I’d say this touches on it a bit:

http://www.tvhe.co.nz/2013/07/31/lucas-critique-and-dsge-models/

A couple of things I’d keep in mind though:

1) Micro “works” sure – but what this means depends on the question being asked! For “macro” questions, micro doesn’t provide unique answers.

2) Macro variables “supervenes on” microeconomics. Individual agents are making choices, but with reference to aggregates. In this context, when asking macro (and even a broader set of) questions we need to consider how this use and evolution of aggregates impacts upon behaviour. One of the key channels of analysis is by thinking of these things as a form of “coordination game”.

I’ve been thinking, while blissing out in holiday time out reading Skidelsky and Sheehan on Keynes, that if micro works, why did Keynes invent macro?

True! Economists have many models, with differing assumptions, begging for elements of the individual to be observable, measurable, and thereby making them testable.

But they aren’t, so instead we just have to enjoy making fun micro models, yus! 😀

After reading all that I’m still not convinced that economists have any good models that explain social interactions 😉

http://www.igmchicago.org/igm-economic-experts-panel/poll-results?SurveyID=SV_1z4X7kmHnVYo28d

However, we also shouldn’t forget that sometimes people self report wanting cash:

http://www.tvhe.co.nz/2010/12/17/all-i-want-for-christmas-is-cash/

And that even with the signal and non-pecuniary benefits mentioned, we also need to think of the opportunity cost – there are other means to provide these same benefits that aren’t tied up in a costly “arms race” 😉 :

http://infometrics.co.nz/Forecasting/5203/901/An-economist's-Christmas-present