Pei-Yu Sharon Chen

Measuring Switching Costs and the Determinants of Customer Retention in Internet-Enabled Businesses: A Study of the Online Brokerage Industry

The ability to retain and lock in customers in the face of competition is a major concern for online businesses, especially those that invest heavily in advertising and customer acquisition. In this paper, we develop and implement an approach for measuring the magnitudes of switching costs and brand loyalty for online service providers based on the random utility modeling framework. We then examine how systems usage, service design, and other firm and individual-level factors affect switching and retention. Using data on the online brokerage industry, we find significant variation (as much as a factor of two) in measured switching costs. We find that customer demographic characteristics have little effect on switching, but that systems usage measures and systems quality are associated with reduced switching. We also find that firm characteristics such as product line breadth and quality reduce switching and may also reduce customer attrition. Overall, we conclude that online brokerage firms appear to have different abilities in retaining customers and have considerable control over their switching costs.

Bundling with Customer Self-Selection: A Simple Approach to Bundling Low-Marginal-Cost Goods

With declining costs of distributing digital products comes renewed interest in strategies for pricing goods with low marginal costs. In this paper, we evaluate customized bundling, a pricing strategy that gives consumers the right to choose up to a quantity M of goods drawn from a larger pool of N different goods for a fixed price. We show that the complex mixed-bundle problem can be reduced to the customized-bundle problem under some commonly used assumptions. We also show that, for a monopoly seller of low marginal cost goods, this strategy outperforms individual selling (M = 1) and pure bundling (M = N) when goods have a positive marginal cost or when customers have heterogeneous preferences over goods. Comparative statics results also show that the optimal bundle size for customized bundling decreases in both heterogeneity of consumer preferences over different goods and marginal costs of production. We further explore how the customized-bundle solution is affected by factors such as the nature of distribution functions in which valuations are drawn, the correlations of values across goods, and the complementarity or substitutability among products. Altogether, our results suggest that customized bundling has a number of advantages-both in theory and practice-over other bundling strategies in many relevant settings.

Versioning and Piracy Control for Digital Information Goods

Technological advances in digitalization and communications technologies have aggravated the information goods piracy problem. In contrast to previous literature which mainly considers solutions, such as law enforcement or technology protection that work on increasing individual piracy costs to alleviate the piracy problem, we consider using versioning as a potential instrument to fight piracy. We show that while a single version is the optimal strategy for an information goods provider absent piracy, the presence of piracy may lead firms to offer more than one quality, and versioning can be an effective and profitable instrument to fight piracy for digital information goods under some conditions. Our results indicate that the incentive to version is greater when the piracy cost is in a certain range. This strategic interaction between piracy cost and product line design has an interesting implication on optimal investment in piracy control. We also find that versioning can act as both a strategic substitute and a strategic complement to other instruments that increase consumer piracy costs. We further provide a general model, as a nonlinear mixed-integer program, that would assist an information goods provider in determining how many versions to offer, at what quality levels and prices.

The Impact and Implications of On-Demand Services on Market Structure

This paper considers on-demand services and its impact on market structure, firm profitability, and consumer welfare. The unique properties of on-demand services are the conversion of fixed costs to variable costs, removal of capacity constraint, and fast setup time (which enables quick entry by any firm at any time when there is opportunity), whereas privacy and security concerns and switching costs have been noted as the biggest barriers from adopting on-demand services. With a stylized model capturing these benefits and barriers to using on-demand services, we establish several results. First, we show that conversion of fixed cost to variable cost enables new and small firms to enter existing markets and leads to the creation of new markets. Second, we show that competition and the threat of new entrants can be an important driver of a firms decision to switch to on-demand services. In addition, a firms barriers to using on-demand services can influence another firms entry decision. Third, we show t...

Value of Multi-Dimensional Rating Systems: An Information Transfer View

This paper empirically examines the value of multi-dimensional online rating system (versus single-dimensional online rating system) from an information transfer perspective. Our key identification strategy hinges on a natural experiment that took place on TripAdvisor.com that allows us to identify the causal effect with a difference-in-difference approach. Our key findings, first show that consumers’ ratings for the same restaurants are significantly higher in TripAdvisor after its adoption of the multidimensional rating system. Second, we show that restaurants with lower price level benefit more from rating system change. Third, we show that the ratings in single-dimensional rating system are similar to the lowest dimension in the multi-dimensional system. The results demonstrate the information value of multi-dimensional ratings. Our study provides important implications for a better design of online WOM systems to help consumers match their preferences with product/service attributes.