The impacts of personalised pricing
Introduction
Traditionally markets have operated under the assumption that identical prices will be offered to all consumers for identical products regardless of data collected on that consumer. However developments in artificial intelligence and increase in actionable data held by firms has allowed the start of prices tailored to consumers. Personalised pricing is the practice of offering individuals different prices based on perceived willingness to pay which can be gathered from consumer data such as purchase history , online behaviour and personal circumstances. Whilst it can offer benefits to consumers through greater market transparency and increased access for lower income households, the information asymmetry needed for successful implementation and the benefits to firms in terms of efficiency and reduced consumer surplus means that a widespread use of personalised pricing will benefit firms over consumers.
Firms' profits
A benefit personalised pricing offers to firms is the potential to increase their profits. It creates an incentive for firms to offer lower prices for lower income consumers enlarging their target audience whilst retaining the profitability of higher income consumers. Whilst the lower prices certain groups will receive will lead to a decrease in profit per good, the increased supply of otherwise wasted goods will increase overall profit (Li and Derdenger, 2025). However this change will differ depending on the firm's access to information and the level of competition in the market.
If a firm has actionable data about its consumers (demographics, propensity to consume) it can engage in first degree price discrimination where each consumer is charged its maximum willingness to pay. As firms will have different access to information there will be large variation in the effect of the normalisation of personalised pricing on profits. A company lacking data will struggle to put forward the optimum price, if priced too high this will result in a loss of potential sales, and if too low a loss in potential profit. For monopolist firms which tend to have more granular consumer data, Acquisti and Varian (2005) found that personalized pricing increases profits. The dominance of these firms means there is a lack of alternatives so even with issues with fairness raised by consumers the introduction of personalised pricing will have a limited impact on the demand from higher end consumers allowing the maximisation of overall profits.
The impact on profits for a firm in a competitive market is less predictable. Even with accurate data and optimum prices if other firms begin offering the same prices the effect on profit is unclear. If all firms have access to the same information on consumers the introduction of personalised pricing can result in increased competitiveness as each firms attempts to undercut each other (which personalised pricing facilitates by increasing prices for certain consumers allowing significantly lower ones for others) or minimal change is pricing algorithms provide similar outputs leading to uniform prices across firms.
Uber
We can look at the effects of personalized pricing through a study into Uber. Uber can be considered to exist in a competitive market, whilst it is the dominant market player, the emergence of rivals such as Lyft and Bolt means we must consider its pricing decisions in the way we would a firm operating in a competitive market. Uber first introduced personalized pricing in 2023, before these prices were set based on distance and time. As seen in figure three, whilst the number of users opening the app dramatically increased the number of requests remained relatively stagnant, this is a result of consumers with a lower willingness to pay being deterred by the high prices prioritizing those who valued Uber more in that moment and were subsequently willing to pay a higher price. This is beneficial for both the consumer and the firm, allowing the firm to receive greater profit per unit and allowing users with greater need to access the service. The difference in the red and blue lines on the graph can be interpreted as a measure of this productive efficiency. Uber requires incentives for drivers, an increase in price when demand is high will encourage supply (due to the potential increase in earnings) leading to a reduction in the gap between supply and demand resulting in fewer shortages and reduced waiting time (Mercatus Center, 2025) thereby creating more favorable outcomes for consumer and driver as well as increasing firm profits.
Market transparency
Personalised pricing can result in an increase of market transparency. This is only the case if the personalised pricing is communicated to the consumer. The department for business energy and industrial strategy suggests there are three main forms of disclosure of personalised pricing. (a) informing consumers that the price is personalised, (b) explaining to consumers how the price is personalised, and (c) facilitating comparison between personalised prices amongst consumers (Acquisti and Varian, 2005). Personalised pricing requires profiling and automated decision making (setting a price based on data), this increases the transparency required by the firm. Information right act WP29 (applicable in the UK) states that data subjects need to be aware their data is being used for profiling and that decisions are being made based on the profile created (de Streel and Jacques, 2019). The subject must also be given access to all the data used to create the profile and related information including the category they are placed into (this category being used to determine the price they are given). By this nature this increases transparency by forcing the provider to share its data reducing information asymmetry.
However personalised pricing can make price comparison more difficult, even if a firm aims to provide this by pursuing policy C to achieve comparison they would require cooperation from all providers in the market. Most firms tend to avoid option C as it exposes price discrimination and allows competitors to undercut them more easily so the likelihood of it being achieved in a given market is low. Firms can also avoid disclosing personalised pricing, for example offering discounts by emails to certain groups, in this case the personalised pricing is not disclosed to the general consumer and the market becomes more opaque.
Consumer trust
A loss in consumer trust can be expected with an increase in the commonality of personalised pricing. Personalised pricing results in winners and losers which is seen as socially undesirable, and can result in feelings of exploitation. If consumers lack confidence that they are receiving a good or fair price they may reduce their demand or cease participation in the market leading to a long term reduction in demand causing a drop in profits and even firm exits (Richards et al., 2016).However the extent to which this occurs will partially depend on the customers' perceptions of price fairness (Department for Business, Energy & Industrial Strategy, 2018), as the “losers” in personalised pricing are often wealthier consumers; it can be seen as a way to increase equity in the market so the loss in trust will be limited. Personalised pricing brings with it concerns about consumer privacy which can damage trust. Personalised pricing requires detailed data including demographic and purchasing and behavioral data which can be seen as invasive. It can cause people to interact less with the firm online such as via its website in an attempt to limit ability to track their data.
Welfare
The impact of personalized pricing on consumer welfare is unclear,and will be dependent on the level of price discrimination. If there is strong first degree price discrimination then consumer surplus will be reduced to 0 representing a loss in consumer welfare. However this level of discrimination is purely theoretical, even with data tracking a consumer's purchases over a long period of time, researchers observe a substantial amount of randomness in consumers choices ( Rossi, McCulloch, and Allenby, 1996) making it impossible to predict the consumers willingness to pay for each marginal unit to a suitable degree of accuracy to achieve a solely producer surplus. This effect will still be seen with third-degree price discrimination, whilst not eradicating consumers surplus it will result in a lower level. Consumer welfare increases for those with a low willingness or ability to pay, personalized pricing allows firms to charge them lower prices boosting welfare by broadening access to a product (market expansion). This effect means for partial personalized pricing which still allows for consumer surplus personalized pricing can increase consumer welfare. However a theoretical perfect price personalisation would result in consumers being charged their maximum willing price decreasing consumer welfare.
Collusion
Personalised pricing uses algorithms which can lead to increased collusion between firms. Algorithmic pricing allows firms to use the same algorithm reducing the need for continuous communication between colluding parties. This makes the collusion (which is illegal) harder to detect and more efficient. However collusion only poses a benefit to the firm if they have imperfect information on their consumer base. With imperfect information, collusion is beneficial for firms as it allows them to reduce uncertainty and the risk of losing customers to undercutting and Bertrand competition. For collusion to be facilitated firms need to have information on close and far consumers in the market to allow them to set a joint price which will optimize profits for both firms, this information is also needed to allow them to detect if the colluding firm is sticking to the agreed price or undercutting. If the firm has perfect information on its market then the benefits offered by collusion would be limited. The firms would know how their consumers will respond to their prices , their loyalty and likelihood of switching brands and can offer the optimal price. Therefore the price offered by a rival will be limited by its effect on the firm's sales. Collusion can increase due to personalised pricing but only when firms have imperfect information about their own consumer base but some knowledge of the bases of the other firms in the cartel.
In the case of uber, its rivals also use dynamic pricing models so the consumer is unlikely to find a suitable alternative with a major price difference and as uber retains significant brand loyalty dynamic pricing allows it to maximise its profits and efficiency. As personalised pricing becomes more widely available to firms we can expect the spread of this model into other areas, such as product delivery where shorter wait times are prioritised, this will lead to more options in shipping compared to the standard vs express that we experience currently.
Conclusion
In conclusion, the widespread implementation of personalized pricing would alter the ways in which consumers and firms interact. Whilst advances in personalised pricing may increase productive efficiency and allow the entrance of lower income or price-sensitive consumers into certain markets, this benefit will be realized by the firm through increased profit. By allowing higher levels of price discrimination, personalised pricing allows firms to capitalize on increased information advantage to reduce consumer surplus. Concerns around privacy and data exploitation increase feelings of inequity and distrust in consumers. Although regulation around transparency required in pricing for firms can help reduce concerns and increase market transparency for consumers, the current lack of guidelines around personalised pricing means that if we were to see an increase in prevalence we would not see the expected increase in transparency. As markets become more dependent on data and we see an increase in personalised pricing the extent to which the burden falls on consumers will depend on whether law changes sufficiently in response.
By Freya Brolly
Sources
https://assets.publishing.service.gov.uk/media/5bbb2384ed915d238f9cc2e7/Algorithms_econ_report.pdf
https://ueaeprints.uea.ac.uk/id/eprint/80677/1/Published_Version.pdf
https://www.econstor.eu/bitstream/10419/205221/1/de-Streel-Jacques.pdf
