On Behalf Of Sean Brooks
While economists may debate the existence of true public goods, there is no doubt that there are partial public goods. These goods have high capital and fixed costs in relation to their costs due to individual users. In fact most goods have at least a portion of goods that are not attributable to users.
We speak of charging user fees for roads, but the 'cost' of allowing a car to drive on a road is nearly negligible vs. the cost of allowing a heavy truck to drive on it, and very little compared to the capital costs and the damage done by weather. The individual driver in a passenger aute 'takes' very little from the public by driving down it. This changes slightly when the road gets croweded, and there's not enough capacity, and not as good, to go around. Paying for these road fairly would involve a mix of local LVTs, user fees on heavy trucks, and congestion fees for popular routes. Whether or not bonds are used, they must still be paid back from some source.
Many of the functions of local government are similar. Most aspects of Public Safety aren't directly attributable to individual people, and in any case, a great deal of the costs of providing public safety are relatively fixed. You may be able to find that benefits accrue to property owners in terms of their reduced insurance cost (which then may improve the value of land). You may find that commercial rents increase due to a safe pedestrian shopping area. Neither of these belie the justification for using LVT to pay for them. The fire department may be of intense benefit to the owner who's property was saved from a kitchen fire, but you'd probably not want to deal with the potential distortions created by a fee-for-use fire department, when general benefit can be attributed to the fire department.
Schools, which often comprise the bulk of local spending, are another issue altogether. Certainly no function of government seems to have an effect on property values more than the quality of the school system. It may be feasible to attribute a portion of the costs as variable costs, such that each student should pay a tuition. I disagree with this notion, in that once the capacity is built and paid for (E.g. the school district has X classroom spaces and Y teachers), individual students do not add appreciably to the cost of providing schooling. Furthermore, attendance is cumpulsory, so attempting to charge tuition-as-a-congestion-fee wouldn't work.
Rail transit has enormous fixed costs in relation to their variable costs, such that once a good level of service is offered, (e.g. trains in each direction every 7 1/2 minutes) it costs very little to add an additional rider. In fact the time cost in collecting such a trivial portion of the fee may severely outweigh the benefit. One possible exception would be a rush hour congestion fee, which, again, may or may not be worth it - especially in view of the tremendous increases in local land values due to rail infrastructure. (in fact, and in reply to the the .pdf regarding the
As for buses, their benefit depends on their use. If they are regularly used and provide all day service - in other words if they can be counted on as a means of getting around an area - they probably benefit property values quite a bit. For them to be counted on as all-day transportation, they have to run all day, regardless of ridership - which makes the variable costs of ridership moot.
On the contrary, if the buses are run in such a manner that the number of buses in service is a function of ridership, then the costs are attributable to the riders. This would be the case for intercity coaches, as well for very popular bus routes - where this fee would be indistinguishable from a congestion charge.
It is my opinion that when public transit is effective enough that a portion of people can do without cars, or a portion of families can do without a second or third car, then that nonspent spending capacity is generally reflected in local economies, and raises land values.