This website uses cookies.

By clicking the "Accept" button or continuing to browse our site, you agree to first-party and session-only cookies being stored on your device to enhance site navigation and analyze site performance and traffic. For more information on our use of cookies, please see our Privacy Policy.

  1. Home
  2. ASSA Annual Meeting
  3. 2026
  4. Program
  5. E-Commerce
« Back to Results

E-Commerce

Paper Session

Sunday, Jan. 4, 2026 10:15 AM - 12:15 PM (EST)

Philadelphia Convention Center, 203-B
Hosted By: American Economic Association
  • Chair: Leon Musolff, University of Pennsylvania

Algorithmic Risk Aversion and Recommendation-Mediated Demand

Andreas Alexander Haupt
,
Stanford University
Aroon Narayanan
,
Massachusetts Institute of Technology

Abstract

Recommendation algorithms shape market outcomes by determining which offers are made salient to consumers. These systems learn from repeated interactions using online learning algorithms. This paper identifies an algorithmic property with important implications for market behavior. We show that learning algorithms are sensitive not only to the expected quality of observable signals of match quality, but also to the noise in those signals. Focusing on a widely used sequential learning mechanism that shares key features with modern recommender systems---$\varepsilon$-Greedy---we demonstrate the emergence of risk aversion at the algorithmic level. Specifically, the algorithm systematically favors actions that generate less volatile feedback. When faced with alternatives that have identical expected returns, and under a broad set of conditions, $\varepsilon$-Greedy selects the lower-variance action with probability approaching one. This emergent risk aversion arises from asymmetric exploration in the learning process. Finally, we show that introducing optimism or applying a particular reweighting of past observations restores risk-neutral behavior.

Ranking Algorithms and Equilibrium Prices

Yufeng Huang
,
University of Rochester
Malika Korganbekova
,
University of Chicago
Aliya Korganbekova
,
Harvard University

Abstract

We examine the equilibrium implications of ranking algorithms in e-commerce platforms. A more favorable ranking may increase price due to higher and less-elastic demand, and it may decrease price due to forward-looking sellers' incentive to maintain such a ranking using low prices. Collaborating with a major U.S. e-commerce platform, we utilize an experiment that "boosts" new products' rankings and show that a more favorable ranking leads to lower prices on average. By directly observing components of the ranking algorithm, we demonstrate that the lowering of prices occurs predominantly among products with higher algorithmic uncertainty, which are capable of improving future ranking using lower prices, and among high-quality products, which benefit long-term from the ranking improvements. We estimate a structural model with limited consumer consideration sets to examine alternative platform algorithm designs. We find that the current ranking algorithm--a common ranking algorithm used by Google and other e-commerce and search platforms--puts a downward pressure on stationary equilibrium prices, leading to significant consumer welfare gains.

Unmasking the Deception: The Interplay between Fake Reviews, Ratings Discrepancy, and Consumer Demand

Yunhao Huang
,
University of California-Berkeley
J. Miguel Villas-Boas
,
University of California-Berkeley
Mingduo Zhao
,
University of California-Berkeley

Abstract

In online marketplaces, consumers rely on reviews to make informed purchase decisions, making the presence of fake reviews detrimental. Previous literature implies that products with fake reviews can display some patterns in review distribution, such as a higher discrepancy in ratings. Consumers might take this pattern into account when making their purchase decisions. In this paper, we explore the interplay between fake reviews and ratings discrepancy, and their impact on consumer demand, while controlling for average product ratings. First, using a data set with fake review labels, we find that product ratings discrepancy is positively correlated with the probability that the product has fake reviews. Second, through an identification strategy exploiting ratings discrepancy changes due to rating distribution rounding, we find evidence consistent with a negative causal impact of ratings discrepancy on consumer demand. Then, we conduct two experiments to establish and quantify the mechanism of the impact of ratings discrepancy on consumer demand through consumer suspicion of fake reviews. The first experiment shows that higher ratings discrepancy increases consumer suspicion of fake reviews, and the second experiment shows that heightened suspicion reduces consumer willingness to pay (WTP). Together, these findings reveal that consumers use ratings discrepancy as a signal of fake reviews, and this suspicion significantly impacts their purchase decisions. The findings highlight the importance of understanding the relationship between fake reviews, rating discrepancies, and consumer demand in online marketplaces.

Oligopoly Competition in Fake Reviews

Yuta Yasui
,
Kochi University of Technology
Hisayuki Yoshimoto
,
University of Glasgow

Abstract

How do fake reviews alter oligopoly market outcomes? We model fake reviews in strategic quantity and price competitions. Each firm has private information about its product quality, which consumers need to infer before purchasing. Consumers observe a noisy review rating for each firm, which is the combination of authentic and fake reviews, thus the subject of strategic manipulation. A firm’s fake review writing action, though costly, could inflate the rating of its product, increase consumers’ willingness to pay, and upwardly shift the firm’s demand function. By focusing on a linear strategy with private information, we establish a monotone equilibrium fake review strategy. When consumers rationally conjecture firms’ costly fake review generation, expected prices and quantities are invariant to fake reviews, and severe competition diminishes the amount of fake reviews. Furthermore, expected consumer surplus increases with fake reviews due to their signaling role. By contrast, when some portions of consumers naively believe observed ratings are genuine, strategic substitutability emerges among firms’ fake review actions, and equilibrium oligopoly market outcomes are distorted by fake reviews. With naive consumers, severe competition can increase the amount of fake reviews and damage consumer surplus.

Preference Measurement Error, Concentration in Recommendation Systems,and Persuasion

Andreas Alexander Haupt
,
Stanford University

Abstract

Algorithmic recommendation based on noisy preference measurement is prevalent in recommendation systems. This paper discusses the consequences of such recommendation on market concentration and inequality. Binary types denoting a statistical majority and minority are noisily revealed through a statistical experiment. The achievable utilities and recommendation shares for the two groups can be analyzed as a Bayesian Persuasion problem. While under arbitrary noise structures, effects on concentration compared to a full-information market are undetermined, if noise is symmetric, concentration increases and consumer welfare becomes more unequally distributed. We define symmetric statistical experiments and analyze persuasion under a restriction to such experiments, which may be of independent interest.

Tacit Collusion by Pricing Algorithm with Rule-Based Rivals

Yongzhi Xu
,
Texas A&M University

Abstract

Pricing algorithms, particularly reinforcement learning algorithms, have been increasingly used by firms in competitive markets, helping them capture more information about the market and their rivals. While prior work has shown that reinforcement learning algorithms can lead to supracompetitive prices in the absence of communication between firms, existing studies largely assume simultaneous adoption by competing firms. Within a framework of price competition between two firms both initially using rule-based strategies, we provide theoretical and simulation evidence that the prices of both firms weakly increase when one firm adopts an algorithm. We also find that the firm using a rule can ``free ride" and benefit more from the other firm's adoption. Our findings contribute to the literature by highlighting the importance of the order of algorithm adoption and the transition from rule-based strategies to learning-based algorithms, and demonstrate how tacit collusion can occur in a broader set of circumstances.
JEL Classifications
  • L8 - Industry Studies: Services