Networks, social information and compliance

Borzino, Natalia (2016) Networks, social information and compliance. Doctoral thesis, University of East Anglia.

Download (5MB) | Preview


    The work developed in my PhD on “Networks, Social Information and
    Compliance” focuses on compliance (voluntary and non), diffusion of information
    and spillovers in diverse network structures. More specifically, we test
    experimentally how minimal social information is diffused through different
    network structures and its role on increasing the level of efficiency along with its
    positive effect on voluntary compliance of emergent social norms and tax
    compliance. In the first two chapters, we implement a networked version of the
    trust game with two senders and one receiver. We manipulate in a minimal way
    the social information available in the network and, the novelty of our work
    consists in the introduction of non-binding suggestion about the level of trust and
    trustworthiness, which is totally fair and (partially) efficient. It is also manipulated
    in the two studies the selection mechanism of the roles in the game by introducing
    social status. Our findings suggest that social information has a positive and
    significant effect on increasing the level of trust in the network. The non-binding
    suggestion has also a positive and significant effect on individual decisions. In the
    last chapter, we study in a laboratory experiment how tax compliance information
    is diffused in a fixed-six-nodes circle network. The game has four information
    conditions: No Info, Full Info, Positive Info and Negative Info. In the No info
    treatment, subjects get individual information about whether they were audited,
    the outcome of it and her final payoff. In the Positive Info (Negative Info)
    treatment, participants get information whether adjacent connected nodes were
    audited and found compliant (noncompliant). In the Full Info, participants get
    both positive and negative signals. We control for the effect of signals on
    participants’ beliefs on the ex-ante fixed and unknown audit probability by an
    incentive compatible mechanism. The tax rate and fine rate are fixed and known
    by the subjects. Our findings suggest that positive and negative signals have a
    significant effect in the levels of reporting and compliance at individual level.
    Indeed, diffusion of non-strong negative signal (one bad example) has a negative
    effect on individuals’ tax compliance. The diffusion of strong positive signals
    (two good examples) is required to generate any increase in compliance decisions
    within networks.

    Item Type: Thesis (Doctoral)
    Faculty \ School: Faculty of Social Sciences > School of Economics
    Depositing User: Katie Miller
    Date Deposited: 21 Jul 2017 14:49
    Last Modified: 21 Jul 2017 14:49

    Actions (login required)

    View Item