Current Version Slides ArXiv: 2403.13983 Online Appendix
How weak incentives (such as slight moral concerns) affect strategic communication
I study communication in settings where the sender's and receiver's preferences are almost independent. Previous literature studies the case where preferences are precisely independent, where communication is theoretically possible, but incompatible with any realistic uncertainty about agent preferences. This compatibility can be restored if preferences are 'nudged' in a compatible, state-dependent manner — such as moral concerns about lying or misleading the other party. I characterize communication in this setting through graph-theoretic representations of equilibria called communication graphs. Compatibility restricts these communication graphs to be acyclic and connected, and requires compatibility between the graph and the dependence in the sender's preference.
Sympathy (ie. aligning preferences, associated with consequentialist ethics) promotes communication in quantitative settings, but can impede communication in similar 'qualitative' settings — such as a seller possessing vertically differentiated products. Money-burning (e.g. advertising) can be a powerful persuasion tactic in the latter cases.
Current Version Slides ArXiv:2609.06151 Online Appendix
Why moral actions are rational, even in an extremely (but not entirely) selfish world
I model a large anonymous society where individuals repeatedly face a dilemma between acting selfishly or contributing to social surplus. If agents care even slightly about the social surplus then pro-social norms can be sustained through a desire to set an example for others if agents are sufficiently patient.
In contrast to other models of cooperation in this setting, these norms can be robust to a sub-population of purely selfish actors. There are a continuum of robust equilibria, that under weak assumptions can be identified with a hyperplane of distributions over the natural numbers. Equilibria that feature higher strategic homogeneity (more concentrated distributions) are more robust.
More robust equilibria can be found by increasing patience or altruism (which are substitutes in this sense), while increasing the publicity of bad actions — beyond the modest level of one observer on average — has no effect..
Current Version Slides ArXiv:2602.23098
Summary: I show that powerful techniques of analyzing communication and repeated games share the same general structure, making the resulting equilibria incompatible with private preferences. I discuss methods of amending the technique and analyze the resulting constraints on the structure of repeated game equilibria with imperfect monitoring.
This has severe implications for equilibria of repeated games with imperfect monitoring: pure equilibria are necessarily perfect public equilibria, non-PPE equilibria have a 'reputation' structure that I describe, and belief free equilibria do not exist.
Summary: I study how heterogeneous altruism (linked to varying opportunity costs of participating in norms) associated with wealth produce a negative link between inequality and pro-social norms/social trust, and that unbalanced growth can erode norms despite increasing altruism at the individual level.
I characterize several transformations of the wealth distribution that promote social trust: wealth floors always have a positive effect, while wealth caps can have a positive effect on trust in unequal societies (despite reducing the altruism between agents). Uniform wealth and progressive wealth redistribution also both promote trust.
Summary: The presence of other agents with similar preferences over limited options introduces an externality due to the increased likelihood of being outcompeted and thus obtaining a less preferred option. I study how investment (in schools, neighbourhoods, or jobs) distorts preferences, and thus creates such an externality.
In particular, investing in more exclusive objects (according to the market) imposes a negative externality on those matched with less exclusive objects; while funding less exclusive objects has a positive externality on those matched with more exclusive objects.
'Bottom-up' investment in objects, reducing object inequality, is guaranteed to produce a positive externality while 'top-down' funding produces negative externalities. Gentrification specifically occurs when investment is unbalanced, increasing the exclusivity of an option relative to others.
This reasoning suggests that efforts to 'hierarchize' the market (for example, vertically ranking schools, or a culture emphasizing 'match prestige') reduce the efficiency of the market.