That's a tough question right now Dlight, on one hand its always good to look for the best way to maximize the value of your product and one way is by extending the effective life of the product. Other ways are to maximize value with a path to upgrade the technology, and another is to provide the most value now with an expected short life cycle of the product but much lower cost to entry. In my opinion the decision should not be driven by bandwidth requirements. Operators want to maximize the value of their investments too, so thinking these networks will be reserved for only HIGH bandwidth products and out of reach for lower bandwidth applications is shortsighted in my opinion. LTE will offer QoS and by doing so the operators will have flexibility they have never had before, this means more competitive pricing if you are willing to accept a QoS throttled solution, in other words the operator can carve out a bandwidth model that fits your requirement and probably offer it at more competitive pricing than they do today because the transport costs (or cost per bit) to run an LTE network are lower than any other network, why? because the networks are are more efficient. Here is a comment from a study done by Motorola recently
Lower cost per bit
The combination of LTE’s increased spectral efficiency and flexibility, added capacity, and simpler network architecture should offer a very cost effective value proposition. For example, each EUTRAN cell will support up to four times the data and voice capacity when compared to HSxPA Release 6. Combined with the improved coverage resulting from the possible deployment in low frequency bands and the use of advanced antenna systems, LTE networks will provide service providers with a significant improvement on cost per bit delivered.
Another good reference is
my two bytes worth