Investment

Performance on public goods games is odd.  Here is how a standard game works:

• There are a number of players, each with a pot of money;
• The players are not in direct contact with one another;
• Each player has the opportunity to make a donation to a common pot from their personal fund if they wish;
• Once all donations are made the common pot is multiplied by a number greater than 1;
• The new total common pot is divided equally between all players;
• The game can be repeated for a number of iterations.

It would be a rational response not to donate and to sit back and simply receive a portion of the common pot at each round.  This would be to free ride, but if a significant number of individuals did this there would be a tragedy of the commons and the whole system would collapse.  Public goods and collective actions are vulnerable to this destabilizing force.

Whilst a certain proportion of people do free ride the majority do not in public goods games, and indeed they tend to make quite generous donations. And whilst the amount of these donations drops off over a number of iterations it still remains quite high - the source of this decision is something that perplexes scientists.  This paper is just one exploration of the complexities involved.

It is possible that donations in such situations convey social information.  Indeed, Simon Gachter recently commented, in a keynote speech at the European Human Behaviour and Evolution Association, that giving players the opportunity for a minimal social interaction prior to the game increased donations significantly.  This yields the possibility that people think their donations might be judged in some way.  Perhaps they signal commitment, reliability and so forth and in so doing cement social bonds, or initiate new connections, that could prove useful in the future.  So in a time of need having a reputation as someone who was generous and committed resources previously will increase the chances of aid.  In this sense donations act as a stimulus package for future reciprocity.  Behaviour on public goods games might represent a form of future oriented bet hedging.

The preceding comments have not captured a novel thought.  But if it is true it would be interesting to know how increased investment might be encouraged.  Perhaps there are particular ways of indicating future uncertainty that would lead to higher donations?

There are also issues of maintenance.  For example, some employees start a new job and are eager to put their shoulder to the wheel and invest in the institution.  We should predict that this investment will drop off a little over time, if this situation is a public goods game, and we should also predict that any disappointing lack of investment on the part of the institution or employer will lead to a diminishment in donation.  It is invariably the case that the institution will behave like this at some point and effectively reduce the multiplier below zero at times.  Some people, in spite of institutional failings, remain loyal and keep donating at a relatively high level.   Others, as soon as they are 'let down' radically reduce their contributions.  Why is this?

It seems there are different thresholds of tolerance.  Some folk play the long game and are prepared to ride out a fair few dips in the multiplier.  (I suspect they are fairly tolerant of free-riders too.)  Others need less information to make a decision to shift.  In other words, discounting functions differ across people, and many things may contribute to this including personality variables and developmental trajectories.  This kind of bet-hedging should be ecologically sensitive but under selection too.

As I get older I am increasingly intolerant of free-riders and institutional inadequacies.  It could be that I only have so many donations left in me and I am running out of time.  But as I am in my early middle age and have more time left in work than I have so far contributed I doubt this is the full answer.  What I fear is that I have factor analysed my experience, realized that there are limited types of fellow player and strategy available, and I now need less information to ascertain just what I am dealing with.  As I have no power I cannot effectively punish to improve my odds.  This leaves free-riding or a partial reduction of contribution... but that feels wrong...