|
on Microeconomics |
| By: | Eliaz, Kfir; Eilat, Ran |
| Abstract: | We incorporate negative reciprocity into strategic information transmission, examining how a receiver's response to perceived manipulation influences optimal information design. In one setting, greater signal inaccuracy increases the likelihood that the receiver disregards it; in another, inaccuracy shifts the receiver's preferences unfavorably for the sender. In both cases, the revelation principle fails, yet we characterize the set of posterior beliefs on which an optimal signal is supported. While full revelation may be optimal in one setting, it is never so in the other. Our findings connect information design with behavioral economics, highlighting the implications of design-dependent preferences. |
| Keywords: | Negative-reciprocity; Information design |
| JEL: | D82 D91 |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:cpr:ceprdp:20295 |
| By: | Hattori, Keisuke |
| Abstract: | This paper studies costly information sharing in teams where members may duplicate one another's work. Members first decide whether to share their intended task assignments and then choose productive effort. Sharing allows a teammate to direct effort toward nonoverlapping tasks, while the sharer benefits only through complementary team production. This creates strategic complementarity in sharing and can sustain both a low-sharing, low-effort equilibrium and a high-sharing, high-effort equilibrium. The incentive to initiate sharing is strongest at an intermediate level of duplication risk: when duplication is limited, sharing prevents little wasted work; when duplication is severe, a lone sharer retains too little effective output to gain much from the teammate's response. Yet the value of coordinated sharing rises with duplication risk. A leader who shares first can select the high-sharing equilibrium, while larger teams may require a critical mass of initial sharers. The main equilibrium structure also survives with continuous sharing. The analysis identifies when communication platforms must be supplemented by visible first movers or coordinated initial participation to prevent duplicate work. |
| Keywords: | information sharing, duplicate work, team production, coordination failure, leadership |
| JEL: | C72 D23 J24 M12 M54 |
| Date: | 2026 |
| URL: | https://d.repec.org/n?u=RePEc:zbw:esprep:342325 |
| By: | John Mori |
| Abstract: | A representative democracy is immune to gerrymandering if it satisfies Chambers' (2008) representative consistency. We examine preference aggregation and show that representative consistency is mutually inconsistent with three other normative desiderata -- efficiency, anonymity, and neutrality. We show this impossibility result both in a setting with ordinal preferences and a setting with expected utility preferences. |
| Date: | 2026–07 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2607.05660 |
| By: | Ahmadzadeh, Amirreza |
| Abstract: | I study a dynamic principal–agent relationship in which an agent must exert costly effort to learn a privately observed bi-nary state before taking an action. The principal wants to match the action with the state, while the agent is biased toward one action, generating both a moral hazard (effort choice) and an adverse selection (action choice after learning the state) problem. The principal disciplines the agent through verification (at a cost), reduced workload and termination. We show reduced workload is always a valuable instrument, even when the cost of verification is small and the loss from shirking is large. By promising a reduced workload in the future, the principal can lower verification costs across multiple periods. For high biases, verification and reduced workload are insufficient instruments, and the principal must rely on firing along the equilibrium path. The threat of future firing complements verification and saves verification costs over time. |
| Date: | 2026–07 |
| URL: | https://d.repec.org/n?u=RePEc:tse:wpaper:132028 |
| By: | Do, Jihwan; Miklos-Thal, Jeanine |
| Abstract: | The cloud services industry, which is currently dominated by a few large providers, has come under scrutiny from antitrust authorities worldwide. One concern is that ``egress fees"—charges for transferring data out of a provider’s cloud—could harm competition and welfare by discouraging multi-clouding, whereby a user combines services from several providers. Motivated by this policy concern, we analyze the effects of banning price discrimination against multi-stop shoppers in a market where multi-product firms sell complementary goods to buyers with elastic demands, and multi-stop shoppers impose higher service costs than one-stop shoppers. We find that if buyers are locked into a specific product combination, then a ban on price discrimination against multi-stop shoppers raises social welfare for a wide range of demand functions. If product choices are endogenous and buyers' product preferences are weak, however, then a ban on price discrimination tends to harm social welfare. |
| JEL: | D43 L13 L40 |
| Date: | 2025–04 |
| URL: | https://d.repec.org/n?u=RePEc:cpr:ceprdp:20101 |
| By: | Enache, Andreea; Rhodes, Andrew |
| Abstract: | We consider a setting in which a platform matches buyers and sellers, who then wish to transact with each other multiple times. The platform charges fees for hosting transactions, but also offers convenience benefits. We consider two scenarios. In one scenario, all transactions must occur on the platform; in the other scenario, buyers and sellers can disintermediate the platform after the first transaction, and do subsequent transactions offline. We find that the platform reacts to disintermediation by using a ``front-loaded'' pricing scheme, whereby it charges more for earlier transactions. We also show that sometimes the platform is better off when disintermediation is possible---because it can use disintermediation to screen users' private information about their convenience benefits. Buyers are not necessarily better off when they can disintermediate, due to the way in which the platform adjusts its fees. |
| Keywords: | Platforms |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:cpr:ceprdp:20298 |
| By: | Bergemann, Dirk; Bojko, Marek; Duetting, Paul; Paes Leme, Renato; Xu, Haifeng; Zuo, Song |
| Abstract: | We study mechanism design when agents have private preferences and private information about a common payoff-relevant state. We show that standard message-driven mechanisms cannot implement socially efficient allocations when agents have multidimensional types, even under favorable conditions. To overcome this limitation, we propose data-driven mechanisms that leverage additional post-allocation information, modeled as an estimator of the payoff-relevant state. Our data-driven mechanisms extend the classic Vickrey-Clarke-Groves class. We show that hey achieve exact implementation in posterior equilibrium when the state is either fully revealed or the utility is affine in an unbiased estimator. We also show that they achieve approximate implementation with a consistent estimator, converging to exact implementation as the estimator converges, and present bounds on the convergence rate. We demonstrate applications to digital advertising auctions and large language model (LLM)-based mechanisms, where user engagement naturally reveals relevant information. |
| Keywords: | Large Language Models |
| JEL: | D47 D82 D83 |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:cpr:ceprdp:20227 |
| By: | Andrew B. Choi; Christoph Schlom; Chengyang Zhu |
| Abstract: | We introduce a moral hazard model in which public information about a payoff-relevant state arrives over time, an agent decides when to make an irreversible investment, and a principal commits to a state-contingent policy to incentivize investment. To discourage the agent from waiting for more information, the principal's optimal policy provides certainty, reducing the degree to which the agent's payoff depends on the state. This is inefficient -- both players would be better off with less certainty. We study when the agent receives positive rent, and when moral hazard delays investment. Our results apply to environmental subsidies and R&D incentives. |
| Date: | 2026–06 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2606.28583 |
| By: | Florian Brandl |
| Abstract: | We characterize correlated equilibrium in finite normal-form games. Interpreting correlated strategies as action recommendations, we show that correlated equilibrium is the unique solution concept that never recommends a pure-strategy dominated action, treats payoff-equivalent actions interchangeably, and respects the sure-thing principle under uncertainty about payoffs and the correlation device. A parallel characterization identifies coarse correlated equilibrium among solution concepts that recommend dominant actions whenever they exist and treat payoff-equivalent actions as strongly interchangeable. |
| Date: | 2026–07 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2607.06282 |
| By: | Florian Brandl; Warut Suksompong; Nicholas Teh |
| Abstract: | We consider the fair allocation of indivisible goods with binary valuations. In this setting, the maximum Nash welfare rule, the leximin rule, and all additive welfarist rules with a strictly concave function coincide. We show that for any number of agents, this rule is the only rule that satisfies envy-freeness up to one good, strategyproofness, neutrality, minimal completeness, and invariance under disapproving unassigned goods (IDU). Moreover, we present an alternative characterization for two agents, where we replace IDU with non-redundancy and resource-monotonicity. In both characterizations, all axioms are necessary. |
| Date: | 2026–07 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2607.10064 |
| By: | Fernando Payró Chew; Norio Takeoka; Jianming Xia |
| Abstract: | This paper provides axiomatic foundations for a model in which ambiguity attitudes are endogenously determined through cognitive optimization. The decision maker evaluates acts using alternative non-additive aggregation rules and optimally trades off the benefits of less ambiguity-averse evaluation against cognitive cost. The resulting framework generalizes Choquet expected utility and accounts for preference reversals identified by Machina (2009). The model is characterized by a novel axiom, Comonotonic Convexity, which regulates the evaluation of mixtures by requiring the decision maker to avoid hedging whenever it yields no benefit. We interpret this axiom as reflecting aversion to unnecessary complexity in the evaluation of acts. |
| Keywords: | ambiguity attitude, Choquet expected utility, cognitive optimization, comonotonicity, Machina’s paradox |
| JEL: | D11 D81 |
| Date: | 2026–07 |
| URL: | https://d.repec.org/n?u=RePEc:bge:wpaper:1587 |
| By: | Schmutzler, Armin |
| Abstract: | This paper provides a simple unified discrete-choice framework for analyzing differentiated duopolies. This framework nests models of horizontal and vertical differentiation, including standard textbook models (Hotelling and Shaked-Sutton). Contrary to these models, it also applies to economic environments where horizontal differentiation coincides with positive correlation of product valuations across consumers, and environments where vertical differentiation coincides with negative correlation. The paper provides an equilibrium characterization that is applicable independently of the type of differentiation and the sign of the valuation correlation. |
| Keywords: | Duopoly; Differentiated products; Price competition |
| JEL: | D43 L13 |
| Date: | 2025–04 |
| URL: | https://d.repec.org/n?u=RePEc:cpr:ceprdp:20171 |
| By: | Davide Bosco; Mario Gilli |
| Abstract: | When information is complete, symmetric group-contest games `a la Tullock are known to display within-group-symmetric (WGS) pure-strategy equilibria. Far less is known about the possible existence and structure of their asymmetric counterparts—pure-strategy equilibria where, symmetry notwithstanding, members of the same group exert different effort levels. This note fills the gap by proving that within-group-asymmetric (WGA) pure-strategy equilibria indeed exist, but only for a specific subset of incentivisation schemes. All such equilibria display a peculiar ‘group-splitting’ structure: a subgroup of members free-ride exerting null effort, whereas the other members provide the same positive effort level. Because of this symmetry-in-asymmetry, the game displays the same aggregative structure identified in the context of WGS equilibria. Accordingly, an explicit characterisation is feasible also for their WGA counterparts. |
| Keywords: | (A)Symmetry, Conflict, Group contests, Sharing rules, Strategic complements and substitutes |
| JEL: | C72 D71 D74 |
| Date: | 2026–07 |
| URL: | https://d.repec.org/n?u=RePEc:mib:wpaper:579 |
| By: | Bauch, Gerrit (Center for Mathematical Economics, Bielefeld University) |
| Abstract: | I provide an axiomatization of expected utility in which topological continuity is replaced by a geometric axiom. The axiom requires a finite set of indifferent lotteries that span a hyperplane. In the case of three prizes, two indifferent lotteries suffice. The axiom is weaker than Solvability, as well as logically independent of Weak Continuity and the Archimedean axiom. |
| Keywords: | Expected utility representation, finite continuity axiom |
| Date: | 2026–07–27 |
| URL: | https://d.repec.org/n?u=RePEc:bie:wpaper:768 |
| By: | Antonio Cabrales; Wenhao Cheng |
| Abstract: | This paper studies how organizations should jointly design evaluation rules and assign workers when performance depends on both effort and non-discretionary advantage. Agents choose effort in positions linked by a competition network, while their effective advantage depends on own type and spillovers through a second network. The planner chooses both the assignment and the effort weight in evaluation. Equilibrium effort rises with a position's Katz-Bonacich centrality and falls with effective advantage. The optimal evaluation rule generally differs from true output. When effort is more important in production, the planner lowers the effort weight and uses negative assortative assignment to strengthen incentives. When advantage is more important, the planner raises the effort weight and uses positive assortative assignment to exploit spillovers. We also study a constraint requiring assignments to be pairwise stable, which creates an output loss depending on the intensity of competition. |
| Keywords: | relative performance evaluation, worker assignment, organizational design, incentives; contests, network games, peer effects, spillovers, assortative matching |
| JEL: | D23 D85 C72 J33 M52 |
| Date: | 2026 |
| URL: | https://d.repec.org/n?u=RePEc:ces:ceswps:_12816 |
| By: | Bervoets, Sebastian; Faure, Mathieu; Renou, Ludovic |
| Abstract: | Deviations from Bayesian updating are traditionally categorized as biases, errors, or fallacies, thus implying their inherent ``sub-optimality.'' We offer a more nuanced view. We demonstrate that, in learning problems with misspecified models, non-Bayesian updating can outperform Bayesian updating. |
| Keywords: | Learning |
| JEL: | C72 D83 |
| Date: | 2025–04 |
| URL: | https://d.repec.org/n?u=RePEc:cpr:ceprdp:20114 |
| By: | Sonin, Konstantin |
| Abstract: | Information manipulation is a powerful tool in the hands of any authoritarian leader. Dictators block independent media, censor news, pay influencers, and control citizens' social connections. In our model, citizens acquire information from censored sources or through social networks. Naturally, information manipulation has less impact when consuming news is costly and percolation in the network is low. Less intuitively, it might be optimal for the regime to target peripheral, rather than centrally connected citizens, and the propaganda's maximum impact is when percolation of information is close to zero (the society is atomized) or close to one, but not in-between. |
| Keywords: | Authoritarian regime |
| JEL: | P00 D85 L82 |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:cpr:ceprdp:20320 |
| By: | Bergemann, Dirk; Bonatti, Alessandro; Smolin, Alex |
| Abstract: | We develop an economic framework to analyze the optimal pricing and product design of Large Language Models (LLM). Our framework captures several key features of LLMs: variable operational costs of processing input and output tokens; the ability to customize models through fine-tuning; and high-dimensional user heterogeneity in terms of task requirements and error sensitivity. In our model, a monopolistic seller offers multiple versions of LLMs through a menu of products. The optimal pricing structure depends on whether token allocation across tasks is contractible and whether users face scale constraints. Users with similar aggregate value-scale characteristics choose similar levels of fine-tuning and token consumption. The optimal mechanism can be implemented through menus of two-part tariffs, with higher markups for more intensive users. Our results rationalize observed industry practices such as tiered pricing based on model customization and usage levels. |
| Keywords: | Large Language Models |
| JEL: | D47 D82 D83 |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:cpr:ceprdp:20226 |
| By: | Onuchic, Paula; Ramos, João |
| Abstract: | This paper introduces a model of group communication, in which a group of senders with conflicting interests collectively communicate with a receiver through the disclosure or non-disclosure of information about a relevant state. Collective disclosure decisions are reached via the aggregation of group members’ disclosure recommendations via a pre-determined deliberation procedure. In contrast with classic results from single-agent disclosure, (sequential) equilibria of the group disclosure game typically do not involve full disclosure. We investigate the relation between the group’s deliberation procedure and features of equilibrium communication. In particular, we characterize changes in the deliberation procedure that increase a group’s informativeness; and show that the receiver interprets group messages less favorably for group members who have relatively more power. |
| Keywords: | Disclosure; Groups; Organizations; Communication |
| JEL: | D7 D8 |
| Date: | 2025–03 |
| URL: | https://d.repec.org/n?u=RePEc:cpr:ceprdp:20057 |
| By: | Yuan Deng; Yilin Li; Wei Tang; Hanrui Zhang |
| Abstract: | We study the limits of third-degree price discrimination when the production cost is Bayesian and private to the seller, generalizing the seminal work of Bergemann, Brooks and Morris (2015). The rough setup is the following: A monopoly seller sets different prices for buyers in different "segments" of the market so as to maximize seller surplus. Different ways in which the aggregate market is decomposed into segments lead to different welfare outcomes, i.e., (seller surplus, buyer surplus) pairs. When the production cost is Bayesian, the region of achievable welfare outcomes can exhibit complex shapes beyond the clean characterization by Bergemann, Brooks and Morris for the case with a fixed cost. We show that with a Bayesian cost, this region coincides with a proper projection of a polytope defined by a polynomial number of linear constraints, the essential ones of which correspond to flow conservation in a "discounted" flow network. As a result, we give a polynomial-time algorithm that computes optimal market segmentations in terms of any linear combination of the seller surplus and the buyer surplus. En route, we establish the following structural property: Any market can be written as a convex combination of "extremal markets" in a way preserving the seller surplus and the buyer surplus. These extremal markets are piecewise equal-surplus with respect to different possible costs, generalizing a similar notion introduced by Bergemann, Brooks and Morris when the cost is fixed. |
| Date: | 2026–07 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2607.12615 |
| By: | Bergemann, Dirk; Bonatti, Alessandro; Wu, Nick |
| Abstract: | In digital advertising, auctions determine the allocation of sponsored search, sponsored product, or display advertisements. The bids in these auctions for attention are largely generated by auto-bidding algorithms that are driven by platform-provided data. We analyze the equilibrium properties of a sequence of increasingly sophisticated auto-bidding algorithms. First, we consider the equilibrium bidding behavior of an individual advertiser who controls the auto-bidding algorithm through the choice of their budget. Second, we examine the interaction when all bidders use budget-controlled bidding algorithms. Finally, we derive the bidding algorithm that maximizes the platform revenue while ensuring that all advertisers continue to participate. |
| Keywords: | Data; Advertising; Competition; Auctions |
| JEL: | D44 D82 D83 |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:cpr:ceprdp:20263 |
| By: | Charles Angelucci; Roi Orzach |
| Abstract: | We analyze collaborative experimentation across multiple independent domains. Each domain contains infinitely many potential projects with asymmetric benefits. In each period and in each domain, two players can idle, jointly explore a new project, or jointly exploit a known one, with voluntary transfers. For intermediate discount factors, treating domains as independent during experimentation is suboptimal. The optimal experimentation policy exhibits common features of collaborative experimentation: lengthy exploration, temporary project exploitation, recall of past projects, and inefficient initial or terminal idling within certain domains. We connect these findings to research on buyer-supplier dynamics and persistent productivity differences. |
| JEL: | D21 D70 D83 L25 |
| Date: | 2026–07 |
| URL: | https://d.repec.org/n?u=RePEc:nbr:nberwo:35460 |
| By: | Furkan Sezer |
| Abstract: | We study information design when a designer with commitment shapes the information of strategically interacting, far-sighted agents whose actions drive a persistent, controlled Markov state. We introduce the Markov Bayes correlated equilibrium (Markov BCE), the controlled-Markov generalisation of the BCE of Bergemann and Morris (2016), characterised by a dynamic obedience condition that adds a continuation-value term to the static one and reduces to it when actions cannot move the state. Recommending actions is without loss; the designer's problem is recursive in the agents' promised continuation utilities and is solved by a set-valued backward-induction algorithm whose optimum exists and lies between the no-disclosure and first-best values. For linear-quadratic-Gaussian payoffs the obedience condition becomes a covariance condition with a modified interaction matrix, and the stationary case reduces to an algebraic Riccati equation. When agents instead learn the transition, we identify the rent an agent earns from a model of the dynamics sharper than the designer anticipates: it is non-negative, zero at the known-dynamics benchmark, and deterred only by building slack into obedience. Under persistent excitation the cumulative rent grows logarithmically as heterogeneous agents' estimates converge. Two worked examples, in congestion and resource coordination, together with a numerical study illustrate the theory. |
| Date: | 2026–07 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2607.04308 |
| By: | Wei Ma |
| Abstract: | We present a revealed preference characterization of the discounted expected utility model with a concave utility function. The characterization offers a nonparametric test of the model. We apply the test to an experimental data set in the literature and find that the model is almost always rejected even when all payments involved are subject to risk. |
| Date: | 2026–06 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2606.29779 |
| By: | Matthew Cashman |
| Abstract: | Widely held views can go unspoken when speaking out alone invites retaliation. I recast such silence as a problem of safe coalition formation. When safety comes in numbers, there is a largest group that could speak safely. Open organizing must stay safe every step the way, making for a cascade of speaking up that stops short of that largest group. Social assurance contracts need safety only at the destination, and get there by keeping commitments private until enough people have joined: a tunnel to the largest safe group. If no member may ever regret joining, such failure privacy is necessary, not merely sufficient. |
| Date: | 2026–07 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2607.05802 |
| By: | Le Yaouanq, Yves; Schwardmann, Peter; van der Weele, Joël |
| Abstract: | We present a self- and social-signaling model formalizing findings in political psychology that moral and political judgments stem primarily from intuition and emotion, while reasoning serves to rationalize these intuitions to maintain an image of impartiality. In social interactions, agents’ rationalizations are strategic complements: others’ rationalizations weaken their ability to judge critically and make their actions less revealing of (inconvenient) truths. When agents are naive about their own rationalizations, our model predicts ideological and affective polarization, with each side assigning inappropriate motives to the other. Cross-partisan exchanges of narratives reduce polarization but are avoided by the agents. In within-group exchanges agents favor skilled speakers, whose narratives worsen polarization. Our model explains partisan disagreements over policy consequences, aligns with empirical polarization trends, and offers insights into efforts to disrupt echo chambers. |
| JEL: | D72 D83 D91 P16 |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:cpr:ceprdp:20259 |
| By: | Chengcheng Wang; Zexin Ye |
| Abstract: | As firms increasingly adopt AI-powered pricing algorithms, a key and urgent policy concern is how to regulate the potential algorithmic collusion. This paper approaches the regulatory question through the lens of information design and examines how different disclosure rules, committed to by a third-party intermediary, shape learning outcomes when firms delegate pricing to Q-learning algorithms under stochastic demand. We analyze three disclosure rules: no disclosure, full disclosure, and upper censorship. Upper censorship, which truthfully reveals low-demand states while pooling high-demand ones, delivers higher profits than full disclosure, consistent with theoretical predictions. However, we uncover a profit reversal: when the discount factor is high, no disclosure yields higher profits than full disclosure, whereas when the discount factor is low, full disclosure performs better. This pattern is exactly the opposite of what classical collusion theory predicts. Overall, these findings show that Q-learning agents respond systematically to the information structure and further suggest that restricting information sharing may backfire when algorithms are sufficiently patient, highlighting the need to reassess regulatory approaches in AI-mediated markets. |
| Date: | 2026–07 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2607.04345 |
| By: | Eliaz, Kfir; Spiegler, Ran |
| Abstract: | A monopolist curates a database for users seeking to learn a parameter's value: "nowcasters" focus on its current value, while "forecasters" target its long-run value. The monopolist designs a menu of contracts described by fees and data-access levels, balancing revenue and data-storage costs. The optimal menu offers full access to historical data, while current data is fully provided to nowcasters but may be withheld from forecasters. Compared to the social optimum, the monopolist oversupplies historical data, undersupplies current data, and may provide excessive data overall. |
| Keywords: | Data markets |
| JEL: | D42 |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:cpr:ceprdp:20299 |