You raise some legitimate objections to GPS-based road pricing, including implementation costs and privacy concerns. But you may be too quick to dismiss the concept. Although it may seem unnecessary now, it has the potential of more efficient and equitable taxing of vehicle use which deserves serious consideration. Let me describe some potential benefits.
Motor vehicle travel imposes a number of extenal costs, including costs for roads and parking facilities, congestion impacts imposed on other road users, accident risk imposed on other road users, air an noise pollution, and external costs associated with petroleum production and use. A basic principle of economics is that efficiency and equity require prices that reflect marginal costs (unless a subsidy is specifically justified for other reasons). Most costs imposed by motor vehicles vary by time and location, so optimal pricing requires a mechanism that reflects when and where a vehicle is driven. For example, a congestion charge should only be imposed for vehicles driven under peak-period, urban conditions. Similarly, the air pollution costs imposed by a vehicle driving in a highly congested area (such as downtown Houston) are an order of magnitude higher than the same vehicle driven in rural Montana. It would be inefficient to impose a constant per-mile air pollution charge on that vehicle without regard to where it is driven.
If you consider external costs associated with fuel consumption (externalities associated with fuel production, importation and CO2 emissions) to be the most significant problem of vehicle travel, and so increased fuel efficiency is the primary planning objective, then you are entirely right: fuel taxes are better then GPS-based fees. But if you are interested in charging motor vehicles taxes and fees that accurately refect their road wear impacts, congestion impacts, accident extenalities or pollution costs, then GPS-based pricing is better, because it can more accurately reflect the costs imposed by a particular trip.
Transportation economists have long known that a weight-distance charge is more accurate than fuel taxes in reflecting the roadway costs of each vehicle. In particular, current fuel taxes tend to under-charge heavy trucks, and therefore fail to give shippers an incentive to use trucks that minimize road damages (see FHWA, 1997 Federal Highway Cost Allocation Study Final Report (and Addendum), Federal Highway Administration, , 1997 and 2000.)
Similarly, GPS-based pricing of vehicle emissions would give people who drive frequently in the most polluted locations a strong incentive to choose low-emitting vehicles without imposing unnecessary costs and design restrictions on motorists who mainly drive in areas where pollution problems are much lower.
GPS-based pricing has already been implemented for trucks in some European countries (see Swiss GPS-Based Freight Truck Fees, and Facts About The Toll System For Heavy Goods Vehicles (HGVs): Distance-Related HGV Toll Ensures Allocation Of Infrastructure Costs In Line With The User Pays Principle) The use of this technolgy is expanding, because it is the most efficient and equitable way to charge for road use.
Transprotation economists have long argued that congestion pricing is the only practical way to reduce traffic congestion problems; any other approach is a second-best solution that provides only limited and teporary relief and may result in more costs than benefits to soceity overall. Some cities are beginning to implement some form of congestion pricing (most recently London), and this will probably become more common in the future. However, current pricing systems are inefficient, requiring expensive hardware along roadsides (either transmitters or cameras) where pricing is implemented. If congestion pricing is to be widely implemented, it would be far more cost effective to extablish a standard system that can function in different cities and does not require roadside equipment. GPS-based systems appear to be most practical.
