|
on Industrial Organization |
| By: | Hangcheng Zhao; Ron Berman |
| Abstract: | Online sellers have been adopting AI learning algorithms to automatically make product pricing and advertising decisions on e-commerce platforms. When sellers compete using such algorithms, one concern is that of tacit collusion - the algorithms learn to coordinate on higher than competitive. We empirically investigate whether these concerns are valid when sellers make pricing and advertising decisions together, i.e., two-dimensional decisions. Our empirical strategy is to analyze competition with multi-agent reinforcement learning, which we calibrate to a large-scale dataset collected from Amazon.com products. Our first contribution is to find conditions under which learning algorithms can facilitate win-win-win outcomes that are beneficial for consumers, sellers, and even the platform, when consumers have high search costs. In these cases the algorithms learn to coordinate on prices that are lower than competitive prices. The intuition is that the algorithms learn to coordinate on lower advertising bids, which lower advertising costs, leading to lower prices. Our second contribution is an analysis of a large-scale, high-frequency keyword-product dataset for more than 2 million products on Amazon.com. Our estimates of consumer search costs show a wide range of costs for different product keywords. We generate an algorithm usage and find a negative interaction between the estimated consumer search costs and the algorithm usage index, providing empirical evidence of beneficial collusion. Finally, we analyze the platform's strategic response. We find that reserve price adjustments will not increase profits for the platform, but commission adjustments will. Our analyses help alleviate some worries about the potentially harmful effects of competing learning algorithms, and can help sellers, platforms and policymakers to decide on whether to adopt or regulate such algorithms. |
| Date: | 2025–08 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2508.08325 |
| By: | Cesare Carissimo; Fryderyk Falniowski; Siavash Rahimi; Heinrich Nax |
| Abstract: | This paper proposes a fresh `meta-game' perspective on the problem of algorithmic collusion in pricing games a la Bertrand. Economists have interpreted the fact that algorithms can learn to price collusively as tacit collusion. We argue instead that the co-parametrization of algorithms -- that we show is necessary to obtain algorithmic collusion -- requires algorithm designer(s) to engage in explicit collusion by algorithm orchestration. To highlight this, we model a meta-game of algorithm parametrization that is played by algorithm designers, and the relevant strategic analyses at that level reveal new equilibrium and collusion phenomena. |
| Date: | 2025–08 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2508.14766 |
| By: | Askitas, Nikos (IZA) |
| Abstract: | Generative AI (GenAI) and Large Language Models (LLMs) are moving into domains once seen as uniquely human: reasoning, synthesis, abstraction, and rhetoric. Addressed to labor economists and informed readers, this paper clarifies what is truly new about LLMs, what is not, and why it matters. Using an analogy to auto-regressive models from economics, we explain their stochastic nature, whose fluency is often mistaken for agency. We place LLMs in the longer history of human–machine outsourcing, from digestion to cognition, and examine disruptive effects on white-collar labor, institutions, and epistemic norms. Risks emerge when synthetic content becomes both product and input, creating feedback loops that erode originality and reliability. Grounding the discussion in conceptual clarity over hype, we argue that while GenAI may substitute for some labor, statistical limits will, probably but not without major disruption, preserve a key role for human judgment. The question is not only how these tools are used, but which tasks we relinquish and how we reallocate expertise in a new division of cognitive labor. |
| Keywords: | automation and outsourcing, technological change, labor economics, autoregressive models, Large Language Models, Generative Artificial Intelligence, human-machine collaboration knowledge work, epistemic norms, digital transformation |
| JEL: | J24 O33 O31 J22 D83 L86 J44 O38 |
| Date: | 2025–08 |
| URL: | https://d.repec.org/n?u=RePEc:iza:izadps:dp18070 |
| By: | Soheil Ghili; K. Sudhir; Nitish Jain; Ankur Garg |
| Abstract: | We build on theoretical results from the mechanism design literature to analyze empirical models of second-degree price discrimination (2PD). We show that for a random-coefficients discrete choice ("BLP") model to be suitable for studying 2PD, it must capture the covariance between two key random effects: (i) the "baseline" willingness to pay (affecting all product versions), and (ii) the perceived differentiation between versions. We then develop an experimental design that, among other features, identifies this covariance under common data constraints in 2PD environments. We implement this experiment in the field in collaboration with an international airline. Estimating the theoretically motivated empirical model on the experimental data, we demonstrate its applicability to 2PD decisions. We also show that test statistics from our design can enable qualitative inference on optimal 2PD policy even before estimating a demand model. Our methodology applies broadly across second-degree price discrimination settings. |
| Date: | 2025–07 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2507.13426 |
| By: | Serguey Braguinsky; Joonkyu Choi; Yuheng Ding; Karam Jo; Seula Kim |
| Abstract: | We provide evidence that mega firms have played an increasingly important role in shaping new technological trajectories in recent years. While the share of novel patents---defined as patents introducing new combinations of technological components---produced by mega firms declined until around 2000, it has rebounded sharply since then. Furthermore, we find that the technological impact and knowledge diffusion of novel patents by mega firms have grown relative to those by non-mega firms after 2001. We also explore potential drivers of this trend, presenting evidence that the rise in novel patenting by mega firms is tied to their disproportionate increase in cash holdings and the expansion of their technological scope. Our findings highlight an overlooked positive role of mega firms in the economywide innovation process. |
| Keywords: | Mega Firms; Innovation; Novel Patents; Knowledge Diffusion |
| JEL: | O31 O33 O34 L11 L25 |
| Date: | 2025–08–06 |
| URL: | https://d.repec.org/n?u=RePEc:fip:fedgfe:2025-60 |
| By: | Randall Lewis; Florian Zettelmeyer; Brett R. Gordon; Cristobal Garib; Johannes Hermle; Mike Perry; Henrique Romero; German Schnaidt |
| Abstract: | Amazon's new Multi-Touch Attribution (MTA) solution allows advertisers to measure how each touchpoint across the marketing funnel contributes to a conversion. This gives advertisers a more comprehensive view of their Amazon Ads performance across objectives when multiple ads influence shopping decisions. Amazon MTA uses a combination of randomized controlled trials (RCTs) and machine learning (ML) models to allocate credit for Amazon conversions across Amazon Ads touchpoints in proportion to their value, i.e., their likely contribution to shopping decisions. ML models trained purely on observational data are easy to scale and can yield precise predictions, but the models might produce biased estimates of ad effects. RCTs yield unbiased ad effects but can be noisy. Our MTA methodology combines experiments, ML models, and Amazon's shopping signals in a thoughtful manner to inform attribution credit allocation. |
| Date: | 2025–08 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2508.08209 |
| By: | Jeremy Proz; Martin Huber |
| Abstract: | Collusion and capacity withholding in electricity wholesale markets are important mechanisms of market manipulation. This study applies a refined machine learning-based cartel detection algorithm to two cartel cases in the Italian electricity market and evaluates its out-of-sample performance. Specifically, we consider an ensemble machine learning method that uses statistical screens constructed from the offer price distribution as predictors for the incidence of collusion among electricity providers in specific regions. We propose novel screens related to the capacity-withholding behavior of electricity providers and find that including such screens derived from the day-ahead spot market as predictors can improve cartel detection. We find that, under complete cartels - where collusion in a tender presumably involves all suppliers - the method correctly classifies up to roughly 95% of tenders in our data as collusive or competitive, improving classification accuracy compared to using only previously available screens. However, when trained on larger datasets including non-cartel members and applying algorithms tailored to detect incomplete cartels, the previously existing screens are sufficient to achieve 98% accuracy, and the addition of our newly proposed capacity-withholding screens does not further improve performance. Overall, this study highlights the promising potential of supervised machine learning techniques for detecting and dismantling cartels in electricity markets. |
| Date: | 2025–08 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2508.09885 |