Social Mobility, Social Exclusion and Social Networks – SOSOSO

We explored four areas of research.

Axis 1: Networks shape individual decisions. In many circumstances, the efforts of individuals depend on the efforts of their neighbours. Education is an important example, where school effort is a key element of academic success, which depends in a complex way on the dynamics of the classroom and friends. We approach this axis from two angles: the individual angle, to understand how an individual's relationships directly affect his or her decisions, and the social planner's angle, to better understand how public policies can affect individuals differently depending on the structure of their networks.

Axis 2: Networks insure against risks. A vast empirical literature has shown the role played by informal insurance in helping poor households and communities facing risks, particularly in developing countries. It is also well known that risk-taking during investments is essential to the economic development of a group. In this axis, we wanted to understand how the structure of the links between individuals can influence their decisions to take risks on the one hand, and to be insured in the event of a negative shock on the other.

Axis 3: Networks as providers of opportunities. Here, we want to understand how two individuals in two different networks can succeed differently because of the opportunities offered by their networks. Some examples are access to the labour market, the possibility of migration, the ease of meeting individuals from communities other than one's own.

Axis 4: Learning in networks. In this last axis, we will explore repeated interactions, to understand how individuals adjust their behaviours over time, according to the structure of the network in which they operate.

Axis 1: We have produced 7 articles in this axis. We have shown how individual decisions in terms of education can depend on the network in which these individuals are born, but also how a public policy promoting social diversity can have major effects on social mobility. We also analyzed the problem of a social planner who could shape the network in which individuals live. We have shown that networks that are optimal from his point of view can lead to the exclusion of some members.
Axis 2: We have produced 4 articles in this axis. We considered the introduction of altruistic considerations among individuals linked by a network, who can make transfers to provide financial support to those who need it. We have shown that a positive shock on an individual's income benefits all members of the network, even if they are not directly linked to each other. On the other hand, a negative shock on one will negatively affect all the others.
Axis 3: We have produced 6 articles in this axis. We explored different themes that illustrate how networks can affect the opportunities available to individuals: on migration opportunities, on their ability to integrate, on employment, on their ability to interact with other communities. In each case, the structure of the links appears to be a major determinant of why some individuals are more successful than others.
Axis 4: We have produced 3 articles in this axis. In these articles, we considered individuals who had to make decisions at every step of the way, and adjusted their choices based on the success of their previous decisions. We have shown how such learning behaviour leads to a situation in which the individuals' situation depends directly on the structure of the network in which they operate.

As is often the case in economic models, there is a gap between the intentions when writing a model and the model that is finally written and analyzed. This is a necessary process from which we have not escaped. However, a great satisfaction of this project lies in the fact that our ideas on the initial themes could be explored. While we were not able to answer all the questions we had asked in the project, we sometimes answered questions that we had not asked ourselves when we wrote the project and that proved to be interesting.

The four years we have devoted to this project have confirmed the idea that networks, the structure of links between economic actors, have a major explanatory role in the situation of individuals. Whether in terms of social mobility or social exclusion, being in the «right« network or the «wrong« network has first-order consequences. Even if these intuitions are natural, this project has made it possible to more accurately understand the mechanisms by which these networks act, on our decision-making, on the opportunities available to us, on the level of risk we can take, on the way in which public policies will affect us. These teachings are rich in two ways. First, they provide a better understanding of why some individuals do better than others, without having to their individual characteristics. Secondly, they make it possible to imagine potentially very effective public policies (i. e. that improve a situation without being too costly).
However, there remains an essential question to explore, which will undoubtedly be the subject of much future research: how to implement these public policies based on changes in the structure of relationships between individuals?

The project has a total of 20 articles published, under revision or submitted. I share the view that artificial inflation of publications is an easy but harmful practice. I therefore stick only to the papers that come directly from the project, by members of the project, on the themes of the project.
At the time of writing the project, it was planned to reach 17 publications. We have 20 of them, which is a sign of success.
I would also like to point out that all papers are published, or submitted exclusively in journals of rank 1* or rank 1. This indicator is crucial, since in economics the impact of a publication is directly related to the rank of the journal. The publication of 20 articles in 4 years in journals of rank 1* or 1 is in my opinion the best indicator that this project has been a success.

Finally, each of these papers has been presented several times at conferences and seminars. The dissemination of our work has been fairly intensively ensured through this, and continues to be so after the completion of the funding as we all continue to participate regularly in conferences where we present or refer to the work developed as part of this project.

The research project aims at analysing the impact of social networks on two different, but related phenomena: social exclusion and social mobility. The key role played by networks in explaining mobility and exclusion patterns has been largely documented in social sciences, in particular in sociology, but it has been essentially neglected by economists. Our aim is to fill this gap, and the novelty of our approach comes from the use of new concepts and tools developed in the economics of networks.
Social scientists have long tried to understand why individuals with advantaged backgrounds generally end up being economically and socially more successful. The economic literature on social mobility and exclusion has focused on the role played by traits inherited from parents (genes, cultural norms, etc.) but has largely ignored peer effects. Some exceptions have accounted for these effects, but they usually model them as some aggregate characteristic of the community in which individuals live. We believe that models of peer effects should rely on a more detailed description than aggregate characteristics, and that social networks, which contain information on relations between individuals, provide the appropriate framework.

We will study three channels through which networks affect mobility and exclusion, both theoretically and empirically. Our empirical investigations will make use of unique datasets containing detailed information on social networks. Economists have studied very few datasets containing network information, so their exploitation will constitute, as such, a significant contribution to the literature.
In a first set of contributions, we will analyse how individuals, in a given economic environment, can see their chances of success increase or decrease, depending on the network they are embedded in. In particular, we will focus on the way in which social networks can shape differently the level of effort (typically in education) that individuals choose, based on complex interdependences. Whether they come from heterogeneous externalities, from social psychological factors, or from the social planner’s action, these differences constitute powerful explanations of social exclusion and of disparities in social mobility. We will also study these ideas empirically, using data from the National Longitudinal Survey of Adolescent Health, a survey of 132 middle and high schools in the United States.
In a second set of contributions, we will analyse how networks affect the individuals’ economic environment, by protecting them differently against harmful events. In developing countries, coverage against shocks is often implemented through informal insurance, which is known to take place within social networks. We will analyze important issues that the literature has left aside: the relationship between the nature of the risk and the way individuals manage these risks; the impact of formal insurance on the informal arrangements which take place along network lines; the introduction of subsistence considerations. We will test our theory’s predictions on two datasets containing network information: One on 75 villages in South India, the other collected in 48 communities in Lesotho.
Our third objective is to analyze how social networks provide different access to opportunities, and the impact this has on mobility and exclusion. We will investigate how networks facilitate the transmission of human capital or the access to the labor market. We will also introduce the views, accepted in other social sciences, that individuals interact in several networks, and study what this implies in terms of mobility and exclusion. Last, we will empirically examine how social structures affect the ability of a community to provide public goods, using original data collected in 121 villages in rural Mali. This should bring insights on the best way to introduce public policies in a community, as a function of its social structure.