UCLA

This dissertation consists of three independent chapters that discuss social networks (defined in the broad sense) in different angles.

Chapter 1 studies how an agent's propensity to accept bribes depends on the organizational structure, which we model with a broad set of random networks that contains two canonical special cases. In hierarchies, agents' best responses exhibit strategic substitutability, with bribe taking being risker if others accept more bribes, for it is then easier for a corruption investigation to trace through bribe transactions to locate bribe takers. On the contrary, best responses in flat, two-layer networks feature strategic complementarity, as more bribe acceptances better protect criminal subordinates from being caught, reducing the risk of bribe taking. While incentives differ across networks, we show that for any of our random network, in equilibrium, increasing its density always deters agents from accepting bribes. Nevertheless, opposite results for hierarchies and two-layer networks are obtained if we make the number of subordinates each agent monitors more evenly distributed. We use this model to point out a corruption identification problem and propose a remedy to it.

Chapter 2 studies how parochial fairness concerns – a player's incentives to compare wages with those in the same group – affect group deviations. We propose a new theoretical framework based on the transferable-utility cooperative game through extending the utility space to incorporate in players' other-regarding incentives. We then apply in the Fehr-Schmidt utility specification to study how parochial fairness concerns govern income redistribution outcomes after coalitional deviations and the structures of core allocations. We find that while both disadvantageous and advantageous inequality aversion exacerbate income inequality after a coalition deviates, advantageous inequality inclination ameliorates it. In addition, if players are moderately averse to advantageous inequality, the grand-coalition allocation most robust to coalitional deviations is the "tyranny-of-the-majority" allocation that gives the single poorest player indefinitely small amount and equates the other players' incomes.

Chapter 3 studies the rewards-based crowdfunding industry. We ask how a creator of a crowdfunding project optimally designs the pricing strategies for rewards to maximize the total fund raised. We build a structural model based on [Bre87] and estimate the project and reward values and the distribution of backers’ preferences with data on reward prices and the numbers of backers buying each reward. We find that backers’ preference distribution heavily weighs towards the lower side -- most backers are of low types; and that a creator optimally employs a bow-shaped pricing strategy -- they extract most of the surplus from low- and high-type backers and charge those in the middle little to no premiums.