Microeconomic Theory

Ambiguity in Games: Communication with Multiple Priors

Francesco Conti (University of Texas at Austin)

I extend the notions of Correlated and Bayes-Nash Equilibria using the von Neumann-Morgenstern approach for ambiguity. Following the approach to mechanism design in Myerson (1991), I define two versions of incentive compatibility under ambiguity to study general principal-agent problems with ambiguous communication. A version of the Revelation Principle for ambiguous communication is provided.

Persuasion under Subjective Causal Models

Marta Mojoli (University of Zurich)

How can a sender persuade a receiver when information is interpreted through a subjective causal model? I study a Bayesian persuasion problem in which the receiver updates beliefs via a directed acyclic graph, known to the sender, that encodes her perceived causal relationships among state attributes and the sender's signal. Because the receiver applies Bayes rule to a subjective distribution consistent with her causal model rather than to the objective data-generating process, her posteriors may diverge from standard Bayesian beliefs even when prior and experiment are common knowledge. The structure of this wedge depends on whether the signal is downstream or upstream in the receiver's causal model. When the signal is downstream, or when it is upstream but satisfies a mild consistency requirement that prevents prior distortion, any persuasive outcome attainable through an arbitrary experiment can be replicated by one consistent with the receiver's causal model. This No Confusion Principle implies that optimal persuasion reduces to a constrained Bayesian persuasion problem: subjective causal models do not expand, but instead limit the sender's persuasive power relative to the standard benchmark. When the consistency requirement fails, the sender can exploit the receiver's causal model to distort her effective prior, potentially achieving outcomes unattainable under standard Bayesian persuasion. I characterize the resulting feasibility constraints, identify the structure of optimal experiments, and provide conditions under which causal models are irrelevant.

One Truth, Many Datasets: How Context Shapes Model Extraction

Andrea Salvanti (Universitat Pompeu Fabra)

We develop a formal framework for analyzing how decision-making is informed by the extraction of statistical relationships from data. In our model, distinct datasets are generated by a common causal relationship that maps a relevant subset of predictors into a distribution

over outcomes. A decision maker observes a dataset and seeks to extract a decision rule that maps a relevant subset of variables into actions, trading off the rule’s value against its complexity cost. Leveraging the invariance of the underlying causal rule across datasets, we derive novel testable predictions that decouple the rules' value from their costs in predicting both the ex-ante probability of extracting the invariant rule, the substitution with other rules, and the emergence of disagreement. We test these predictions and find supporting evidence through our own experiment. We also demonstrate how our framework can be applied to study the cost argument using data from Kendall and Oprea (2024).

Corruption in Auctions: a Foundation for the Second-Price and Dutch Auctions

Alex Tordjman (Stanford University)

An auctioneer is running a private auction on behalf of a principal, and has to publicly reveal the outcome. The auctioneer seeks to  extract a rent from the auction by engaging some coalition of bidders in corruption. Given an auction, an auctioneer's plan is k - corrupt if for all type profiles, the auctioneer can find a coalition of size k such that his plan i) always weakly improves the sum of the utilities of the coalition members compared to the non-cooperative outcome, ii) that inequality holds strictly for a positive measure of the type space and iii) admits an innocent explanation for bidders that do not belong to the coalition. The corruption-robustness index of an auction is the minimum k such that a k - corrupt plan is feasible. The second-price sealed-bid auction with optimal reserve and its weak dominance equilibrium is maximally corruption-robust within the class of static, symmetric optimal auctions with an index of two. The Dutch auction (with optimal reserve) has an index equal to the number of bidders. 

Risk Aversion in Information Cascades

Elizaveta Zelnitskaia (CERGE-EI)

How do risk preferences influence the speed and character of social learning? I consider this question with theoretical modeling and a laboratory experiment, extending classic information cascade settings (Banerjee, 1992; Bikhchandani et al., 1992) by assuming that agents differ in risk preferences and choose between asymmetrically risky options. Theory predictions show that heterogeneity of agents’ risk preferences may slow down convergence speed and increase the frequency of converging to incorrect choices. The experiment shows that preference heterogeneity per se neither slows cascades nor affects the likelihood of convergence to incorrect choices. However, when individuals know the exact risk preference types of others – rather than just their population frequencies – the likelihood of converging to incorrect choices increases significantly. This suggests that detailed information on heterogeneous preferences can undermine social learning, likely because its perceived complexity prompts reliance on behavioral heuristics and introduces imprecision in information processing – a result not previously documented. These insights are relevant for insurance, financial, and business contexts, where heterogeneous risk preferences may reduce the accuracy of short-run social learning – and, surprisingly, knowing others’ preferences can amplify, rather than mitigate, this effect.

Urban Economics 

No Single Market for Electricity: Aggregate Productivity Costs of Price Dispersion

Paula Patzelt (London School of Economics)

Electricity prices differ by several multiples across European countries, reflecting both natural advantages in resource endowments and policy choices about generation mix, market design, and cross-border transmission. I ask how much aggregate productivity is lost because this dispersion shifts production away from its efficient spatial allocation. A structural model of firm location choices across segmented electricity markets characterises when price dispersion is efficient versus distortionary, and introduces an electricity-congestion channel depending on the marginal generation source. I estimate production activity elasticities that govern the aggregate TFP loss using a quasi-natural experiment from Norway, where uneven exposure to continental European prices drove persistent within-country price differentials, and corroborate them in a European cross-country panel. Firm activity responds to electricity prices at the intensive margin with effects scaling in electricity intensity, while extensive margin responses are small, pointing to substantial mobility frictions. The calibrated model decomposes the productivity cost of policy-induced dispersion into shares correctable by market integration and generation policy, with realistic transmission expansion as the achievable correction and announced national renewable trajectories as the forward-looking scenario.

Family-led Structural Change

Luca Looser (Universitat Pompeu Fabra)

Structural transformation does not unfold in a social vacuum. As workers leave agriculture for new sectors and places, access to those opportunities is mediated by families: some individuals inherit relatives who can open doors to jobs and destinations, while others do not. This paper studies how family ties shaped structural change in the United States during the late nineteenth and early twentieth centuries. I combine full-count census data with population-scale genealogical linkages to build extended kin networks over time and space, and I exploit deaths of sector-specific relatives as shocks to family-based access. These shocks alter occupational choices and propagate within families. Guided by this evidence, I develop and estimate a dynamic spatial overlapping-generations model in which dynastic family capital lowers migration costs to connected places and entry costs into connected sectors. Quantitatively, family ties slow early reallocation out of agricul-ture, later accelerate the shift into nonfarm work, and dampen adjustment to a major agricultural shocks

The Hidden Costs of Urbanization

Mateo Moglia (IP Paris-CREST)

What are the housing supply costs in dense areas? Levering a unique dataset, linking development costs, building permits, and precise land occupation, I provide precise estimates of the housing construction costs by project type. I embed those cost estimates in an option-value forward-looking model of land (re)development, calibrated on the Paris area. Thanks to model estimation, I am able to  precisely separate regulation from construction costs in the local housing supply elasticities. Counterfactual exercises highlight the importance of targeted subsidies and loans to address the housing supply crisis. 

Environmental Transition 

Wired For Change? Clean Technology Adoption and Labor Market Transitions

Guillaume Wald (Mines Paris-PSL)

I investigate the effect of clean technology adoption on labor market outcomes. I leverage a demand-side heat pump subsidy shock in France that triggered supply side adoption by heating firms, creating a natural experiment for studying worker adjustment. Using matched employer-employee data, I find establishment-level adoption increases both job creation and separations, indicating within-firm labor reallocation. Workers experience an average +10% rise in hours worked and +12% rise in earnings, challenging fears of severe adaptation costs. Decomposing by worker type, I find that stayers drive the results, with 20% higher hours and earnings. Both leavers and newcomers face initial losses; however, within one year, leavers are overcompensated and newcomers recover to baseline. Subsidy-driven technology adoption therefore results in low transition costs, avoiding mass displacement and directly updating workers’ skills.

The economic consequences of closing the flood insurance protection gap

Thomas Bézy (Paris School of Economics)

Climate change exacerbates the already substantial flood insurance protection gap, raising challenges for the insurance industry to maintain coverage in flood-prone areas. This paper provides an empirical analysis and welfare evaluation of a mandatory flood insurance pool that successfully closes the flood insurance protection gap while keeping premiums affordable for low-income households in floodplains. Focusing on France where flood insurance is mandatory and premiums are risk-independent, I estimate the efficiency costs of the policy through construction in floodplains using dwelling-level geolocated data. I employ a difference-in-differences strategy comparing risky and safe areas across time. I run the same analysis in Belgium, that also made insurance mandatory in 2007, and find consistent results in the two countries. It appears that efficiency costs are modest: without the policy, total flood damage costs would have decreased by less than 2% between 1981 and 2020 in France. To balance these costs against the distributional benefits of the policy, I evaluate the welfare effects of alternative insurance scenarios using a combination of an insurance demand model, a quantitative spatial model, and an optimal social insurance framework. I use these models to determine the optimal degree of risk-pricing under mandatory insurance. Finally, I compare the welfare implications of three policy scenarios: no government intervention, partial risk-sharing (similar to the US system), and mandatory insurance pooling.

Firm Dynamics 

Creation and Destruction in Networks: The Market Value of Technological Innovations

Abhijit Tagade (London School of Economics)

We integrate production and innovation networks into a unified framework to study the creative versus redistributive general equilibrium effects of technological innovations. We characterize a static profit multiplier that captures propagation through supply chains and product-market competition under endogenous markups, and a dynamic value multiplier that embeds knowledge spillovers and shows that the innovation network redistributes value sensitivity, with upstream firms creating externalities that compound over time. We derive a firm-level innovation wedge defined by the ratio of external to private value that serves as a sufficient statistic for the market-implied socially optimal R&D subsidy, and develop an event-study estimator that recovers it from stock returns.

Monetary Tightening and Divergence in Firm Innovation: Who Cuts back and Who Pushes Forward? 

Shuwen Wang (Columbia Business School)

I document that monetary policy exerts heterogeneous impacts on firm innovation efforts, and could contribute to market and R&D concentration. Controlling for financial constraints and various firm-level characteristics, a 100 basis point rate hike leads firms with 1% higher existing patent values (innovative firms) to reduce their R&D spending by 4% less than firms without patent values (non-innovative firms). This 4% gap slightly narrows but remains significant at 2% level after 12 quarters. Frontier firms—those generating the highest patent values within their sector—even increase their R&D spending by 2.2%, while their peers cut back R&D by 6% one quarter after the tightening shock, resulting in a clear divergence in R&D responses. Their sustained R&D activity is financed through additional equity issuance following the tightening. This divergence primarily stems from innovative firms’ high marginal value of R&D investment. Additionally, innovative firms are particularly sensitive to R&D cuts, as such reductions may signal them as being non-innovative. When monetary policy raises interest rates, less innovative firms tend to adopt existing technologies due to higher marginal costs, whereas innovative firms persist in developing new ideas. Such monetary tightening further exacerbates the divergence, potentially widening the innovation gap in the future.

Corporate Taxation and Firm Dynamics

Gemma Harris (Paris School of Economics)

We propose that lower corporate tax rates can contribute to falling business dynamism by favoring large incumbent firms over potential entrants. Using firm balance-sheet data from France, we first document significant heterogeneity in implicit corporate tax rates across the firm size distribution and show that the corporate tax becomes regressive at the very top. Within industries, a tax rate of 1percentage point (pp) lower on the largest firms by is associated with firm creation rates between 2 and 3 pp lower. We then use the 2017 corporate tax reform in France to estimate the dynamic effects of a corporate tax cut on firm entry at the industry level. Finally, we build a theoretical model of firm dynamics that accounts for heterogeneous responses across firms to changes in the statutory rate. While a corporate tax cut raises the expected after-tax profits of potential entrants, it also disproportionately benefits large incumbents, allowing them to increase their relative size and thus mark-up and market share. Average price increases as markups rise, leading to higher input costs and barriers to entry. When this indirect effect raising barriers to entry is strong enough, it outweighs the direct effect of higher expected after-tax profits for entrants and firm creation falls.

Demand-Driven Stagnation: Customer Acquisition and Persistent Firm Growth Slowdowns

Thomas Lazarowicz (UCL)

This paper shows that weak demand has played a critical and lasting role in the United Kingdom’s post-financial crisis slowdown. Exploiting local variation in the 2010 austerity program, I document that young firms, traditionally major contributors to job creation and productivity growth, saw their growth stall in regions hit harder by spending reductions.

Young firms cut marketing expenditure by 15% and saw output growth slow, while employment and other production inputs were unaffected. To explain why demand shocks disproportionately scar young firms, I develop a dynamic general equilibrium model where firms accumulate customers with heterogeneous income. Two features are key. First, non-homothetic preferences generate higher income elasticities for low-income households. Second, low-income customers churn faster, making them over-represented in the acquisition pool. Together, these forces create a “churn trap”: young firms inherit low income-biased customer bases and face disproportionately volatile demand. When household income falls, the expected value of acquiring new customers drops sharply, causing young firms to cut marketing - their primary investment in customer acquisition. The calibrated model replicates key empirical patterns and attributes an upper-bound estimate of approximately one-quarter of the UK’s post-crisis productivity shortfall to these demand-side forces.

Public Policy

Motivated or Frustrated? Aspirations and Optimal Taxation

Thomas Lloyd (University of Michigan)

Standard optimal tax formulas treat the equity-efficiency tradeoff as independent of how taxation reshapes social comparisons and the social incentives to provide effort. I show this omits an important mechanism: when aspirations are socially determined, tax-induced changes to the income distribution alter individuals' reference points, generating behavioral responses beyond the classic labor-leisure tradeoff. Embedding preferences with endogenous aspirational thresholds into a Mirrlees model, I derive sufficient statistics formulas for optimal linear and nonlinear income taxes. These formulas require three new objects: (i) the aspiration formation function, which specifies whose incomes shape reference points; (ii) the disutility from higher aspirations; and (iii) the behavioral response of taxable income to changes in aspirations. The sign of this behavioral response across the income distribution determines whether higher aspirations motivate (or frustrate) labor supply, which in turn calls for lower (or higher) optimal tax rates. The aspiration formation function determines who contributes to the aspiration externality, whether the corrective motive is uniform or concentrated at the top of the income distribution, and the structure of a feedback-driven aspiration multiplier that amplifies or dampens the aggregate elasticity of taxable income. Using a randomized information provision experiment combined with hypothetical choice scenarios, I estimate the behavioral response among US respondents and find an income-weighted average of 0.26, with aspirations predominantly motivating effort across the income distribution. Consistent with this evidence, standard formulas can overstate optimal progressivity.

The Impact of Unemployment Insurance on Job Search

Pierre Rousseaux (IP Paris-CREST)

This paper studies how shortening unemployment insurance duration affects jobseekers’ search behavior over the unemployment spell. I exploit a 2023 French reform that reduced potential benefit duration by 25 percent for new claimants, and leverage the universe of online applications submitted on the French public employment service job board, matched with vacancy characteristics. Using a within-spell difference-in-differences strategy that compares pre- and post-reform cohorts around February 1, 2023 with the same difference around February 1, 2022, I show that jobseekers submit a similar total number of applications over a shorter benefit period, with search effort peaking around the new exhaustion date. Search selectivity, accounting for duration dependence, shows no detectable change in targeted wages, contract types, hours worked, or skill match, although jobseekers broaden their occupational search earlier than pre-reform claimants. A directed search model rationalizes pre-reform behavior but fails to predict post-reform dynamics, suggesting that the observed response is difficult to reconcile with a standard rational job search model. I then extend the model to account for these dynamics.

The Supply-Side Effects and Incidence of Investment Stimulus Policies

Louis de Lachapelle (IP Paris-CREST)

Who benefits from investment subsidies: the firms that invest, or the capital-goods producers that supply them? We study the supply-side effects and incidence of the French suramortissement (2015–2017), a temporary extra-depreciation allowance that mechanically reduced the user cost of eligible industrial equipment by about 13%. We first confirm that beneficiary firms significantly increase their stock of subsidized capital in response to the policy. Turning to the supply side, we show that domestic capital-goods producers respond to the resulting demand shock by raising prices by 3% on average, offsetting slightly less than one quarter of the mechanical cost reduction. This also implies that only about three quarters of the value-based response reflects real capital accumulation. Their domestic sales increase by 11%, which, after accounting for the endogenous price response, implies an elasticity of capital flows with respect to the user cost slightly lower than unity. Their exports are unaffected, confirming that their expansion is driven by the domestic demand shock due to the subsidy. We embed these findings in a supply-chain incidence framework with imperfect competition and show that at least 33% of the surplus going to firms is captured by capital-goods producers. Their employment grows by 4% on average and dividend payouts rise substantially, while wages remain flat, suggesting that the upstream gains accrue primarily to shareholders rather than to workers.

International Trade

The Effect of Intangible Capital on Offshoring

Jiyoung Lee (University of Washington & Bank of Korea)

This paper analyzes the impact of intangible capital on firms' offshoring decisions, aggregate productivity, and external competitiveness. I develop a two-country offshoring model with endogenous intangible investment that captures its unique scalability in a framework featuring heterogeneous firm entry and exit. The model successfully replicates the key post-GFC U.S. empirical regularities—the coexistence of aggregate productivity gains and real exchange rate appreciation. Following a positive productivity shock, the entry of domestic firms lowers average firm productivity and raises domestic prices, thereby amplifying the Harrod–Balassa–Samuelson effect. Furthermore, incorporating intangibles yields substantially larger aggregate productivity gains than non-intangible models by improving resource allocation. This research contributes by endogenously modeling the scalability of intangible capital—a crucial yet previously understudied factor in offshoring models.

Do Economic Sanctions Stop Wars? An Empirical Analysis

Ramy Shaban (Université Paris-Dauphine-PSL)

This paper studies whether economic sanctions reduce interstate military conflicts. I build a country–year panel covering the period 1950–2014 and measure sanction intensity as the share of a target country’s imports, exports, and international financial exposure subject to sanctions. Because sanctions are imposed precisely when targets are politically isolated or militarily aggressive, I instrument sanction exposure with gravity-predicted exposure to potential sanctioning countries’ leave-one-out propensity to impose sanctions.  The outcome is the number of countries with which the target is involved in an initiated interstate military conflict reaching the use of military force. The estimates reveal a sharp difference across economic linkages: sanctions on imports reduce initiated conflict involvement, whereas sanctions on exports and financial sanctions do not have a statistically significant effect. In the preferred 2SLS specification, the coefficient implies that sanctioning about 30.1% of a target’s imports is associated, under the linear approximation, with one fewer active initiated military conflict.  The results suggest that sanctions are most effective when they constrain access to foreign inputs, rather than when they primarily reduce export revenue.

Asymmetric Welfare Costs of Semiconductor Export Restrictions

Wei Guo (Purdue University)

This paper examines the welfare consequences of U.S. export controls on advanced semi-conductors to China. Using monthly product-level trade data, I estimate a triple-difference specification around the October 2022 export control expansion, comparing advanced and legacy chip categories across countries and time. The estimates show a sharp decline in U.S. exports of advanced logic chips to China, while mature-node semiconductor products are much less affected. I then discipline a three-country quantitative general equilibrium model in which advanced semiconductors are essential Leontief inputs for China’s high-technology production. The model implies strongly asymmetric welfare effects: China bears large welfare losses because advanced chips have limited short-run substitutes, while US gains through trade diversion and capital accumulation. The welfare decomposition shows that bilateral trade losses alone substantially understate the total cost, because semiconductor input shortages propagate through downstream production networks. The paper highlights how export controls on technologically differentiated intermediate inputs can generate welfare effects far larger than those suggested by direct trade exposure alone.