Competitive reasons for the Name-Your-Own-Price channel

Hann and Terwiesch (2003) estimate the frictional costs (in euros) for three products, PDAs, CD-RW drives, and MP3 players. The median frictional costs for bids on the three products are 6.08, 4.29, and 3.54, respectively. Standard deviations of costs are 7.57, 4.42, and 3.81. The numbers in the text are averages across these three products converted into dollars (at a rate of euro 1 = 0.90$, which was the exchange rate when the original data were collected).

Other non-posted price mechanisms may also be able to mitigate price competition in this context. However, it is beyond the scope of the current paper to compare the NYOP format to all possible pricing mechanisms.

Fay (2008) and Fay and Xie (2008) provide additional discussion of the role of opacity in market interactions.

There is some evidence that current NYOP sellers have indeed committed to such a decision structure. For example, Priceline determines whether a given bid is accepted using a complicated computer formula which includes a random element (Segan 2005) and employs a “randomizer” program for deciding whether to accept a bid for a hotel room (Malhotra and Desira 2002; Haussman 2001). In particular, rather than setting the threshold price equal to the lowest rate offered by the entire set of hotels that have rooms available, Priceline only compares the bidder’s offer to the rates set by two randomly selected hotels. Such an action is consistent with the speculation in Kannan and Praveen (2001) that Priceline may “deliberately forgo a successful transaction” in order to influence consumers’ expectations, i.e., persuade consumers to bid higher in the future.

Notice that if the second type also believed that the threshold price was drawn from U[0, 101], they would not bid 150, but instead bid 101 or less.

Sequential format choices are not required for the core results of this paper. However, with simultaneous format decisions, multiple equilibria may arise—one in which firm A selects the NYOP format and firm B selects posted prices and one in which firm A selects the posted-price format while firm B selects the NYOP format. Thus, coordination issues (i.e., which equilibria is played) may arise in a simultaneous-choice setting.

While assuming a common reservation value dramatically simplifies the analysis, it is not essential for the paper’s main qualitative results. A more subtle point is that variation in frictional costs is essential for appropriately segmenting consumers. The NYOP mechanism is an effective means of targeting consumers with low frictional costs but not those with low valuations. Thus, to receive the price discrimination effect touted in the literature (e.g., Hann and Terwiesch 2003), variation in frictional costs is essential, but variation in reservation values is not.

In particular, for \(0 < \bar c < \frac{1}{4}\) and \(\bar c > \frac{1}{2}\), the key idea of Proposition 1 continues to hold, i.e., one firm will choose the NYOP form if there is “sufficiently” little brand loyalty. However, the amount of brand loyalty that is “sufficient” to generate this result depends on \(\bar c\). For \(\bar c < \frac{1}{4}\), it is possible for the NYOP format to be chosen in equilibrium even when λ exceeds \(\frac{1}{9}\) (since the cost of bidding on the NYOP channel is not as onerous, on average). But, for \(\bar c > \frac{1}{2}\), the sufficient condition for the NYOP format to be chosen in equilibrium is tighter (since fewer consumers are willing to bid at an NYOP site).