Imagine a network of rail/bus/street/streetcar/parking nexi, distributed over a metro or semi-metro area.

Each node in the network is connected to at least one other node, so that the network is completely routable; there might not be direct connections from every node to every other node, but there is at least one indirect path from each node to every other node.

To access the system, individual travelers (and possibly cargo shipments with handlers) access a node by whatever means they wish – parking their car there (for a fee), and registering at a computer kiosk similar to the ticket sales machines and booths common to current conventional mass transit systems. Each traveler indicates (on a computerized system) their intended destination, and their bid for the trip – i.e., how much they will pay for the trip. They could also signal mode preferences (I’d rather take a cab over the river than the train) or book future travel (I know I will want to get on the train back home every day around 5:15.) The transit authority would set the minimum prices and maximum prices, or might choose to allow free bidding. Passengers could also access the system from home or public computers with Internet access.

The owners and operators of the conveyances on the network – whether trains, buses, cars, streetcars, hovercraft, or what have you – act as independent contractors on the network. The transit authority will no doubt have a registration, screening, and inspection role to play in the supply- side access to the market. Each contractor will monitor, in real time, the expressed demand within the network and route their capacity accordingly; the real-time monitoring will permit conveyances to signal acceptance of conveyance terms to groups of people at a time and “snap up” those customers – first come first serve. Once a passenger accepts the terms (“a train will take you to the Napoleon Hill station and will arrive within four minutes; the fare is $2.00″), the kiosk charges their credit card or accepts cash and makes the sale. The negotiation process could take place at the kiosk, or via the traveler’s own independent communications device. The kiosks themselves would constantly show the average prices for various trips on the network, allowing travelers to be aware of the going market rates for the travel they wished to buy.

The experience from the point of view of a traveller would be something like this. John enters the Union Station transit terminal and hurries over to the nearest free kios. He has a hot date, and wants to get to the Lincoln Tunnel terminal – just one block from his girlfriend’s apartment – as quickly as he can. Punching in “Lincoln Tunnel, and make it snappy.” He sees that the train ride most recently has gone for $3.50, the train will arrive in seven minutes, and the transit time will be half an hour. Not good enough – but a cab company offers to pick him up immediately (they have a car already at the station) and get him there in twenty four minutes or less for $27.00. He accepts, swipes his card, grabs the receipt that tells him to go to cab stand B and get in cab 974, and hurries away from the kiosk.

The next customer at the kiosk, Mary, arrives with her daughter and a crated dog in tow. Mary is heading for the veterinarian, and has an appointment there in an hour. She’s not in a big hurry, and she doesn’t want to spend a lot of money. On the downside, the vet is four blocks from the nearest terminal, and Mary does not want to carry this dog four blocks in the July sun. There isn’t a train service to her vet’s terminal (the tracks haven’t been laid yet by the rail companies – but Mary does check the box to indicate that she’d be interested in rail service if someone decides to make the option available) but the bus takes just 40 minutes and costs only $2.00 for Mary, $1.00 for her daughter (kids ride for half price on most conveyances, a marketing tactic almost universally adopted), and an additional $2.00 fee for the dog crate. Mary books the bus ticket and splurges on an additional $5.00 to cover the quick cab ride from the destination terminal to the vet’s office.

(In one alternative timeline, when Mary and family return from the vet the total fare ends up being $17.00, because now it’s rush hour and the bidding for rides is hotter. In another timeline, the dog has to stay with the vet overnight so Mary and her daughter take in a movie and dinner out to avoid the rush hour, and come home much later for $5.00 total.)

A third customer – a married couple this time, Phyllis and Phil, arrive for their regular commute. It’s mid-afternoon, but both Phyllis and Phil work an odd shift as CD duplicators at the local software company. They punch in their standard commute settings on their iPhones and quickly confirm a $3.00 bus-ride to work, and hasten on their way – nothing fancy, they do this every day and the fare is always about $3.00. Today there’s one change, though – Phil also remembers to buy a ticket for his dad, who’s coming to visit for Saturday dinner the next day. He locks in a great deal on the 60-minute train ride for his father, with a bus ride to cover the six blocks between Phillis and Phil’s apartment and their home terminal, and e-mails the ticket information and confirmation code to his dad before taking the escalator up to the bus loading area.

By utilizing a bidding model, and publishing the average bid amount information, both passengers and conveyance providers have exact and accurate information of the demand picture at any given time. Given most urban infrastructures, it would seem a logical first assumption that much, if not all, demand could be profitably met by transit companies. Full buses and trains make money, and the classic problem of mass transit has been the need to have large quantities of the equipment running at low capacity in order to provide service at a particular time of day. Under a bidding model, people wishing to take trips that other people don’t want to take will have to bear the costs of their relatively expensive choices or needs. Alternatively, people with those needs could trade time for money, by waiting longer periods in order to accumulate more bidders for a particular ride. (“There might not be anyone else needing to go over to the little-used Junktown terminal, but if I wait a half an hour, I bet a few more people will show up.”)

This would also be a highly profitable model for the conveyance companies, while also promoting vigorous competition and eliminating the kind of ill treatment of passengers that “you’re stuck with us” mass transit authorities are notorious for providing. If train company A will not provide good service to the people wishing to make a trip, then bus company B is certain to spot the opportunity (“look, a whole wad of unmet bids from Johnson Street – let’s snag those and send a bus over”) and scoop them up. Customers equal money in this model – and customers who want to represent more money in order to get better levels of service are free to do so.

I would anticipate the development of several tiers of service providers, in fact – from luxury cabs to utilitarian jitney bus companies – in order to capture the entire transit market. There is so much diversity of demand and desire that a similar diversity of providers would spring into existence, funded by the real demand for real service.

The main peril that I see in such a system is that established conveyance companies would engage in regulatory capture of the transit authority and try to keep out new competition. I would mitigate that risk by building free competition principles into the charter and structure of the transit authorities themselves – for example, by establishing an “open audition” policy by which any comer could offer services on a trial basis within the system (indicated clearly as such), so that the more adventurous travel patrons could “take a chance” on the newcomer and quickly establish a bona fide track record. There are other ways of keeping competition open, but all of them in the end rely on a demand for free competition from the transit customers themselves.

Municipalities or areas that wished to subsidize travel for their poorer citizens would, of course, be free to do so. However, my suspicion is that this model would drive down the costs of travel for nearly all travelers – with the exception of those who under current systems enjoy high level of service in areas or times of low demand. Those folks, it should be acknowledged, would end up paying more or enduring worse service.