Why LightSquared failed: It was science, not politics.
Good summary of the battle between LightSquared, and GPS users/makers. On the political/interest group side, it shows that established industries have ability to protect their regulatory interests, even against a well-financed lobbying campaign. (LightSquared) Perhaps it also shows that the FCC is also able to make technically appropriate decisions, at least when the two sides are approximately balanced in political power.
Of course it might have been “socially better” to enforce GPS receivers to limit their frequency use, from the very beginning, thus avoiding the conflict with LS or anyone else proposing to use those frequencies. But there’s no gradual trajectory that will get us there now – the article suggests that it would have been too complex for the FCC to design and regulate an evolutionary path.
On the flip side, proposed users of new unlicensed spectrum still get little attention or lobbying clout. A casual cost/benefit analysis suggests that per MHz, Wi-Fi has had the highest social value of any recent spectrum uses. I’d love to see someone analyze this.
Art and Science in Technology - Roger Bohn's Blog 
I’d also be curious about the CBA on slices of wireless spectrum.
On a purely intuitive and speculative level, I want to guess that WiFi would appear so valuable because of its limited range. Is it really that valuable when every household in a neighborhood is creating redundant networks in the same spectrum in the name of security and privacy?
And while we’re at it: the average person spends maybe $50 a year on WiFi gear, but $250 a year for their 3G spectrum if you count just a basic smartphone data plan. Does that say something about social value?