Sorry for the late reply. It's true that unions tend to have more success forming in monopsony employer situations, usually because of the labour force abuses that tend to follow. The other context where unions have high coverage (and tend to be unusually militant) are industries where the job carries a high degree of risk for the worker.
Historically this risk has almost always taken the form of physical danger (e.g. miners, steelworkers, construction workers etc.). Whether intellectual labourers will be motivated to organise by less proximate, but no less real, risks (e.g. the risk of being automated out of a job, being denied ownership of one's own IP/thoughts etc.) remains to be seen.
The other interesting factor is that there seem to be parts of the tech industry where a worker produces orders of magnitude more surplus value for their employer than they are paid for their efforts. I suspect this may have been a factor in the unionisation of screen entertainment workers and professional sports people (particularly when radio and TV turned those markets into 'winner-takes-all' markets).
And I'm not sure I'd personally characterise many of the markets in the tech industry as competitive. The most lucrative markets seem to be either dominated by a de facto monopolist (e.g. search, social networking) or duopolists (e.g. Windows vs OSX, GFX cards and CPUs, the android and iOS platforms etc.).
