My last post pointed out a fundamental distinction between effects of diversity on trust and effects of demography on trust:

If, as in Putnam’s original story, diversity per se were really bad for growth, segregation would sharply raise average trust.  Indeed, segregating two communities could conceivably raise trust in both
communities.  This is what makes diversity a special social variable. 
If diversity in and of itself has bad effects, so does integration –
regardless of the characteristics of the mingling populations.  If the
effects of diversity are demographic effects in disguise, however,
integration has distributional effects, but is zero-sum overall.

Is there any way diversity could end up being a social positive, rather than merely zero-sum?  Sure.  The top mechanism to consider:

Standard economic theory says that good things have decreasing marginal utility and bad things have increasing marginal disutility.  Doubling the quantity of food less than doubles the social value of food.  Doubling the quantity of pollution more than doubles the social harm of pollution.  Now suppose you can either have a segregated society, where half the population has 25% trust and half has 75%, or an integrated society, where the whole population has 50%.  While average trust is the same in both scenarios, the social effects of trust should be, on net, better in the integrated society.  Why?  Because the net benefits of moving from 50% to 75% trust will, by standard economic logic, be smaller than the net benefits of moving from 25% to 50% trust. 

One variant on this story: Innovation itself might vary non-linearly with trust.  Very low trust might choke off innovation entirely, while moderate trust provides a solid foundation for dynamism.  Returning to the 25%/75% scenario, homogeneity allows growth only in the high-trust enclave, while diversity allows growth in both.

Are such effects genuine?  Unfortunately, I’ve seen few papers that test for non-linear benefits of trust, except for this paper finding negative marginal effects of high trust on GDP per capita.*  Anything I’ve missed?

* Nor should we forget the historic crimes that homogeneous societies like Germany and Japan have inflicted on out-groups and dissidents.