Seller Reputation, Buyer Informativeness and Trust in the Market
How does healthiness of an Online Market Impact Price Dispersion

Dr. Juan Feng
Associate Professor of the Department of Information Systems
Contact: juafeng@cityu.edu.hk
Written by Dr. Juan Feng

Price dispersion refers to the phenomenon that an identical product is sold at different prices by different sellers. It is commonly believed that price dispersion is due to the fact that buyers do not have the capability to do thorough search before making purchase. With the development of Internet economy, it is curious to understand whether price dispersion is still pervasive compared to the physical world. Because internet market features fierce competitions due to an increased number of sellers, price transparency and reduced buyer search costs, it is claimed to be a frictionless market since consumers can effortlessly locate the best deal on the Internet. According to the classical economic model, all sellers should set the same price in the “frictionless” market —the “law of one price” (LOP). However, contrary to the theoretical prediction, researchers find substantial price dispersion in online markets. In addition, there are also contradictory findings about whether a seller with higher reputation level should set a higher price.

In this project we try to reconcile the contradiction between empirical findings and theoretical predictions. We understand that Internet markets inherit much higher uncertainties compared to physical markets (for example, sellers are often hidden under the masks of meaningless electronic IDs; it is hard to inspect the product before payment; at the same time, the payment and delivery for the products are also separated, etc.). From the seller side, reputation systems developed for the internet markets greatly help alleviate such uncertainties, which also affecting the price dispersion we observe in practice. From the buyer side, whether or not a buyer knows how to search for an ideal deal determines whether or not this buyer will buy a product at a high price, which in turn affects the pricing strategy of sellers. For example, a low-reputation seller may completely give up the consumers who know how to search and compare, and sets a high price just for those uninformed buyers. However, by doing so, such low-reputation sellers forego the opportunity to sell to more buyers and they may always remain to be low-reputation sellers.

To capture this dynamic feature of reputation growth in an Internet market, we include the healthiness of an Internet market in our model. No matter how perfect the existing reputation system is, it cannot capture the true “quality” of a new seller who just joins the market. A “healthy” market should help those “high-quality” new sellers with low reputation under the current reputation system to develop into a high-reputation seller. The chance that a low reputation seller may be a “good” seller gives confidence for some “informed” buyers to buy from a low-reputation new seller. This in turn gives incentive for a low-reputation seller to grow into a high reputation one, which also boosts the “healthiness” of the market. How likely a low-reputation seller is a “good” seller is determined by the internet market, and this in turn determines how likely an “informed” buyer will buy from a low-reputation seller.

Our model finds that online price dispersion depends on both the seller reputation, buyer informativeness, as well as the healthiness of the Internet market: (1) A high-reputation seller may set a lower price compared to a low-reputation seller when there are many informative buyers; (2) online price dispersion reduces when more buyers are informative who can search and compare between multiple offers; (3) online price dispersion is smaller among sellers with higher reputation; (4) online price dispersion reduces when the internet market is more healthy, that is, when there is less chance that a low-reputation seller is a “bad” one, or, when the internet market attracts less “bad” sellers.

Research Studies by

Dr. Yuewen Liu, PhD of City University of Hong Kong and China University of Technology Joint Program; and Dr. Juan Feng

[ Back ]