EURO 2024 Copenhagen
Abstract Submission

EURO-Online login

4263. Algorithmic pricing and collusion on a price comparison website

Contributed abstract in session MA-31: Recommender systems, stream Analytics.

Monday, 8:30-10:00
Room: 046 (building: 208)

Authors (first author is the speaker)

1. Charlie Lindgren
Economics, Dalarna University
2. Kenneth Carling
Dalarna University
3. Ross May
School of Information and Engineering, Dalarna University
4. Niklas Rudholm

Abstract

Empirical evidence identifies a large proportion of retailers using algorithmic pricing for making price decisions. This opens several questions surrounding tacit collusion and its detrimental effects on competition and consequently consumers in the form of supracompetitive prices. As an initial step towards answering whether or not tacit collusion is happening among pricing algorithms, one need to first identify which retailers are using such technology. We have defined two new statistical markers for identifying pricing algorithms in low-resolution data. One of the markers adapts counting the number of price changes by instead computing the average rate of price change per retailer over all its products both within and across countries. The other, novel, marker computes the persistence of a retailer in the marketplace. Using robust standardization of the two markers and applying a heuristic based on appropriate thresholds of these standardized markers, we have identified varying numbers of retailers within and between countries for the product categories employing algorithmic pricing. We have therefore created a new heuristic for identifying pricing algorithms in low-resolution data sets. Furthermore, this heuristic allows for further research in empirically identifying whether or not tacit collusion exists among algorithms and the effects such pricing has on the profitability of the retailers who employ such technology compared to those who do not.

Keywords

Status: accepted

Back to the list of papers