Fake reviews in digital markets: analysis and solutions

The research group (FakeInDigitalMarkets) is a Thyssen Foundation-funded research project on fake valuations in digital markets. Researchers from TU Clausthal, Helmut Schmidt University Hamburg (HSU) and the University of Cologne have joined forces to investigate two central aspects using empirical-experimental methods. On the one hand, the state of knowledge on the scope of fake ratings is to be expanded, i.e. how is fake rated, when is fake rated and to what extent is fake rated. On the other hand, it is about the development and verification of measures to counteract and reduce fake reviews.

Background, relevance and problem

In the last decade, rating systems have established themselves in digital markets and on online platforms. Every day, millions of reviews of products, services, institutions or actors are published on Amazon, ebay, Google, Facebook or Airbnb. Today, users of online services can rate almost anything and anyone. Rating systems are also playing an increasingly important role in classic "offline" markets. For example, employers can be rated on glassdoor.com, restaurants on yelp.com and doctors on arzt- auskunft.de. Based on the information provided by rating systems, countless decisions are made every day. Many decision-makers rely on the ratings provided. Rating systems are believed to be able to solve or at least reduce the problem of asymmetric information in digital markets.

Recent reports as well as academic work suggest that manipulated valuations are widespread and likely to increase in scope (including Mayzlin et al. 2014, Luca and Zervas 2016, Bundeskartellamt 2020 ). Public reporting has also prominently addressed the issue (e.g., Süddeutsche Zeitung 08.06.2022, Frankfurter Allgemeine Sonntagszeitung 14.10.2022 or New York Times 18.6.2021). Manipulated reviews, also referred to in common parlance as "fake reviews," are reviews that are posted illegally on platforms either by the user or via third parties, usually without any transactions behind them. The aim of such reviews is either to improve the average rating of one's own products and services or to worsen the reputation of competitors. Manipulating reviews or shopping for good reviews is fraud, potentially leads to inefficiency, and thus poses an essential challenge to the functioning of digital markets. The problem also appears to have intensified, during the Covid 19 pandemic, as much activity has shifted to online markets and digital information seeking has increased overall due to reduced face-to-face interaction.

A growing literature shows that rating systems in digital markets and online platforms are able to reduce risks for customers, increase cooperation, and enable efficiency gains (including Tadelis 2016, Greiff and Paetzel 2020). In addition to the problem of manipulated ratings, extensive research has already been conducted on "selection bias," ratings written in particular by users who have had a particularly good or particularly bad experience (Dellarocas and Wood 2008, Hu et al. 2009, Bolton et al. 2019, among others) and reciprocal rating in sequential ratings (Bolton et al. 2013 and Fradkin et al. 2021, among others).

Research Agenda

The scientific project "Fake valuations in digital markets: Analysis and Solutions (FakeInDigitalMarkets)" uses laboratory and online experiments to systematically investigate who uses fake reviews, why, and to what extent, to what extent fake reviews hinder the functioning of review systems, and how to succeed in reducing the volume of fake reviews on online portals. A deeper insight into the functioning here includes the analysis of the extent to which rating behavior, expectation formation, and decisions depend on fake opportunities and the rating system itself. In a second step, solutions for reducing manipulated ratings will be explored. The goal is to identify solutions for protecting the functioning of rating systems and reducing potential inefficiencies created by fake ratings in digital markets.

Research questions

Building on the previously mentioned established experiments on rating systems (Greiff and Paetzel 2015, 2016, 2020, Bolton et al. 2019, Dorner et al. 2020), as well as the current preliminary study Krügel and Paetzel (2021), fake ratings are explored with regard to the functioning of the rating system. The following guiding questions exemplify the focus of the research:

1. Which users use fake ratings in which situation?
2. How exactly is the functioning of rating systems influenced by fake ratings?
3. How do fake ratings affect the overall reliability of ratings and rating systems?
4. Do fake ratings lead to inefficiencies in online markets, in that users no longer trust the ratings and thus transactions that would have occurred with real ratings fail to occur?
5. What mechanisms work best to reduce the extent of manipulated reviews and the harm caused by fake reviews?
6. How can the display of reviews be optimized so that incentives to create fake reviews can be reduced?

Bibliography

Bolton, G., B. Greiner, and A. Ockenfels (2013): 'Engineering Trust: Reciprocity in the Production of Reputation Information', Management Science 59(2), 265-285.

Bolton, G.E., D.J. Kusterer, and J. Mans (2019): 'Inflated reputations: Uncertainty, leniency, and moral wiggle room in trader feedback systems', Management Science 65(11), 5371-5391.

Dellarocas, C. and C.A. Wood (2008): 'The Sound of Silence in Online Feedback: Estimating Trading Risks in the Presence of Reporting Bias', Management Science 54(3), 460-476.

Fradkin, A., E. Grewal, and D. Holtz (2021): 'Reciprocity and Unveiling in Two-Sided Reputation Systems: Evidence from an Experiment on Airbnb', Marketing Science 40(6), 1009-1216.

Greiff, M. and F. Paetzel (2020): 'Information about average evaluations spurs cooperation: an experiment on noisy reputation systems', Journal of Economic Behavior & Organization 180, 334-356.

Hu, N., J. Zhang, and P.A. Pavlou (2009): 'Overcoming the J-shaped distribution of product reviews', Communications of the ACM 52(10), 144-147.

Krügel, J. P. and Paetzel, F. (2022): 'The Impact of Fraud on Reputation Systems', Available at SSRN: https://ssrn.com/abstract=3710843

Luca, M. and G. Zervas (2016): 'Fake it till you make it: reputation, competition, and yelp review fraud', Management Science. 62(12), 3412-3427.

Mayzlin, D., Y. Dover, and J. Chevalier (2014): 'Promotional reviews: An empirical investigation of online review manipulation', American Economic Review 104(8), 2421-2455.

Tadelis, S. (2016): 'Reputation and Feedback Systems in Online Platform Markets', Annual Review of Economics 8(1), 321-340.