This is a good overview of why these methods are used. Unfortunately, a lot of people in engineering fields seem to assume that these particular tactics are th

This is a good overview of why these methods are used. Unfortunately, a lot of people in engineering fields seem to assume that these particular tactics are the only way a union can be run. Usually, the reaction to the idea that engineers should really unionise is causes an immediately hostile reaction. A union is what you (the members) want it to be, and a union that only addressed certain topics[1] would probably stop some problems from happening at all. There's no reason to authorize a union to interfere with the more mundane problems unless you want those feature and design the union to have those powers. The key thing to remember - which the usual knee-jerk anti-union rhetoric ignores - is that problems start to grow when too much power accumulates in any one place. Neither labor or capital can be allowed to grow to powerful, as it is the conflict between them that keeps each side restrained. A few decades ago we saw the problems of labor getting too powerful, and before that the problems of capital (business owners) having too much power was obvious. Like any control system, you have to design feedback[2] into the control mechanisms. [1] Issues involving collusion like the recent wage-fixing agreement between Apple/Google/Intel/{...} or the "working recent grads to death" (13 hours/day, 7 days/week) that EA used to require. [2] If only we had a sociological version of a PID controller.