Seminars

Next Seminars

Monday, November 19th 2024

11h00 - E5.208

Marion Leroutier

“The Cost of Air Pollution for Workers and Firms”

This paper shows that even moderate levels of air pollution, such as those found in Europe, harm the economy by decreasing firm performance. We estimate the causal effect of fine particulate matter pollution (PM2.5) on monthly firms’ sales and worker absenteeism in France, using administrative data on 160,000 firms representing half of the country’s GDP. We exploit within-municipality variation in air pollution induced by changes in monthly wind direction. We find that a 10 percent increase in firms’ monthly PM2.5 exposure decreases sales in the following two months by 0.4 percent on average, with heterogeneous effects across economic sectors. Concurrently, sick leave increases by 1 percent, highlighting the negative effects of air pollution on workers’ health. Yet sales losses are an order of magnitude larger than we would expect if pollution-induced worker absenteeism was the main underlying channel. We provide suggestive evidence that air pollution also affects firm performance via a decrease in the productivity of non-absent workers and in local demand. Our results suggest that reducing air pollution in line with the World Health Organization’s guidelines would generates economic benefits largely exceeding the cost of environmental regulation in Europe, and on par with the monetized benefits from reduced mortality.

 Calendar

• November, 19th 2024 : Marion Leroutier (CREST) - “The Cost of Air Pollution for Workers and Firms”, with Hélène Ollivier (PSE, CNRS)
• December, 10th 2024 : Eugénie Dugoua (LSE)

Organization 

Organized on Tuesdays from 11:00 am to 12:15 pm. 
Location : Room E2.508 
Organiszer : Can Askan Mavi - INRAE - can-askan.mavi@inrae.fr

Latest Archives

• November, 18th 2024 : Ines Chiadmi (PSAE) - "Digging in Flemish Farmers' Preferences and Satisfaction with Soil and Water Quality Extension Services"

The efficacy of information provision policy instruments is proven to complement traditional environmental policies in pollution control. In the agricultural sector, information is notably disseminated by means of agricultural extension services. In this study, we focus on the advisory services targeting the improvement of soil and water quality in Flanders, Belgium. Using a Discrete Choice Experiment (DCE), we elicit Flemish farmers' preferences for advisory strategies in terms of the group size, the interaction format, the provenance of the advisors, the theme of guidance activities, the frequency and the cost of participation. The results of the conditional logit analysis reveal an aversion to advisors representing government institutions and to the cost attribute. We are currently exploring how farmers' population heterogeneity influences the outcomes of our DCE. In addition, we are investigating the interaction between our DCE results and covariates representing farmers' satisfaction with existing advisory activities and their tendencies toward information avoidance.

• November, 14th 2024 : Pierre Uginet (Ecole d'Economie de Paris) - "Meltwater, seasonality and economic activity in South Asia”

While climate change is projected to profoundly alter the processes of solid water melt, the connections between seasonal snowpack melt, glacier shrinkage, and downstream economic activities remain unclear. In this paper, through an econometric analysis of meltwater shocks and overall economic activity at sub-national levels, I aim to enhance our understanding of whether and how hydrological conditions affect economic development. I construct a novel dataset of pixel-level monthly meltwater shocks based on snow and ice coverage and temperature, which I combine with data on nighttime light (NTL) intensity across the entire region of Himalayan-fed watersheds. The dataset spans from 2000 to 2013. I estimate the month-specific annual change in economic activity in a watershed as a function of meltwater shocks. This shock measure accounts for all upstream watersheds’ contributions. The main findings suggest that (i) economic development suffers from both excess and shortage in meltwater volumes, (ii) agriculture serves as the primary channel through which impacts of negative meltwater shocks propagate through the economy, (iii) positive meltwater shocks create torrential floods that act negatively on both the level and the growth rate of economic activities and (iv) groundwater pumping can mitigate the effects of meltwater shocks. The main implications for economic research are that upstream hydrological conditions causally impact downstream economic development.

• November, 7th 2024 : Frédéric Ang (Wageningen University) - "Accounting for animal health in efficiency analysis:An application to Swedish dairy farms”

Poor animal health is a central concern in modern livestock production. Despite the necessity to incorporate animal health in agricultural production analysis, the theoretical and empirical developments are limited on this subject. This article appropriately characterizes the axiomatic properties of animal health within a production framework. We treat animal health as a weakly disposable output with characteristics of an input and an output, which permits computing animal health-adjusted efficiency measures and shadow prices of animal health. The application considers 980 observations of Swedish dairy farms over the years 2009-2016. We use a Benefit-of-the-Doubt approach for assessing animal health, which captures its multiple dimensions with weights being optimized for each farmer. Applying a random sample-split procedure within a Data Envelopment Analysis framework, we statistically verify the extent to which inclusion of animal health in the production framework changes the efficiency estimates. The results show that including animal health in production analysis increases the average efficiency estimates from 0.900 to 0.973, a finding that largely also holds in a statistical sense. The shadow prices of animal health are mostly positive, which indicates a general willingness-to-pay for animal health. Our findings suggest that taking into account non-marketed decision variables such as animal health is important for production analysis.

• November, 5th 2024 : Bernard Villeneuve (Université Paris-Dauphine) - “Don't bet the Farm on Crop Insurance Subsidies: A Marginal Treatment Effect Analysis of French Farms”, with Céline Grislain-Letrémy (Banque de France) and Marc Yeterian (U. Paris Dauphine)

Crop insurance is one of the most important protections against climate-related risks for farmers. Despite being heavily subsidized, insurance take-up in France remains surprisingly low. The goal of this paper is twofold; first, we explain this paradox by analyzing the heterogeneous effects of taking up crop insurance, and second, we provide concrete welfare-enhancing policy recommendations to increase insurance take-up. Using a micro-level panel of 17,000 French farmers over 20 years, we first use a moment-based regression to identify the local average treatment effects (LATE) of insurance on expected revenues and variance. Then we investigate the factors causing the heterogeneity in these effects, both observable through interaction terms and unobservable through a marginal treatment effect (MTE) design. We conclude that insurance subsidies have very little impact on crop insurance demand, especially for those who would benefit the most. Finally, we suggest cost-efficient ways to increase insurance take-up based on administrative simplification, information and imitation.

• October, 15th 2024 : Xavier D'Haultfoeuille (CREST-ENSAE) - “Two-way Fixed Effects and Differences-in-Differences in Heterogeneous Adoption Designs without Stayers”, with Clément de Chaisemartin (Sciences Po), Diego Ciccia (Northwestern University) and Felix Knau (LMU)

We consider treatment-effect estimation under a parallel trends assumption, in designs where no unit is treated at period one, all units receive a strictly positive dose at period two, and the dose varies across units. There are therefore no true control groups in such cases. First, we develop a test of the assumption that the treatment effect is mean independent of the treatment, under which the commonly-used two-way-fixed-effects estimator is consistent. When this test is rejected or lacks power, we propose alternative estimators, robust to heterogeneous effects. If there are units with a period-two treatment arbitrarily close to zero, the robust estimator is a difference-in-difference using units with a period-two treatment below a bandwidth as controls. Without such units, we propose non-parametric bounds, and an estimator relying on a parametric specification of treatment-effect heterogeneity. We use our results to revisit Pierce and Schott (2016) and Enikolopov et al. (2011).

• October, 8th 2024 : Niko Jaakola (University of Bologna) - “Differential Games of Public Investment”

We define a differential game of dynamic public investment with a discontinuous Markovian strategy space. The best response correspondence for the game is well-behaved: best responses exist and uniquely map almost all profiles of opponents’ strategies back to the strategy space. Our chosen strategy space thus makes the differential game well-formed, resolving a long-standing open problem and allowing the analysis of a wider class of differential games and Markov-perfect equilibria. We provide a ‘cookbook’ necessary and sufficient condition for constructing the best response, and demonstrate its use with a canonical model of non-cooperative mitigation of climate change. Our approach provides novel, economically important results: we obtain the entire set of symmetric Markov-perfect Nash equilibria, and demonstrate that the best equilibria can yield a substantial welfare improvement over the equilibrium which previous literature has focused on. Our methods do not require specific functional forms.

• September, 24th 2024 : Lionel Wilner - Compensating against fuel inflation: Price subsidies or transfers?”, avec Odran Bonnet (Insee), Etienne Fize (IPP, PSE) et Tristan Loisel (CREST)

Compensating agents against substantial and unexpected shocks requires both targeting tax policies and taking behavioral responses into account. Based on transaction-level data from France, this article exploits quasi-experimental variation provided by 2022 fuel inflation and excise tax cuts. After disentangling anticipation from price effects, we estimate a price elasticity of fuel demand of -0.31, on average, which varies little with respect to income and location but substantially decreases with fuel spending, in absolute value. Using targeted transfers only achieves imperfect compensation, yet a budget-constrained policy-maker seeking to alleviate excessive losses relative to income prefers income-based transfers to price subsidies.

• September, 19th 2024 : Christophe Gouel (PSAE) -  “AI tools for research”

This  presentation provides an overview of AI tools, particularly Large Language Models (LLMs), and their applications in research. Focusing on the capabilities of models like GPT-4, it covers their potential use  cases, including writing assistance, coding, entity recognition, and  literature review. The presentation highlights how LLMs can enhance  productivity in tasks such as drafting, editing, and idea generation, while also addressing limitations like  hallucinations and plagiarism concerns. Practical demonstrations of  interacting with LLMs through chat interfaces, APIs, and AI copilots are  provided, emphasizing the importance of prompt engineering and customization. The talk concludes by noting the rapid advancements  in AI tools and their growing role in academic research.

• August, 29th 2024 : Clara Berestycki (Columbia University) present “Behavioral adaptation to wildfire-induced air pollution”, joint with Keith Chen (UCLA Anderson)

The intensity and frequency of wildfires and their associated toxic plumes are increasing in the United States, a dynamic in part driven by global warming. In this paper, we use novel, high-frequency mobility data to explore individual behavioral responses to large air pollution shocks caused by wildfires in California. Using geo-localized cell-phone pings, we measure individual time use throughout the day. Leveraging this unique dataset, we can estimate individual dose-response functions to wildfire-induced PM 2.5 air pollution for different metrics. As expected, time spent at work and outside decreases with pollution, and conversely time spent at home and inside increases with air pollution, with some non-linearities for extreme pollution levels. We benchmark the magnitude of this response by estimating similar dose-response functions for rainfall and find that the increase in time spent at home during an extreme pollution event is equivalent to approximately four times that of a rainy day. We also explore heterogeneity in behavioral responses to air pollution across demographic groups. Building on the growing literature documenting inequalities in climate adaptation, we find suggestive evidence that Black and Hispanic residents adapt less to air pollution shocks than White residents.

• June, 25th 2024 : Howard Smith (Oxford University) - “ The Rise of Discounters and its Impact on Concentration, Market Power and Welfare ”

In this paper we quantify changes to concentration, market power, and welfare in grocery retailing in Great Britain over 2002-2019. We document that during this period the rapid expansion of two retailers, known as discounters, was associated with a reduction in market concentration at retail and manufacturer levels across the large majority of narrowly-defined product categories.  We develop an equilibrium model that embeds consumer choice over retailer and product and uses a Nash-in-Nash bargaining framework to capture manufacturer-retailer relations, and apply it to the market for breakfast cereals. We show that gains in discounters' productivity, and expansion in their store network and product ranges, lowered market concentration and average prices and raised consumer and total surplus and that lower income households especially benefited.

• June, 20th 2024 : Berk Oktem (Université de Pau et des Pays de l'Adour) - “Illicit crops and forests: Exploring the relation between opium poppy demand shock and deforestation in Mexico”

By and large, agricultural expansion is the main cause of deforestation, and crop prices are found to be an important driver. Yet, we know very little about how crop prices influence illicit crop cultivation and deforestation. By examining the heroin demand shock from the United States, I investigate the association between the rising demand for opium poppy, higher prices, and their impact on land use decisions and deforestation in Mexico. I employ an empirical strategy borrowed from Daniele et al. (2023), based on exploiting geographic variation in poppy suitability and the timing of the OxyContin reformulation in 2010. The latter led to a substitution between OxyContin, a widely used recreational painkiller, and street heroin in the United States, increasing opium poppy demand and prices in Mexico—the main supplier of heroin to the United States. My primary findings suggest that a positive demand shock for an illegal crop can reduce agricultural activity for legal crops and deforestation. These results differ from the literature that examines the relationship between crop prices and deforestation. However, I argue that this relationship holds true due to the very specific characteristics of the opium poppy plant.

• June, 18th 2024 : Laura Khoury (Paris Dauphine-PSL)  -  “Peer effects in prison”,  avec Julian Johnsen (University of Bergen)

Peer interactions likely play an especially important role in the criminal sector due to its secretive nature and lack of formal institutions. Crucially, a large part of criminal peer exposure happens in prison, and is thus directly under the influence of policy makers and prison administrators.
This paper provides a more comprehensive understanding of how peer effects shape criminal behavior among prison inmates, focusing on how co-inmates influence recidivism and the formation of criminal networks post-incarceration. Our research design causally identifies peer effects in prison using rich Norwegian register data on over 140,000 prison spells and leveraging within-prison facility variation in peers over time. Our analysis reveals several novel findings. First, exposure to more experienced co-inmates increases recidivism. Second, exposure to ``top criminals'' (i.e. those with extreme levels of criminal experience) plays a distinctive role in shaping these recidivism patterns. Third, we provide novel documentation of criminal network formation among prison co-inmates, which could serve as a critical mechanism through which peer influences operate within prisons. Fourth, our results underscore the role of homophily in accentuating these peer and network effects. Together, these findings not only contribute to the theoretical understanding of peer influences in criminal activities but also offer practical insights for policy-makers aiming to reduce recidivism through strategic inmate grouping and prison management policies.

• June, 13th 2024 : Chloé Antoine - “Agricultural productivity and biodiversity effects: theory and evidence”, joint with Jean-Marc Bourgeon (PSAE) and José De Sousa (Université de Paris Panthéon-Assas)

Agricultural specialization on the most suitable crop allows farmers to make the most of local land characteristics, but it also amplifies the proliferation of pests, thereby requiring heavy pesticide use. Given the negative externalities associated with these chemicals, ecological theory suggests reducing specialization to leverage natural pest control. This paper examines the economic desirability of such proposal by quantifying the productivity gains achievable through diversification. We develop a model of endogenous crop resilience that incorporates concentration effects and derive the equilibrium distribution of land among crops resulting from individual farmers’ choices. Using disaggregated data and novel instrumental variables, we then estimate the reduced-form equation for crop productivity provided by the model on a global scale. Results indicate that both same-crop and inter-crop effects matter. Notably, we demonstrate that agricultural diversity has a positive and large impact on yields of major crops like maize, rice, and wheat.

• May, 23 th 2024 : Arnaud Dragicevic (Chulalongkorn University) -“Towards Sustainable Agriculture: Implementing a Market for Payments for Environmental Services in Agri-Food Systems”, with Serge Garcia (BETA) and Jean-Christophe Pereau (BSE)

This study evaluates the impact of a Payments for Environmental Services (PES) market on the sustainability of an agri-food supply chain. It employs a variational inequality approach within a multi-criteria decision-making framework. The theoretical findings are validated through numerical simulations using a neural network-based machine learning algorithm. They conclusively demonstrate the potential to not only achieve but also surpass objectives related to carbon and biodiversity neutrality targets. This study highlights the effectiveness of combining market valorisation of environmental services with government subsidies for ecological transition as a powerful strategy to significantly reduce the global ecological footprint. It shows that actors within the supply chain are inclined to intensify their offsets towards achieving environmental neutrality if they are initially compensated for their efforts to mitigate their environmental impacts.

• May, 21th 2024 : Clément Bellet (Erasmus School of Economics) - “How Sticky are Consumption stereotypes? Evidence from the Meat Gender Gap”

"Using consumer surveys and supermarket purchase data, this study reveals a persistent gap in red meat consumption between single men and women in the US. Investigating whether this disparity stems from gender stereotypes, we collect survey data to assess the influence of attitudes, beliefs, and implicit biases on meat consumption. Findings suggest that the gap is largely due to preferences and perceived needs rather than differences regarding beliefs about environmental, health, or ethical impacts. Moreover, an Implicit Association Test uncovers a strong bias linking meat with masculinity. The study further examines the stickiness of these consumption stereotypes through an experiment with identity priming and a de-biasing treatment, analyzing their effects on meat consumption expectations and conjoint analysis outcomes."

• May, 16th 2024 : Aude Pommeret (Savoie Mont Blanc U.) - “Carbon Taxes and Tariffs, Financial Frictions, and International Spillovers”

Ambitious climate policy, coupled with financial frictions, has the potential to create macrofinancial stability risk. Such stability risk may expand beyond the economy implementing climate policy, potentially catching other countries off guard. International spillovers may occur because of trade and financial channels. Hence, we study the design and effects of climate policies in the world economy with international trade and financial flows. We develop a two-sector, two-country, dynamic general equilibrium model with financial frictions, climate policies, including carbon tariffs, and macroprudential policies. Using the calibrated model, we evaluate spillovers from unilateral domestic carbon pricing to foreign economies and back. We also examine more ambitious climate architectures involving carbon tariffs or a global carbon price. We find that accounting for cross-border financial flows and frictions in credit markets is crucial to understand the effects of climate policies and to guide the implementation of macroprudential policies at the global scale aimed at minimizing transition risk and paving the way for ambitious climate policy.

• May, 2nd 2024 : Valentin Cocco (PSAE) - “Guilty or scapegoat? Land consolidation and the hedgerow decline”

Land consolidation is a standard policy instrument which aims to reduce the fragmentation of farmland through spatial redistribution of land ownership. While the reform's primary goal was to improve agricultural productivity, its potential adverse effects on the landscape raise questions about its environmental sustainability. We investigate the landscape impact of the French consolidation program in the second half of the 20th century. Often blamed for the drastic decline of hedgerows observed in the countryside, social scientists discuss its responsibility. Using a staggered difference-in-differences strategy on a longitudinal survey in Lower Normandy, France (1972-2020), we bring the first causal evidence on this debate and show that consolidation decreased hedgerow density by -14.3m/ha (standard error: 2.36), which represents 17.6% of the total decline observed in consolidated municipalities. Our analyses also suggest a diminishing impact with the consolidation period and the time elapsed since consolidation, and an increasing impact with the initial hedgerow density. Our nuanced findings challenge the prevailing narrative that consolidation is the leading cause of the hedgerow decline and call for considering land consolidation in a broader context of political, social, and market drivers of landscape dynamics.

• April, 30th 2024 : Biliana Yontcheva (DICE) - “ Chain formation and consumer welfare on the retail pharmacy market ”

This paper evaluates the effect of deregulated ownership and horizontal integration on the retail pharmacy market. Using data on the full population of reimbursed prescriptions in Slovakia in 2017, we examine whether outlets of pharmacy chains perform better than their independent counterparts in terms of consumer preferences and operating costs. Our preliminary findings indicate that consumers perceive pharmacy chains as having higher quality on average than independent outlets, although there is substantial heterogeneity in the effects, both across chain brands and across consumer types. We do not find evidence for substantial productivity gains due to chain affiliation, suggesting that growth in the sector is mainly driven by selective takeovers in locations with high economies of scale.

• April, 25th 2024 : Pauline Rossi (CREST) - “The Negligible Effect of Free Contraception on Fertility: Experimental Evidence from Burkina Faso”, with Pascaline Dupas (Princeton), Seema Jayachandran (Princeton) and Adriana Lleras-Muney (UCLA)

We conducted a randomized trial among 14,545 households in rural Burkina Faso to test the oft-cited hypothesis that limited access to contraception is an important driver of high fertility rates in West Africa. We do not find support for this hypothesis. Women who were given free access to medical contraception for three years did not have lower birth rates; we can reject even modest effects. We cross-randomized additional interventions to address possible inefficiencies leading to low demand for contraception, specifically misperceptions about the child mortality rate, limited exposure to opposing views about family size and contraception, and social pressure. Free contraception did not influence fertility even in combination with these other interventions.

• April, 4th 2024 : Clément Nedoncelle (PSAE) - “Soybean exports, market power, and deforestation”, with Philippe Delacote (INRAE-CEC) and Léa Crépin (PSAE-CEC)

This paper investigates the impact of market power exerted by large soy exporters on soy production and its relationship with deforestation in Brazil between 2004 and 2017. We hypothesize that large exporters leverage their power to markdown soybean prices from producers. Empirical analysis reveals that municipalities with large exporters experience lower agricultural revenues per unit of soybean produced, supporting this markdown hypothesis. Theoretical modeling shows that larger exporters pay lower prices, leading to reduced soy production and consequently lower deforestation rates. These findings are empirically confirmed. These findings highlight the critical role of market structure in shaping environmental outcomes in agricultural supply chains.

• April, 2nd 2024 : Francesco Ricci (University of Montpellier) - “Efficient recycling”

We elaborate a model of the market economy, where material inputs are produced by the primary sector as virgin resources or produced by the recycling sector as secondary resources,  emphasizing the supply-side linkages between virgin and recycled materials. We posit a representative structure of the recycling value chain and use it to analyze how market mechanisms can explain the heterogeneity in recycling rates across materials and sectors. Two situations can emerge: either part of recyclable materials is disposed of as definite waste, or there is a lasting scarcity of recyclable materials.  If the possibility of recycling end-of-life products is socially valuable, the additional value is transmitted to the primary resource producers through prices.  In the presence of market failure due to negative externalities from waste disposal, the optimal policies can involve a Pigouvian tax on waste disposal, a combination of virgin output tax and recycling subsidy (the deposit-refund system), or a minimum recycled content regulation. However, we show that when recyclable materials are scarce, the optimal policy mix can imply a tax on recycling. When it comes to market failure due to natural monopoly power at the stage of recovery of EOL products, we show that the widespread average cost pricing regulation of waste management utilities threatens the efficient transmission of the value chain. From the perspective of increasing the use of mineral resources for energy transition, our paper highlights the crucial role recycling can play and how policy tools should be flexibly adapted to address these challenges.

• March, 28th 2024 : Manon Desjardins (Skema Business School and U. Côte d'Azur) -“Internal Carbon Pricing in the Multidivisional Firm”, with Bernard Sinclair-Desgagné (Skema Business School)

In response to environmental concerns, multidivisional firms running internal markets often seek to have the associated internal (or transfer) prices incorporate their social cost of pollution. We develop a theoretical rationale and framework for this phenomenon. Modeling two vertically related subsidiaries located in different jurisdictions, we examine how transfer prices would convey mandatory carbon fares, and consider the impact on each subsidiary’s production and emissions abatement. Thereby, we also show how the firm’s environmental strategy might interact with fiscal compliance. Through transfer pricing, finally, a carbon fare in one jurisdiction can have an incidence on the other jurisdiction’s subsidiary; implications for environmental governance are discussed.

• March, 26th 2024 : Victoire Girard (Nova SBE) - “ Artisanal mining in Africa: Green for gold? ”

The livelihoods of 130 to 270 million people depend on artisanal mining, a labor-intensive and often informal extractive activity. We levy research in geology, and two sources of exogenous time variation, to build the first proxy of artisanal gold mining in Africa -- the main form of artisanal mining. We first document the impact of a change in the potential value of artisanal mining. We establish that the historical increase in the gold price accounts for 8% of deforestation continent-wide, 28% in gold areas. In parallel, artisanal gold mining increases local economic wealth, but these economic impacts are of smaller magnitude. Second, we turn to weather shocks. Artisanal mining may provide an alternative livelihood when a weather shock jeopardizes agricultural output. We confirm the heterogeneous effects of adverse weather : compared to the average area, deforestation is magnified in gold-suitable areas, while losses in night-light emissions are compensated for. The primary driver of deforestation appears to be the mining activity itself, while mining-induced changes in human settlements or local demand have limited impacts.

• March, 14th 2024 : Mathilde Degois - “Does Agricultural Productivity leads to Human Capital Accumulation or Re-Location? Evidence from Long-Term Irrigation in India”

We estimate and compare the effects of long term canals irrigation access on education in India. To do so, we perform a spatial RDD at the border of irrigated areas. We find that irrigation access increases women's education for landowners households in rural areas. This occurs due to assortative matching in the marriage market drives the results, with the selective migration of better-educated women to irrigated areas. Overall, these results highlights that agricultural productivity and technological innovation can lead to a geographical re-location of human capital with long-term consequences.

• March, 12th 2024 : Raphaële Préget (INRAE-SupAgro) - “Innovativeness, innovation adoption, and priming: nudging farmers in a large-scale randomized experiment in France”

This article is an empirical contribution on measuring farmers' ability to innovate, and on the effectiveness of a nudge-type non-monetary incentive on their (stated) intention to adopt an innovation such as the French “Label bas carbone”, a voluntary scheme that certifies carbon credits. We propose an original methodology for measuring farmers' capacity to innovate (“innovativeness”), adapting the scale of Goldsmith and Hofacker (1991) to the specificities of farmers' decisions in a professional setting. Based on an online survey with more than 6,000 responses from French farmers, we validate this scale and evaluate with a randomized experiment included in the questionnaire the net impact of a priming nudge targeting the most innovative farmers. The results indicate that the nudge tested has no significant or detectable impact on the surveyed sample, leading us to discuss the effectiveness of nudges when trying to influence high-stakes decisions.

• March, 7th 2024 : Nathalie Lavoie (UMass) - “Competing with fad products: erroneous beliefs and market outcomes”, with Christoph Bauner (UMass)

We study how erroneous nutrition assumptions affect manufacturers’ profits and consumer surplus and how the government could intervene to improve welfare. In our model, two manufacturers produce a conventional product and a fad version misperceived to bring health benefits. We compare the laissez-faire outcome to two outcomes: one without false beliefs and the other with information provision reducing the false belief’s prevalence. We find that false beliefs about the health benefits of fad products lower consumer surplus and total welfare under some conditions. Information provision generally increases total welfare, but, in some situations, this occurs at the expense of consumer surplus.

• March, 6th 2024, Ph-D student seminar
  • Maxime Tranchard, "Assessing the effect of sin taxes: How does design matter?"
  • Ondine Berland will "Junk Food, Leisure, and Income Inequalities"

Many studies have documented that high-income households tend to adopt diets of higher nutritional quality. Guided by the fact that the 'liking for sweet taste is both innate and universal’ (Drewnowski, 1997) and differences in supply conditions only explain a small share of diet inequalities (Allcott et al., 2019), we investigate a driver of food preferences that has not been explored yet: income enables households to substitute the pleasure derived from tasty (but unhealthy) foods with expensive leisure activities. We start by leveraging the latest wave of the French Consumer Expenditure Survey data (Budget des Familles, 2017) to describe the income gradient in demand for leisure and healthy food. We document a positive correlation between income, diet quality, and non-durable out-of-home leisure. In an effort to evidence causality, we turn to transaction data (Kantar 2019-2020) to explore the effects of 2020 COVID-19-related lockdowns that can be seen as a negative and very large shock on out-of-home leisure, affecting disproportionately rich households (who demand more leisure). This unique exogenous variation will allow for a more direct analysis of the impact of leisure consumption on diet quality.

• March, 5th 2024 : Martina Bozzola - “ Internal and External Validity of Farmers’ Risk and Uncertainty Preferences Elicited Using Contextualized Field Experiments: Evidence from Zimbabwe ”

Farming is a risky and uncertain activity and hence risk and uncertainty preferences are important determinants of farmers’ decisions. This paper explores the internal and external validity of farmers’ risk and uncertainty preferences elicited using contextualized field experiments in Zimbabwe. Each farmer is exposed to contextualized and non-contextualized lottery tasks. Contextualization is implemented for different crops to investigate the robustness of our findings. Results suggest that risk and uncertainty preferences elicited using contextualized tasks are not always consistent with those elicited via non-contextualized tasks. Non-contextualized tasks tend to overestimate the degree of risk and uncertainty aversion when compared to contextualized tasks. Results vary across crop contextualization and this finding is related the relevance of the crop to farmers. Preferences elicited using relevant crop contextualization are more internally valid than those elicited using non-contextualized tasks. Preferences are barely correlated with past and future farm decisions, regardless of whether tasks are contextualized or not. This implies a low level of external validity.

• February, 29th 2024 : François Bareille - “Strategic pesticide applications in organic-conventional mixed landscapes”

Organic and conventional farmers face the same pests  but have heterogeneous tools and incentives to control them. This paper aims to theoretically and empirically characterise the strategic interactions to control pests that occur between these two types of farmers. Specifically, we develop a game theory model with two heterogeneous agents facing the same public bad to derive theoretically-consistent propositions related to the impacts of their specific characteristics on their respective pest control efforts. Our analyses indicate that organic farmers are likely to free-ride on the contributions of the conventional farmers when they manage a smaller part of the landscape, and vice versa. These strategic interactions are theoretically exacerbated when the exposure to pest abundance decreases, or when the heterogeneity of treatment efficacy or treatment expected payoffs between the two agents increase. Using exhaustive French data on insecticide purchases against the vector of a particular vine disease (Flavescence dorée), we provide empirical supports for all these propositions. Our preferred estimates indicate that organic farmers free-ride on conventional farmers' efforts until they represent about 6% of the landscape. Beyond this threshold, conventional farmers' efforts decrease and are partially substituted by those of organic farmers (by a factor 0.8 for 1). Conventional farmers eventually start to free-ride when organic farmers manage 72%  of the landscape. These results are robust to several sensitivity analyses. Finally, we empirically show that farmers' free riding behaviors almost disappear for high pest abundance, and that the heterogeneity of treatment efficacy and expected payoffs between farmers do affect their strategic interactions.

• February, 27th 2024 : Isis Durrmeyer - “ The Welfare Consequences of Urban Traffic Regulations ”

We develop a structural model to represent individual transportation decisions, the equilibrium road traffic levels, and speeds inside a city. The model is micro-founded and incorporates a high level of heterogeneity: individuals differ in access to transportation modes, values of travel time, and schedule constraints; road congestion technologies vary within the city. We apply our model to the Paris metropolitan area and estimate the model parameters from publicly available data. We predict the road traffic equilibria under driving restrictions and road tolls and measure the policy consequences on the different welfare components: individual surplus, tax revenues, and cost of emissions.

• February, 6th 2024 : Clément de Chaisemartin (Sciences-Po) - “Difference-in-Differences for Continuous Treatments and Instruments with Stayers”

We propose difference-in-differences estimators for continuous treatments with heterogeneous effects. We assume that between consecutive periods, the treatment of some units, the switchers, changes, while the treatment of other units does not change. We show that under a parallel trends assumption, an unweighted and a weighted average of the slopes of switchers' potential outcomes can be estimated. While the former parameter may be more intuitive, the latter can be used for cost-benefit analysis, and it can often be estimated more precisely. We generalize our estimators to the instrumental-variable case. We use our results to estimate the price-elasticity of gasoline consumption.

• February, 1st 2024 : Rafael Schütz (PSE) - “Does it take a stick to eat a carrot? Food consumption, Pigouvian pricing, and the trade-off between the environment and animal welfare”

The production of animal products, especially meat, accounts for 15 percent of global greenhouse gas emissions. It also endangers the welfare of billions of animals raised for food each year. Any intervention to reduce meat consumption thus faces a potential trade-off in producing a given amount of meat: high emissions (e.g. beef) vs. large number of animals reared and slaughtered (poultry). This paper investigates this trade-off empirically in the context of Pigouvian price interventions targeting animal source foods. Harmonizing 40 years of household expenditure survey and aggregate supply data from the UK, I first quantify food consumers' animal welfare footprints in a Distributional National Accounts framework. Regardless of the animal welfare indicator used, the footprints vary widely across the population, but are only weakly associated with income. Despite the increase in poultry meat consumption, mainly replacing beef and sheep, the distribution of footprints is relatively stable over time, with only a slight downward shift in recent years. Simulations predict that both a carbon price on food and an animal welfare levy will reduce both the environmental and animal welfare impacts of food consumption. This effect holds for demand-system and machine learning-based approaches.

• January 18th 2024 : Chloé Beaudet (PSAE) “Spatially allocating preferences derived from a choice experiment: a comparison of two methods”

Discrete choice experiments (DCE) are useful tools in environmental economics to guide policy making. However, DCE’s results are usually presented as average preferences among a population or a group, which can be of limited usefulness for spatial planning policies in which spatial heterogeneity is strong. In this paper, we propose two methods for spatializing the preferences derived from a DCE, that we call the “one-step” and the “two-step” methods. Both methods use the influence of socio-economic and spatial variables on the general preferences of the population to predict average willingness to pay (WTPs) for different attributes at small geographic scales. We aim to compare these methods to give guidelines and best practices to spatialize the results of DCEs. To do so, we employ Monte Carlo simulations, making different assumptions on the data generating process, to determine which method performs better in theory. Then, we apply the methods' comparison to the case of light pollution mitigation policies in Montpellier Metropolitan Area. Preliminary results lead us to prefer the one-step method over the two-step method, although the latter shows practical aspects that can be of interest to applied researchers.

• January 23th 2024 : Emmanuelle Taugourdeau (CREST,CNRS) - “Does leadership in policy setting reduce pollution and make countries better off?”

We analyse whether there exists a first mover and/or a second mover (dis)advantage in a game where a more industrialized country -the leader- and a less industrialized country -the follower- decide on two policy instruments (profit taxation and environmental regulation) to raise their tax income while limiting the damage from pollution. We show that when governments care about world emissions, they can use a sequential game to reach their national goals and reduce the level of world emissions, even if industrial leakage occurs. This is no longer the case when emissions are considered solely as local.

• January 16th 2024 : Nicolas Treich (TSE,INRAe) - “The Animal-Welfare Levy”

We provide a non-anthropocentric rationale for implementing a levy on meat consumption due to animal-welfare considerations. It operates as a Pigouvian tax and addresses externalities on farmed animals. Under total utilitarianism, the levy is a subsidy when an animal’s life is worth living, and a tax when it is not. Under average utilitarianism, it is always a tax when human welfare exceeds animal welfare. Even under conservative assumptions, calibrated tax levels are substantial and would make most-intensive animal farms unprofitable. Taxes are significantly higher for chickens and pigs than for cows, in contrast to the taxation of other meat externalities.

• January 9th 2024 : Enrique Ide (IESE) -Cross-Market Mergers with Common Customers: When (and Why) Do They Increase Negotiated Prices?”

I examine the implications of cross-market mergers of suppliers to intermediaries that bundle products for consumers. These mergers are controversial. Some argue that suppliers’ products will be substitutes for intermediaries, despite not being substitutes for consumers. Others contend that because bundling makes products complements for consumers, products must be complements for intermediaries. I contribute to this debate by showing that two products can be complements for consumers but substitutes for intermediaries when the products serve a similar role in attracting consumers to purchase the bundle. This result leads to new recommendations and helps explain why cross-market hospital mergers raise prices.