Study reveals limits of ride-hailing pricing

Algorithmic pricing has helped scale ride-hailing across cities worldwide, but new research suggests its limits are becoming more visible at the margins of urban transport networks, particularly during periods of stress.

That is the message from new analysis by Oxford Economics, conducted in collaboration with inDrive, which examines how in-app fare negotiation affects efficiency, affordability and access in on-demand mobility markets.

Speaking to Cities Today, Anubhav Mohanty, Director at Oxford Economics, said the key insight for European and North American cities is that pricing flexibility still matters, even in higher-income, well-regulated environments.

“European and North American markets also experience thin or stressed conditions, such as late-night travel, outer suburbs, bad weather, and periods of driver shortage,” said Mohanty. “A single posted price in such contexts can prevent trips that riders and drivers would otherwise be willing to take.”

The study focuses on emerging markets, drawing on survey data from riders and drivers across seven countries. However, Oxford Economics argues that the underlying dynamics apply more broadly. Algorithmic pricing systems are typically designed to optimise for average market conditions, which can leave gaps when trips fall outside the norm.

According to Mohanty, these inefficiencies tend to surface in dense cities where demand is generally strong, but highly variable at specific times or locations.

“In dense and regulated cities like London or New York, algorithmic pricing is highly effective on average, but it may struggle with local and situational variation,” he said. “These are not failures of the algorithm per se, but the limits of centralised optimisation in environments with high micro-level heterogeneity.”

The research identifies longer trips to less well-served areas, short but inconvenient journeys, unsocial hours and routes perceived as slower or riskier as common scenarios where centrally set prices can fail to clear the market. In these cases, trips may not take place even when both riders and drivers would accept a slightly different fare.

Oxford Economics examined in-app fare negotiation as one way to address this challenge. In this model, algorithms continue to provide matching at scale and suggest a baseline price, but riders and drivers can adjust fares to reflect individual constraints and preferences.

From a city perspective, Mohanty said this flexibility has implications for affordability and first-and-last-mile access, particularly where public transport coverage is weaker.

“When riders and drivers can agree on prices that reflect their individual needs and constraints, some marginal trips can happen organically at more affordable fares, rather than being enabled by platform-funded discounts or subsidies,” he said. “Fare negotiation thus complements, rather than replaces, the algorithmic pricing system.”

Findings from the study show that fare negotiation was widely used in the markets analysed and was associated with higher completed trip volumes, improved access in harder-to-reach locations and reduced reliance on discounts. The report frames this as a shift towards human–algorithm collaboration, rather than a rollback of automation.

The research suggests hybrid pricing models may offer a way to improve responsiveness at the edges of the transport network while preserving the efficiency benefits of algorithmic systems.

Oxford Economics concludes that as cities face growing fare pressure, driver shortages and coverage gaps, particularly outside peak periods, limited pricing flexibility could play a role in strengthening on-demand mobility without increasing subsidy dependence.

Image: Sarayuth Punnasuriyaporn | Dreamstime.com