... How plausible is that? Dunno. In a racially-diverse society, especially a poorly integrated one, people of different races might well tend not to treat one another in ways worthy of trust. But it's complicated: they might also be more reliable towards others of the same race. In such a society, anyway, there'd be no such thing as "the degree to which someone should actually be trusted" because it would depend on who's considering trusting whom. Not to mention, e.g., that in such a society there could be a lot of divergence between how much you "should" trust someone else for your own sake (less) and how much you "should" trust them to help fix the messed-up state of society (more), etc., etc., etc.
Some things that aren't clear to me about Putnam's work, and that seem important: In racially diverse societies, is it only inter-race levels of trust that are lower? Putnam's work, IIUC, is concerned with the USA; are things different in places where there isn't a history of slavery? or where racial diversity on something like equal terms has been going on for longer? What actually happens over time in these less-trusting more-diverse places? Is it specifically highly-visible racial diversity (black, white, ...) that is associated with these differences in trust, or do you get the same thing with religious or cultural diversity that doesn't have quite such visible markers?
Oh, I can answer one of those. According to http://abstractnonsense.wordpress.com/2006/10/09/ethnic-dive... Putnam found that intraracial trust is less in more-diverse settings, as well as interracial trust. That's a shame.