This concept is not as bad as it sounds from a tower utilization perspective. The WCDMA standard is wasteful in spectrum terms, so this will actually help make up for it. Since it will be splitting the load, it would be easy to use this to throttle a phone to a neighboring tower that may not be as busy, while not diminishing overly much the quality of service. However, in situations where towers are ALL full, the effect will be minimal. Ala, San Francisco / New York.
This is all just a band-aid any way. GSM carriers in north america have completely missed the first LTE boat, which is the future of wireless. That said, it is also royally screwed in north america as the spectrum is so heavily divided, and there isnt even a standardization of the exact technology being used.
Thus, LTE will be more fragmented than any other wireless standard in the US, and as such will likely end up with Verizon owning the market by attrition from others having not jumped in early.
That doesnt benefit us as consumers however.
Frankly, as much as I am opposed to this, cellular networks need to be controlled by a single body at the spectrum level, and let the carriers do what they need to in order to provide service and still be competitive. With even 4 big carriers all trying to use a very limited amount of spectrum, serving large metropolitan areas well is hard, especially with WCDMA.