Research

Working Papers

Harvesting Ratings with Johannes Johnen
Revise and Resubmit, Management Science.
    Download Paper     Web Appendix

Abstract Ratings play a crucial role in online marketplaces, shaping consumer decisions and firm strategies. We investigate how firms strategically use pricing to influence ratings, and how this undermines ratings as signals of product quality. We develop a two-period model of price competition between an established firm and a potentially high- or low-quality entrant, capturing the challenge high-quality newcomers face in building reputation. Consumers rate based on value-for-money, but cannot distinguish whether positive ratings result from genuine quality or discounted prices. Low-quality entrants take advantage of this and may offer low prices to harvest good ratings in the future, or mimic high prices to signal high quality. We show that ratings harvesting inflates positive ratings, reduces their informativeness, and exacerbates the cold-start problem, discouraging high-quality entry. Our results mirror empirical patterns and generate implications for how rating design affects market outcomes: reducing effort-costs to rate induces more but less-informative ratings, and discourages entry. Policy implications include discouraging excessive discounts for new sellers, incorporating price-paid into rating displays, and balancing rating effort-costs to preserve informativeness. While the effects of individual entrants' harvesting may appear temporary, harvesting hinders high-quality entrants from building reputation, discouraging entry and causing lasting distortions.
Cite as: Johnen, J. and Ng, R. (2024). Harvesting Ratings. CRC TR 224 Discussion Paper No. 509.

Free and Open-Source Software: Coordination and Competition
    Download Paper

Abstract Free and Open-Source Software (FOSS) are developed by a community of developers led by a coordinator. To attract more developers, coordinators may increase FOSS quality by improving coordination. However, more developers create more diverse views, and permissive FOSS licenses mean additional developers drive FOSS characteristics away from the preferences of existing developers. Permissive FOSS licenses induce competition, leading to lower prices, but lower the incentive to coordinate FOSS, leading to lower quality FOSS serving only niche markets. I study coordinators with different motives: self-interested Founders; volunteering Altruists; profit-driven Managers, showing that permissive licenses only favour Managers.
Cite as: Ng, R. (2024). Free and Open-Source Software: Coordination and Competition. CRC TR 224 Discussion Paper No. 585.

Ratings with Heterogeneous Preferences with Jonathan Lafky
    Download Paper

Abstract We examine how ratings are interpreted in the presence of heterogeneous preferences among both raters and consumers. Raters with altruistic motives should rate for the benefit of future consumers, however an ambiguity arises when preferences are heterogeneous. Multiple equilibria exist in which ratings reflect either the preferences of raters or the preferences of future consumers. In an online experiment, we examine how ratings are selected by raters and interpreted by consumers, and how two types of information influence equilibrium selection. We show how both raters and consumers update their evaluation of what a rating represents in each environment, doing so in similar ways.
Cite as: Lafky, J. and Ng, R. (2024). Ratings with Heterogeneous Preferences. CRC TR 224 Discussion Paper No. 594.

Moderating Content-Hosting Platforms with Greg Taylor
    Download Paper

Abstract We study how content moderation facilitates communication on online platforms. A sender transmits information to a receiver, exerting effort to signal their truthfulness. Communication fails without moderation because the effort required is prohibitive. Moderation resolves this problem by making effort a more powerful signal of veracity. However, moderation crowds-out sender effort, decreasing content quality on the platform. A socially optimal or profit-maximizing policy may therefore involve limited moderation. We study the choice between being a platform or broadcaster, how moderation influences competition for attention, and the effects of misinformation actors, AI-generated content, and moderator errors on the sustainability of communication.
Cite as: Ng, R. and Taylor,G. (2025). Moderating Content-Hosting Platforms. CRC TR 224 Discussion Paper No. 698.


Some Works In Progress

Competition through Recommendations     Draft available, do reach out!

Abstract This paper examines how a two-sided platform designs its recommender system to be informative about value-for-money and how competition affects this design. More informative recommendations generate ranking and screening effects: they steer demand toward high value-for-money products, intensifying price competition between firms, which drives out lower-quality firms. Hence, more informative recommendations benefit consumers but reduce per-transaction revenue. A monopolist platform still prefers informative recommendations due to demand expansion. Competing platforms choose even more informative recommendations than a monopolist. Moreover, if entrant platforms cannot design informative systems, incumbents may strategically reduce informativeness. This illustrates how market power can explain 'ensh*ttification'.

Expectations Management with Philipp Brunner