Understanding income distribution through a public goods game on a dynamical network

Pathak, Spandan (2018) Understanding income distribution through a public goods game on a dynamical network. Masters thesis, Indian Institute of Science Education and Research Kolkata.

PDF (MS dissertation of Spandan Pathak (13MS130)) 13MS130.pdf - Submitted Version Restricted to Repository staff only Download (8MB)

Abstract

Human beings, being social creatures, have always had a preference towards fairness which in turn is correlated to largely equal distribution of resources and mutual cooperation. Any individual deviating distinctly from such scenario is expected to be punished by the existing social structures through different regulatory mechanisms. However, throughout the course of history, large and ever-increasing inequality in wealth has always been a sad reality. To understand the causes and impact of this wealth inequality in a society, numerous experiments have been performed in a laboratory setup involving networked public goods game where the players have the same incentives of losing or gaining wealth as in real-world economic scenarios. These studies have investigated the role of underlying network topology, network fluidity, punishment and signaling mechanisms, participation cost etc. on the sustenance of cooperation within the population. Recently, Nishi et al. have carried out an interesting study highlighting the effects of visibility and invisibility of wealth on the evolution of income-inequality of a population whose underlying structure is described by an evolving network. Their work indicates that it is the visibility of wealth, but not the inequality itself, which leads to higher inequality, lesser wealth accumulation and lower connectivity within the network. Our aim in this thesis is to explore in detail the underlying principles behind such empirical observations. We do so by formulating decision-making rules that can explain the empirical results. We further extend the analysis to understand how individual decisions affect network restructuring and eventually global metrics like average wealth and income inequality in a variety of situations. We find that the underlying spatial structure, inherent tendency of cooperation at the beginning of the game along with weightedness of the network has very little effect on the final equilibrium-like state of wealth distribution. As an interesting contrast to previous results that higher network rewiring rate can sustain the higher level of cooperation, we find that beyond an optimal level of network fluidity, inequality increases again. Finally, we investigated the consequences of implementing taxation within the system and tried to obtain the conditions under which income equality reduces. We observe that a higher tax-rate leads to greater average wealth in the population, but breeds higher inequality as well. We further show that global wealth redistribution by a central tax-collection authority along with the introduction of differential tax slabs for different wealth groups reduces inequality without a decrease in cooperation rate, connectivity and any significant associated reduction in average wealth.

Actions (login required)

View Item