Nearly Right

When AI decides your pound is worth less than your neighbour's

How surveillance algorithms quietly rigged the market against consumers whilst regulators looked away

Within days of becoming Federal Trade Commission chair, Andrew Ferguson quietly withdrew his predecessor's investigation into surveillance pricing. The January 2025 decision ended the brief period when this systematic repricing of consumer markets faced genuine regulatory scrutiny. Now, as companies accelerate deployment of algorithmic pricing infrastructure across thousands of retail locations, the American public remains largely unaware that a fundamental transformation of market mechanics is occurring beneath their daily shopping routines.

The surveillance pricing system operates through carefully constructed invisibility. When Delta Airlines faces accusations of using AI to "jack up" prices based on personal information, their response follows a familiar pattern: deny individual targeting whilst acknowledging the use of "dynamic pricing models streamlined by new technology." This linguistic precision obscures the reality that companies have built sophisticated infrastructure to charge different customers different prices for identical goods based on algorithmic assessment of their willingness to pay.

The architecture of algorithmic extraction

The FTC's now-abandoned investigation revealed the scope of this infrastructure. Eight companies—including financial giants Mastercard and JPMorgan Chase, consulting behemoths McKinsey and Accenture, and specialised pricing firms like PROS and Revionics—serve at least 250 clients across grocery, retail, travel, and financial services. These "pricing middlemen" process vast datasets including customer location, browsing history, credit information, and real-time behavioural patterns to generate individualised prices.

The scale is vast. Kroger's data subsidiary 84.51° leverages purchasing information from over 62 million US households. Task Software processes 300 million customer interactions daily across its platform ecosystem. Consultancy McKinsey promises clients "two to seven percentage points of sustained margin improvement" through surveillance pricing implementation, whilst industry analysts declare personalised pricing "a cornerstone of modern business strategy."

This represents a quantum leap beyond traditional dynamic pricing. Airlines have practiced yield management since the 1980s, with American Airlines' Robert Crandall generating $500 million annually by adjusting fares based on booking timing and route demand. But surveillance pricing targets individual consumers based on personal characteristics and behavioural analysis rather than market-wide supply and demand dynamics.

The exploitation becomes visible

Real cases demonstrate how this system functions in practice. Target charged customers $100 more for televisions when geofencing technology detected them in store parking lots, calculating that physical presence indicated higher willingness to pay. The company ultimately paid $5 million in civil penalties for these deceptive practices. The Princeton Review charged systematically higher prices to customers in zip codes with larger Asian populations. Uber reportedly analysed user phone battery levels, charging higher fares to customers with low battery who might be more desperate for rides.

These aren't isolated incidents but systematic patterns. Research shows broadband providers consistently charge higher prices in lower-income, non-white neighbourhoods. Studies of ridesharing reveal different pricing based on payment method selection and destination analysis. The pattern extends internationally: travel booking sites charged customers from affluent cities up to $500 more per night for identical hotel rooms.

Dr. Zephyr Teachout at Fordham Law School, who helped develop anti-price-gouging regulations, argues this undermines fundamental market principles: "Public pricing is foundational to economic liberty. When companies can charge each person a different price based on algorithmic assessment of their desperation, we lose the ability to make rational economic decisions."

The efficiency illusion

Proponents frame surveillance pricing as efficiency enhancement. University of Chicago economists Sanjog Misra and Jean-Pierre Dubé conducted field experiments showing that personalised pricing improved company revenue by 86% over uniform pricing whilst providing lower prices to 60% of customers. This research, conducted with employment website ZipRecruiter, demonstrates the system's capacity for both profit extraction and selective consumer benefits.

But this analysis obscures the fundamental mechanism: wealth transfer from consumers to companies through information asymmetry. As Lee Hepner of the American Economic Liberties Project explains: "The literature acknowledges that personalised pricing is a transfer of wealth from consumer to seller. The goal is to maximise revenue by exploiting what companies know about individual desperation and ability to pay."

The efficiency argument breaks down under systematic analysis. Unlike traditional price discrimination that might lower prices during off-peak demand, surveillance pricing requires comprehensive data collection infrastructure that adds costs rather than reducing them. Companies invest in facial recognition systems, electronic shelf labels, algorithmic processing, and data broker partnerships not to improve efficiency but to extract maximum value from existing transactions.

The infrastructure acceleration

Whilst regulatory attention waned, infrastructure deployment accelerated. Walmart plans electronic shelf labels in 2,300 stores by 2026; Kroger targets 2,600 stores with the same technology. These digital price tags enable real-time price adjustments based on customer characteristics, time of day, inventory levels, and competitive analysis.

The technology extends beyond simple price changes. Kroger deploys cameras at digital displays using facial recognition to determine customer gender and age for targeted offers. Companies like Plexure, which works with McDonald's and other chains, promise to increase order frequency by 30% and order size by 35% by analysing customer data including recent paychecks and purchase timing.

Even traditional grocers have transformed into media companies. Kroger's acquisition of data firm 84.51° (named for the coordinates of Kroger's Cincinnati headquarters) created partnerships with streaming services to deliver targeted advertising based on purchasing data. The proposed Kroger-Albertsons merger would combine datasets covering more than 100 million households, creating unprecedented consumer surveillance capabilities.

The democratic pricing problem

Beyond economic harm, surveillance pricing fundamentally alters the relationship between consumers and markets. Democratic societies rely on transparent price mechanisms that enable collective evaluation of value and comparative decision-making. When every transaction becomes an individualised negotiation mediated by algorithmic assessment, markets cease functioning as democratic institutions.

The isolation of consumers from one another's pricing experiences enables this system. Unlike physical stores where customers might compare receipt totals, online and app-based transactions occur in private digital spaces where price discrimination remains invisible. Companies exploit this isolation to implement systematic wealth extraction that would trigger immediate backlash if conducted transparently.

Former FTC Chair Lina Khan captured this systemic concern: "Firms that harvest Americans' personal data can put people's privacy at risk. Now firms could be exploiting this vast trove of personal information to charge people higher prices, fundamentally upending how consumers buy products and how companies compete."

The regulatory retreat

The brief window of regulatory scrutiny has closed. Ferguson's withdrawal of the surveillance pricing investigation signals that companies can proceed with infrastructure deployment without federal oversight. This occurs precisely as the system reaches critical mass—the point where surveillance pricing becomes embedded in routine market interactions rather than experimental edge cases.

The timing reveals sophisticated regulatory arbitrage. Companies built data collection capabilities during privacy law delays, implemented pricing infrastructure during pandemic distraction, and now consolidate gains during regulatory transition. The pattern suggests coordinated strategy rather than coincidental innovation timing.

Some state-level resistance emerges. New York's proposed Preventing Algorithmic Pricing Discrimination Act would restrict automated pricing based on personal characteristics. But federal preemption and interstate commerce complexities limit state authority over national pricing systems operated by technology platforms.

The systemic transformation

What surveillance pricing represents extends beyond individual company practices to systematic infrastructure capture. Traditional markets develop infrastructure to reduce transaction costs and improve price discovery—stock exchanges, commodity markets, standardised contracts. Surveillance pricing creates infrastructure to increase transaction costs for consumers whilst improving extraction efficiency for sellers.

This infrastructure inversion has broader implications. Electronic shelf labels, facial recognition systems, and algorithmic intermediaries function as value capture rather than value creation mechanisms. The system institutionalises information inequality whilst claiming technological progress.

Companies understand the reputational risks. Wendy's retreated from dynamic pricing announcements after customer backlash. Retailers systematically deny surge pricing intentions whilst installing the enabling infrastructure. The gap between capability and implementation suggests recognition of fundamental unfairness that public revelation would expose.

The surveillance pricing transformation succeeds through technological mediation that obscures accountability whilst concentrating benefits among sophisticated actors who understand the system's mechanics. As regulatory attention turns elsewhere and infrastructure deployment accelerates, this systematic repricing of consumer markets may become irreversibly embedded in daily economic life—fundamentally altering market democracy whilst remaining largely invisible to those it exploits.

#artificial intelligence