تجزیه و تحلیل تجربی از هزینه مناقصه در مزایده های قیمت مالکیت به نام خودتان
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|7295||2010||14 صفحه PDF||سفارش دهید||11162 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Journal of Interactive Marketing, Volume 24, Issue 4, November 2010, Pages 283–296
Interactive pricing mechanisms integrate customers into the price-setting process by letting them submit bids. Name-your-own-price auctions are such an interactive pricing mechanism, where buyers' bids denote the final price of a product or service in case they surpass a secret threshold price set by the seller. If buyers are given the flexibility to bid repeatedly, they might try to incrementally bid up to the threshold. In this case, charging fees for the option to place additional bids could generate extra revenue and reduce incremental bidding behavior. Based on an economic model of consumer bidding behavior in name-your-own-price auctions and two empirical studies, we analytically and empirically investigate the effects bidding fees have on buyers' bidding behavior. Moreover, we analyze the impact of bidding fees on seller revenue and profit based on our empirical results.
Interactive pricing mechanisms that integrate customers into the price-setting process are an essential part of the Web economy and electronic commerce (Bapna et al., 2003, Jap & Naik, 2004, Pan et al., 2004 and Ratchford, 2009). Online marketplaces apply a broad variety of interactive pricing mechanisms, such as ascending bid auctions (e.g., eBay), reverse auctions (e.g., MyHammer), or name-your-own-price auctions (e.g., Priceline). As opposed to posted prices set by the seller or the retailer, buyers can interactively influence the final price of a product through the submission of bids or the exchange of messages with a seller through the auction interface. Given certain consumer and market characteristics, the specific design of interactive pricing mechanisms determines consumers' bidding behavior and thus seller revenue and profit. Previous research (for an overview see Bajari & Hortaçsu, 2004 and Jap, 2007) derives design recommendations for standard online auction mechanisms. However, the Internet has also given rise to new interactive pricing mechanisms such as name-your-own-price auctions, which have generated considerable research interest (e.g., Fay, 2004, Hann & Terwiesch, 2003, Spann et al., 2004, Wang et al., 2009 and Wolk & Spann, 2008). Name-your-own-price auctions were pioneered by Priceline, which sells travel services such as airline tickets, hotel rooms or car rentals on its electronic platform. Broadening Priceline's concept, where many firms vie to make a sale to a particular consumer (Pinker, Seidmann, and Vakrat 2003), other companies use name-your-own-price auctions to sell products or services in B2C-markets (e.g., low-cost airlines (www.germanwings.com) or software sellers (www.ashampoo.com)). eBay's “Best Offer”-feature is yet another example of this mechanism. At the outset of a name-your-own-price auction, a seller sets a secret threshold price indicating her minimum acceptable price. A buyer's bid then determines the price of the product if it at least equals the seller's threshold price. Hence, buyers in name-your-own-price auctions always pay the price denoted by their bid. Moreover, information about the seller's threshold price or other buyers' bids is never published. Name-your-own-price auctions differ from standard auctions in that bidders do not compete with each other based on their bid amount but only have to surpass the secret threshold price set by the seller in order for a transaction to occur. As buyers at Priceline are typically allowed to place one bid for a specific product (e.g., airline tickets) within 24 h, Priceline's implementation of a name-your-own-price auction has been referred to as a single-bid policy (Hann and Terwiesch 2003). Despite this restriction to a single bid, bidders may place multiple bids, for example, by the illegitimate but practicable use of multiple credit cards (Fay 2004). Thus, perfect enforcement of the single-bid policy may not be feasible. This is part of the reason why some name-your-own-price sellers–including Priceline for some product categories such as calling capacity–have used a multiple-bidding policy to allow their customers to engage in “online haggling” (Hann and Terwiesch 2003) and raise their bids if an initial bid did not surpass the secret threshold price. Given this flexibility to place additional bids, buyers generate a higher expected consumer surplus due to the possibility of transacting at lower prices when they start bidding sequences at lower values. Further, buyers might bid closer to their reservation price with a multiple-bidding policy (Spann, Skiera, and Schäfers 2004). This, in turn, could lead to a higher number of successful bids and increase seller profit. Nevertheless, buyers could exploit their ability to place multiple bids and–starting at the lowest price–increase their bids with minimum increments. Using such an “epsilon strategy” (Hann and Terwiesch 2003), bidders could ensure that they purchase the product for the minimum price accepted by the seller (i.e., the seller's threshold price). Simultaneously, sellers can employ a number of constraints to avoid such minimum increment bidding behavior. For example, sellers could ask buyers to pay a small monetary fee per bid and thus charge buyers for the additional flexibility they gain over the single-bid format. Given such bidding fees are solely imposed upon the rejection of an initial bid free of charge, they constitute a form of a name-your-own-price auction in-between the single-bid format (i.e., virtually “infinite” bidding fees following the initial bid) and a multiple-bidding policy with no bidding fees. We distinguish bidding fees from both costs of entry (McAfee & McMillan, 1987 and Samuelson, 1985) and frictional costs (also “search costs” or “bid evaluation costs”) such as the opportunity cost of time necessary to log on the bidding site and to place the bid and the cost of the mental effort to determine optimal bid values (Bakos, 2003, Carr, 2003, Hann & Terwiesch, 2003, Shugan, 1980, Snir & Hitt, 2003 and Stigler, 1961). Yet, unlike these frictional costs, bidding fees are explicitly charged by the seller and have to be paid by the buyer on top of her successful bid. Therefore, sellers could use bidding fees to increase profits in name-your-own-price auctions with a multiple-bidding policy. In this case, the profit-increasing effect of bidding fees for a seller depends on the additional profit from these fees and their effect on bidding behavior. Moreover, the use of bidding fees has recently gained momentum with the rise of entertainment shopping platforms such as Swoopo (www.swoopo.com) and DubLi (www.dubli.com). Factors potentially influencing bidding behavior suggest ambiguous implications for seller revenue and profit. Since bidding fees reduce expected consumer surplus, fewer buyers will have an incentive to engage in a name-your-own-price auction. Moreover, bidding fees could diminish maximum bid values if bidders account for the decreasing consumer surplus, thus reducing the likelihood that bids surpass the seller's threshold price. Both effects could reduce the number of products sold and seller revenue. However, bidding fees could encourage bidders to increase consecutive bid values by higher amounts if they want to enhance the likelihood of success with a lower number of bids. This, in turn, could result in a higher degree of price discrimination when some buyers “overbid” the seller's uniform threshold price by a larger amount than other buyers would in a multiple-bidding scenario. Apart from the monetary amount of the bidding fees, employing this strategy could generate higher profit margins and thus increase seller profit. As a result of this potential benefit, bidding fees as a design option in a name-your-own-price auction require a thorough study of their effect on bidding behavior, revenue and seller profit. Therefore, the aim of this study is to analytically and empirically investigate the effects that bidding fees have on buyers' bidding behavior. The contribution of our paper is fourfold. First, we develop an economic model of bidding behavior to derive theoretical explanations for the effects of bidding fees on bidding behavior in a name-your-own-price auction. Second, we empirically compare multiple-bidding with single-bid policies. Third, we analyze the effects of bidding fees on bidding behavior in two empirical studies: a laboratory experiment with induced valuations and a field study with real-world transactions over the Internet. Fourth, we derive revenue and profit implications from the use of bidding fees based on the data gathered in the field study. Our paper is structured as follows. In Analytical Model of Bidding Behavior with Bidding Fees, we develop an economic model of bidding behavior in a name-your-own-price auction with bidding fees charged by a seller. Building on this model, we provide theoretical hypotheses for bidding behavior in such name-your-own-price auctions. In Empirical Studies of Bidding Behavior with Bidding Fees, we test our hypotheses in two empirical studies: a laboratory experiment and a field experiment. Further, we use the results of the field experiment to draw revenue and profit implications on the usage of bidding fees for name-your-own-price sellers. Conclusion concludes the paper with final implications and directions for future research.
نتیجه گیری انگلیسی
We analyze the effects of bidding fees on bidding behavior in name-your-own-price auctions, where the payment of the bidding fees is contingent on a successful bid. We base our analyses on an economic model of bidding behavior and two distinct data sets: a laboratory experiment with induced valuations and a field experiment involving real-world transactions. Despite some name-your-own-price sellers' current enforcement of a single-bid policy, our results indicate that consumers bid up to higher values if they are given the flexibility to bid repeatedly. Furthermore, our analytical predictions and our empirical data both indicate the strong influence bidding fees have on bidding behavior in name-your-own-price auctions. The use of bidding fees leads to a lower number of bids and higher bidding increments. Especially when we control for consumers' valuations and minimize confounding effects in the laboratory, bidding fees result in consumers starting their bidding sequences at much higher values. Interestingly, we find an asymmetric effect of bidding fees on bidding behavior, indicating that consumers' perceived impact of bidding fees is dominated by their existence rather than their actual level. We find in our empirical studies that bidding fees increase seller profit because consumers bid by higher increments when bidding fees are present. Thereby, bidding fees provide additional profit when consumers overbid the unknown threshold price by a higher amount than they would without bidding fees. This allows sellers to more effectively price-discriminate. Moreover, we find no significant effect of bidding fees on consumers' entry decision in the field experiment, suggesting consumer acceptance is not a crucial factor for the chosen levels of bidding fees and the chosen design when the initial bid is always free. Additionally, we control for entry in our laboratory experiment and generate results consistent with the field experiment. However, we find that fewer units are sold when bidding fees exist because consumers submit fewer bids compared to a situation with multiple bidding and no bidding fees. Therefore, sellers who are interested in maximizing transaction revenue–perhaps because they receive a fixed commission from this revenue–are better off applying a multiple bidding policy without bidding fees. Consequently, the heretofore neglected use of bidding fees in name-your-own-price auctions can be beneficial for sellers. However, it depends on the seller's business model whether bidding fees are beneficial. Our results on the impact of bidding fees on average increments, the number of bids placed, the percentage of successful bidding sequences and selling prices support sellers in their decision on whether to use bidding fees in name-your-own-price auctions. Further, our empirical results could be used to augment analytical models of optimal pricing mechanism design (Spann et al., in press). Further, our economic model of bidding behavior in name-your-own-price auctions with bidding fees can be used to simulate seller revenue and profit as well as the optimal threshold price and bidding fee levels for different demand conditions (i.e., distributions of consumers' willingness-to-pay, frictional costs and beliefs) and supply conditions (e.g., capacity and costs). Such a simulation analysis can be the basis of a decision support system for sellers on the use of bidding fees and related threshold price levels. We have to acknowledge several limitations, which provide avenues for future research. First, our profit and revenue results are limited to the specific values of bidding fees applied in our field study and to the chosen values of the threshold prices. Second, we cannot analyze the long-term effects of multiple bidding and bidding fees on seller revenue and profit. Although we find in our analysis that allowing multiple bidding is advantageous with and without bidding fees, this cannot be generalized to all competitive situations. Fay (2009) shows that multiple bidding may be disadvantageous if a posted-price rival reacts and chooses a lower price than the single-bid policy of the name-your-own-price seller. Therefore, the long-term viability of a multiple-bidding policy may require that the name-your-own-price seller is small compared to the posted price rival. Further, fewer successful transactions when bidding fees exist can have adverse effects on repeat purchases, word-of-mouth and the consumer acceptance of a name-your-own-price seller. Third, factors such as seller ratings (Bruce et al. 2004), collusion and information sharing on the Internet could alter average profit margins over time. Consumers may learn about the threshold prices from other consumers' bidding success via information diffusion (Hinz and Spann 2008), which can influence optimal threshold price levels and may require additional differentiation of products to make them and their threshold prices less comparable. Fourth, an interesting challenge for future research will be to better understand the mental processes that drive consumers' reactions to bidding fees.