Selected work in progress
Disseminating market information via mobile phones to cashew producers: an impact evaluation in Guinea-Bissau, with Giulio Schinaia, Sebastian Schaber, Dayvikson Tavares and Adewusi Mendonça
In collaboration with ANCA, the Ministry of Economy and Finance, the Ministry of Agriculture, Nitidae and MTN Guiné-Bissau, and financial support by the MEF and two generous and competitive grants, given by PEP and PEDL, we are implementing a cluster randomized controlled trial with 2000 Bissau-Guinean families spread across nearly 300 villages. The objective of this project is to analyze how communicating information on prices and market dynamics through Market Information Systems, introducing the n'kalo service into the country, can increase the prices obtained by cashew farmers. Increased transparency and richer information should increase producers' negotiating capacity and facilitate them to better choose when to sell their crops. This information might be a particularly important support given the uncertainty derived from the COVID-19 pandemic.
Dissemination and media coverage: PEDL C-19 piece, National Conference in Bissau, PEP policy brief, Radio France International
Does group size and gender matter for performance in skill diverse teams? Experimental evidence, with Shan Aman-Rana
In this paper we investigate whether the effect of group diversity on performance depends on group size. We carried out a lab experiment with university nursing and economics students in Guinea-Bissau. Students were randomly assigned into groups with different size and gender composition to solve tests as a team. We find that in groups with homogenous skills, size of the group does not matter for performance. Coherently with our theoretical model, for students in skill diverse groups the increase in performance from being in a large group is 60% higher than when being in a small one. However, this large gain from introducing skills diversity in teams only materializes for male students. While the gender composition of groups does not matter for men, it plays an important role for female students. In a large skill diverse group, the performance of female student improves when matched with other female students, as the theory suggests, but it reduces when matched with male ones. Our findings suggest that both group size and gender play a central role in determining how skills diversity affects performance of workers.
Beliefs Systems and Health Behaviors in Guinea-Bissau, with Alexander Coutts, Teresa Molina Millán and Pedro Vicente
Gender, information exchange and choice over co-workers: experimental evidence, with Shan Aman-Rana, Clement Minaudier and Shamyla Chaudry
Selected Publications
Loss of structural balance in stock markets, with E. Ferreira, S. Orbe, J. Ascorbebeitia and E. Estrada, Scientific reports 11.1 (2021): 1-10.
We use rank correlations as distance functions to establish the interconnectivity between stock returns, building weighted signed networks for the stocks of seven European countries, the US and Japan. We establish the theoretical relationship between the level of balance in a network and stock predictability, studying its evolution from 2005 to the third quarter of 2020. We find a clear balance–unbalance transition for six of the nine countries, following the August 2011 Black Monday in the US, when the Economic Policy Uncertainty index for this country reached its highest monthly level before the COVID-19 crisis. This sudden loss of balance is mainly caused by a reorganization of the market networks triggered by a group of low capitalization stocks belonging to the non-financial sector. After the transition, the stocks of companies in these groups become all negatively correlated between them and with most of the rest of the stocks in the market. The implied change in the network topology is directly related to a decrease in stock predictability, a finding with novel important implications for asset allocation and portfolio hedging strategies.
Waving goodbye? The determinants of autonomism and secessionism in Western Europe, with M. Portos and J. Vourdas, Regional Studies 52.2 (2018), 197-211
This paper sheds light on the main aggregate-level determinants of electoral support for regionalist parties across 10 Western European countries. A region being relatively richer than the country to which it belongs is associated with higher electoral support for regionalist parties only to the extent that the region is culturally differentiated. This hypothesis is substantiated theoretically, tested empirically and found to hold in the form of a strong and significant interaction effect between cultural and economic variables. This result, omitted in previous studies, implies a profound change in the interpretation of the role of income and cultural differences in explaining support for regionalism, for both autonomist and separatist parties.
The Best of Both Worlds: Developing Equation-Based Models into Agent-Based Models, with Gräbner, C., Bale, C. S., Furtado, B. A, Gentile, J. E., Henderson, H., Lipari, F. Computational Economics (2017): 1-20
We argue that building agent-based and equation-based versions of the same theoretical model is a fruitful way of developing rigorous and powerful complementary insights into real-world questions. We use the epistemological concept of “models as isolations and surrogate systems” (MISS) as the philosophical underpinning of this argument. In particular, we show that agentbased and equation-based approaches align well when used simultaneously and, contrary to some common misconceptions, should be considered complements rather than substitutes. We illustrate the usefulness of the approach by examining a model of the long-run relationship
between economic development and inequality (i.e. the Kuznets hypothesis).
Working papers
Social Institutions and Economic Inequality: Modeling the onset of the Kuznets curve, with C. Bale, B. Alves Furtado, J.E. Gentile, C. Gräbner, H. Henderson and F. Lipari. Institute of Applied Economic Research (2015), Discussion Paper 204
Theoretical models of the Kuznets Curve have been purely analytical with little contribution towards an understanding of the timing of the process and the presence of additional mechanisms affecting its timing. This paper proposes an agent-based version of Acemoglu and Robinson’s model of the Kuznets Curve. In extending their analytical framework we include heterogeneity of agents’ income and a mating mechanism that together represent elements of social mobility. These two simple changes proved to be enough to shed light on the length and timing before high inequality implies regime change. Thus, this work may contribute to an effective empirical assessment of the Kuznets curve as it explicitly considers the time dimension of the process and the effects of considering social dynamics.
Santa Fe Institute, 2014 Complex Systems Summer School Proceedings
Disseminating market information via mobile phones to cashew producers: an impact evaluation in Guinea-Bissau, with Giulio Schinaia, Sebastian Schaber, Dayvikson Tavares and Adewusi Mendonça
In collaboration with ANCA, the Ministry of Economy and Finance, the Ministry of Agriculture, Nitidae and MTN Guiné-Bissau, and financial support by the MEF and two generous and competitive grants, given by PEP and PEDL, we are implementing a cluster randomized controlled trial with 2000 Bissau-Guinean families spread across nearly 300 villages. The objective of this project is to analyze how communicating information on prices and market dynamics through Market Information Systems, introducing the n'kalo service into the country, can increase the prices obtained by cashew farmers. Increased transparency and richer information should increase producers' negotiating capacity and facilitate them to better choose when to sell their crops. This information might be a particularly important support given the uncertainty derived from the COVID-19 pandemic.
Dissemination and media coverage: PEDL C-19 piece, National Conference in Bissau, PEP policy brief, Radio France International
Does group size and gender matter for performance in skill diverse teams? Experimental evidence, with Shan Aman-Rana
In this paper we investigate whether the effect of group diversity on performance depends on group size. We carried out a lab experiment with university nursing and economics students in Guinea-Bissau. Students were randomly assigned into groups with different size and gender composition to solve tests as a team. We find that in groups with homogenous skills, size of the group does not matter for performance. Coherently with our theoretical model, for students in skill diverse groups the increase in performance from being in a large group is 60% higher than when being in a small one. However, this large gain from introducing skills diversity in teams only materializes for male students. While the gender composition of groups does not matter for men, it plays an important role for female students. In a large skill diverse group, the performance of female student improves when matched with other female students, as the theory suggests, but it reduces when matched with male ones. Our findings suggest that both group size and gender play a central role in determining how skills diversity affects performance of workers.
Beliefs Systems and Health Behaviors in Guinea-Bissau, with Alexander Coutts, Teresa Molina Millán and Pedro Vicente
Gender, information exchange and choice over co-workers: experimental evidence, with Shan Aman-Rana, Clement Minaudier and Shamyla Chaudry
Selected Publications
Loss of structural balance in stock markets, with E. Ferreira, S. Orbe, J. Ascorbebeitia and E. Estrada, Scientific reports 11.1 (2021): 1-10.
We use rank correlations as distance functions to establish the interconnectivity between stock returns, building weighted signed networks for the stocks of seven European countries, the US and Japan. We establish the theoretical relationship between the level of balance in a network and stock predictability, studying its evolution from 2005 to the third quarter of 2020. We find a clear balance–unbalance transition for six of the nine countries, following the August 2011 Black Monday in the US, when the Economic Policy Uncertainty index for this country reached its highest monthly level before the COVID-19 crisis. This sudden loss of balance is mainly caused by a reorganization of the market networks triggered by a group of low capitalization stocks belonging to the non-financial sector. After the transition, the stocks of companies in these groups become all negatively correlated between them and with most of the rest of the stocks in the market. The implied change in the network topology is directly related to a decrease in stock predictability, a finding with novel important implications for asset allocation and portfolio hedging strategies.
Waving goodbye? The determinants of autonomism and secessionism in Western Europe, with M. Portos and J. Vourdas, Regional Studies 52.2 (2018), 197-211
This paper sheds light on the main aggregate-level determinants of electoral support for regionalist parties across 10 Western European countries. A region being relatively richer than the country to which it belongs is associated with higher electoral support for regionalist parties only to the extent that the region is culturally differentiated. This hypothesis is substantiated theoretically, tested empirically and found to hold in the form of a strong and significant interaction effect between cultural and economic variables. This result, omitted in previous studies, implies a profound change in the interpretation of the role of income and cultural differences in explaining support for regionalism, for both autonomist and separatist parties.
The Best of Both Worlds: Developing Equation-Based Models into Agent-Based Models, with Gräbner, C., Bale, C. S., Furtado, B. A, Gentile, J. E., Henderson, H., Lipari, F. Computational Economics (2017): 1-20
We argue that building agent-based and equation-based versions of the same theoretical model is a fruitful way of developing rigorous and powerful complementary insights into real-world questions. We use the epistemological concept of “models as isolations and surrogate systems” (MISS) as the philosophical underpinning of this argument. In particular, we show that agentbased and equation-based approaches align well when used simultaneously and, contrary to some common misconceptions, should be considered complements rather than substitutes. We illustrate the usefulness of the approach by examining a model of the long-run relationship
between economic development and inequality (i.e. the Kuznets hypothesis).
Working papers
Social Institutions and Economic Inequality: Modeling the onset of the Kuznets curve, with C. Bale, B. Alves Furtado, J.E. Gentile, C. Gräbner, H. Henderson and F. Lipari. Institute of Applied Economic Research (2015), Discussion Paper 204
Theoretical models of the Kuznets Curve have been purely analytical with little contribution towards an understanding of the timing of the process and the presence of additional mechanisms affecting its timing. This paper proposes an agent-based version of Acemoglu and Robinson’s model of the Kuznets Curve. In extending their analytical framework we include heterogeneity of agents’ income and a mating mechanism that together represent elements of social mobility. These two simple changes proved to be enough to shed light on the length and timing before high inequality implies regime change. Thus, this work may contribute to an effective empirical assessment of the Kuznets curve as it explicitly considers the time dimension of the process and the effects of considering social dynamics.
Santa Fe Institute, 2014 Complex Systems Summer School Proceedings
