Ongoing research on Covid-19 at AMSE

  • Research


«Learning in pandemic times: Anxiety, cognitive capacities and decision-making of university students»

Timothée Demont, Eva Raiber, Daniela Horta Saenz.

This research project aims at better understanding the impact from the major, deeply uncertain, and society-wide shock of the ongoing Covid-19 pandemic on university students. By definition, students are on the threshold of professional life and need substantial cognitive resources to make the most of their university training. Through an online experiment with 1600 students at Aix-Marseille University, we investigate the following question: Does the anxiety generated by the Covid-19 pandemic affect the learning and decision-making abilities of students and, if so, by how much and through which channels? Our design will quantify the extent of concern about future labor-market opportunities and impairment of social life and network, test the role of competitive pressure, and perform a variety of heterogeneity analyses to identify particularly vulnerable groups such as female, low economic background and depression-prone students.


«Covid-19, Lockdowns and Well-Being: Evidence from Google Trends»

Abel Brodeur, Andrew E. Clark, Sarah Flèche, Nattavudh Powdthavee.

The Covid-19 pandemic and government intervention such as lockdowns may be severely affecting people’s mental health. We use Google Trends data to test whether Covid-19 and the associated lockdowns implemented in Europe and America led to changes in the search-terms used for well-beingrelated topics. Using difference-in-differences and a regression discontinuity design, we find a substantial increase in search intensity regarding boredom in both Europe and the US. We also find a significant increase in searches regarding loneliness, worry and sadness, while searches on stress, suicide and divorce declined. Our results suggest that people’s mental health may already have been severely affected by the pandemic and lockdowns.


«Covid-19 crisis, vulnerability, and the French labour market: analysis of a decree issued in the acute phase of the epidemic»

Bruno Ducouture, Florence Jusot, Pierre Madec, Mathieu Plane, Xavier Timbeau, Thomas Renaud, Bruno Ventelou, Jerome Wittwer.

This working-paper aims to quantify the extent of ‘vulnerability to Covid-19’ in the French labor market, as defined by the decree of May 5, 2020 and to study the socioeconomic determinants of the effective use of this decree by eligible parties. The decree offers a partial-activity option (with financial compensation) for workers who have an underlying medical condition increasing the risk of severe Covid-19 illness. We used two surveys, the EHIS/ESPS (European Health Interview Survey – Enquête Santé et Protection Sociale) as well as the Continuous Employment Survey, to estimate the number of “vulnerable” people in non-teleworkable jobs in the French working population. The prevalence of “risk” pathologies turns out to be high even at working ages: 4.8 million people are vulnerable, or 17.5% of those in employment. While teleworking is theoretically possible for some, 3.5 million employees are unable to work remotely, including 2.8 million who would qualify for the social-protection benefit as offered in the decree. Using an additional declarative survey in May 2020 (COCONEL), we find that non-take-up of the partial-activity compensation by those eligible is prevalent (around 80%). We discuss the reasons: ignorance of the system, reduced exposure to risks, or “forced presenteeism”, for example due to the fragility of the employee’s situation vis-à-vis her/his employer.



«Tracking the Dynamics and Allocating Tests for Covid-19 in Real-Time: An Acceleration Index with an Application to French Age Groups and Départements»

Christelle Baunez, Mickael Degoulet, Stéphane Luchini, Patrick Pintus, Miriam Teschl.

An acceleration index is proposed as a novel indicator to track the dynamics of the Covid-19 pandemic in real time. Using French data on confirmed cases and tests for the period following the first lockdown - from May 13, 2020, onwards - our acceleration index shows that the ongoing pandemic resurgence can be dated back to around July 7. It reveals that pandemic acceleration has, since early September, been stronger than national average for the [59 − 68] and [69 − 78] age groups, the latter showing the strongest acceleration index as of October 25. In contrast, acceleration is weakest among the [19 − 28] age group, about half that of the [69 − 78] group as of October 25. In addition, we propose an algorithm to allocate tests among French départements based on both the acceleration of the pandemic and the feedback effect of testing. Our acceleration-based allocation differs significantly from the current French territorial distribution system, which is population-based. We argue that both our acceleration index and our allocation algorithm are useful tools to guide public health policies as France enters a second lockdown period of indeterminate duration.


«Liberté, Egalité, Fraternité... Contaminé? Estimating the impact of French municipal elections on Covid-19 spread in France»

Guilhem Cassan, Marc Sangnier.

On March 15, about 20,000,000 voters cast their vote for the first round of the 2020 French municipal elections. We investigate the extent to which this event contributed to the Covid-19 epidemics in France. To this end, we first predict each département’s own dynamics using information up to the election to calibrate a standard logistic model. We then take advantage of electoral turnout differences between départements to distinguish the impact of the election on prediction errors in hospitalizations from that of simultaneously implemented anti-contagion policies. We report a detrimental effect of the election in locations that were at relatively advanced stages of the epidemics by the time of the election. In contrast, we show that the election did not contribute to the epidemics in départements with lower infection levels up to March 15. All in all, our estimates suggest that elections accounted for about 4,000 excess hospitalizations by the end of March, which represents 15% of all hospitalizations up to this time. They also suggest that holding elections in June might not have been as detrimental.


«Population preferences for inclusive COVID-19 policy responses: evidence from a Discrete Choice Experiment (DCE)»

Thierry Blayac, Dimitri Dubois, Sebastien Duchêne, Phu Nguyen-Van, Bruno Ventelou, Marc Willinger.

Our Discrete Choice Experiment study examines population preferences on various “menus” of Covid-19 epidemic control policies. Preference-ranking analysis is applied to the whole population, and also differentiated by subgroups (different age-groups; those at risk). Masks, limited transport, and (even) digital tracking are well accepted. Restaurant closures and excessive restrictions on leisure travel are not. Subpopulation analyses also demonstrate that the acceptability of some strategies depends on personal characteristics: the young population stands out quite strongly in their preferences on anti-covid policies, particularly in their demand for monetary compensation. Knowing how people rank the various Covid-19 prophylactic measures is a prerequisite for designing sets of suitable epidemiccontrol programmes likely to be collectively accepted by the population. The marked difference in the attitude of the young population suggests the need for a tailored menu of anti-covid policies.


«Experimental evidence on the behavior of youth-that-risk in the context of the Covid-19 crisis in two large metropolitan areas of Côte d’Ivoire»

Maria Laura Alzua, Habiba Djebbari, Assi Kimou.

Our objective is to shed light on the behavior of the vulnerable young people living in the street during the Covid-19 pandemic. For this purpose, we collect data from youth-at-risk living in two large metropolitan centers in Côte d’Ivoire to document knowledge, attitudes, and behavior in relation to the disease. We also design and conduct experimental games to understand the motives behind their choices, in particular whether strategic ignorance may explain lack of adherence to regulations that are not appropriate to their circumstances.


«U.S. Churches’ Response to Covid-19: Results from Facebook»

Eva Raiber, Paul Seabright.

This study investigates U.S. churches’ response to the SARS-CoV-2 pandemic by looking at their public Facebook posts. For religious organizations, in-person gatherings are at the heart of practice. Yet religious in-person gatherings have been identified as some of the early hot spots of the new coronavirus. Our sample contains information on church characteristics and their Facebook posts for nearly 4000 churches that posted at least once in 2020. We find that the share of churches posting at least one video on a given Sunday more than doubled from before the pandemic (before March 2020) to April 2020 (the peak in terms of deaths in the U.S. so far), and posts have since remained well above baseline levels. We examine the extent to which churches’ prior characteristics and behaviors (such as their size and worship style) are associated with their response to the Covid-19 shock.


«The impact of Covid-19 on farmers in Rural Bénin»

Esther Adimi, Maria Laura Alzua, Habiba Djebbari, Rosaine Yegbemey.

The food production sector has been largely sheltered from the Covid-19 crisis. In Bénin, as in other countries of the region, the authorities cordoned off urban centers as the main policy response to contain the Covid-19 epidemic. The cordon is designed to allow for trade in goods but to limit population mobility. In this study, our objective is to assess, using geographical regression discontinuity, the impact of Covid-19 on farmers’ production decisions and their expectations for the future. We also experimentally investigate their demand for contract farming as a means to cope with future crises.


«Shutdown policies and worldwide conflict»

Nicolas Berman, Mathieu Couttenier, Nathalie Monnet, Rohit Ticku.

We provide real-time evidence of the impact of Covid-19 restrictions on conflicts globally. Combining daily information on conflict events and government policy responses to limit the spread of the coronavirus, we explore how conflict levels vary following shutdown and lockdown policies. We use the staggered implementation of restriction policies across countries to identify the size and duration of their effect on conflict intensity.


«Preventing a European Banking and Financial Crisis after the Covid-19 Health Crisis: Lessons from the Last Decade»

Marie-Hélène Gagnon, Céline Gimet.

This paper investigates how the European Central Bank can reduce financial and banking fragmentation, a stated objective in crisis periods. We use regional SVAR models and national GVAR models to study the impact of interest rates, quantitative easing (QE), and long-term refinancing operations (LTROs) on price and volume indicators of fragmentation through the main channels of transmission of monetary policies. Using data from the last decade, we find that LTROs reduce banking price dispersion, but have no effect on credit volume indicators. The positive signal generated by QE measures increases credit volume in the most fragile countries, where an unbalanced repurchase of government bonds is needed to decrease long-term interest rate spreads.


«CoVaPred : new Covid-19 vaccine acceptance in France»

François Alla, Pierre Arwidson, Stéphane Luchini, Michael Schwarzinger, Verity Watson.

Although Covid-19 vaccine hesitancy may represent a major barrier to reaching herd immunity, opinion polls suggesting that hesitancy is increasing worldwide do not necessarily consider specific new vaccines and related acceptance by a specific population. Under the CoVaPred project, we conducted a randomized experiment in July 2020 to assess acceptance of Covid-19 vaccines with different characteristics in France. Repeated discrete choice data were collected from a representative sample of the French working age population without prior SARS-CoV-2 infection (n=1942) and analyzed using a two-part model disentangling outright vaccination refusal (regardless of vaccine characteristics) from vaccine hesitancy. The randomized experiment demonstrates outright vaccination refusal, as of early July 2020, by 29.4% of the French population of working age. In addition, Covid-19 vaccine acceptance in the French population of working age varies from 27.4% to 61.3% for new vaccines manufactured in China vs. the European Union with respectively 50% vs. 90% efficacy and a risk of serious side effects of 1/10,000 vs. 1/100,000. Vaccine acceptance increases when respondents are informed about herd immunity benefits. This means that herd immunity from vaccination may only be achieved in France with Covid-19 vaccines made in the European Union and a communication campaign that also stresses herd immunity benefits.


«ANR-funded project “ECOVID-19”: Economic Epidemiology of Covid-19»

Jérôme Adda, Raouf Boucekkine, Josselin Thuilliez.

ECOVID-19 aims to provide a comparative costeffectiveness analysis of public policies undertaken in real-life conditions during the Covid-19 epidemic in France. We compare three main measures: i) confinement measures ii) testing with a specific focus on undocumented infections iii) reminders on the so-called “barrier gestures”. To do so, we build a theoretical model of disease diffusion and test the model using quasi-experimental variation. 


«After Covid-19: What lessons can be learned in terms of growth opportunities, the emergence of new value chains and Mediterranean integration?»

Report CMI, FEMISE, coordinated by Patricia Augier.

The report will be available in 2021, with a focus on Algeria, Egypt, Jordan, Morocco and Tunisia. This reports aims to (i) identify the growth and job opportunities of the “post-Covid” period for Mediterranean countries (ii) explore different paths towards better regional integration. Chapter 1. What are the implications in terms of food security ? Chapter 2. What new challenges for value chains and economic integration in Mediterranean countries? Chapter 3. What opportunities can the health sector offer?


©  Made with Khartis, basemap from Natural Earth (CCO).


→ This article was issued in AMSE Newletter, Fall 2020.