RESEARCH

DOI: 10.1080/10196780802420752 Copyright ß 2008 Electronic Markets Volume 18 (4): 345-361. www.electronicmarkets.org

Downloaded By: [Schmelich, Volker] At: 14:54 24 March 2010

A

b

s

t

r

a

c

t

Academic interest in the popularity and success of online auctions has been increasing. Although much research has been carried out in an attempt to understand online auctions, little effort has been made to integrate the findings of previous research and evaluate the status of the research in this area. The objective of this study is to explore the intellectual development of consumer behaviour in online auction research through a meta-analysis of the published auction research. The findings of this study are based on an analysis of 83 articles on this topic published mainly in information systems (IS) journals between 1998 and 2007. The results indicate that the consumer behaviour research on online auctions can be categorized into three major areas – facilitating factors, consumer behaviour and auction outcomes. Based on this literature review, directions for future research on eauction consumer behaviour are discussed, including potential new constructs, unexplored relationships and new definitions and measurements, and suggestions for methodological improvements are made. Keywords: online auction, e-auction, online consumer behaviour

A

u

t

h

o

r

s

Xiling Cui ([email protected]) is a PhD candidate in MIS at the Chinese University of Hong Kong. Her research interests focus on electronic commerce, online auctions, and IT investment analysis. Vincent S. Lai ([email protected]) is a Professor in MIS at the Chinese University of Hong Kong. His Research focuses on IS adoption and diffusion, virtual collaboration, electronic commerce, and global IS strategy. His articles have been published in CACM, J MIS, DSS, I&Mt, EJIS, EJOR, JIT, among others. Connie K. W. Liu ([email protected]) is a PhD student at the Chinese University of Hong Kong. Her current research focuses on institutionalization, ERP, imitation and IS adoption. She has previously published in Journal of Database Marketing & Customer Strategy Management.

Research on Consumer Behaviour in Online Auctions: Insights from a Critical Literature Review XILING CUI, VINCENT S. LAI AND CONNIE K. W. LIU

INTRODUCTION Online auctions have become an increasingly popular and efficient ecommerce method of facilitating the participation of Internet users in trading activities through flexible pricing processes, convenient access and the availability of a large variety of products. Although online auctions are not appropriate for every type of purchase, and they operate in a manner different to traditional sales channels (Bapna et al. 2001b, 2003b, 2004), researchers (such as Dai and Kauffman 2002 and Pinker Seidmann and Vakrat 2003) believe that they can be used as an effective tool to reduce purchase prices, save time, streamline the bidding process and enable a worldwide selection of suppliers and products. However, if online auctions are not used judiciously and their risks are not carefully assessed, then these types of cyberspace purchases can have adverse effects, such as damaging supplier relationships, choosing incapable suppliers, driving out qualified suppliers and underestimating the total costs associated with the lower purchase price (Hartley et al. 2006, Jap 2003). In recent years, online auctions have received wider acceptance, particularly among larger organizations

(Hannon 2004), and more researchers are devoting effort to the area. Among the various interesting online auction topics, consumer behaviour is fast becoming one of the most popular research areas, and many studies in this area have already been carried out. For example, research efforts (Bichler et al. 2001, Leloup and Deveaux 2001, Townsend and Bennett 2003) have been dedicated to investigating the fundamentals and general development of Internet auctions. There is also more specific online auction research, such as studies on technology or service adoption (Bosnjak et al. 2006, Hu et al. 2004, Stafford and Stern 2002, Zhang and Li 2006), bidding behaviour (Ba et al. 2003, Bapna et al. 2001b, 2004, Borle et al. 2006, Easley and Tenorio 2004, Namazi and Schadschneider 2006, Roth and Ockenfels 2002), reputation or trust (Ba et al. 2003, Gregg and Scott 2006, MacInnes et al. 2005, Melnik and Alm 2002, Standifird 2001, 2002), the winner’s curse (Bajari and Hortacsu 2003, Oh 2002), and auction mechanisms (Budish and Takeyama 2001, Hann and Terwiesch 2003, Kauffman and Wang 2001, Mathews 2004, Mathews and Katzman 2006, Spann et al. 2004, Spann and Tellis 2006, Standifird et al. 2004–5).

Downloaded By: [Schmelich, Volker] At: 14:54 24 March 2010

346

Xiling Cui, Vincent S. Lai and Connie K.W. Liu & Consumer Behaviour in Online Auctions: A Review

However, most of these previous studies of consumer behaviour in online auctions have been carried out largely on a piecemeal basis, thus inhibiting the emergence of cumulative findings. Fortunately, a few recent studies have brought together some of the earlier investigations on online auctions. For example, Pinker et al. (2003) reviewed the current business and research issues in online auctions and made a plea for more research into B2B auctions. Chakravarti et al. (2002) performed a literature review of pre-Internet auctions that can provide insight into online auction research. Baker and Song (2007) reviewed the existing studies of single-item online auctions and proposed a model for future online auction research. Despite all of these efforts, it is still critical that researchers develop a unifying framework and synthesize the current findings on consumer behaviour in online auctions to provide a structure for the state of the knowledge that exists in this field. The identification of the controversial points in these research findings and a distillation of the managerial implications of this new area of knowledge (McKinney et al. 2002) are also essential. As a first step towards the synthesis of the research findings on consumer behaviour in online auctions, and to deepen and develop the theoretical foundations of this area of research, it is critical that we examine previous efforts to identify trends for new research opportunities. This paper aims to present a more comprehensive overview of the current research literature on consumer behaviour in online auctions in the hope that more valid research topics and directions can be derived. This literature review includes multiple-item auctions with an emphasis on consumer behaviour. The research questions posed by this paper address the following. 1. What types of consumer behaviour have been investigated in online auctions? Can these findings be synthesized to provide a structure for the state of the knowledge in this domain? 2. What bidding strategies, if any, have been systematically and successfully adopted in online auctions? 3. Based on the literature review, can research gaps be identified for possible future research endeavours?

FUNDAMENTALS OF ONLINE AUCTIONS Four basic online auction mechanisms have been adopted from traditional offline auctions – English auction, Dutch auction, sealed first-price auction, and sealed second-price auction. An English auction begins with the lowest acceptable price and then increases the bids until no bidder will increase it further. The winner is the one with the highest bid, if price is the only criteria. A Dutch auction, in contrast, begins with a high asking price that is gradually reduced until someone is willing to accept it. Usually the Dutch auction is designed for multiple

identical items and has been used for perishable commodities such as flowers and fish. A sealed first-price auction accepts bids in a concealed fashion, and the bidder with the highest bid wins the auction. A sealed second-price auction, which is also known as a Vickrey auction, is similar to a sealed first-price auction, except that the highest bidder wins, but pays only the second-highest bid. This auction type is designed to alleviate bidders’ concern of overpaying for the products they bid for. In online settings, multi-unit auctions may extend the sealed second-price mechanism to third-price or even further down the line, and the bids are not necessarily sealed. Like in the multi-unit auctions in eBay and Taobao.com, bidders do not only bid on prices but also on the quantity they want to buy. All winning bidders pay the same price per item, the lowest successful bid price. This is also called Dutch Internet Auction. In addition to these mechanisms, new auction mechanisms have also emerged to facilitate online auctions. For example, name-your-own-price (NYOP) is a variant of the sealed first-price auction mechanism, in which a bidder submits a bid for an auctioned item, and the sellers respond by either accepting or rejecting it. The group buying auction mechanism has also developed rapidly due to the ease of Internet access. In group buying auctions, bidders submit an order bid rather than a price bid. The selling price drops as more buyers place their orders, which aggregates the power of buyers to gain volume discounts (Chen et al. 2006, Kauffman and Wang 2001). Other than these mechanisms, online auctions have new features added during the derivation process from their traditional counterparts. These new features fall into three categories, auction setting, bidding behaviour and service facilities, as shown in Table 1. In an auction setting, the ending rules of an online auction can either be ‘hard close’ or ‘soft close’. With the former, the auction ends at a fixed time that has been made known to all bidders (Namazi and Schadschneider 2006). In a soft-close English auction, in contrast, the auction can be extended for a set period of time with each new bid. Probably due to the hard-close rules and the longer duration of online auctions, a unique and interesting sniping phenomenon has also emerged in online auctions. In this strategy, bidders wait until nearly the end of the auction to submit their bids (Borle et al. 2006, Roth and Ockenfels 2002). In addition, online auctions also offer unique services that are only possible with Internet technologies. For example, reputation systems provide sellers and buyers with the opportunity to rate each other and leave comments after each transaction; reference prices are provided by the online auction intermediary (or the sellers in some cases) to help buyers compare the different prices on offer; and buyout options are provided by sellers to give buyers an alternative method of acquiring an item for a fixed price without having to wait for the entire process to be completed. In addition, online auction sites also offer escrow services through a

Electronic Markets Vol. 18 No 4

347

Table 1. Comparison between online auctions and traditional auctions Traditional auction Auction setting

Bidding behaviour

Downloaded By: [Schmelich, Volker] At: 14:54 24 March 2010

Service facility

Online auction

Auction space

Limited

Auction types Auction duration Auction ending

Many Short Soft close

Entry to an auction Perception of potential bidders Multiple-item bidding Snipe bidding Reputation systems Escrow services Reference prices Buyout option

Not easy Easy

Almost unlimited More Long Mainly hard close Easy Difficult

No No No No No No

Yes Yes Yes Yes Yes Yes

neutral trusted-third-party (TTP) to guarantee the transaction security of buyers and sellers. With this service, the buyer deposits money into the safekeeping of a TTP, pending on the satisfaction of the contractual conditions. Once these conditions have been fulfilled, the TTP sends the money to the seller.

LITERATURE REVIEW RESEARCH METHODOLOGY Although the research on consumer behaviour in online auctions has been carried out across several disciplines, namely, information systems (IS), marketing, economics and psychology, the aim of this paper is to provide a review of the literature on consumer behaviour research on online auctions within the IS discipline. A two-stage process was followed in searching the available pool of articles – article identification and article analysis. In the article identification stage, a thorough search was carried out on the ProQuest database for the 10-year period from 1998 to 2007, using ‘electronic auction’, ‘e-auction’, ‘online auction’ and ‘Internet auction’ as the search keywords. The abstracts of the top 10 MIS journals (Rainer and Miller 2005) were also manually searched for the same period in case not all of the articles were found in ProQuest. Only articles that were available in full were retrieved. Comments, notes, columns, book reviews, conference papers, working papers and news reports were excluded, as were papers that address auction theory, technical algorithms, B2B auctions (because businesses exhibit different types of behaviour than do consumers) and auction design. At the article analysis stage, two authors were independently involved in selecting the relevant papers from the

article pool. This process began by reading the abstract of a paper to determine its relevance to the research topic. The papers that were selected individually were then compared and discussed, and a final list was compiled after much deliberation. The selection process took nearly two months and resulted in a final research pool of 83 papers, as shown in Table 2. These papers were then reviewed separately by the two authors, although weekly meetings were conducted to consolidate the findings.

RESEARCH RESULTS The review of the selected articles suggests that the study of consumer online auction behaviour diverges widely in terms of both its scope and focus. It is therefore impossible to summarize and synthesize all of the findings of prior studies. As anticipated, many of these studies produced contradictory or inconclusive findings, thus increasing the difficulties the authors faced in synthesizing the literature. Consequently, our survey is not an exhaustive one. We intend merely to pinpoint the major findings of the prior studies to highlight some of the critical efforts that have been undertaken in the pursuit of knowledge on online auctions. Of the 83 online auction papers on consumer behaviour reviewed, the research foci and topics can be summarized and categorized into three main categories: 1) facilitating factors; 2) consumer behaviour; and 3) auction outcomes, with the first category affecting the second two. The following sections of the paper provide a detailed discussion of these three categories.

Facilitating factors Facilitating factors refer to the factors that facilitate online auctions and affect consumer behaviour and the auction outcome. Of the studies reviewed, the facilitating factors identified can be further classified into three categories, based on the different perspectives of the intermediary, the seller and the buyer.

Intermediary

factors. Auction intermediaries can manipulate many variables to enhance their competitiveness and achieve greater success in their auction operations. Of these variables, the technology property of their online platforms is clearly the most crucial variable. For example, the effects of the site architecture and the informational content and design of auction sites have been found to be critical to buyers’ ratings and perceptions of usability and to their subsequent overall site preferences (Kwon et al. 2002). In addition, researchers have also applied the Theory of Reasoned Action (TRA) and the Technology Adoption Model (TAM) to validate this correlation between technological characteristics and auction site adoption (Bosnjak

348

Xiling Cui, Vincent S. Lai and Connie K.W. Liu & Consumer Behaviour in Online Auctions: A Review

Table 2. Papers on consumer behaviour in online auctions Year 1999 2000

2001

Downloaded By: [Schmelich, Volker] At: 14:54 24 March 2010

2002

2003

2004

2004–5 2005

Journal American Economic Review Information Technology and Management Journal of Industrial Economics Marketing Letters Quarterly Journal of Electronic Commerce Communications of the ACM Decision Support Systems Economics Letters Electronic Commerce Research Journal of Management Journal of Management Information Systems Marketing Letters American Economic Review Artificial Intelligence Magazine Behaviour and Information Technology Economic Inquiry Electronic Markets European Journal of Information Systems International Journal of Electronic Commerce International Journal of Information Management Journal of Industrial Economics MIS Quarterly Brazilian Electronic Journal of Economics Communications of the ACM Decision Support Systems Management Science Information Systems Research Journal of the Operational Research Society Journal of Consumer Psychology Journal of the Academy of Marketing Science Journal of Computer Information Systems Psychology and Marketing Rand Journal of Economics Applied Economics Letters Electronic Markets Information Systems Research Journal of Consumer Research Journal of Economics Journal of Interactive Marketing Management Science MIS Quarterly International Journal of Electronic Commerce Electronic Commerce Research and Applications Electronic Markets International Journal of Consumer Studies International Journal of Electronic Commerce Internet Research Journal of Interactive Marketing Journal of Marketing Research Management Science Marketing Science Organizational Behaviour and Human Decision Processes The Rand Journal of Economics

Reference Lucking-Reiley 1999 Bapna et al. 2000 Lucking-Reiley 2000 Wilcox 2000 Herschlag and Zwick 2000 Bapna et al. 2001b Bapna et al. 2001a Budish and Takeyama 2001 Bichler et al. 2001; Leloup and Deveaux 2001 Standifird 2001 Kauffman and Wang 2001 Dholakia and Soltysinski 2001 Roth and Ockenfels 2002 Ockenfels and Roth 2002 Kwon et al. 2002 McDonald and Slawson 2002 Standifird 2002 Rafaeli and Noy 2002 Oh 2002; Stafford and Stern 2002; Ward and Clark 2002 Wang et al. 2002 Melnik and Alm 2002 Ba and Pavlou 2002 Mathews 2003 Bapna 2003; Townsend and Bennett 2003 Ba et al. 2003 Bapna et al. 2003a; Hann and Terwiesch 2003; Pinker et al. 2003 Bapna et al. 2003b Brint 2003 Ariely and Simonson 2003 Hartley et al. 2006; Jap 2003 Ottaway et al. 2003 Gilkeson and Reynolds 2003 Bajari and Hortacsu 2003 Dodonova and Khoroshilov 2004 Mollenberg 2004 Hu et al. 2004; Pavlou and Gefen 2004 Kamins et al. 2004 Mathews 2004 Bruce et al. 2004; Heyman et al. 2004; Spann et al. 2004 Easley and Tenorio 2004 Bapna et al. 2004 Standifird et al. 2004–5 Kauffman and Wood 2005 Rafaeli and Noy 2005 Cameron and Galloway 2005 MacInnes et al. 2005 Yang 2005 Dholakia 2005 Park and Bradlow 2005 Carare and Rothkopf 2005; Ding et al. 2005; Terwiesch et al. 2005 Dholakia and Simonson 2005 Ku et al. 2005 Ariely et al. 2005

Electronic Markets Vol. 18 No 4

349

Table 2. Continued Year 2006

Downloaded By: [Schmelich, Volker] At: 14:54 24 March 2010

2007 2008

Journal Economic Theory Decision Support Systems Information & Management Information Technology and Management International Journal of Electronic Commerce International Journal of Operations and Production Management International Journal of Modern Physics Journal of Consumer Behaviour Journal of Economics & Management Strategy Journal of Marketing Statistical Science Journal of Electronic Commerce in Organizations Decision Support Systems

et al. 2006, Stafford and Stern 2002) and have confirmed it to be significant. In addition to technology properties, auction intermediaries can also manipulate their auction sites by offering different auction mechanisms, for example, the traditional English auction, the Dutch auction or certain innovative online auction mechanisms, such as nameyour-own-price (NYOP) and group buying auctions. We find that English auctions have been investigated most frequently, encompassing research into bidding strategies (Bapna et al. 2004), revenue comparison (Bapna et al. 2001a), bidding behaviour (Ariely and Simonson 2003) and fraudulent acts (Kauffman and Wood 2005). Dutch auctions have received far less attention because they are far less common among the popular auction sites. However, contrary to intuition, an interesting study that applied a decision theory model and a game theory model to investigate bidding behaviour suggests that bidders in Dutch auctions prefer to purchase objects sooner and at a higher price to economize on the cost of return when it is positive (Carare and Rothkopf 2005). In the NYOP research, experienced bidders are found to exhibit lower frictional costs, which are the implicit and explicit costs associated with market transactions, than are inexperienced bidders (Hann and Terwiesch 2003). Research also suggests that a large number of NYOP consumers do not exhibit rational decision-making, as predicted by an economic model (Spann and Tellis 2006). Consumers who place many bids with rather long inter-bid times are more likely to bid irrationally. However, if they are willing to haggle extensively, then they can, on average, obtain a product at a lower price (Terwiesch et al. 2005). Research on group buying, another new mechanism in online auctions, has also generated many interesting findings. Kauffman et al. (2001), for example, have

Reference Mathews and Katzman 2006 Antony et al. 2006; Chen et al. 2006; Lin et al. 2006; Onur and Tomak 2006; Zhang and Li 2006; Zhang 2006; Amyx and Luehlfing 2006 Kauffman and Wood 2006 Gregg and Scott 2006 Hartley et al. 2006 Namazi and Schadschneider 2006 Stern and Stafford 2006 Houser and Wooders 2006 Spann and Tellis 2006 Borle et al. 2006 Baker and Song 2007 Bapna et al. 2008

identified three effects that exist simultaneously in group buying: the ‘positive externality effect’, the ‘price drop effect’ and the ‘ending effect’. The externality effect postulates that existing orders can stimulate new orders, which means that a greater number of existing orders can potentially generate more new orders, whereas the price drop effect suggests that prices are more likely to fall when they are closer to the next discount level, and the ending effect means that bidders are more likely to place orders during the last three hours of an auction. Chen, Chen and Song (2006) have argued that when economies of scale are considered, group buying auctions outperform fixed price mechanisms if the seller is risk-seeking. In addition to technology properties and auction mechanisms, intermediaries can also manipulate their service facilities to improve their service and performance levels. The types of facilities that have been researched include TTPs, reference points, reputation systems and buyout options. The digital certificates issued by TTPs have been found to be effective in forcing market participants to behave honestly (Ba et al. 2003). Effective TTPs have also been found to engender trust, not only in a few reputable sellers, but in the entire community (Pavlou and Gefen 2004). In the reference point research, Dholakia et al. (2005) found that participants submit fewer, lower and later bids when reference points are provided. Reference prices have also been found to have a negative association with overbidding (Amyx and Luehlfing 2006) and a direct impact on the final bid price (Kamins et al. 2004). The reputation systems research suggests that the negative feedback posted in the reputation systems of online auction sites can predict future incidences of auction fraud and that experienced bidders are better at using this information to avoid such fraud (Gregg and

350

Xiling Cui, Vincent S. Lai and Connie K.W. Liu & Consumer Behaviour in Online Auctions: A Review

Scott 2006). The research on the buyout option has also provided some interesting results. For example, it has been found that a buyout price is appealing to risk-averse bidders (Budish and Takeyama 2001). Nevertheless, if the seller and the bidder are risk-neutral, then the former will choose a buyout price that is sufficiently high to ensure that the option will never be exercised (Mathews and Katzman 2006). However, time impatience on either side of the transaction can motivate the seller to offer an option price that is low enough to be exercised when bidders do not discount future transactions (Mathews 2004). Empirical observation has also confirmed that time-impatient bidders are likely to choose the buyout options (Standifird et al. 2004–5). Table 3 summarizes this discussion of facilitating factors.

Downloaded By: [Schmelich, Volker] At: 14:54 24 March 2010

Seller factors. There are two research sub-streams related to sellers: auction settings and the seller’s reputation in online auctions. Auction settings usually refer to product characteristics (such as the product picture and description), starting price, bid increments, auction duration and auction time. In general, research on auction products has already confirmed the correlation between product characteristics and auction site ratings (Kwon et al. 2002). The inclusion of product pictures and descriptions on auction sites has also been confirmed to

increase the perceived utility of auction products (Kauffman and Wood 2006), although other research has found that product pictures are not critical (Ottaway et al. 2003). The research on starting prices suggests that a low starting value can draw in more bidders, but that their bids are likely to be relatively low, thus eventually leading to a low final price – a phenomenon called the ‘anchoring effect’ (Ariely and Simonson 2003). Anchoring effect means that people’s valuation usually starts from some starting value, named ‘anchor’, and is adjusted up or down based on their assessment or beliefs (Tversky and Kahneman 1974). In online auctions, the starting price often works as the anchor and then has certain impact on the bidders’ bidding behaviour and the final price, which has also been empirically confirmed through a field experiment with Bidz.com (Dodonova and Khoroshilov 2004). Researchers have also found that bid increments have a positive effect on bidder participation when the increment is below a certain value, but to have a negative effect when the increment is above that value (Bapna, Goes and Gupta 2001b). As for auction duration, some researchers found it to be positively associated with the final price (Brint 2003, Lucking-Reiley 1999) while others did not find such an effect in their studies (Gilkeson and Reynolds 2003, McDonald and Slawson 2002). Interestingly, some even obtained contrary

Table 3. Intermediary factors affecting bidding behaviour Sub-stream

Factors

Research findings

References

Technology properties

Theory of Reasoned Action/Technology Adoption Model (TRA/TAM) Information design

Bosnjak et al. 2006; Stafford and Stern 2002 Kwon et al. 2002

Auction mechanism

English auction

Generalized to the online auction setting Related to the buyers’ overall preferences Most research is conducted in this context Bidders prefer to bid sooner Experienced bidders are more rational

Dutch auction Name-Your-Own-Price (NYOP)

Group buying Service facility

Trusted-Third-Party (TTP) Reference point

Reputation system Buyout option

Network effect/price drop effect/ending effect Enhances bidders’ honest/active behaviour Reduces bidding fever; increases sniping; mediates starting price on final price Helps to predict future auction fraud Insure risk-aversion/impatient bidders

Carare and Rothkopf 2005 Hann and Terwiesch 2003; Spann et al. 2004; Spann and Tellis 2006; Terwiesch et al. 2005 Chen et al. 2006; Kauffman and Wang 2001 Ba et al. 2003; Pavlou and Gefen 2004 Amyx and Luehlfing 2006; Ariely and Simonson 2003; Brint 2003; Dholakia and Simonson 2005; Kamins et al. 2004 Gregg and Scott 2006 Budish and Takeyama 2001; Mathews 2004; Mathews and Katzman 2006; Standifird et al. 2004–5

Downloaded By: [Schmelich, Volker] At: 14:54 24 March 2010

Electronic Markets Vol. 18 No 4

findings that auction duration is negatively correlated with the final price (Ariely and Simonson 2003). They argued that it is probably due to the trumping of the ‘arrival process effect’ by the ‘waiting time effect’. The arrival process effect means that the longer an auction lasts, the greater the likelihood of the auction site being hit by new bidders, thus resulting in the accumulation of more bidders and an increase in the competitive pressure on the auction’s closing price. The waiting time effect, in contrast, is the utility of delay during the auction period and results in a low final auction price. Another possible explanation for the conflicting results was found in a recent study that the relationship between the auction duration and the final price may be impacted by sellers’ reputation (Bapna et al. 2008). A consumer may prefer a short auction to a long one, because it may result in fewer search, delay and monitoring costs during the course of the auction (Kauffman and Wood 2006). Research also suggests that the timing of an auction can have a significant effect on the total number of bids and hence on the final price. A ‘weekend effect’ has also been identified. This effect suggests that bidders are willing to pay a higher price for an auction item if the end of the auction coincides with a weekend (Kauffman and Wood 2006). Studies on the seller’s reputation have also generated quite a few interesting research findings. An analysis of the reputation data in eBay’s feedback forum suggested that sellers’ reputations follow a lognormal distribution (Lin et al. 2006). Empirical studies have identified that they have a consistent, statistically significant and positive impact on the final prices achieved at auctions (Ba and Pavlou 2002, Houser and Wooders 2006, Melnik and Alm 2002, Ottaway et al. 2003) and that this area deserves to be investigated further (McDonald and Slawson 2002). Standifird (2001), along the same lines of research, found that a positive reputation on the part of a seller has only a mild influence on the final bid price, but that a negative reputation rating has a critical influence on it. In addition to its effect on the final bid price, other research has indicated that a seller’s reputation can also have a significant, negative impact on the likelihood of a dispute in online auction transactions (MacInnes et al. 2005). In addition to reputation ratings, Standifird (2002) has also investigated reputation types by comparing the auctions conducted through three online auction venues, CNET, eBay and Amazon, which represent three different types of reputation schemes. The results indicated that a much higher final bid price was likely to be achieved when computer-related items were auctioned through CNET or eBay rather than through Amazon, demonstrating that different reputation types may engender different auction outcomes. Table 4 summarizes our discussion of the seller factors.

351

Bidder factors. The two major research sub-streams related to bidders that we have identified are the bidder’s own attributes and the social influence of other bidders. Bidder attributes mainly concern the bidder’s reputation, experience and emotions. Researchers have found that a bidder’s reputation has an impact on the likelihood of a dispute, but that this effect is less significant than the seller’s reputation (Houser and Wooders 2006, MacInnes et al. 2005). Although some online auctions may ask the buyer and the seller to rate each other’s reputation after each transaction, most sellers tend to be lenient in their feedback, which eventually results in unreliable ratings for bidder reputations. Probably due to this phenomenon, research has found that bidders’ reputation ratings are rarely used as a reference for bidding behaviour in online auctions (Dholakia 2005). However, experience is considered important in determining bidders’ behaviour and the strategies they adopt during auctions. In other words, bidders gain experience from previous auctions, which allows them to learn the bidding strategies that are more likely to be successful. However, when strategic dominance is weak, bidders’ experience has a weaker impact on their bidding behaviour (Wilcox 2000). On the emotional front, research has reported that bidders experience strong emotions in terms of their excitement at winning and frustration at losing, and that these emotions change dynamically as a function of the outcomes of their previous bids (Ding et al. 2005). Another popular research stream from the bidder’s perspective is social influence, which investigates the social effect of submitted bids on the bidding intentions of others. Kauffman et al. (2006), for example, found that bidders tend to pay more for items that other bidders are also interested in. This is called the ‘herd effect’. Heyman et al. (2004) found that a greater number of participants in the early stages of an auction will lead to more active bidding at a later stage, and they called this the ‘quasi-endowment effect’. This effect contributes to the sense of ownership that bidders develop during an auction, even though they do not yet have legal claim to the item in question. In addition to these herd and quasi-endowment effects, researchers have also explored the effect of social facilitation, which is the tendency for a bidder to perform better in terms of bidding behaviour and performance in the presence of other bidders. Rafaeli et al. (2005), for example, identified the presence in virtual environments of social facilitation, which motivates a stronger purchasing decision among customers. They also indicated that the social presence of other bidders in online auctions affects not only the price, the timing and the number of bids, but also the auction’s market outcomes. Table 5 summarizes our discussion of the buyer factors.

352

Xiling Cui, Vincent S. Lai and Connie K.W. Liu & Consumer Behaviour in Online Auctions: A Review

Table 4. Sellers’ factors affecting bidding behaviour Sub-stream Auction setting

Factors Product description/Image

Starting price

Downloaded By: [Schmelich, Volker] At: 14:54 24 March 2010

Bid increment

Duration

End timing

Sellers’ attributes

Sellers’ reputation

Research findings Enhances bidding intention

No image effect found Positively associated with final price

Negatively associated with the # of bids; mediated by price comparison/anchoring effect Negatively associated with auction success Positive effect on active participation As it increases, it first increases the auctioneer’s revenue, then reduces it Negatively related to the # of bids/final price Positively associated with final price No effect on auction success or final price The upper relationship is impacted by sellers’ reputation Weekend effect, midnight effect No effect Positively related to intention to bid/final price

Moderated by the item price No effect on the final price Positive and negative (or high and low) rankings

Consumer behaviour. Consumer behaviour is the aspect that we want to highlight in this literature review. We have identified a large body of research related to consumer behaviour, which suggests the importance and wide acceptance of this research topic in online auction research. Due to the wide range of research issues explored in these publications, we have classified them into four groups – adoption, entry decision, bidding behaviour and bidding strategy.

Online auction adoption. Researchers have already identified the general reasons for the adoption of online auctions. Lower transaction costs, lower purchasing prices and the inclusion of a wider pool of suppliers are some of the primary factors that support the adoption of these auctions (Carter et al. 2004). Some

References Kauffman and Wood 2006; Kwon et al. 2002; Standifird et al. 2004–5; Zhang and Li 2006 Ottaway et al. 2003 Ariely and Simonson 2003; Brint 2003; Gilkeson and Reynolds 2003; McDonald and Slawson 2002 Ariely and Simonson 2003; McDonald and Slawson 2002 Gilkeson and Reynolds 2003 Bapna et al. 2003b Bapna et al. 2003a; Bapna et al. 2001b; Bapna et al. 2003b Ariely and Simonson 2003 Brint 2003; Lucking-Reiley 1999 Gilkeson and Reynolds 2003; McDonald and Slawson 2002 Bapna et al. 2008 Kauffman and Wood 2006; McDonald and Slawson 2002 Brint 2003 Ba and Pavlou 2002; Houser and Wooders 2006; Kauffman and Wood 2006; McDonald and Slawson 2002; Melnik and Alm 2002; Ottaway et al. 2003; Standifird 2001; Standifird et al. 2004–5 Bruce et al. 2004 Ariely and Simonson 2003; Gilkeson and Reynolds 2003 McDonald and Slawson 2002; Standifird 2001

researchers have applied adoption theories, including TRA and TAM, to evaluate the practical utility of online auctions, and their findings have validated the applicability of these theories to explaining and predicting the subsequent adoption of the technology by bidders (Bosnjak et al. 2006). Other researchers, such as Stafford and Stern (2002), have integrated a set of different theories, including TAM, affinity theory and involvement theory, to predict consumer inclinations to use auction sites. Their empirical results, too, have proved the theoretical validity and applicability of the integrated model in their clear and consistent explanation of this adoption phenomenon. In addition to the antecedents of online auction adoption, the barriers to its adoption have also been investigated. These findings suggest that inadequate auction knowledge and

Electronic Markets Vol. 18 No 4

353

Table 5. Bidders’ factors affecting bidding behaviour Sub-stream Bidders’ attributes

Factors Reputation Experience

Emotion Social influence

Network effect

Downloaded By: [Schmelich, Volker] At: 14:54 24 March 2010

Social facilitation

Research findings Less effect than the seller’s reputation No effect on final bid price Causes more strategic bidding Positively related to auction success Excitement of winning and frustration with loss Others’ bids stimulate bidding, even when shilling occurs Affects the price/market outcomes

concerns over information security are two of the major barriers to online auctions for potential buyers (Hartle et al. 2006). Researchers have not only used the general technology adoption models to explore the adoption of online auctions, but have also investigated specific auction technology services, such as payment choice and escrow services, and their subsequent adoption and use. For example, Zhang and Li (2006) studied payment choice in online auctions by investigating product attributes, trader characteristics and payment attributes. Their results indicated that product attributes have a stronger influence on payment choice than do trader characteristics, while a seller’s reputation has no significant influence on the actual payment choice. In contrast, Antony et al. (2006) investigated the determinants of escrow service adoption and showed that, in an experimental setting, such market factors as the fraud rate and the product price are important determinants of the adoption of escrow services. The seller’s reputation also had an indirect effect through the buyer’s risk perception of the adoption of escrow services.

Entry decision. The first decision a consumer needs to make before an auction is whether to participate. This decision could be influenced by his or her interest in the auction item or in the auction site and most probably involves a value assessment (Ariely and Simonson 2003). An entry decision may be made at different stages of the auction – at the beginning, in the middle or near the end – depending on the arrival time of the bidders. However, the entry decision is intrinsic and difficult to observe, because bidders do not have to act immediately even after they have decided to enter an auction; therefore, some researchers have used the bidder’s first bid or subsequent bids during the auction as the convenient proxy for an entry decision (Bajari and Hortacsu 2003). There is very little research that focuses exclusively on the entry decision. Most of the studies that address it are combined with research on either online auction

Reference MacInnes et al. 2005 Ottaway et al. 2003 Borle et al. 2006; Wilcox 2000 Gilkeson and Reynolds 2003 Ding et al. 2005 Dholakia and Soltysinski 2001; Kauffman and Wood 2006; Kauffman and Wood 2005; Ku et al. 2005; Stern and Stafford 2006 Rafaeli and Noy 2005

technology adoption (Kwon et al. 2002) or bidding behaviour (Kauffman and Wood 2006).

Bidding behaviour. In this paper, bidding behaviour denotes general, aggregated behaviour across all phases of the auction, rather than behaviour that is confined to a specific phase of the auction (Ariely and Simonson 2003). In this line of research, researchers (Ariely and Simonson 2003, Kamins et al. 2004, Kauffman and Wood 2006, Pavlou and Gefen 2004) have used both the number of bids and the final bid prices as proxies to measure general bidding behaviour, mostly because of their simplicity and convenience. With few exceptions, the research on online auctions has presumed bidder rationality. However, the empirical findings on the online behaviour of bidders have suggested that many of them bid in excess of their pre-set limits, which is called ‘bidding fever’. A ‘winner’s curse’ has also been identified in the online setting. This is the tendency for the winner to incur a loss because of his or her inaccurate assessment of an object’s resale value (Ku et al. 2005). Although both of these phenomena cause bidders to bid in excess of their limits, auction fever is primarily emotional in nature, while the winner’s curse stems largely from uncertainty over the value of an object. Theoretical validation of the rational choice, escalation of commitment and competitive arousal models have also demonstrated that irrationality plays a critical role in bidding behaviour (Ku et al. 2005). In addition to irrational bidding behaviour, researchers have also turned their attention to such fraudulent bidding conduct as shill bidding – the seller bidding on his or her own auctioned item using secondary registrations or associates to artificially drive up the bid price (Namazi and Schadschneider 2006). Kauffman et al. (2005), for example, believe that fake bids submitted by sellers or their associates facilitate the herd effect. They explain this from the valuation signal perspective, that is, bidders may view a large number of bids as a signal that an item is worth more than what they were originally willing to pay for it. Dholakia and Simonson (2005)

354

Xiling Cui, Vincent S. Lai and Connie K.W. Liu & Consumer Behaviour in Online Auctions: A Review

have suggested that explicit reference points can be used by online auction participants to help them avoid falling into the shill bidding trap caused by herd behaviour bias by encouraging them to be more cautious and riskaverse and to avoid a bidding frenzy. In the presence of explicit reference points, bidders may make less risky choices and adopt more controlled bidding strategies, thus leading to a lower probability of overpaying and a greater incidence of sniping.

Downloaded By: [Schmelich, Volker] At: 14:54 24 March 2010

Bidding strategies. A bidding strategy represents a series of interrelated types of bidding behaviour that a bidder adopts with a certain purpose or tactic in mind across different phases of an online auction (Ariely and Simonson 2003). Park and Bradlow (2005) have developed an integrated model that incorporates four key components of the bidding process – whether people win an auction, (and if so) who bids, when they bid and how much they bid. Bapna et al. (2001b, 2003b, 2000, 2004), in contrast, classify bidders into five types based on the number of bids and the entry and exit times: evaluators (early or middle), participators, agent bidders, opportunists and sip-and-dippers. Evaluators refer to bidders who place just one bid at an early stage (or sometimes in the middle stages) of an auction; participators are bidders who bid throughout the auction, thus increasing the price by the minimum bid increment; agent bidders use online bidding agents to bid at the minimum level required to outbid the current highest bidder until the bid exceeds the reserve price they set earlier; opportunists are late bidders who act only towards the end of an auction; and sip-anddippers are a combination of evaluators and opportunists. These five bidder types tend to apply unique strategies in the bidding process. Evaluators, for example, employ a jump bidding strategy (Easley and Tenorio 2004) to submit bids that are higher than required and usually only once. This strategy can minimize the time and cost of monitoring (Bapna et al. 2004) and is effective in auctions with fewer overall bids (Avery 1998, Bapna et al. 2004, Easley and Tenorio 2004). Research has also found that experienced bidders tend to use a jump bidding strategy in the early stages of an auction (Borle et al. 2006). Participators, in contrast, normally adopt a ratchet bidding strategy (sometimes called participatory bidding, successive bidding or pedestrian bidding) to bid by the minimum bid increment (Brint 2003). Although ratchet bidding has seldom been specifically focused on in studies of online auctions, Borle et al. (2006) consider it to be of strategic importance in online auctions. Agent bidders prefer to use an agent bidding strategy that takes advantage of the automatic bidding software provided by auction intermediaries to place their bids. On most online auction sites, this software allows bidders to set a reserve price and then follow the minimum bid increments, when necessary, in the course of the bidding process (Bapna et al. 2004). This strategy can save the time and cost of tracing and monitoring,

and it is an attractive alternative for participants who value their time greatly (Bapna et al. 2004). However, agent bidding is still not very popular, perhaps because bidders do not fully understand the process or because they like to retain the flexibility to revise their reservation price at all times (Easley and Tenorio 2004). Opportunists frequently adopt a sniping or snipe bidding strategy (Borle et al. 2006, Roth and Ockenfels 2002). This means that they watch a timed auction and then place a winning bid just before it ends to prevent other bidders from outbidding them or driving the price higher (Namazi and Schadschneider 2006). Sip-anddippers, who practice a special variant of sniping, place one early bid to establish their time priority and then snipe during the auction’s closing stages to compete at the margin. This strategy is normally adopted in special multi-unit auctions (auctions that include multiple items of the same product) that require a time priority. In addition to the five aforementioned bidding strategies, Lucking-Reiley (2000) also reported a bid shielding strategy in which one (or more) bidder(s) aim(s) to deceive other bidders. In this strategy, the bidder offers a low bid on an item and then follows it up with another bid – using a false identity – that is high enough to keep rivals from entering the auction. Then, just before the end of the auction, the bidder retracts the high bid that was made using the false identity, and the low bid wins.

Auction outcome. Many online auction studies adopt auction outcome as a dependent variable in their empirical or theoretical enquiries, using either an economic or non-economic perspective. The economic perspective normally explores the final auction price, the auctioneer’s revenue, the winner’s curse and consumer surplus, while the non-economic perspective investigates auction success and the likelihood of winning.

Economic perspective. Economic auction outcomes can be further classified by whether they take the seller’s or the buyer’s perspective. Researchers have explored a wide spectrum of determinants of auction outcomes from the seller’s perspective, ranging from the auction setting and participant attributes to bidding behaviour. By and large, these research findings demonstrate that starting price, the number of bids, auction duration and the seller’s reputation may have effects on the final price (Ariely and Simonson 2003, Gilkeson and Reynolds 2003). The auctioneer’s revenue, another economic outcome that concerns sellers, has been found to be positively related to the insurance premiums paid (Ward and Clark 2002), the shipping price paid, the auction length, the number of bids (Onur and Tomak 2006) and the bid increment (Bapna et al. 2001b, 2003a, 2003b, Onur and Tomak 2006). It has also been found to be marginally affected by the auction’s opening bid (Bapna et al. 2001b, 2003a, 2003b). In research that considers

Downloaded By: [Schmelich, Volker] At: 14:54 24 March 2010

Electronic Markets Vol. 18 No 4

the two ending rules, the hard-close and the soft-close, hard-close auctions have been found to yield lower revenue, especially when the item value is low (Onur and Tomak 2006). From the buyer’s perspective, the common phenomenon of the winner’s curse in traditional auctions also exists in online auctions (Amyx and Luehlfing 2006, Bajari and Hortacsu 2003). Research suggests that the occurrence of the winner’s curse is positively associated with retail price dispersion and auction type, but not with product value (Oh 2002). In multi-unit auctions, researchers identified the existence of consumer surplus (Bapna et al. 2001b, 2004), which is the excess price a consumer would be willing to pay for a product rather than go without it, and it is operationalized as the difference between the highest bid (price which a bidder would be willing to pay) and the winning bid (the price they actually paid). In the studies of Bapna et al. (2001b, 2004), different bidding strategies were also found to result in different economic outcomes. Evaluators fare worst, participators fare best, and opportunists lie somewhere in between in terms of consumer surplus. Agent bidders, who appear later than other types of bidder, have been found to be the best at maximizing surplus and are even better than participators.

Non-economic perspective. The non-economic perspective investigates auction outcomes using auction success as an indicator. Success means that the auction results in a deal – that is, it draws at least one bid and that the final price is higher than the reserve price, if any (LuckingReiley 2000). Gilkeson et al. (2003) studied the determinants of auction success, and their results show that both the starting price (due to the anchoring effect) and the reserve price of the seller have negative effects on auction success, while the number of unexpected bids and bidder experience have a positive impact. The likelihood of winning is a non-economic outcome variable that has been investigated from the bidder perspective. Bapna et al. (2004) indicated that opportunists and sip-and-dippers are generally more eager to win, and thus have a higher likelihood of winning, while others tend to be more cautious.

Summary of research findings. This extensive literature review of consumer behaviour in online auctions suggests that the prior research endeavours in this field can be categorized into three sub-streams: facilitating factors; consumer behaviour; and auction outcome. A detailed analysis of these three sub-streams reveals that they have certain intended or unintended relationships that form themselves into an integrated framework, as depicted in Figure 1. The facilitating factors, which are explored from the intermediary, seller and bidder perspectives, explain the effects of technology properties, service facilities, auction mechanisms, auction setting, sellers’ attributes, bidders’ attributes and social influence

355

on consumer behaviour in online auctions. The varieties of behaviour displayed, including technology adoption, entry decision, bidding behaviour and bidding strategy, in turn, can also be used to explain some of the studies that have focused on auction outcomes. In other words, our literature review suggests that the current research on consumer behaviour in online auctions has already formed itself into a structure that can be used to guide future investigations. Although a cumulative tradition has not yet emerged, the existing body of research provides a foundation to guide and enhance future research endeavours on online auctions.

FUTURE RESEARCH DIRECTIONS Although our review of the online auction literature has covered a vast area and has sometimes cut across disciplines, the work in this area is still being developed, and plenty of research opportunities remain to be explored. By examining and analysing previous studies, we have been able to identify some gaps in the knowledge, as depicted in Figure 2, including potential new constructs, unexplored relationships between constructs and new measurements and definitions of these constructs. Moreover, theoretical and conceptual developments are crucial for this relatively new area of work, and a resolution of the sometimes conflicting findings will contribute substantially to the accumulation of knowledge. This section provides a detailed discussion of some potential future directions for research on consumer behaviour in online auctions, based on the three main categories.

Knowledge gaps related to facilitating factors. The services that are unique to the online environment require further exploration. For example, escrow services, which are also called TTP services, are usually provided by intermediaries to enhance the building of trust and transactional equity between sellers and buyers, and they are believed to have a significant effect on consumer behaviour in online auctions. However, only a few papers have addressed the impact of these services on bidding behaviour (Pavlou and Gefen 2004), the research findings, of which, remain to be confirmed by separate investigations in order to make them generalizable to different areas of application. Because trust is an important issue in the online environment, in which there is no face-to-face interaction and fraud is commonplace, it is worth devoting more resources to considering ways to enhance the level of trust between participants and to investigating the strength and fragility of trust. From the perspective of sellers, product type could well affect the bidding behaviour of consumers. People participate in auctions for different reasons: some believe they can save money, while others look for rare items

356

Xiling Cui, Vincent S. Lai and Connie K.W. Liu & Consumer Behaviour in Online Auctions: A Review

Downloaded By: [Schmelich, Volker] At: 14:54 24 March 2010

Figure 1. Research framework for online bidding behaviour

among the vast range of goods offered online. Therefore, it is not difficult to imagine that consumers may employ different strategies when they acquire different types of products, such as commodities (where saving money is the sole purpose of bidding) or collectables (where obtaining a rare item is the only concern). An investigation of these different product types and the bidding behaviour associated with each of them would thus be an important step in synthesizing consumer behaviour in different contexts and scenarios. In addition, new products and second-hand products may also engender different types

Figure 2. Possible research opportunities. Note: Italics: suggested research areas.

of consumer behaviour, thus giving researchers another direction to explore. From the perspective of bidders, product value can also affect auction participant behaviour. Generally speaking, when they bid on an expensive item online, people tend to be more cautious and perhaps less prone to the herd effect, whereas they may be more impulsive when it comes to bidding on a less expensive item. However, not many auction studies have addressed this issue. Finally, because online auctions attract people from all over the world, it is very likely that cultural factors or

Electronic Markets Vol. 18 No 4

differences play a part in shaping the online behaviour of consumers. It is possible that certain facilitating factors may be regarded as quality factors in the East, while they may be regarded as non-essential in the West (for example, the provision of a socializing platform within the auction site). Research in this area would not only enrich our knowledge of consumer behaviour in online auctions, but would also extend the scope of cultural research. It would also be interesting to examine how people from collectivist cultures respond to online auctions in comparison to offline auctions, as the online environment is more individualistic in nature.

Downloaded By: [Schmelich, Volker] At: 14:54 24 March 2010

Mediating and moderating effects on consumer behaviour. This line of research is still relatively new; therefore, the discovery of new constructs is essential to its growth, both in depth and in breadth. Although research has addressed the direct effects on consumer behaviour, no paper has yet explored the possibility that different types of relationships, such as mediating and moderating effects, may exist. It is possible that a new construct, such as motivation, may mediate the relationship between certain facilitating factors and consumer behaviour, because external causes do sometimes have the ability to affect attitudes towards certain issues and, in turn, alter the behaviour that is adopted. Similarly, such personality traits as risk preference may have a moderating effect on the relationship between certain facilitating factors and consumer behaviour. In fact, certain bidder attributes that are currently thought to have a direct effect on consumer behaviour may more accurately be regarded as moderating factors. This is because certain attributes, such as experience, which clearly has some kind of impact on consumer behaviour, have resulted in conflicting findings in the literature. Therefore, the impact of such attributes may or may not be direct, and thus they warrant further investigation.

Research opportunities related to consumer behaviour. New research areas within the consumer behaviour context include the establishment of a clear definition of the ‘entry decision’. This is essential before future research can be built upon it. As previously discussed, the entry decision may not necessarily equate to a bidder’s first bid, although the alternative is difficult to measure and can be problematic when it comes to verifying the results. Nevertheless, the early decisionmaking process, which ultimately determines a bidder’s subsequent behaviour, requires more attention from researchers. Attention to this issue may even open the door to a whole new area of research, including conceptual studies, the development of models and the measurement of creation and verification. This new research area would require extensive background knowledge and the ability to utilize theories and resources from a wide range of disciplines, including psychology, sociology, IS, marketing and management.

357

Another research gap in the consumer behaviour context is the possible antecedents of bidding strategy. While many researchers have explored this topic, most of them have focused solely on the patterns of bidding strategies or combined such analysis with the effects of these strategies on the auction outcome. Unfortunately, the existing research on the factors that may affect the adoption of different bidding strategies is much weaker. Some researchers have discussed possible factors, but none so far has conducted empirical work to validate their claims. Moreover, bidding strategies that aim to defraud, such as shill bidding and bid shielding, need to be explored in more depth to prevent or reduce their occurrence.

The measurement of auction outcome. The variety of constructs for auction outcome needs to be broadened. Previous research has used different measures and definitions for auction outcome. For example, some studies have used auctioneer revenue to gauge the success of the outcome, while others have tried to measure this construct by looking at the benefit to the consumer. In addition to using the dividing line of economic and non-economic outcomes, online auction research can be investigated in an alternative way by using three levels of outcomes: intermediary-, transaction- and participant-level outcomes. Intermediary-level outcomes refer to outcomes that are directly related to the intermediary or the auctioneer. Transaction-level outcomes are related to the transaction process of one auction, including its final price and the total number of bids in it. Participant-level outcomes are the outcomes from the bidder’s perspective, such as customer gain/ loss, customer satisfaction and entertainment fulfilment. This three-level classification system for outcomes, combined with classification based on the economic versus the non-economic perspective, forms a twodimensional 263 matrix framework (Table 6) that may be used in future auction outcome research. The differentiation of these research alternatives may lead to the compilation of a more organized and comprehensive list of variables and constructs to represent and measure auction outcome. In fact, the auction outcome research topics that have been studied to date could be classified into this framework to help identify those topics that have already been researched and those that require additional study. As indicated in Table 6, the non-economic perspective has received less attention than has the economic perspective. Some of the topics that could be considered for future research include intermediary/auctioneer satisfaction, bidder satisfaction and the bidder’s entertainment fulfilment.

Insights into research methodology. In addition to the knowledge gaps that exist in each part of this integrated framework, there are also methodological imbalances that could easily be improved. First, the research

Downloaded By: [Schmelich, Volker] At: 14:54 24 March 2010

358

Xiling Cui, Vincent S. Lai and Connie K.W. Liu & Consumer Behaviour in Online Auctions: A Review

methodologies used could be extended in scope. To date, most studies have been conducted by analysing real data from auction sites. Such objective data, although they are highly valued in empirical research, also have certain limitations. For example, such personal attributes as intrinsic thought cannot be reflected in observed data. Moreover, most of the observed data provided by online auction houses have been collected as individual auction units of data, meaning that, although these data can clearly describe the patterns of each auction, they cannot track the overall behaviour of each individual. These limitations could easily be solved through the use of surveys, a tool that has been widely used to capture individual-level variables and constructs in other research areas, to make the research on online bidding behaviour more comprehensive and convincing. Another important area of research is the establishment of theoretical models. It is clear from the literature review carried out here that data-driven research is still the main research format in online auction studies. Only a few papers have developed a theoretical framework or model first and then followed it up with confirmatory tests. Clearly, data-driven studies are important for any research topic, particularly when it is in its infancy. However, additional confirmatory studies are needed once a research topic has grown and matured. Widely-accepted, welltested theories from more mature fields, such as marketing, management, psychology or economics, can be borrowed, applied or generalized to establish a theoretical backbone for research in this area, thus making it more sophisticated and credible. The theories that have been employed in online auction research to date include image theory (Dholakia and Soltysinski 2001), the TAM (Bosnjak et al. 2006, Stafford and Stern 2002), the TRA (Bosnjak et al. 2006), affinity theory (Stafford and Stern 2002), involvement theory (Stafford and Stern 2002), information cascades theory (Dholakia and Soltysinski 2001), signalling theory (Gilkeson and Reynolds 2003), prospect theory (Ward and Clark 2002), the winner’s curse (Amyx and Luehlfing 2006, Bajari and Hortacsu 2003, Oh 2002), and the anchoring effect (Ariely and Simonson 2003), among others. In addition to the separate application of these theories in online auction research, there is a need for a parsimonious theory that can link all of the various theories (Baker and Song 2007).

Insights into auction design. Although auction design is not the main focus of this review, we can still draw some research directions that may result in the maximization of the economic and non-economic outcomes for all players, especially buyers. For example, to reduce the winner’s curse in online auctions, Vickrey auctions could be used, particularly for certain common value auctions. NYOP auctions may be useful in counteracting anchoring effects. To investigate these topics, future (laboratory) studies, such as experiments, could be used to test the effects of these modifications, both individually and in combination,

Table 6. Auction outcome classifications Economic Intermediary-level Revenue of the intermediary/ auctioneer Transaction-level Final auction price Participant-level Winner’s curse Consumer surplus

Non-economic Satisfaction of the intermediary/auctioneer* The number of bids Auction success Winning likelihood Customer satisfaction* Entertainment fulfilment*

Italics*: suggested research areas

on the verbal thought processes and objective responses of participants.

CONCLUSION This paper attempts to make a contribution to the research on consumer behaviour in online auctions. It reviews the extended literature on online bidding behaviour and classifies the studies into three different categories. Within each category, these studies are summarized and compared. Explanations and clarifications are given where possible. A research framework for online bidding behaviour, based on previous studies, is presented in the hope of guiding future research. Knowledge gaps and research opportunities have been derived from this framework, and suggestions for future research directions are made. As Zahedi (2002) concluded, synthesis papers are ‘a needed scholarly undertaking’; it is unfortunate that they have been largely overlooked by IS researchers. We hope that the review presented in this paper provides a comprehensive summary of the research on consumer behaviour in online auctions, that our synthesis has been thorough and thoughtful, and that our presentation of the issues and research gaps has been carried out in an unbiased manner, bearing in mind that the proposed definitions, variables, constructs and methodologies are by no means exhaustive. The aim of this paper is to provide readers with an overall picture of what has been done in this research area to date, how the various studies can be related to one another, and what can be done to advance this research in future.

References Amyx, D. A. and Luehlfing, M. S. (2006) ‘Winner’s Curse and Parallel Sales Channels – Online Auctions Linked Within e-Tail Websites’, Information & Management 43(8): 919–27. Antony, S., Lin, Z. and Xu, B. (2006) ‘Determinants of Escrow Service Adoption in Consumer-to-consumer Online

Downloaded By: [Schmelich, Volker] At: 14:54 24 March 2010

Electronic Markets Vol. 18 No 4 Auction Market: An Experimental Study’, Decision Support Systems 42(3): 1889–900. Ariely, D., Ockenfels, A. and Roth, A. E. (2005) ‘An Experimental Analysis of Ending Rules in Internet Auctions’, The Rand Journal of Economics 36(4): 890–907. Ariely, D. and Simonson, I. (2003) ‘Buying, Bidding, Playing, or Competing? Value Assessment and Decision Dynamics in Online Auction’, Journal of Consumer Psychology 13(1&2): 113–23. Avery, C. R. (1998) ‘Strategic Jumping Bidding in English Auctions’, Review of Economic Studies 65(2): 185–210. Ba, S. and Pavlou, P. A. (2002) ‘Evidence of the Effect of Trust Building Technology in Electronic Markets: Price Premiums and Buyer Behavior’, MIS Quarterly 23(3): 243–68. Ba, S., Whinston, A. B. and Zhang, H. (2003) ‘Building Trust Into Online Auction Markets through an Economic Incentive Mechanism’, Decision Support Systems 35(3): 273–86. Bajari, P. and Hortacsu, A. (2003) ‘Winner’s Curse, Reserve Prices and Endogenous Entry: Empirical insights from eBay Auctions’, Rand Journal of Economics 34: 329–55. Baker, J. and Song, J. (2007) ‘A Review of Single-item Internet Auction Literature and a Model of Future Research’, Journal of Electronic Commerce in Organizations 5(1): 42–68. Bapna, R. (2003) ‘When Snipers become Predators: Can Mechanism Design Save Online Auctions?’ Communications of the ACM 46(12): 152–8. Bapna, R., Goes, P. and Gupta, A. (2000) ‘A Theoretical and Empirical Investigation of Multi-item On-line Auctions’, Information Technology and Management 1(1–2): 1–23. Bapna, R., Goes, P. and Gupta, A. (2001a) ‘Comparative Analysis of Multi-item Online Auctions: Evidence from the Laboratory’, Decision Support Systems 32(2): 135–53. Bapna, R., Goes, P. and Gupta, A. (2001b) ‘Insights and Analyses of Online Auctions’, Communications of the ACM 44(11): 42–50. Bapna, R., Goes, P. and Gupta, A. (2003a) ‘Analysis and Design of Business-to-Consumer Online Auctions’, Management Science 49(1): 85–101. Bapna, R., Goes, P. and Gupta, A. (2003b) ‘Replicating Online Yankee Auctions to Analyze Auctioneers’ and Bidders’ Strategies’, Information Systems Research 14(3): 244–68. Bapna, R., Goes, P., Gupta, A. and Jin, Y. (2004) ‘User Heterogeneity and its Impact on Electronic Auction Market Design: An Empirical Exploration’, MIS Quarterly 28(1): 21–43. Bapna, R., Jank, W. and Shmueli, G. (2008) ‘Price Formation and its Dynamics in Online Auctions’, Decision Support Systems 44(3): 641–56. Bichler, M., Field, S. and Werthner, H. (2001) ‘Introduction: Theory and Application of Electronic Market Design’, Electronic Commerce Research 1(3): 215–20. Borle, S., Boatwright, P. and Kadane, J. B. (2006) ‘The Timing of Bid Placement and Extent of Multiple Bidding:

359 An Empirical Investigation Using eBay Online Auction’, Statistical Science 21(2): 194–205. Bosnjak, M., Obermeier, D. and Tuten, T. L. (2006) ‘Predicting and Explaining the Propensity to Bid in Online Auctions: A Comparison of Two Action-theoretical Models’, Journal of Consumer Behaviour 5(2): 102–16. Brint, A. T. (2003) ‘Investigating Buyer and Seller Strategies in Online Auctions’, Journal of the Operational Research Society 54(11): 1177–88. Bruce, N., Haruvy, E. and Rao, R. (2004) ‘Seller Rating, Price, and Default in Online Auctions’, Journal of Interactive Marketing 18(4): 37–50. Budish, E. B. and Takeyama, L. N. (2001) ‘Buy Prices in Online Auctions: Irrationality on the Internet?’, Economics Letters 72(3): 325–33. Cameron, D. D. and Galloway, A. (2005) ‘Consumer Motivations and Concerns in Online Auctions: An Exploratory Study’, International Journal of Consumer Studies 29(3): 181–92. Carare, O. and Rothkopf, M. (2005) ‘Slow Dutch Auctions’, Management Science 51(3): 365–73. Carter, C., Kaufmann, L., Beall, S., Carter, P., Hendrick, T. and Petersen, K. (2004) ‘Reverse Auctions – Grounded Theory from the Buyer and Supplier Perspective’, Transportation Research 40(3): 229–54. Chakravarti, D., Greenleaf, E., Sinha, A., Cheema, A., Cox, J. C., Priedman, D., Ho, T. H., Issac, R. M., Mitchell, A. A., Rapoport, A., Rothkopf, M. H., Srivastava, J. and Zwick, R. (2002) ‘Auctions: Research Opportunities in Marketing’, Marketing Letters 13(3): 281–96. Chen, J., Chen, X. and Song, X. (2006) ‘Comparison of the Group-buying Auction and the Fixed Pricing Mechanism’, Decision Support Systems 43(2): 445–59. Dai, Q. and Kauffman, R. (2002) ‘Business Models for Internet-based B2B Electronic Markets’, International Journal of Electronic Commerce 6(4): 41–72. Dholakia, U. M. (2005) ‘The Usefulness of Bidders’ Reputation Ratings to Sellers in Online Auctions’, Journal of Interactive Marketing 19(1): 31–40. Dholakia, U. M. and Soltysinski, K. (2001) ‘Coveted or Overlooked? The Psychology of Bidding for Comparable Listings in Digital Auctions’, Marketing Letters 12(3): 225–37. Dholakia, U. M. and Simonson, I. (2005) ‘The Effect of Explicit Reference Points on Consumer Choice and Online Bidding Behaviour’, Marketing Science 24(2): 206–17. Ding, M., Eliashberg, J., Huber, J. and Saini, R. (2005) ‘Emotional Bidders – An Analytical and Experimental Examination of Consumers’ Behaviour in a Priceline-like Reverse Auction’, Management Science 51(3): 352–64. Dodonova, A. and Khoroshilov, Y. (2004) ‘Anchoring and Transaction Utility: Evidence from On-line Auctions’, Applied Economics Letters 11(5): 307–10. Easley, R. F. and Tenorio, R. (2004) ‘Jumping Bidding Strategies in Internet Auctions’, Management Science 50(10): 1407–19.

Downloaded By: [Schmelich, Volker] At: 14:54 24 March 2010

360

Xiling Cui, Vincent S. Lai and Connie K.W. Liu & Consumer Behaviour in Online Auctions: A Review

Gilkeson, J. H. and Reynolds, K. (2003) ‘Determinants of Internet Auction Success and Closing Price: An Exploratory Study’, Psychology and Marketing 20(6): 537–66. Gregg, D. G. and Scott, J. E. (2006) ‘The Role of Reputation Systems in Reducing Online Auction Fraud’, International Journal of Electronic Commerce 10(3): 95–120. Hann, I.-H. and Terwiesch, C. (2003) ‘Measuring the Frictional Costs of Online Transactions: The Case of a Name-YourOwn-Price Channel’, Management Science 49(11): 1563–79. Hannon, D. (2004) ‘Online Buying Gathers Steam, One Buyer at a Time’, Purchasing 130(16): 40–3. Hartley, J. L., Lane, M. D. and Duplaga, E. A. (2006) ‘Exploring the Barriers to the Adoption of e-Auctions for Sourcing’, International Journal of Operations and Production Management 26(2): 202–21. Herschlag, M. and Zwick, R. (2000) ‘Internet Auctions: Popular and Professional Literature Review’, Quarterly Journal of Electronic Commerce 1(2): 161–86. Heyman, J. E., Orhun, Y. and Ariely, D. (2004) ‘Auction Fever: The Effect of Opponents and Quasi-endowment on Product Valuations’, Journal of Interactive Marketing 18(4): 7–21. Houser, D. and Wooders, J. (2006) ‘Reputation in Auctions: Theory, and Evidence from eBay’, Journal of Economics & Management Strategy 15(2): 353–69. Hu, X., Lin, Z., Whinston, A. B. and Zhang, H. (2004) ‘Hope or Hype: On the Viability of Escrow Services as Trusted Third Parties in Online Auction Environments’, Information Systems Research 15(3): 236–49. Jap, S. (2003) ‘An Exploratory Study of the Introduction of Online Reverse Auctions’, Journal of the Academy of Marketing Science 67(3): 96–108. Kamins, M. A., Dreze, X. and Folkes, V. S. (2004) ‘Effects of Seller-supplied Prices on Buyers’ Product Evaluations: Reference Prices in an Internet Auction Context’, Journal of Consumer Research 30(4): 622–8. Kauffman, R. J. and Wang, B. (2001) ‘New Buyers’ Arrival under Dynamic Pricing Market Microstructure: The Case of Group-buying Discounts on the Internet’, Journal of Management Information Systems 18(2): 157–88. Kauffman, R. J. and Wood, C. A. (2005) ‘The Effects of Shilling on Final Bid Prices in Online Auctions’, Electronic Commerce Research and Applications 4(1): 21–34. Kauffman, R. J. and Wood, C. A. (2006) ‘Doing their Bidding: An Empirical Examination of Factors that Affect a Buyer’s Utility in Internet Auctions’, Information Technology and Management 7(3): 171–90. Ku, G., Malhotra, D. and Murnighan, J. K. (2005) ‘Towards a Competitive Arousal Model of Decision-making: A Study of Auction Fever in Live and Internet Auctions’, Organizational Behaviour and Human Decision Processes 96(2): 89–103. Kwon, O., Byung, K., Choong-Ryuhn and Lee, E. J. (2002) ‘Impact of Website Information Design Factors on Consumer Ratings of Web-based Auction Sites’, Behaviour and Information Technology 21(6): 387–402. Leloup, B. and Deveaux, L. (2001) ‘Dynamic Pricing on the Internet: Theory and Simulations’, Electronic Commerce Research 1(3): 265–76.

Lin, Z., Li, D., Janamanchi, B. and Huang, W. (2006) ‘Reputation Distribution and Consumer-to-consumer Online Auction Market Structure: An Exploratory Study’, Decision Support Systems 41(2): 435–48. Lucking-Reiley, D. (1999) ‘Using Field Experiments to Test Equivalence Between Auction Formats: Magic on the Internet’, American Economic Review 89(5): 1063–80. Lucking-Reiley, D. (2000) ‘Auctions on the Internet: What’s Being Auctioned, and How?’, Journal of Industrial Economics 48(3): 227–52. MacInnes, I., Li, Y. and Yurcik, W. (2005) ‘Reputation and Dispute in eBay Transactions’, International Journal of Electronic Commerce 10(1): 27–54. Mathews, T. (2003) ‘A Risk Averse Seller in a Continuous Time Auction with a Buyout Option’, Brazilian Electronic Journal of Economics 5(2). Mathews, T. (2004) ‘The Impact of Discounting on an Auction with a Buyout Option: A Theoretical Analysis Motivated by eBay’s Buy-it-Now Feature’, Journal of Economics 81(1): 25–52. Mathews, T. and Katzman, B. (2006) ‘The Role of Varying Risk Attitudes in an Auction with a Buyout Option’, Economic Theory 27(3): 597–613. McDonald, C. G. and Slawson, V. C. (2002) ‘Reputation in an Internet Auction Market’, Economic Inquiry 40(3): 633–50. McKinney, V., Yoon, K. and Zahedi, F. ‘‘M.’’. (2002) ‘The Measurement of Web-customer Satisfaction: An Expectation and Disconfirmation Approach’, Information Systems Research 13(3): 296–315. Melnik, M. I. and Alm, J. (2002) ‘Does a Seller’s Ecommerce Reputation Matter? Evidence from eBay Auctions’, The Journal of Industrial Economics 50(3): 337–49. Mollenberg, A. (2004) ‘Internet Auctions in Marketing: The Consumer Perspective’, Electronic Markets 14(4): 360–71. Namazi, A. and Schadschneider, A. S. (2006) ‘Statistical Properties of Online Auctions’, International Journal of Modern Physics 17(10): 1485–93. Ockenfels, A. and Roth, A. E. (2002) ‘The Timing of Bids in Internet Auctions: Market Design, Bidder Behaviour, and Artificial Agents’, Artificial Intelligence Magazine 23(3): 79–87. Oh, W. (2002) ‘C2C Versus B2C: A Comparison of the Winner’s Curse in Two Types of Electronic Auctions’, International Journal of Electronic Commerce 6(4): 115–38. Onur, I. and Tomak, K. (2006) ‘Impact of Ending Rules in Online Auctions: The Case of Yahoo.com’, Decision Support Systems 42(3): 1835–42. Ottaway, T. A., Bruneau, C. L. and Evans, G. E. (2003) ‘The Impact of Auction Item Image and Buyer/seller Feedback Rating on Electronic Auctions’, The Journal of Computer Information Systems 43(3): 56–60. Park, Y.-H. and Bradlow, E. T. (2005) ‘An Integrated Model for Bidding Behaviour in Internet Auctions: Whether, Who, When, and How Much’, Journal of Marketing Research 42(4): 470–82. Pavlou, P. A. and Gefen, D. (2004) ‘Building Effective Online Marketplaces with Institution-based Trust’, Information Systems Research 15(1): 37–59.

Downloaded By: [Schmelich, Volker] At: 14:54 24 March 2010

Electronic Markets Vol. 18 No 4 Pinker, E. J., Seidmann, A. and Vakrat, Y. (2003) ‘Managing Online Auctions: Current Business and Research Issues’, Management Science 49(11): 1457–84. Rafaeli, S. and Noy, A. (2002) ‘Online Auctions, Messaging, Communication and Social Facilitation: A Simulation and Experimental Evidence’, European Journal of Information Systems 11(3): 196–207. Rafaeli,S.andNoy,A.(2005)‘SocialPresence:InfluenceonBidders in Internet Auctions’, Electronic Markets 15(2): 158–75. Rainer, K. and Miller, M. (2005) ‘Examining Differences Across Journal Rankings’, Communications of the ACM 48(2): 91–4. Roth, A. E. and Ockenfels, A. (2002) ‘Last-minute Bidding and the Rules for Ending Second-price Auctions: Evidence from eBay and Amazon Auctions on the Internet’, American Economic Review 92(4): 1093–103. Spann, M. and Tellis, G. J. (2006) ‘Does the Internet Promote Better Consumer Decisions? The Case of Name-YourOwn-Price Auctions’, Journal of Marketing 70(1): 65–78. Spann, M., Skiera, B. and Schafers, B. (2004) ‘Measuring Individual Frictional Costs and Willingness-to-Pay via Name-Your-Own-Price Mechanisms’, Journal of Interactive Marketing 18(4): 22–36. Stafford, M. R. and Stern, B. (2002) ‘Consumer Bidding Behaviour on Internet Auction Sites’, International Journal of Electronic Commerce 7(1): 135–50. Standifird, S. S. (2001) ‘Reputation and e-Commerce: eBay Auction and the Asymmetrical Impact of Positive and Negative Ratings’, Journal of Management 27(3): 279–95. Standifird, S. S. (2002) ‘Online Auctions and the Importance of Reputation Type’, Electronic Markets 12(1): 58–62. Standifird, S. S., Roelofs, M. R. and Durham, Y. (2004–5) ‘The Impact of eBay’s Buy–It–Now Function on Bidder

361 Behaviour’, International Journal of Electronic Commerce 9(2): 167–76. Stern, B. B. and Stafford, M. R. (2006) ‘Individual and Social Determinants of Winning Bids in Online Auctions’, Journal of Consumer Behaviour 5(1): 43–55. Terwiesch, C., Savin, S. and Hann, I.-H. (2005) ‘Online Haggling at a Name-Your-Own-Price Retailer: Theory and Application’, Management Science 51(3): 339–51. Townsend, A. M. and Bennett, J. T. (2003) ‘Living and Bidding in an Auction Economy’, Communications of the ACM 46(12): 351–3. Tversky, A. and Kahneman, D. (1974) ‘Judgment under Uncertainty: Heuristics and Biases’, Science 185(4157): 1124–31. Wang, K., Wang, E. T. G. and Tai, C.-F. (2002) ‘A Study of Online Auction Sites in Taiwan: Product, Auction Rule, and Trading Type’, International Journal of Information Management 22(2): 127–42. Ward, S. G. and Clark, J. M. (2002) ‘Bidding Behaviour in On-line Auctions: An Examination of the eBay Pokemon Card Market’, International Journal of Electronic Commerce 6(4): 139–55. Wilcox, R. T. (2000) ‘Experts and Amateurs: The Role of Experience in Internet Auctions’, Marketing Letters 11(4): 363–74. Yang, K. C. C. (2005) ‘Consumers’ Attitudes Toward Regulation of Internet Auction Sites: A Third-person Effect Perspective’, Internet Research 15(4): 359–77. Zhang, J. (2006) ‘The Roles of Players and Reputation: Evidence from eBay Online Auctions’, Decision Support Systems 42(3): 1800–18. Zhang, H. and Li, H. (2006) ‘Factors Affecting Payment Choices in Online Auctions: A Study of eBay Traders’, Decision Support Systems 42(2): 1076–88.