Optimal Digital Content Distribution Strategy in the Presence of the Consumer-to-Consumer Channel

although the online business-to-consumer (B2C) channel is the primary selling channel for digital content (e.g., videos, images, and music), modern digital technology has made possible the legal dissemination of such content over the consumer-to-consumer (C2C) channel through personal computing devices, such as PCs, mobile phones, and portable media players. This paper investigates the optimal channel structure and the corresponding pricing and service strategies for digital content distribution in order to understand the business value of introducing the C2C channel alongside the prevailing B2C channel. We identify conditions under which it is more profitable to use both B2C and C2C channels simultaneously (i.e., the dual-channel distribution). In such cases, the seller performs price discrimination among consumers but provides them with a higher level of service. Our analysis further characterizes the benefits of service provision. We show that service provision can increase the dual-channel pricing flexibility, reduce the seller’s B2C channel dependence, and allow the seller to tolerate higher C2C channel redistribution costs. finally, in examining the effect of a competitively determined B2C channel price on optimal channel strategy, we find that the seller prefers a dual-channel distribution under higher B2C channel prices. ∗ all authors contributed equally to this paper. 242 fENg, guO, aND ChIaNg KeY Words and phrases: B2C, C2C, channel strategy, digital content, dual channel, pricing, service. in a reCent intervieW, Douglas Merrill, former google chief information officer (CIO) and now president of EMI digital business division, said, “I think the industry as a whole has got some really interesting experiments in what the future world is. . . . We don’t know yet what the real business model is going to be. We have to do more experiments, try more things to see what works” [17]. at a time when the music industry is in flux, Merrill’s move from google to EMI aims to help the company form a digital business strategy to compete on the Internet. It is yet to be seen what innovation is needed in a realm where all music companies have had limited success. It is clear, however, that integration of technology innovation into the firm’s digital business strategy is being taken seriously. The Web has created unprecedented opportunities to distribute digital content (or information goods) such as music, videos, e-books, and software. anyone who owns the product can become an effective marketer in the distribution chain due to easy reproduction and redistribution. On the other hand, this capability can easily be outweighed by increasing online infringement. In contrast to conventional methods such as tight law enforcement, and technology protection that increases the costs to pirates, of copyright violation, an alternative solution invites consumers to a broader participation in the distribution chain so they can be hosts for legitimate business as easily as for copyright infringement [10]. This new approach rewards consumers who share the digital content with others. Because unlimited copies of the digital content can be created and distributed, the reward could be large enough to cover the original purchase price. The reward provides an economic incentive for consumers to purchase and distribute digital files legitimately. as new technological tools can meet both business and consumer needs that were impossible before, allowing consumers’ active participation in the distribution chain is now recognized as an innovative digital business strategy for distributing information goods [24]. Several business initiatives with regard to the digital content distribution have been proposed in various technological environments. for instance, Brilliant Digital Entertainment subsidiary altnet formed an alliance with KaZaa to kick off a peer-topeer (P2P) marketing campaign [10]. KaZaa users who search for content see results displaying altnet files with a gold icon, indicating that the files are available for lawful download and use. users who share the searched files with others can accumulate gold Points to redeem rewards. More recently, a novel platform was proposed in the mobile commerce environment for the free trade of digital content where ordinary users are allowed to market and resell copies of digital content to neighbors in their wireless devices vicinity [5, 12]. Transactions among users can first be performed in an offline P2P manner without the immediate assistance of any central entity. later when users are connected to the Internet, sales are recorded and parties involved in OPTIMal DIgITal CONTENT DISTrIBuTION STraTEgY 243 the transactions are credited to their own accounts accordingly. resellers share part of the proceeds with the original copyright owner as both incentive rewards for their marketing efforts and compensations for their contribution of personal resources for the offline transaction. Such a process has been shown to be technically feasible [15]. The underlying technology that makes the decentralized distribution architecture work is digital rights management (DrM) [9]. DrM is an umbrella term including any digital protection system on any type of digital media. Common types of DrM include encrypted codes on DVDs and CDs, digital watermarking, and product activation. DrM aims at protecting ownership and copyright of digital content by restricting what actions an authorized recipient may take with respect to that content. for example, apple’s iPod uses its own DrM system called “fairPlay,” which encrypts songs and limits the playback of music purchased on iTunes, thereby protecting musicians’ copyrights [3]. While music labels continue their battle against illegal downloading in the courts, they have increasingly shifted their focus toward promoting legitimate online distribution. Embracing and leveraging DrM technology is essential to this effort. So far, a healthy number of DrM-codec platforms have been approved for use by mobile operators to explore this mobile music market opportunity [31]. Two other factors also contribute to the development of this new form of paid-content, consumer-to-consumer (C2C) channel distribution. first, increasingly popular social interactions among taste-sharing user groups can easily transform fans into effective marketers in the Internet environment. This model exploits social interactions to promote digital product distribution and consumption. Second, newly available information technologies (IT) such as micropayments provide trading platforms to support distributed information processing and market transactions among consumers. This secured trading environment makes it technically feasible for the copyright owner to effectively track sales and monitor user accounts. Both social networks and new trading platforms help create the most cost-effective model for legitimate C2C distribution. Yet although the concept is intuitively appealing and technically feasible, its underlying business model is neither well defined nor widely adopted. Today, the dominant trading platform for digital products is based on the online business-to-consumer (B2C) model. There are over 500 legitimate online music services in over 40 countries. iTunes, the leader in online downloads, has sold over 2 billion tracks since its launch in april 2003 and more than 1 billion in 2006 alone [9]. The recent innovation of the incentive-based C2C channel distribution represents a new business strategy addressing the emerging new form of digital content delivery. along with the huge potential, growing pains can occur when established business models are confronted with emerging technological advances. Businesses must carefully assess the potential benefits and involved trade-offs when integrating the new C2C channel distribution into their existing B2C channel strategy. This paper aims to address these fundamental issues in the dual-channel design and distribution. We have developed economic models to study the business potential of adopting a dual-channel distribution for a digital product with heterogeneous consumers whose willingness to pay depends on both their inherent valuation of the product and the 244 fENg, guO, aND ChIaNg associated service provision provided by the product seller/creator. under various market scenarios, we identify conditions under which it is more profitable to use both B2C and C2C channels simultaneously. We find that when the dual-channel distribution is optimal, the seller will always set a higher B2C channel price and offer a higher level of service in comparison to the optimal single-channel distribution strategy. Our further analysis also indicates that service provision increases the dual-channel pricing flexibility, reduces the seller’s B2C channel dependence, and allows the seller to tolerate a higher C2C channel redistribution cost. We also explore a likely scenario where the digital product has a competitively determined price in the B2C channel. We find that the higher the B2C channel price, the less the seller’s B2C channel dependence in adopting an optimal dual-channel strategy. literature review tWo streams of literature are partiCularlY relevant to our study—dual-channel strategies in the supply chain and economics models for digital content distribution. a large body of work on the supply-chain structures over the Internet has focused on the dual-channel distribution models, which typically consist of a manufacturerowned direct sales channel and an intermediated retail channel. Channel conflict and coordination have been widely studied in the literature, often with a focus on pricing [29]. under various scenarios, dual-channel design has been shown to have important strategic implications in firms’ optimal distribution strategy [18]. Besides pricing, service has been viewed as a nonprice attribute that positively affects demand [14, 28, 30]. Strategic consumers who either have heterogeneous valuation [8] or heterogeneous preference toward channel selection [4] have also been considered in modeling channel competition. however, most works in this area assume the dualchannel distribution structure of selling physical products in a B2C environment. few insights are available for managing the sales of digital products in an environment that can be characterized by extensive C2C interaction. With the increasingly sophisticated Internet delivery channel and the ever-increasing digitalization of products and services, economics of digital goods have received substantial research attention in recent years [19, 23, 25]. Due to the popularity of P2P technologies and free content shared and exchanged in the network, piracy is often considered as a significant threat on the copyright owners’ profitability [6, 16]. Some studies assume that pirate copies are low-quality alternatives to original products. It is shown that, under some conditions, unauthorized reproduction can help the copyright owner price discriminate different classes of consumers [26]. alternatively, piracy can be viewed as an opportunity for consumers to try out the product before making a legitimate purchase. Therefore, sampling can be a potential piracy-mitigating strategy [7]. Much research effort in this area shows the potential of using economic rather than purely technical solutions to fight piracy. In addition to piracy, free riding [27] is another widely observed phenomenon. Various incentive mechanisms and payment schemes have been proposed to encourage users to share files in P2P networks [11]. One way to ensure participants are comOPTIMal DIgITal CONTENT DISTrIBuTION STraTEgY 245 pensated is cascading payments, where royalties and commissions are determined by the flow of distribution through the system [1]. another solution, dynamic referral strategy, provides payment to users who distribute content to others depending on the sufficient diffusion of digital media in the P2P network [13]. however, none of these proposed methods have yet been successfully implemented. recent work also shows that in the competitive interaction between the centralized client-server structure and the decentralized P2P networks, the coexistence of a P2P network with a competing centralized architecture can be mutually beneficial [2]. The same study suggests that the impact of P2P networks needs to be carefully considered when pricing digital legal downloads in the B2C market. Prior work, while interesting, does not consider an integrated model that can be supported by a platform structure taking advantage of both the centralized (e.g., iTunes) and decentralized (e.g., C2C) distribution strategies. This paper aims to provide a first step toward understanding the consumer-oriented dual-channel distribution of digital content. a Dual-Channel Model with C2C Distribution in this seCtion, We present tWo general models. The benchmark model has a digital product seller (or copyright owner) operating an online B2C channel that sells directly to customers. We compare this single-channel model with a dual-channel model in which the seller allows consumers who have already purchased the product from the online B2C channel to legally resell to other consumers (i.e., using both the B2C direct channel and the C2C indirect channel simultaneously). first, let us focus on the single-channel model. Consider a seller who sells a digital product to a market with heterogeneous consumers through the online B2C direct channel. We characterize each consumer by a parameter, n ∈ [0, a], representing his or her basic valuation for the digital product. following a conventional treatment (e.g., [21, 22, 27]), we assume that the basic valuation is uniformly distributed with the population density normalized to one, which, without further loss of generality, captures heterogeneity in necessities, preferences, tastes, and purchasing power, and so on, for the digital product. We further assume that a consumer’s willingness to pay also depends on a product-specific service, denoted by s, that is provided by the seller. The service can be understood as a nonprice attribute associated with the product offering, such as product quality, after sales supports, software updates, warranties, and so on. The service can also be viewed as terms and conditions that are defined by the access control technologies in terms of usage, transfer, or storage of the digital product. for example, enhancing the quality of graphical user interface (guI) or incorporating additional features of a given computer software can typically increase consumers’ willingness to pay for the product. Similarly, a song with unlimited playbacks is more valuable than the same song with a limited number of playbacks, and a video file that can be played in most commonly used digital media devices is more attractive than the same video file that can only be played on some restricted codec platforms. Therefore, we define the reservation price for the consumer whose basic valuation is n as U(n, s) = n + fs, where f is a positive constant that measures the service effect. 246 fENg, guO, aND ChIaNg let p denote the price charged by the seller. assume that each consumer, who demands at most one unit, will buy the product if the resulting surplus is nonnegative. Then, the valuation of the consumer who is indifferent about buying from the B2C channel is defined as n[: n[ = p – fs. (1) Because all customers whose valuation satisfies n ∈ [n[, a] will buy the product, the total number of buyers will be a – n[, that is, D(p, s) = a – p + fs. (2) Obviously, the market demand is affected negatively by market price p, but positively by the seller’s service effort s. alternatively, the seller can consider a dual-channel model consisting of both a B2C direct channel and a C2C indirect channel. Suppose the original seller charges a royalty fee, denoted by R, to the reseller for successfully redistributing the digital product to a customer through the C2C channel. let d be the cost incurred by the resellers to redistribute the product to a customer. The redistribution cost d reflects the sellers’ time and the effort spent in the redistribution processes, which include identifying potential customers, interacting with them, differentiating among them, and redistributing the product to them through the C2C technology. To account for the rich interactive patterns and unique C2C distribution nature in the C2C network, assume that the C2C channel pricing decision is delegated to the consumer resellers, who may charge a high/low price to high-/low-valuation consumers. Therefore, unlike that in the B2C channel, the price in the C2C channel is not a decision variable for the original seller. The C2C channel prices will be negotiated between the resellers and the buyers such that the resellers have a nonnegative surplus for making a sale and the buyers have a higher surplus to buy from the resellers (rather than to buy from the B2C channel with price p). Thus, given the B2C price p, the royalty fee R, and the redistribution cost d, all possible prices in the C2C channel fall on the interval [R + d, p]. This implies that no sales will occur in the C2C channel if R + d > p. When R + d ≤ p, the marginal consumer who is offered the lowest possible price R + d is indifferent to buying from the C2C channel or nothing at all. The marginal consumer’s valuation n] is expressed as n] = R + d – fs. (3) When there is an opportunity to buy the product in the C2C channel, there is a channel cannibalization effect on the B2C channel. To model this effect, we define q, 0 < q ≤ 1, as the proportion of innovators (or loyal consumers) who always buy from the B2C channel and become the early adopters of the product provided that the price p does not exceed their reservation prices. If the innovators cannot afford to buy from the B2C channel (p is higher than their reservation prices), they would buy from the C2C channel so long as the offered price in the C2C channel is lower than or equal to their reservation prices. accordingly, those innovators whose valuation satisfies n ∈ [n[, a] (Segment 1 in figure 1) will buy from the online B2C channel, while those OPTIMal DIgITal CONTENT DISTrIBuTION STraTEgY 247 with n ∈ [n] , n[ ] (Segment 2 in figure 1) will buy from the C2C channel at an affordable price. all other customers whose valuation satisfies n ∈ [n], a] (Segment 3 in figure 1) will buy from the C2C channel with a higher consumer surplus. as a result, the total number of customers in Segments 1, 2, and 3, respectively, can be spelled out as q(a – p + fs), q(p – R – d), and (1 – q)(a – R – d + fs). In summary, the total demand in the B2C channel (Segment 1) is given by D B (p, s) = q(a – p + fs), (4) while the total demand in the C2C channel (Segments 2 and 3) is specified as D C (p, R, s) = (1 – q)(a – p + fs) + p – R – d. (5) It is straightforward to verify that Equation (5) reflects the empirically most likely scenario where the demand in the C2C channel increases in the B2C channel price p and the service level s, but decreases in the royalty fee R. Note that the parameter a can be interpreted as the potential market size. When price and service are offered such that all consumers make their purchase, we say the market has full coverage. Otherwise, the market has partial coverage. To capture the diminishing returns on service expenditure, following a common assumption in the literature (e.g., [14, 28]), let the service cost incurred by the seller be a quadratic function of the service effort, s/2. When only the B2C channel strategy is adopted, the seller sets the price and the service level that maximize its total profit given by