Socially-Optimal Design of Service Exchange Platforms with Imperfect Monitoring

Reviewer: Cecilia G. Manrique

The dilemma of rating mechanisms reminds one of ratings of professors according to their students' evaluation of their performance in the classroom. Very often, once the students who do not like their professor make their entries, the professor is stuck with such a rating forever. Thus, many try not to put much stock in rating mechanisms. But when it comes to services being provided, such ratings have been found to be very useful when taken with a grain of salt. Thus, ratings of services provided by hotels, catering services, and housing rentals using TripAdvisor or Angie's List are very useful to consumers who want to make sure that they get the most out of their expenditures on such items. According to the paper, "In a typical service exchange platform, a user plays a dual role: as a client, who requests services, and as a server, who chooses to provide high-quality or low-quality services." Some of the common features of many service exchange platforms-such as the large user population and the anonymity of users-complicate matters because the optimal recommended behavior depends not only on the current rating distribution, but also on the history of past rating distributions. As a result, each user interacts with a randomly matched partner without knowing the partner's identity. The absence of a fixed partner and the anonymity of the users create problems in the form of "free-riders," who may receive high-quality services from others as a client, while providing low-quality services as a server. Thus, the authors recommend a rating mechanism that consists of a rating update rule, helping to alleviate the impact of past low ratings. (Oh, were that to be used for the student evaluation of instruction!) With the use of equations, graphs, tables, and figures, the authors make the case for the optimal design of a social exchange platform that undergoes imperfect monitoring. They use game theory and a system model to design their platform. They admit that keeping track of all the users' ratings and recommending rewards or punishments for high-quality or low-quality services is a monstrous task. Errors arise from inaccurate assessments or from some system or network errors. Sometimes, whitewashing can occur whereby users with low ratings can register as new users with the hope of clearing their history of bad behaviors that warrant low ratings. The paper instead proposes a design framework using simple binary rating mechanisms that can achieve the social optimum in the presence of rating update errors. The bottom line is this: the ratings game is a waiting game. It requires patience in anticipating more ratings that can offset poor previous ratings. Online Computing Reviews Service