Algorithms Are Setting Our Prices. We Don’t Have to Let Them Set the Rules.
Effective oversight of algorithmic pricing demands both deep technical expertise and nuanced regulatory frameworks.
By Angelina Freeman, Isabella Coronado Doria, Elijah Maubert, and Husein Pumaya Yakubu
Whether you’re a frustrated Uber ride-seeker or a Taylor Swift fan attempting to buy tickets, you’re familiar with algorithmic pricing — the use of automated decision-making in the setting of prices. With the rise of artificial intelligence, algorithmic pricing has become a powerful tool in modern commerce, as firms leverage AI to integrate information on consumers and market conditions. With potential benefits for both firms and consumers, algorithmic pricing also opens the door to new and subtle forms of anti-competitive behaviour that traditional laws and enforcement mechanisms are not fully prepared to handle. A non-legislative, innovative approach is required to address this issue without overstraining bureaucratic capacity or stifling innovation.
For many Canadians facing rising rent and grocery bills, the influence of algorithm-driven pricing is becoming impossible to ignore. The White House estimated that anti-competitive behaviour on the behalf of algorithm-using landlords cost American renters approximately $70 a month, for a total cost of at least $3.8B in 2023 alone. The Canadian Competition Bureau, too, is investigating the use of algorithms by landlords. Similar algorithms are being used to set prices on products across Canada. The costs to Canadians could be enormous if the anti-competitive implications are not properly addressed.
AI-enabled pricing algorithms are trained on immense amounts of real-time market and consumer data to output prices more quickly and more precisely than human-based methods. Algorithmic pricing allows firms to better segment consumers and often customize prices, functioning as a more advanced way to price discriminate, a legal practice that allows firms to maximize profits. This can benefit consumers, such as those who receive tailored discounts of cheaper prices in off-peak times. Algorithmic pricing can also be used to enforce illegal price fixing schemes and cartels.
Most concerningly, algorithmic pricing can facilitate tacit collusion, a growing problem that is not even addressed under current competition law. This occurs when firms reach a collusive outcome without having an agreement, such as when multiple firms feed data into the same algorithm. Giant corporations can more easily afford the data, and algorithmic pricing risks entrenching the dominant position of incumbent firms.
Throughout their decision-making processes, pricing algorithms rely on immense amounts of consumer data, raising myriad privacy concerns. The increased importance of data also brings competition considerations: the 2022 amendments to the Competition Act expanded the list of factors to determine an impact on competition to include network effects and non-price competition, such as consumer privacy.
The very features that make algorithmic pricing efficient — speed, complexity, and autonomy — also make it harder to detect and regulate when misused. While Canada’s Competition Bureau has recognized the risks of algorithmic pricing, it remains in a study-first mode to avoid undermining the benefits competition of the technology. Under current competition law, tacit collusion is not explicitly illegal, hampering enforcement efforts. Regardless, monitoring algorithmic pricing and identifying potential anti-competitive behaviour presents an immense challenge due to the “black box” nature of algorithmic pricing, desire of firms to maintain ownership over their code and other IP that generates their competitive edge, and the high level of technical expertise needed to audit algorithms.
Addressing these complexities requires careful consideration. Effective oversight of algorithmic pricing demands both deep technical expertise and nuanced regulatory frameworks capable of balancing innovation, competition, and consumer protection. While public entities and non-profits grapple with these issues, much of the necessary expertise resides in the private sector, creating a dynamic interplay between industry, regulators, and consumer advocates as they seek effective governance solutions.
Given the rapid evolution and complexity of AI technologies, particularly in pricing systems, traditional legislative routes may struggle to keep pace with emerging challenges. In the context of Canada’s competition ecosystem, a more effective approach could lie in fostering collaborative frameworks that engage regulators, industry stakeholders, and civil society. Such adaptive, multi-stakeholder models may offer the agility and technical depth needed to respond to the unique risks posed by algorithmic pricing — without stifling innovation or overburdening regulatory institutions.
Canada is now at a critical juncture, facing significant implications for market fairness and consumer well-being. While finding effective solutions that balance competitive fairness and innovation is challenging, the urgency of addressing these concerns grows daily. These dynamics will define Canada's ability to foster competitive markets that protect consumers and sustain technological advancement in the era of AI-driven commerce.
Angelina Freeman, Isabella Coronado Doria, Elijah Maubert, and Husein Pumaya Yakubu are students at the Max Bell School of Public Policy. This article is informed by research conducted for their Policy Lab project at the Max Bell School of Public Policy. A full report detailing the methodology and findings will be published on the school’s website upon completion of the project.