value on perceptions of shipping fees as a profit generator and offer evaluations. While people believe that retailers are entitled to a certain level of profit on the price of the product (Kahneman et al. 1986), the same entitlement does not hold for shipping fees. However, people recognize that retailers incur delivery costs and are entitled to recover the “cost of doing business” (Schindler et al. 2005). The implication is that consumer inferences of shipping fees as a profit generator and evaluations of online offers may be more favorable when retailers link the shipping fee structure to their actual shipping cost. When a justification for a negative outcome is provided, the outcome is perceived as fairer than when the same outcome arises with no justification offered (Bies and Shapiro 1988). We suggest that when a credible statement with a rationale for the shipping fee structure (that links the shipping fee to the actual cost of shipping) is provided, the effect of shipping fee structure on consumer perceptions of shipping fees as a profit generator and offer evaluations is likely to be attenuated. For example, a retailer may include a statement on its website claiming that the shipping fees correspond to what the shipping company is actually charging. Thus, for order values below the threshold, consumers provided with such a justification will no longer view the shipping fee as a penalty but as a cost of doing business. As such, regardless of order value, consumer perceptions of shipping fees as a profit generator and evaluations of the online offers are not likely to differ across the flat rate and threshold-based free shipping structures when the rationale for shipping fees is clearly communicated. Study 2 examines the moderating role of justification for shipping fee structure on consumer perceptions of shipping fees as a profit generator and evaluations of online offers. H2a: Compared to an online offer with flat rate shipping, a threshold-based free shipping offer will lead to higher (lower) perceptions of shipping fees as a profit generator when the order value is below (above) the threshold for free shipping. H2b: When a justification for the shipping fee is present, perceptions of shipping fees as a profit generator and offer evaluations will not differ across flat rate and threshold-based free shipping structures, regardless of order value.

764


Study 1a: order value and an alternative important referent Study 1a tests H1, which predicts that compared with flat rate shipping, an online offer with threshold-based free shipping will be evaluated less (more) favorably when the order value is below (above) the free shipping threshold. However, the effect will be attenuated in the presence of an important alternative referent. The rationale is that consumers will focus on the important referent, drawing attention away from the shipping fee and specifically the natural referent of free shipping associated with threshold-based free shipping, regardless of order value. Method One hundred and seventy-nine undergraduate students (47% female, age ranging from 18 to 26 years, mean age =20.1 years) from a U.S. university participated in the study for extra credit. Participants were assigned at random to one of eight conditions of a 2 (shipping fee structure: flat rate and threshold-based free shipping) x 2 (order value: $34.99 and $44.99) x 2 (alternative referent: none and $2 price discount) between-subjects design. Participants imagined that they were buying a new flash drive. They were told that they came across an online store that had suitable flash drives and the model they liked was described in detail. The online store either charged a flat rate shipping fee of $6 for all order values, or shipped free for order values above $40 and $6 otherwise. The flash drive was priced either below or above the threshold of $40 for free shipping at $34.99 and $44.99, respectively. When the alternative referent was present, it was stated that $34.99 ($44.99) was a discounted price, and the regular price was $36.99 ($46.99). Another condition was added where the flash drive was priced at $50.99 with a referent of $52.99 as the regular price and threshold-based free shipping. The total price was the same as the high order value with flat rate shipping condition (44.99+6.00=$50.99). Offer evaluations were measured by averaging responses to three seven-point items (α=.93; “The online store’s offer is” 1 = Unattractive/Undesirable/Unreasonable; 7 = Attrac-

tive/Desirable/Reasonable). Perceptions of quality (α=.83; “The flash drive is:” 1 = Low quality/Worse than others/Not appealing; 7 = High quality/Better than others/Appealing) and price perceptions (r=.63; “The price that the online store is asking for the flash drive is” 1 = Low/Unacceptable; 7 = High/Acceptable) were also measured. Results and discussion Offer evaluations A 2x2x2 ANOVA (the ninth condition was excluded) revealed a significant two-way interaction between shipping structure and order value (F(1, 152)= 7.45, p.13).

Table 1 Study 1a: means (standard deviations) of dependent measures Shipping fee structure

Threshold-based free shipping

Order value (Total price)

$34.99 ($40.99)

$44.99 ($44.99)

4.02 (1.00)

4.72 (1.61)

4.75 (1.24)

4.73 (1.28)

No alternative referent Offer evaluations Alternative referent Offer evaluations

Flat rate shipping $50.99 ($50.99)

4.74 (1.54)

$34.99 ($40.99)

$44.99 ($50.99)

4.95 (1.35)

3.37 (1.37)

4.46 (1.34)

4.39 (1.58)


Discussion Study 1a supports H1 by demonstrating that order value moderates the effect of shipping fee structure on consumer evaluations. Online offers were evaluated less (more) favorably with threshold-based free shipping compared to flat rate shipping when the order value was below (above) the threshold for free shipping. The findings are consistent with the argument that the referent used to evaluate the fee under threshold-based free shipping structure is free shipping. A follow-up study with 50 participants from an online paid panel provides further support for this premise. Using a scenario similar to that used in the main study, except for an offer, participants were randomly assigned to either flat rate or threshold-based free shipping conditions. Participants were asked to respond to the following questions: “In evaluating the retailer’s shipping fee policy, you compared the current shipping fee to the shipping fees you have typically paid in the past” and “In evaluating the retailer’s shipping fee policy, you compared the current shipping fee to free shipping” (1 = Strongly disagree; 7 = Strongly agree). The results show that participants in the threshold-based free shipping condition disagreed more with the first statement (M’s= 4.12 and 5.34; F(1, 48)=3.45, p