Relationships and economic transactions

The Leitmotive of my dissertation is the role that the networks of relations existing among economic agents play in their behavior and, thus, in economic outcomes. While networks play a crucial role on many dimensions, one of the most prominent and investigated features is that they are privileged channels of information diffusion. The first and second chapter of this thesis explore instances where social networks are information channels. They both belong to the stream of network literature focusing on links at the individual level and their relevance with respect to labor outcomes. The first chapter focuses on the hiring stage and the potential wage differential imputable to being hired through social networks. The chapter develops a theoretical framework that takes into account the effects of moral hazard concerns and the diversity of networks and cultures on the choice of hiring channels. This model rationalizes the emergence of either informal or formal hiring channels and of either positive or negative wage differentials for workers hired through informal channels, depending on circumstances. In particular, the prevalent way of perceiving the use of networks as hiring channel is determinant. Indeed, a worker hired through a personal contact may either feel indebted and grateful or, on the contrary, she may think that she is somehow favored and does not really need to exert as much effort as the other workers. Conditional on being employed, in contexts where favoritism is prevalent, social networks are likely to be adopted as hiring channels for unskilled jobs and to result in wage penalties and even more, the stronger the ties. In contexts where gratitude predominates, however, the opposite happens. The empirical analysis is based on firm and individual matched data from the 2003 Investment Climate Assessment survey of the World Bank on Senegal's manufacturing formal sector. It estimates an endogenous switching model, allowing for both endogenous selection and switching effects of formal and informal hiring channels. The results for Senegal are consistent with the theoretical predictions in case of favoritism: informal hiring channels are preferred to fill unskilled vacancies and are associated with a wage penalty. Moreover, the probability of having been hired through a social network and the absolute value of wage penalties are increasing in the strength of ties. The second chapter focuses on the role played by networks during the career progression, independently of the hiring channel. It takes into account multiple mechanisms whereby networks have an impact on career outcomes and their dynamic coevolution. Indeed, professional links affect salary not only directly if they are valuable to the employer, but also indirectly through mobility since they represent privileged channels of information diffusion about job opportunities. Moreover, professional links have an impact on labor outcomes, but at the same time a worker's network is substantially shaped by career choices. The chapter develops therefore a dynamic framework where the utility of workers is affected by the characteristics of their professional network and by the mobility decisions they make during their career. The optimal value depends on professional networks directly through current utility to the extent that links are valuable to the employer and indirectly through the effect that they have on mobility decisions. The empirical analysis is based on individual and firm matched information provided by BoardEx Ltd, a UK supplier of data to headhunting companies, describing the career history of thousands of Executives in US, UK, France, and Germany. The data support the main hypotheses of the theoretical setting and the insight that professional networks are relevant both because they are valuable for the employer and because they facilitate job mobility. These findings are robust to alternative definitions of career value and specifications accounting for mobility and link endogeneity. We also find that contemporaneous colleagues are not a useful component of a worker's network. Moreover, networks characterized on average by ties between nodes that have been colleagues for a long time have a lower direct and indirect effect on labor outcomes. Finally, networks where many links are represented by workers that have been colleagues a long time before are less valuable to the employer. The third chapter investigates the relations between organizations and suggests that they may provide them with commitment ability. It models risky projects with autocorrelated productivity shocks as creating an option value of investing over time so that later investments benefit from the information revealed by the realization of earlier investments. However, once a firm has invested in a production project, lobbies within it (e.g., some divisions of the firm may have divergent interests from those of the Head Office or Board of Directors) or outside it (e.g., there may be political pressure as well as pressure from upstream or downstream trading partners) may pressurize into paying out early revenues from such investments precisely when the autocorrelation of productivity implies the firm should be reinvesting them in the project. An alliance with one or more other firms characterized by joint ownership of a production project may then provide a commitment mechanism against lobbies' pressure and therefore enable more efficient levels of investment. The same argument applies when decision makers are governments instead of firms, as often the case for infrastructure projects in developing countries. The Business Environment and Enterprises Performance survey data corroborate the model's prediction that organizations under conditions favorable to internal or external lobbying pressure are more likely than other firms to choose joint ventures, a common form of joint ownership, as their corporate governance structure. Moreover, the estimation of an instrumental variable probit model suggests that this effect is actually downward biased, consistently with the theoretical model. Indeed, although joint ventures are more necessary when a firm is under potential lobbying pressure, they also contribute to reducing such pressure. In conclusion, the three chapters of this thesis explore some instances where networks are key to economics, contributing to the shift from atomistic explanations toward a more relational understanding of economic phenomena.