Competition and privacy in the digital economy are shaped by data intermediaries, firms that collect and sell consumer information. Our recent study published in the International Journal of Industrial Organization, “Collecting and Selling Consumer Information: Selling Mechanisms Matter”, uncovers a fundamental insight: the way these firms sell consumer data, i.e., the selling mechanism, directly influences both competition and privacy protection.

Data intermediaries primarily use two distinct mechanisms to sell consumer information: take-it-or-leave-it offers and auctions. Each mechanism generates significantly different economic and welfare outcomes that demand careful consideration. Take-it-or-leave-it offers enhance competition and benefit consumers but can lead to excessive data collection. On the contrary, auctions protect privacy at the cost of reduced market competition. This apparent trade-off can be resolved through coordinated regulations that balance market competitiveness and privacy protection.

Two Selling Mechanisms, Two Outcomes: Competition vs. Privacy

When intermediaries employ take-it-or-leave-it offers, they sell the similar datasets to competing firms. This possibility intensifies competition as companies leverage detailed consumer profiles to compete more aggressively on prices and quality. This results in higher consumer surplus, with more competitive prices for consumers. However, this mechanism also creates strong incentives for intermediaries to collect ever-growing volumes of consumer data. More granular data increases the ability of firms to extract surplus from consumers, leading to over-collection of personal information and raising privacy concerns.

Auctions represent a fundamentally different approach. In equilibrium, the intermediary optimally sells data to a single firm. This reduces incentives to collect large amounts of data and therefore leads to fewer data collected and stronger privacy protection. Yet, this exclusive sale of data also weakens competition in the product market: with only one informed firm, the competitive pressure is reduced, leading to higher prices for consumers.

Hence, we highlight a new trade-off. Under take-it-or-leave-it offers, intermediaries collect substantially more consumer data, consumers benefit from lower prices, but their privacy is reduced as firms gain deeper insights into individual behavior. Auctions reverse this dynamic: by restricting data sales to one firm, auctions limit the incentives to collect data, thereby improving privacy, but softening competition.

The Benefits of Joint Regulation

The solution to this trade-off hinges on coordinated regulation. Two complementary regulatory tools show particular promise. Data minimization rules, implemented in the European GDPR, already limit how much consumer data firms can collect. By enforcing strict data minimization, especially for intermediaries using take-it-or-leave-it offers, regulators can curb excessive data collection while maintaining the competitive benefits of this selling mechanism. Consumers continue to enjoy lower prices and better choices with significantly reduced privacy risks.

Fair, Reasonable, and Non-Discriminatory data-access regimes offer another powerful regulatory solution. FRAND rules ensure that all firms have equal access to consumer data, preventing exclusive deals where a single company gains unfair advantages through exclusive data access. Under data-centered FRAND regulations, competition intensifies as more firms have access to data and compete on a level playing field. This sustains high consumer surplus by preventing intermediaries to over-collect data.

The Way Forward

Data intermediaries are becoming increasingly pivotal in a fast-growing digital economy. Their selling strategies shape not only how firms compete, but also how much personal data is collected. Effective policies must simultaneously protect privacy and preserve market competition — not as competing goals, but as a joint coordinated effort.

Collecting and Selling Consumer Information: Selling Mechanisms Matter: IJIO Volume 103 B, December 2025

La entrada IJIO Research Summary – Collecting and selling consumer information: Selling mechanisms matter se publicó primero en EARIE.

From 2008 to 2017, Apple’s App Store reset an app’s average rating to “No Ratings” whenever developers released a software update. This policy discouraged updating by imposing a reputational penalty on top of software development costs. When Apple unexpectedly removed this policy in September 2017, allowing developers to maintain their ratings across updates, three striking patterns emerged:

First, developers increased their update frequency by 35%—from one update every seven weeks to one every five-and-a-half weeks. This response was immediate and persistent. Second, the distortionary effects of the policy were strongest among the highest quality apps on the store. Apps in the top quintile of the rating distribution increased their updating after the policy change by 54% more than apps in the lowest quintile. Third, the policy affected update content and quality. Feature updates increased by 24%, while bug fix updates saw an even larger response. These changes improved app quality and performance: apps’ average ratings and demand—proxied by the number of new ratings received—increased.

The original rating reset created a reputational penalty that varied systematically across developers. For the most popular apps, ratings matter little for demand, so the penalty was minimal. But for more niche applications dependent on ratings to attract new users, the temporary reputation loss imposed a substantial cost to updating. The policy had an “inverted-U” shaped effect: while the reset policy affected all apps, the strongest distortions occurred among apps for whom the size of the penalty was neither too small nor too large.

The policy’s impact also depended on the rate at which apps could reaccumulate ratings after a reset. Apps in the slowest quintile for new ratings arriving, for whom a reset could have long-lasting effects, were far more responsive to the policy change than those that could quickly rebuild their reputation. For apps that could rapidly rebuild ratings, the estimated effect was statistically indistinguishable from zero.

Importantly, the reduction in updating under the reset policy appears to have reflected genuine lost innovation. While developers might have bundled the same level of improvements into fewer updates to minimize resets, the increases in both ratings and demand following the policy change indicate the original policy suppressed overall development effort.

This research provides empirical evidence that platform design decisions create first-order distortions in competitive behavior. The findings are particularly relevant as regulatory authorities worldwide scrutinize platform practices. While much attention focuses on platforms’ potential to self-preference or discriminate, this study highlights how ostensibly neutral design choices can significantly affect innovation incentives.

The results suggest several lessons. First, platforms should carefully consider how reputation systems and other design choices interact with sellers’ incentives to compete and improve their products. Second, in increasingly dynamic digital markets, policies intended to maintain rating accuracy or provide “fresh starts” may have unintended consequences that outweigh their benefits. Third, design choices often vary systematically across seller types, creating differential impacts that may not be immediately apparent.

Ultimately, in markets where platforms wield substantial power over information flows and competitive dynamics, platform design features can have significant consequences for innovation, quality, and consumer welfare. As platform markets grow in economic importance, understanding these mechanisms becomes increasingly critical for platform governance and competition policy.

Platform design and innovation incentives: Evidence from the product rating system on Apple’s App Store, IJIO Volume 99, March 2025

La entrada IJIO Research Summary – Platform Design and Innovation Incentives: Evidence from the Product Rating System on Apple’s App Store se publicó primero en EARIE.

In a recent article (here), Paolo Ramezzana takes a broader look at the issue by exploring how monopolization can propagate across complementary products and how integration between producers of such products might affect competitive outcomes. The article finds that monopolization of a product can give rise to “domino effects” by increasing the likelihood that the complementary products supplied by other firms will also be monopolized. It also finds that a merger between producers of complementary products can increase the profitability of monopolization, thus making it more likely that monopolization will occur, even when the integrated firm cannot leverage its monopoly power across products. This happens because the pricing efficiencies associated with integration make maintaining monopoly power in all products more profitable, resulting in what some commentators have called an “efficiency offense”.

These issues have become increasingly important in recent decades, particularly in fast-growing sectors that rely on systems of complementary components dominated by a few large firms. This is especially true for tech markets, where scale economies, network effects, and vertical restraints in licensing and distribution contracts create favorable conditions for monopolization. Examples that have attracted antitrust scrutiny include personal computers (e.g., Microsoft’s Windows OS and Intel’s microprocessors), mobile devices (e.g., Google’s Android OS, Qualcomm’s baseband chipsets, and NXP’s near-field communication chipsets), and aircraft manufacturing (e.g., GE’s jet engines and Honeywell’s avionics).

From an analytical standpoint, the article models a setting with two perfect complementary components, each produced by an incumbent monopolist facing an entry threat. Each monopolist can independently fend off such threat by signing up buyers to exclusive contracts.  While such a strategy has obvious benefits for an incumbent (i.e., the maintenance of monopoly profits), it also entails costs, as buyers must be compensated for giving up cheaper alternatives.

When an incumbent monopolizes a component, it lowers the costs for other incumbents of monopolizing complementary components by (a) shrinking the demand for those components, thus reducing the number of buyers that other incumbents need to sign up to exclusivity and (b) reducing the amount of compensation per unit of profit that other incumbents need to pay these buyers. As a result, monopolization of other components becomes more profitable and thus more likely to occur in equilibrium.

If the incumbents merge, they can price their complementary products more efficiently (because of the well-known Cournot effect), which enhances their profitability and strengthens their incentives to pursue costly monopolization.  For intermediate levels of entry costs, this can switch the equilibrium from one with entry to one with monopolization, ultimately harming consumers. However, for higher entry costs, entry does not occur regardless of whether the incumbents are merged. In this case, the merger benefits consumers by yielding lower prices.  Note that, contrary to much of the existing literature, the analysis does not assume that the integrated firm has unchallenged monopoly power in any component and, thus, does not rely on a leverage theory of competitive harm.

These findings have implications for competition policy. From an antitrust perspective, they suggest that early intervention – or a commitment to future intervention – in one or very few crucial components can yield competitive outcomes for an entire system. Regarding merger policy, they suggest that, under certain conditions, a merger between firms producing complementary products can reduce competition even if the merged firm does not have unchallenged monopoly power in any product and cannot thus leverage that power into other products.

Entry deterrence, domino effects and mergers in markets for complements, IJIO Volume 99, March 2025

La entrada IJIO Research Summary – Entry deterrence, domino effects and mergers in markets for complements se publicó primero en EARIE.

La entrada IJIO Special issue in Memory of Dr. Patrick Bajari se publicó primero en EARIE.

• Gaurab Aryal
• Nano Barahona
• Zach Brown
• Austin Drukker
• Luise Eisfeld
• Matteo Escude’
• Chiara Farronato
• Chiara Fumagalli
• Shota Ichihashi
• David Imhof
• Doh-Shin Jeon
• Paul Koh
• Nathan Larson
• Gaston Llanos
• Marleen Marra
• Yuta Toyama
• Wen Wang

for having been awarded the recognition of being a top reviewer of the IJIO for 2024.

Thanks to you all for this important service!

La entrada IJIO Top Reviewers for 2024 se publicó primero en EARIE.

Full details on Award winners here.

La entrada Young Economists’ Essay Award (YEEA) Announced se publicó primero en EARIE.

Best Theory paper (published 2024) goes to Florian Dendorfer

Best Empirical paper (published 2024) goes to Farasat A.S. Bokhari, Franco Mariuzzo, Weijie Yan

Full details on Award here.

La entrada The IJIO Best Paper Award se publicó primero en EARIE.

Collusive biding may take many forms and submitted bids could be designed to be consistent with competitive bidding. Those who engage in collusive bidding on public contracts are often innovative and those who survive are perhaps cleverer than those chasing them. This makes detecting bid rigging from bid data extremely difficult and previous studies estimate that the probability of a cartel being caught and convicted is only around 10 to 20% per year (Harrington and Wei, 2017; Combe et al., 2008; Bryant and Eckard, 1991).

Hence, cartels are problematic as they are hard to detect, reduce competition and increase prices. This has driven the development of statistical methods for detecting cartels to support further investigations. As the collection of complementary information (besides the separate bids) from procurements is costly, time-consuming and in some cases almost impossible, an ideal method would enable to test for collusive bidding behavior based on the bids alone.

Aapo Aaltio, Riku Buri,Antto Jokelainen and Johan Lundberg, in their recent paper, focus on cartel detection tests designed for auctions that require only bid data from the auctions. This contrasts with other methods in the literature that often rely on firms’ cost data, which is much harder to obtain. Since most public procurement databases report bids, methods based on bid distributions are well-suited for automated, large-scale cartel screening. The different tests analyzed in this paper are applied on the bidding behavior of two convicted cartels that operated in the Finnish and Swedish asphalt markets during the 1990s and early 2000s.

The paper has two objectives. First, by comparing bidding patterns before and after then cartel investigations, we estimate how the distribution of bids changed after the collapse of the cartel. We find that during the cartel, a large share of bids are within 10% of the winning bid. This clustering of bids is particularly prevalent in the Finnish market. We also observe that during the cartel period, winning bids are isolated, with losing bids typically being at least 1% higher than the winning bid. Together, the clustering of bids and isolated winning bids result in a bimodal distribution of bids during the cartel period. Both of these features, the clustered bids and the missing mass of nearly tied bids, have been proposed as markers of collusion in previous literature (Chassang et al., 2022; Clark et al., forthcoming; Imhof et al., 2018). After the start of cartel investigations, both in Finland and Sweden, the distribution of bids becomes unimodal and the share of bids within 10% of the winning bid decreases. The change is considerably larger in Finland. To support the causal interpretation of our results, we conduct several robustness checks and also conduct a difference-in-differences analysis using data from a control market.

Second, we test the performance of two cartel detection methods, which can be implemented using only information on the distribution of bids. The first method is a distributional regression approach suggested by Clark et al. (2025). This method is based on the observation that while a cartel might find it optimal to leave a gap between the winning bid and the second lowest bid, it does not have similar incentives to manipulate the difference between the other non-winning bids. The method works by comparing two sets of bid differences, where the bid difference is defined as the difference between a bid and the lowest rival bid. The first set of bid differences is calculated from a sample that includes all the bids, whereas the second set is calculated from a sample where the winning bid is excluded. The null hypothesis is that, with competitive bidding, the two distributions should be similar for bid differences close to zero. In both Finland and Sweden, the null hypothesis is rejected for the cartel period. In both cases, consistent with the intuition of the test, we find that during the cartel period, the full set of bid differences has a much lower density close to zero, indicating that the cartel firms avoided leaving bids very close to the winning bid. The null hypothesis is not rejected for the post-investigation period for either of the countries. However, the results for the post-investigation period in Sweden are not as conclusive as they are for Finland.

The second method, developed by Huber and Imhof (2019), uses machine learning to classify tenders as competitive or collusive. As predictors, the machine learning model uses different statistical screens calculated from the distribution of bids. These include, for example, the standard deviation of the bids and the difference between the winner and the runner-up. When the predictive model is trained using the data from the same country, the model correctly classifies about 90% of the tenders with the Finnish dataset and 74% with the Swedish dataset. When the model is trained using data from another country, the prediction rates decrease for both Finland and Sweden. We also find substantial variation in the prediction rates depending on the model specification and the data used to train the model.

To summarize, our study highlights two key findings. First, the Finnish and Swedish cartels caused significant distortions in bid distributions, reinforcing the idea that such patterns can indicate collusion. Second, the effectiveness of detection methods depends on the bidding behavior of the cartel. The Finnish cartel exhibited stronger collusive patterns, making it easier to detect, whereas the Swedish cartel sometimes resembled a competitive market, reducing detection rates. Together, these results underscore that the effectiveness of statistical methods to detect collusive bidding depends on the specific bidding behavior of the cartel.

The full paper is published in IJIO Volume 98, January 2025.

La entrada IJIO Research Summary – Complementary bidding and cartel detection: Evidence from Nordic asphalt markets se publicó primero en EARIE.

The EARIE – CEPR 3rd joint VIOS workshop will be held on Wednesday, April 23 at 15:00 CET and given by Ryan Kellogg (University of Chicago) on his paper The End of Oil.

Registration is open to all interested in IO. Register here

La entrada EARIE-CEPR 3rd VIOS Seminar se publicó primero en EARIE.

However, the idea that a platform would foreclose third-party rivals is somewhat puzzling. As Etro (2024) notes, it ‘conflicts with the decision of hosting third party sellers to collect commission revenues’ in the first place, ‘especially for a platform [such as Amazon] that […] collects most of its revenues from third parties’ (p.1517).

Florian Dendorfer demonstrates in a recent article that a platform commits to refraining from self-preferencing, provided this commitment is credible to consumers. Generally, the platform profits from selling a first-party product as this reduces double marginalization. By the same token, first-party selling enhances market efficiency because it leads to lower prices for consumers. This is true regardless of whether the platform engages in self-preferencing. If the platform allows the first-party product to compete directly with third-party offerings, rather than shielding it from competition, third-party seller profit margins are squeezed due to increased competitive pressure, and double marginalization is further reduced. Not only do consumers benefit from this but the platform earns more revenue from commission fees. Thus, the platform will not engage in self-preferencing.

This result critically hinges on the platform’s ability to commit to refrain from self-preferencing. At the point of sale, the platform has a strong incentive to unfairly promote the first-party product to maximize its retail profit margin. It can attract additional demand to the market only if it convincingly demonstrates beforehand—when consumers sign up—that it will treat third-party products and its own product equally. This finding suggests that in practice a platform is more likely to engage in self-preferencing if it lacks a reputation for fairness and transparency of search results or if the functioning of its search algorithm is opaque or is not disclosed to the public.

The article’s findings are largely robust to various extensions. Among other things, they hold regardless of whether the platform introduces an entirely new variety or replicates an existing product. Qualitatively, it makes no difference whether the platform’s self-preferencing ability is unrestricted or constrained.

Ultimately, the article contributes to our understanding of self-preferential behavior of digital platforms and challenges the common assumption that these platforms always have an incentive to engage in this practice. It underscores a platform’s motivation to lower consumer prices when selling a first-party product and highlights the critical role of transparency in preventing self-preferencing.

The full paper is published in IJIO Volume 97, December 2024.

References

Chen, Nan and Hsin-Tien Tsai, ‘Steering via algorithmic recommendations’, The RAND Journal of Economics, 2023.

Etro, Federico, ‘e-Commerce platforms and self-preferencing’, Journal of Economic Surveys, 2024, 38 (4), 1516–1543.

La entrada IJIO Research Summary – First-party selling and self-preferencing se publicó primero en EARIE.