The authors study the adoption of technologies in a network setting, where agents are connected to each other through a network and the value of the technology is uncertain, high or low. There’s a monopolist seller who aims to maximize the profits. In today’s technology-driven world, potential consumers can learn about technology either by using it themselves, at some risk as there’s uncertainty attached to the value of the technology. Consumers could also wait to get feedback from others, thus free-riding on the information generated by others. The seller could either induce consumers to adopt early by offering direct discounts or offer them referral incentives, which are rewards to early adopters based on the number of his/her friends who buy the technology in the second period. They found that consumers with high or low degrees may choose to adopt technology in the early period while consumers with average or in-between degrees delay technology adoption. Consumers who adopt technology later, free-ride on the information generated by the early adopters of technology.
From the sellers perspective, the authors review two important pricing policies: inter-temporal price discrimination (discounts to consumers who adopt early) and referral incentives. Referral incentives programs have been used by companies like Airbnb and Amazon to grow their consumer base quickly. They have taken a game-theoretic approach on a consumer network wherein consumers decide to adopt the technology of uncertain quality early-on or after a delay based on feedback from their friends. Not adopting the technology is also a possible action. They have used a mean-field version of the game to simplify the problem a bit, as otherwise for early adoption of the technology, each consumer’s decision would depend on her/his forecast of the consumer’s friends’ adoption decision and so on. They found that inter-temporal pricing discrimination acts as a screening mechanism wherein consumers with lower-degree are the early adopters of technology. And, referral incentives target consumers with higher-degree for early technology adoption.
In their work, the seller is not observing the network and is provided only with information about consumers degree distribution to implement these pricing mechanisms. The authors make a mean-field assumption that “each agent conjectures that the degrees of his or her neighbors are drawn i.i.d according to the edge-perspective degree distribution f ̃(d)=f(d)d/(∑_d'▒f(d^' )d').” (page 571 of M.V. Leduc et al., 2017). As noted by the authors themselves, the assumption could be relaxed by introducing correlation in degrees and this could serve as an extension of their work. The other direction in which their work can be extended is by providing the seller with more network theory information and not just the degree distribution of the consumers. The seller could be provided with information on the degrees and centrality measures of the majority of the consumers. As we are looking at the information flow here, good or bad review of the technology, betweenness centrality would prove to be a good measure to study the diffusion process. The network data could also be related to data on consumer's choices and preferences with respect to technology, so that seller can formulate price-discriminatory policies based on their degrees. This will help sellers to control and promote the diffusion of technology in a better way.
References:
Leduc, Matt V., Matthew O. Jackson, and Ramesh Johari. "Pricing and referrals in diffusion on networks." Games and Economic Behavior 104 (2017): 568-594
The authors study the adoption of technologies in a network setting, where agents are connected to each other through a network and the value of the technology is uncertain, high or low. There’s a monopolist seller who aims to maximize the profits. In today’s technology-driven world, potential consumers can learn about technology either by using it themselves, at some risk as there’s uncertainty attached to the value of the technology. Consumers could also wait to get feedback from others, thus free-riding on the information generated by others. The seller could either induce consumers to adopt early by offering direct discounts or offer them referral incentives, which are rewards to early adopters based on the number of his/her friends who buy the technology in the second period. They found that consumers with high or low degrees may choose to adopt technology in the early period while consumers with average or in-between degrees delay technology adoption. Consumers who adopt technology later, free-ride on the information generated by the early adopters of technology.
From the sellers perspective, the authors review two important pricing policies: inter-temporal price discrimination (discounts to consumers who adopt early) and referral incentives. Referral incentives programs have been used by companies like Airbnb and Amazon to grow their consumer base quickly. They have taken a game-theoretic approach on a consumer network wherein consumers decide to adopt the technology of uncertain quality early-on or after a delay based on feedback from their friends. Not adopting the technology is also a possible action. They have used a mean-field version of the game to simplify the problem a bit, as otherwise for early adoption of the technology, each consumer’s decision would depend on her/his forecast of the consumer’s friends’ adoption decision and so on. They found that inter-temporal pricing discrimination acts as a screening mechanism wherein consumers with lower-degree are the early adopters of technology. And, referral incentives target consumers with higher-degree for early technology adoption.
In their work, the seller is not observing the network and is provided only with information about consumers degree distribution to implement these pricing mechanisms. The authors make a mean-field assumption that “each agent conjectures that the degrees of his or her neighbors are drawn i.i.d according to the edge-perspective degree distribution f ̃(d)=f(d)d/(∑_d'▒f(d^' )d').” (page 571 of M.V. Leduc et al., 2017). As noted by the authors themselves, the assumption could be relaxed by introducing correlation in degrees and this could serve as an extension of their work. The other direction in which their work can be extended is by providing the seller with more network theory information and not just the degree distribution of the consumers. The seller could be provided with information on the degrees and centrality measures of the majority of the consumers. As we are looking at the information flow here, good or bad review of the technology, betweenness centrality would prove to be a good measure to study the diffusion process. The network data could also be related to data on consumer's choices and preferences with respect to technology, so that seller can formulate price-discriminatory policies based on their degrees. This will help sellers to control and promote the diffusion of technology in a better way.
References: Leduc, Matt V., Matthew O. Jackson, and Ramesh Johari. "Pricing and referrals in diffusion on networks." Games and Economic Behavior 104 (2017): 568-594