I'm afraid the work-blogging balance is tilting in favour of the former for a bit whilst I finish off a project. But there's a very interesting paper here (PDF) from about ten years ago by Linda Babcock and George Loewenstein that looks at how biases can lead to stalling pay negotations and even strikes.
The whole paper is worth a read. Although much of it is based on experiments they did also analyse school pay negotiations in the US. Some of the findings will come as no surprise - for example, both unions and employers are self-serving in terms of selecting comparator groups for wage rates. But it does point up (to me anyway) how useful developing an understanding of these types of biases could be to the labour movement. I'd be very interested to hear if any unions haved looked at this or plan to do so.