Skip to main navigation Skip to search Skip to main content

Essays on the Economics of Prosocial Behaviour

  • Derrick Xu

Student thesis: Doctoral ThesisDoctor of Philosophy (PhD)

Abstract

This dissertation examines how external shocks or factors, including environmental disasters, charity scandals, and emotional appeals, influence prosocial behaviours. The first study investigates whether flood exposure increases pro-environmental behaviour. Using donation records from 90,000 donors and longitudinal surveys, I show that individuals directly affected by floods perceive their environmental efforts as insufficient and increase support for environmental charities and the Green Party. However, this effect is absent among those whoseneighbours are affected, even within a 200-metre radius. This suggests that personal relevance is key to behavioural change. Additionally, universalists increase green giving even without direct exposure, highlighting the role of moral values. The second study examines how reputation shocks to one charity influence donations to other organisations. While reputational damage could spread, I find that similar charities benefit from increased donations, suggesting that donors treat them as substitutes. This substitution effect occurs only among charities with closely aligned missions, identified using a novel text-based similarity measure rather than traditional sector classifications. This approach provides a novel way to characterise the charitable market. The third study explores why charities frame their causes to evoke negative emotions. I propose two explanations: first, charities maintain a consistent level of negativity to align with donor preferences. Second, they use negative emotional shocks to drive donations. To test these hypotheses, I trained a deep learning model to analyse the emotions evoked by three million images from charities’ media posts. I find that charities’ emotional tones vary substantially in ways not explained by their missions. Donor preferences also differ by socio-economic background and moral values, supporting the first hypothesis. Furthermore, I show that negative emotional shocks effectively boost donations, their impact diminishes with repeated exposure, consistent with the second hypothesis.
Date of Award30 Sept 2025
Original languageEnglish
Awarding Institution
  • University of Bristol
SupervisorYanos Zylberberg (Supervisor) & Sarah L Smith (Supervisor)

Cite this

'