Use this URL to cite or link to this record in EThOS:
Title: Essays in networks and behaviour
Author: Carpenter, William John Wakefield
ISNI:       0000 0004 7660 9891
Awarding Body: University of Cambridge
Current Institution: University of Cambridge
Date of Award: 2019
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
This dissertation, submitted for the degree of Doctor of Philosophy, is comprised of three essays on networks and human behaviour. The first studies the capacity for some agent in a network, who we term the 'principal', to ensure the robustness of their network to potentially disruptive external shocks. Agents susceptible to shocks may invest in contingencies but have insufficient incentive to do so from the perspective of the principal. We derive results on the cost to the principal of overcoming the resulting moral hazard problem. We focus first on the case where agents can only write contracts with their immediate neighbors, finding that costs are strictly increasing in the longest path in the network away from the principal when the probability of shocks is low. However, we also find that for a fixed longest path away from the principal, a secondary 'responsibility' effect can cause costs to be lower when the average distance away from the principal is higher. We extend our model to allow the principal to contract directly with all agents but with additional information processing costs. We consider applications to supply chain robustness to natural disasters and employee adherence to safety regulations in a firm. The second essay develops a model of demand led innovation in supply chains and studies the general equilibrium effects of demand shocks on R&D expenditure. It is found that when no final good producers share suppliers, a positive demand shock will weakly increase innovation in all firms. However, with common suppliers, negative pecuniary externalities may reduce R&D expenditure in some firms. The final essay was produced in collaboration with Dr. Matthew Elliott. It proposes a new model of individual choice as a Markovian decision process over emotional states. This approach allows for economic decisions to depend upon both an individual's current emotional state as well as introspection about how choices will affect their future emotional state. The generality of this framework is shown formally before demonstrating its efficacy in dealing with specific applications drawn from empirical findings in psychology and neuroscience. Specifically, the role of frustration in gambling is explored, along with the role of angry emotional responses in generating inefficiency and delay in negotiation and bargaining.
Supervisor: Elliott, Mattew Sponsor: Economic and Social Research Council
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
Keywords: Economics ; Networks ; Behaviour