Risk mitigation in evolutionary stag-hare games

Can you absorb the risk of your neighbour not going stag hunting with you? If you can't, then the Nash equilibrium is unstable and the hare-hare strategy dominates. See Bret Devereaux on subsistence farming being mostly built around risk mitigation.

Anil Gupta during a neuroeconomics guest lecture talked about how iterated stag-hare games could be used to model this type of behaviour. IIRC the gist was:

• the risk-adjusted value of your marginal currency unit gets bigger as you get poorer
• among evenly poor populations, this skews the values of the stag-hare game to eliminate the mixed strategy equilibrium
• the stag-stag equilibrium is out of reach unless you have perfect trust between players, i.e. it is not resilient to noise
• ostensibly symmetric stag hunt games between rich and poor people are not actually symmetric