Publications

Normal-form two-player games are categorized by players into K analogy classes so as to minimize the prediction error about the behavior of the opponent. This results in Clustered Analogy-Based Expectation Equilibria in which strategies are analogy-based expectation equilibria given the analogy partitions and analogy partitions minimize the prediction errors given the strategies. We distinguish between environments with self-repelling analogy partitions in which some mixing over partitions is required and environments with self-attractive partitions in which several analogy partitions can arise, thereby suggesting new channels of belief heterogeneity and equilibrium multiplicity. Various economic applications are discussed.

Not all barrels of oil are created equal: their extraction varies in both private cost and carbon intensity. Leveraging a comprehensive micro-dataset on world oil fields, alongside detailed estimates of carbon intensities and private extraction costs, this study quantifies the additional emissions and costs from having extracted the “wrong” deposits. We do so by comparing historical deposit-level supplies to counterfactuals that factor in pollution costs, while keeping annual global consumption unchanged. Between 1992 and 2018, carbon misallocation amounted to at least 11.00 gigatons of CO2-equivalent (GtCO2eq), incurring an environmental cost evaluated at $2.2 trillion (US$ 2018). This translates into a significant supply-side ecological debt for major producers of high-carbon oil. Looking forward, we estimate the gains from making deposit-level extraction socially optimal at about 9.30 GtCO2eq, valued at $1.9 trillion, along a future aggregate demand pathway coherent with the objective of net-zero emissions in 2050, and document unequal reserve stranding across oil nations.

Skilled migration is seen as a solution against skill shortages, but its impact on native workers facing increased competition remains debated. To investigate the longer-term potential for displacement, this paper exploits a 2008 reform in France facilitating the hiring of foreign workers within a list of tight occupations. The analysis relies on administrative data and applies a difference-in-differences approach. Results show that the reform successfully boosted migrant hires without affecting native employment. In addition, the negative pressure on occupational wages is much larger among migrants than natives, suggesting that such policies can be a viable solution in the long run.

This paper offers the first detailed characterization of the labor market in a tax paradise and the first assessment of how a reform aimed at discouraging tax avoidance affected workers. We show that incumbents were highly educated, performed specialized tasks, and benefited from a wage gap, particularly at the top.

Immediately after the reform announcement, several workers exited. Stayers experienced wage increases of around 8% and a higher probability of working for several firms simultaneously. New hires earned, on average, 30% less than incumbents, frequently on temporary contracts. The results offer insights into policies promoting economic substance in low-tax jurisdictions.

This paper examines the role of wealth inequality as a predictor of financial crises, analyzing data from 18 countries between 1870 and 2020. Unlike previous research focused on the effects of financial crises on inequality, we explore whether increases in wealth concentration-specifically in the top 1%-elevate crisis risk. Our findings indicate that, even after accounting for key crisis predictors, a one standard deviation rise in the growth of the top 1% wealth share is associated with a 3 to 8 percentage points increase in crisis probability, with results robust across various crisis lists and empirical approaches. Temporal dynamics reveal that while a credit boom can jeopardize the financial system as early as the following year, it takes several years for an increase in private wealth accumulation and wealth concentration at the top to significantly heighten the risk of a systemic bank run, serving as early signals of potential instability. Furthermore, we find evidence that asset price bubbles can serve as transmission channels, although these relationships vary by asset class and « bubble » definition. Our findings suggest that addressing wealth concentration could reduce inequality while acting as a stabilizing force for financial systems, highlighting the importance of incorporating broader inequality metrics in crisis prediction models and exploring policy mechanisms to mitigate systemic risks.

This study investigates the link between Southwest U.S. border crossings and labor market tightness, measured by the job openings to unemployed ratio, over nearly 25 years (2000 to 2024). Analyzing monthly data, it finds a strong positive correlation, suggesting that increased border crossings strongly align with greater job availability. Exploiting data across different presidential administrations reveals no statistically significant differences in this relationship, regardless of the President’s party. The findings suggest a natural economic adjustment mechanism in which crossings naturally decrease as the labor market cools.

Terrorism globally yields severe consequences for individuals and societies, potentially driving migration within and across borders. Yet, empirical evidence on its causal impact remains limited. The contribution of our paper is twofold. First, we construct various indicators of terrorist activity at a fine level of spatial and temporal granularity, which allow to accurately identify individuals’ exposure to terrorist threat. Second, we use these geo-localised indicators to empirically analyse the role of terrorist attacks in shaping internal and international migration intentions for 133 countries between 2007 and 2015. Our results indicate that terrorist attacks spur both internal and international migration intentions, though the effect is stronger for the latter. The effect on international migration intentions is linked to the intensity of attacks, rather than their frequency. Furthermore, the impact varies based on individual and country characteristics.

European governments, struggling with incorporating diverse immigrant populations, introduced integration contracts. Through language training and compulsory civics courses, these contracts aim to induce new migrants to adopt the host society’s culture, respect its values, and improve their labor market outcomes. Despite their popularity, little empirical evidence exists on whether integration contracts catalyze integration or trigger a backlash. To shed light on this question, we leverage the staggered introduction of France’s integration contract across metropolitan departments between 2003 and 2006 to implement a regression discontinuity design. We use census data, labor force surveys, and our own survey of refugees to estimate the effect of the contract on integration outcomes. We find the integration contract facilitated employment in the short term without backlash but did not translate into long-lasting integration gains.

There is growing interest from both policy and academic communities in understanding why people do not migrate. This article offers the first global analysis of the aspiration to stay, defined here as the preference to stay in one’s country of residence. We make use of the unique Gallup World Polls which provide information on aspirations to stay (as opposed to migrating abroad) as well as on individual characteristics and opinions for 130 countries worldwide between 2010 and 2016. We find staying aspirations are far more common than migration aspirations across the globe and uncover important “retain factors” often overlooked in research on migration drivers — related to social ties, local amenities, trust in community institutions, and life satisfaction. Overall, those who aspire to stay tend to be more content, socially supported and live in communities with stronger institutions and better local amenities. We further explore differences in the relative importance of retain factors for countries at different levels of urbanization, and for different population groups, based on gender, education, rural/urban location, migration history, religiosity, and perceived thriving. Our findings contribute to a more holistic understanding of migration decision-making, illuminating the personal, social, economic, and institutional retain factors countering those that push and pull.

Testing for COVID-19 has been strongly recommended for individuals experiencing COVID-19-like symptoms or those with a close relative who tested positive. In France, tests were free of charge until mid-October 2021 and became widely available after June 2020. Our main objective was to investigate whether access to COVID-19 testing in France was associated with socio-economic conditions, considering gender and ethno-racial status. Methods A random population-based cohort survey was conducted in France in May 2020 and November 2020, including 95,388 participants aged 18 and over. We used logistic regressions to identify how having been tested in 2020 was associated with socio-economic status and exposure factors among two groups of individuals. The first group consisted of individuals who had no close relative test positive but reported experiencing Covid-19-like symptoms, such as cough, fever, dyspnea, or sudden onset of ageusia, dysgeusia, or anosmia (N=12,729). The second group included individuals, with or without symptoms, who reported that a close relative had tested positive for Covid-19 (N=5,360). Findings In both groups, testing was more frequent among individuals living in urban areas. For individuals who had no close relative test positive but reported Covid-19-like symptoms, women were 1.04 (95%CI [1.01-1.06]) more likely than men to be tested, and testing decreased with age up to 35. Individuals holding a university degree were 1.08 [1.04-1.12]) more likely than those who only completed high school to be tested, as well as those in one of the three top income deciles with reference to the bottom decile (OR 1.07, [1.02-1.13] for the top decile). Ethno-racial status was not significantly associated with testing. For respondents who had a close relative test positive, testing was strongly associated with having experienced some symptoms, especially after September 1 (OR 1.34, 95%CI [1.30-1.39]). However, there was no link between testing and income, education, or ethno-racial status. Interpretation When an individual experiences symptoms but has no close relative who is positive, the motivation to test to avoid transmitting the virus to relatives by isolating oneself is particularly strong. This strategy makes sense when there are actual possibilities to isolate, which may explain why higher income and education positively impact the propensity to test. For individuals who had a close relative test positive, most socio-economic variables were no longer related to testing once infection risk was controlled for. The availability of tests at no cost is not sufficient in itself to eliminate socio-economic inequalities in testing.

This article investigates the role immigrants and their native-born children play in shaping native attitudes toward immigrants in the European Union. By exploiting the 2017 Special Eurobarometer on immigrant integration, we show that countries with a relatively high share of immigrants are more likely to believe that immigrants are a burden on the welfare system and worsen crime. In contrast, native opinions on the impact of immigration on culture and the labor market are unrelated to the presence of immigrants. We also find that the effects of second-generation immigrants on pro-immigrant attitudes toward security and fiscal concerns are positive (as opposed to first-generation immigrants). Finally, we find no impact of the immigrant share on the attitudes of natives supporting far-left or left political parties, while it is the most negative among respondents affiliated with far-right parties.

This paper examines the estimation of the distribution of wealth using estates left at death. We establish formal conditions for implementing a simplified version of the classic estate multiplier method, relying solely on minimal information about estates and mortality. These conditions are empirically validated, and the simplified approach is applied to produce new long-run top wealth share series for Belgium, Japan, and South Africa, where estate data have previously been underutilized. This method holds potential for expanding the range of countries and years in which wealth concentration can be estimated, especially where estate data exist but the standard method with heterogeneous multipliers is inapplicable.

Since 1950, agricultural productivity has been increasing even as labourers leave agriculture. However, while average productivity of the sector has been converging, withinsector inequality has been increasing. Agricultural income inequality is still less than overall income inequality, but it measures significantly higher when we use higher-quality and more comprehensive survey data. This means not only to observe the entirety of household farm income, but also to measure the magnitude of capital income and corporate profits in the sector. Given the likely increase in agricultural inequality during the process of structural change, I show also the extent to which social protection programmes are both insufficient a nd poorly targeted for rural populations.

The Coq Community Survey 2022 was an online public survey of users of the Coq proof assistant conducted during February 2022. Broadly, the survey asked about use of Coq features, user interfaces, libraries, plugins, and tools, views on renaming Coq and Coq improvements, and also demographic data such as education and experience with Coq and other proof assistants and programming languages. The survey received 466 submitted responses, making it the largest survey of users of an interactive theorem prover (ITP) so far. We present the design of the survey, a summary of key results, and analysis of answers relevant to ITP technology development and usage. In particular, we analyze user characteristics associated with adoption of tools and libraries and make comparisons to adjacent software communities. Notably, we find that experience has significant impact on Coq user behavior, including on usage of tools, libraries, and integrated development environments (IDEs).

EU policies promoting higher environmental standards in agriculture are often perceived as a challenge to the sector’s economic competitiveness. However, well-designed policies can align the EU’s environmental and economic goals, fostering sustainable and inclusive growth. This policy note examines the case of pesticide-reduction targets and finds that competitiveness trade-offs can be mitigated through complementary trade measures. Our analysis highlights that 44% of pesticide use embedded in EU agricultural consumption comes from imports, despite them representing only 16% of the consumption. Particularly striking, a substantial amount of the pesticide use embedded in imports is represented by banned pesticides, exposing a blind spot in current trade policies. Without appropriate safeguards, stricter EU pesticide regulations can shift production to less-regulated markets, undermining global pesticide reduction efforts while disadvantaging EU agriculture. Analysing the potential for policy solutions, we consider different border-adjustment mechanisms, drawing parallels with the Carbon Border Adjustment Mechanism (CBAM). Our findings indicate that such measures preserve EU agricultural competitiveness without compromising on environmental ambition. Aligning trade and environmental policies is therefore not only feasible but essential for effectively reducing global pesticide use while safeguarding EU agriculture.

This policy note provides a revenue estimate of how much European Member States could raise with a minimum tax of 2% or 3% on the wealth of people owning more than €100 million or €1 billion in wealth – the scenarios considered in the report commissioned by the G20 presidency.

This paper examines wealth inequality dynamics in Chile from 2007 to 2021, focusing on two key macroeconomic events: the sharp rise in housing prices after the introduction of a real estate value-added tax in 2016 and the substantial liquidation of pension assets through early withdrawals during the pandemic. We introduce a methodological innovation that aims to improve the measurement of wealth inequality by integrating administrative pension fund records into household wealth surveys using machine learning techniques. Our results reveal extreme levels of wealth concentration, with the top 10% holding approximately two-thirds of national private wealth. However, inequality slightly declined over the period, particularly after 2016, as the outcome of two opposing forces: housing appreciation, which benefited middle-class households, and pension fund withdrawals, which disproportionately reduced wealth at the lower end of the distribution.

The Middle East is among the most unequal regions in the world today, yet little is known about the long-term dynamics of inequality in the region. This paper presents detailed and newly digitized estimates of land inequality in Egypt from 1896 to 2020, drawing on comprehensive records of private agricultural land ownership. This unique dataset enables a systematic analysis of the evolution of land distribution over more than a century and sheds light on the historical forces and redistributive policies that shaped it. To the best of my knowledge, no similarly long-run and granular series on land inequality exists for any other country or land reform context. The results reveal an extreme concentration of land in the early 20 th century, with the top 1% of landowners controlling over 42% of privately owned agricultural land on the eve of Egypt’s 1952 agrarian reform. Subsequent redistribution efforts reduced this figure to 27% by 1980. During the same period, the share held by the next 9% of landowners increased modestly, from 32% to 35%. The most notable gains were achieved by middle landowners (owners of 1 to 5 feddans), whose share increased from 20% to 30%. In contrast, the reform had limited impact on the landless and smallest landowners: the bottom 50% saw only a marginal increase in their land share, from 6% to 8%.

L’enquête « Voices of Ukraine » (VOU), initiée sur les réseaux sociaux en 2022 par une équipe de chercheuses et de chercheurs et un organisme de sondage, donne la parole aux exilé⋅e⋅s ukrainien⋅ne⋅s dispersé⋅e⋅s en Europe. Depuis l’Allemagne, la Pologne ou la France, principalement, les Ukrainien⋅ne⋅s déplacés par la guerre entre leur pays et la Russie reviennent sur leurs expériences du passage des frontières, leur installation dans les pays d’accueil, les obstacles rencontrés pour leur intégration, et leurs craintes et leurs espoirs concernant la guerre en Ukraine.

We propose a two-stage experiment in which people receive feedback about their relative intelligence. This feedback is a noisy message reminded at every stage, so that subjects cannot forget this ego-relevant information. Instead, we exogenously vary whether the informativeness of the message is reminded in the second stage. We investigate how this treatment variation affects the informativeness reported by subjects, and their posterior beliefs about their intelligence. We show that subjects report informativeness in a self-serving way: subjects with negative messages report that these messages are significantly less informative in the absence of reminder than with it. We also show that the lack of reminder about message informativeness allows subjects to keep a better image of themselves. These results are confirmed by complementary treatments in which we decrease messages informativeness: subjects tend to inflate the informativeness of positive messages that should now be interpreted as bad news.

We examine the political costs of taxation in early modern France. We focus on efforts to enforce the salt tax, the rate of which varied across regions. Using a spatial difference-in-discontinuities design, we compare municipalities just inside the high-tax region with those just outside, before and after a reform aimed at curbing illicit salt smuggling. We find that tax enforcement led to a twenty-fold increase in conflicts between taxpayers and the state in municipalities in the high-tax region. This effect persists until the French Revolution, supporting the view that enforcing the salt tax incurred significant political costs. Finally, we document that the likelihood of conflict increases with tax differences between neighboring regions, which we use to derive an upper bound on the political costs of increased tax enforcement in this historical period

Dix-huit Conférences citoyennes régionales (CCR) (13 dans les régions métropolitaines et cinq dans les départements et régions d’outre-mer) et une Conférence citoyenne nationale dédiée à la jeunesse se sont tenues dans la deuxième phase du Grand débat national (GDN) (soit les vendredi et samedi 15-16 mars 2019, soit les vendredi et samedi 22-23 mars 2019). 1 404 personnes y ont participé (1 216 dans les 13 CCR hexagonales, 68 dans la Conférence nationale sur la jeunesse, à partir d’un tirage au sort national dans la base des numéros de téléphones, et 120 dans les cinq CCR des départements et régions d’outre-mer, à partir d’un recrutement spécifique à chacun de ces territoires). Pour la France métropolitaine, il s’agit du plus vaste tirage au sort réalisé pour constituer un mini-public. Ce rapport décrit en premier lieu précisément le processus de recrutement, globalement désigné par les termes de tirage au sort, pour discuter la représentativité des CCR. Puis il décrit le protocole d’observation de leur déroulement, mis en place par l’Observatoire des débats (https://observdebats.hypotheses.org/), ainsi que le processus de passation des questionnaires lors des CCR, d’une part (en entrée et en sortie) par la mission organisatrice, d’autre part (en sortie) par les chercheurs rassemblés par l’Observatoire des débats. Le rapport compare ensuite les caractéristiques socio-démographiques des participants aux CCR avec celles de la population française générale, ce qui nous permet de montrer que les CCR ne sont pas représentatives de la population générale. Finalement, le rapport resitue les réponses données aux questionnaires et les statistiques descriptives associées (c’est-à-dire le nombre de répondants à chaque question et la distribution, la moyenne et l’écart-type des réponses).

The views expressed here are those of the author(s) and not those of the EU Tax Observatory. EU Tax Observatory working papers are circulated for discussion and comment purposes. They have not been subject to peer-review or to the internal review process that accompanies ofBicial EU Tax Observatory publications.

This paper documents the rise of corporate tax-base narrowing measures in the EU using a novel dataset covering both tax rate and tax base reforms implemented between 2014 and 2022. Our findings indicate a shift away from the ‘cut rate -broaden base’ approach, as governments increasingly align corporate taxation with industrial policy objectives. We show that EU tax competition exerts downward pressure on high-tax countries, while the likelihood of tax cuts also varies with the political orientation of governments. Using financial accounts from more than 40,000 affiliates, we find that the average effective tax rate of multinational enterprises in the EU has declined more rapidly than the statutory rate and estimate that tax base reforms account for 24% of this decline. The estimated revenue cost of all reforms combined amounts to 3.5% of total corporate tax revenue collected from the sample firms. These revenue losses should be carefully weighed against the anticipated benefits of tax reforms.