ONGOING PROJECTS

 

Multidimensional measure of fragility

IP: Hannes Mueller

FFO- S05-P-01-German Government

2023-2024


The aim of this project is to support decision-making processes within the German Federal Foreign Office (AA) in a practical way through applications of artificial intelligence (AI) and natural language processing (NLP). The project is developing a multidimensional fragility measure to map fragility on several dimensions (like conflict, economics, gender and climate change) for all countries on a monthly basis. This will enable the foreign office to develop policies based on better foresight regarding these dimensions of fragility. Examples of fragility measures we intend to develop include predicting the collapse of agricultural production from climate catastrophe, predicting the fiscal environment and predicting refugee flows.

Conflict Forecast: Evaluating Intervention Options for Nigeria

IP: Hannes Mueller

FCDO-UK Government

2022-2023


This project has the goal of developing a framework for policy evaluation for the prevention and intervention of conflict in the case of Nigeria in three steps: 

  1. building on the country-level framework developed for the FCDO, we will provide dynamic estimates of the costs (in terms of fatalities, sexual violence, displacement, and economic costs for Nigeria and trade costs, ODA costs, and peacekeeping costs for the donor community, and the UK) 
  2. simulate how different intervention bundles can help Nigeria stabilize and estimate their overall expected likelihood of success and benefits and costs
  3. develop a subnational version of our framework using subnational risk estimates, cost estimates, and a subnational model of policy interventions
  4. simulate the implementation of interventions subnationally, i.e. in specific sectors and regions of the country 

The project will profit from previous work for the FCDO which will be extended along critical dimensions. We will develop this in collaboration with the FCDO’s Nigeria office to reach a tailor-made solution pointing a way out of the armed conflicts.

Development of a model and an indicator to measure the perceived well-being of the population

IP: Ada Ferrer

GGAP-Generalitat de Catalunya

2023


A research team of the FEA led by Ada Ferrer-i-Carbonell will collaborate with the General Directorate of Analysis and Prospective of the Generalitat de Catalunya in the development of a new indicator to evaluate the wellbeing of citizens in Catalonia. The research team will analyze the data from a well-being survey to develop a statistical model based on the data from this survey. The goal of this project is to evaluate the perceived well-being of people in Catalonia.

AXA POSTDOCTORAL FELLOWSHIP

2021-2023


Ursulla Mello has been awarded an AXA post-doctoral fellowship for two years, in order to support her project "Public Policies for Tackling Inequality" at the Fundació d’Economia Analítica and IAE-CSIC. She will study how governmental policy could reduce inequality in access to higher education. The AXA Fellowship is a funding scheme aiming at supporting young promising researchers on a priority topic aligned with AXA and the Society. This support should be transformative for the researcher and the advancement of her research field. Within the framework of the project, the following works have been published:

-Affirmative action and the choice of schools.

-Affirmative action and demand for schooling: evidence from nationwide policies


More information on the project in https://www.axa-research.org/en/project/ursula-mattioli-mello

Aggregation and acquisition of information in markets: positive and normative aspects

PI: Jordi Brandts and Xavier Vives

IESE

2021-2023


IESE and the Fundació d’Economia Analíta have signed a collaboration agreement for the development of research on positive and regulatory aspects of the aggregation and acquisition of information in markets. The aim of the project is to analyze the incentives of economic agents to acquire information when subsequently they have to compete and trade in the market, analyze information acquisition and trading strategies, and the welfare impact on all market participants. The welfare analysis will follow in order to prescribe policy measures. The project aims to contrast the predictions of the theoretical models with experimental evidence in simplified scenarios. In particular, it would aim to validate in the laboratory the derived behavior taxonomy obtained in the theoretical analysis. The results of the research will serve as a basis for the publication of articles on this subject.

Development of indicators of economic, political and economic policy risk, with a global perspective, based on textual analysis, through the use of "machine learning" techniques

PI: Hannes Mueller

Banco de EspaƱa

2020-2023


Political risks are key problem for developing countries and has recently also affected countries in Europe indirectly through forced migration and directly through political challenges to their political order. This has triggered several initiatives in the international donor community that aim at reducing State fragility. The Covid-19 crisis is bound to re-enforce these problems in the mid-term. In the social sciences, research on fragility has long received a lot of attention and, most recently, the issue has also been taken up by researchers outside the social science and, specifically, in machine learning.


This project has three objectives:


1.- Use the developed tools in Mueller and Rauh (2018, 2019) on the full text of millions of newspaper articles that are updated constantly to provide monthly updates of political risk to the Banco de España (BdE) and other public actors and interested researchers.

2.- Tailor the index to needs of the BdE by developing and providing additional forecasts of political and economic risks.

3.- Evaluate the ability of the framework to provide true out-of-sample forecasts.


In relation to this project, the web page https://conflictforecast.org/, has been created with the aim of helping preventive efforts and make the method accessible to a wider audience through state-of-the-art data visualization tools, as well as to make forecasts available to professionals and research teams from all over the world. The researchers involved on the project have produced a paper on “Conflict, social unrest and policy uncertainty measures are useful for macroeconomic forecasting”  where they generate monthly measures of “institutional instability” for Brazil, Colombia and Mexico and use them to predict quarterly GDP using standard methods of macroeconomics. The results clearly indicate that the institutional instability measure used is very useful for making predictions.