Forum Numerica - Social Percolation

To watch in full screen mode, start the video and click on the "UniCApod" logo

This is a project that uses the percolation framework from physics to study diffusion processes in social systems. Understanding diffusion processes is key to market strategies as well as innovation and sustainability policies. In promoting new products and technologies, firms and governments need to understand the conditions favouring successful spread of these products. In a first paper we consider a new product diffusion in a social network through word-of-mouth. Given that consumers differ in their reservation prices, a critical price exists that defines a phase transition from a no-diffusion to a diffusion regime. As consumer surplus is maximised just below a product’s critical price, one can systematically compare the economic efficiency of network structures by investigating their critical price. Networks with low clustering were the most efficient, because clustering leads to redundant information flows hampering effective product diffusion. We further showed that the more equal a society, the more efficient the diffusion process. In a second paper we have used the percolation mechanism for the diffusion of ideas, focusing on the interplay between individual preferences and social influence. The debate on diffusion in social networks has traditionally focused on the structure of the network to understand the efficiency of a network in terms of diffusion. Recently, the role of social reinforcement has been added to the debate, as it has been proposed that simple contagions diffuse better in random networks and complex contagions diffuse better in regular networks. In this paper, we show that individual preferences cannot be overlooked: complex contagions diffuse better in regular networks only if the large majority of the population is biased against adoption. More recent work has focused on the role of homophily among agents in diffusion processes, and on the sequential diffusion of different innovations for understanding societal transitions.

Paolo Zeppini has a master in physics, a master in quantitative finance, a research master (MPhil) in economics and a PhD in economics. In his PhD at the Tinbergen Institute of Amsterdam he studied behavioural aspects of technological change, with focus on bounded rationality, technological variety, lock-in problems, and how behaviours shape technological patterns and have an impact on the natural environment (Thesis: Behavioural models of technological change). In a post-doc at Eindhoven University of Technology (grant “Open Competitie” Path creation and path dependence in a model of recombinant innovation) Paolo has developed models of diffusion on networks and technological transitions. He studied how technological progress can be generated as a coordination phenomenon and cooperation can trigger transitions that circumvent technological lock-in into polluting technologies and foster a paradigm shift towards cleaner technologies and renewable energy. Paolo later worked in Bath (UK) as a research fellow within the GLAMURS EU FP7 project: Green Lifestyles, Alternative Models and Upscaling Regional Sustainability, and later became lecturer in economics. Here he developed and calibrated micro-economic models of consumers’ behaviour for sustainability transitions, extending economic decision making with state-of-the-art psychology theories of social innovation. Paolo carries a number of international collaboration with the University of Amsterdam, Utrecht University, the Autonomous University of Barcelona and LUISS University in Roma. Before joining academia Paolo had a significant experience in the private sector, having working for three years as a financial trader in an investment bank. In the trading room he has been in charge of exotic derivatives pricing, daily market making activity on government bonds, proprietary trading on fixed income assets, interest rates derivatives and currency futures, and the development of automatic trading tools based on multiple regressions of Bloomberg real-time data.