Honesty and Trust

I was recently involved (peripherally) with a business partnership that broke up. The partners did not see eye-to-eye. Their lack of synchronization was not immediately apparent, but it became more so with time. Six months into their venture, they decided to part ways. That got me thinking about honesty, trust and models for those social dynamics. Specifically, I seem to recall that a natural clustering occurs between social entities that act honestly versus those who don't. Naturally, I hit the Web looking for answers.

It turns out to have been the great economist Adam Smith who observed that market interactions lead to "bourgeois social virtues" like honesty. Go figure. The quote is not from Smith (he did not use words like "bourgeois") but from Francis Fukuyama, whose paper at the IMF is referenced below.

Social models of trust seem to be based on incentives and costs (economic models). That approach has worked to a degree, although most economists will readily admit that it doesn't explain significant types of human behavior (e.g. Open Source Software). Not surprisingly, much of the academic literature focuses on either (a) describing trust systems based on costs and incentives or (b) suggesting incentives to make participants in a system more honest.

In the literature of evolution, there are many well-known examples of deception such as hidden ovulation in human women (unlike most other mammals) and uncertainty relating to paternity. There are strong reasons for such deception and so the ability to deceive is actually encoded in our genes! That sheds an interesting light on our social norms, doesn't it? I have come to think of social norms for honesty arising from incentives and costs, but being continued and expanded as a meme - an idea package in a Dawkinsian sense that takes on a life of its own.

There is an interesting paper on "social capital" (of which trust is an effect) and civil societies at the International Monetary Fund. It contains some useful definitions and references:

In anthropology, studies of the great apes have shown that their systems of behavior are similar to ours, but simpler. That provides a nice avenue for study. Rebecca Frank's work at UCLA made sense to me. She discusses social interactions of baboons, where coordination is valued to attain group goals so repeated deceptions become "unprofitable".

Come to think of it, I think I saw that kind of behavior on Animal Planet's Meerkat Manor show. One of the meerkats (Tosca, I think) was tossed out of the group for bearing young when she was not the dominate female. That was an unacceptable social interaction for which she was punished.

There are papers that demonstrate clustering of honest and dishonest people, such as Spatial Structure and the Evolution of Honest Cost-Free Signalling by Krakauer and Pagel. Interestingly, they suggest that islands of honesty can evolve even in situations where there is no cost and a net reward for deception.