Saturday, November 20, 2004

Why Amtrak Isn't Profitable

Why isn't Amtrak profitable?

There are lots of answers but the most basic one is subsidy. People talk all the time about "mass transit" in this country. Yet why hasn't it ever taken off?

The answer is fairly simple. People prefer to drive cars instead of take mass transit. The problem with mass transit is not building it, it's getting people to use it. Why do people drive cars?

Because for thirty some years we've driven down the nominal price of oil using our petro-currency monopoly. Until we recently broke our own petro-currency monopoly when the Fed Funds rate went down the oil price shortly thereafter followed. Our market position as issuing the currency oil was priced in allowed us to have low interest rates and cheap oil.

Right now as organizing most of our suburbs and exurbs are not going to be energy sustainable commutes if the price of oil keeps on going up.

That means we need to start getting serious about mass transit. Prevous attempts were futile and I didn't bother about trying to change things. As long as the petro-dollar ruled it was pretty clear that mass transit wasn't going to work. However things have changed and are changing now.

We need to seriously think about mass-transit now. Otherwise there is going to be a big crunch as people collapse back into urban zones. The price in core urban areas of real estate already high would skyrocket. Now is the historical moment when we really need to start thinking about a serious national mass transit plan.

Do we want it locally owned or operated? How about state or regional? Who pays for the infrastructure? Does the government guarentee the initial developments? Where will we zone the hubs and rails?

It's not just commuters. The food and goods distribution in this country is right now based upon truckers. They will still have a place. However with deisel prices rising even faster than gasoline prices, that is going to become increasingly economically unviable.

What I envision is a dual use commuter and commercial hub/spoke system. It's not the most ideal but we know how to make it work (airlines) and we can charge companies that ride the rails with their cars an infrastructure fee. It could be built relatively quickly by simply expanding the legacy rail system which is basically on the same standard design concept. If we do it fast when the time comes, there will be minimal suburban and exurban implosion and the increased profitability from lowered transport fees will increase the viability of various businesses.

With airline consolidation and increased jet fuel costs, the age of cheap flights is coming to a close. New high speed trains and tracks could be a cost efficient alternative for continental travel. This is the time to start thinking about this stuff, before the big crunch hits.


At November 20, 2004 at 3:53 PM, Anonymous Anonymous said...

I'm a bit in a rush, so I will just remark that rail transportation competes mostly with airplanes, not cars, on the passenger side, and airplanes and trucks on the freight side. The reason for that being, of course, the lower *dollar* efficiency of the currently existing rail system, and its interfaces to the road system. (The train does not drive right into your warehouse. You have to load the freight onto trucks at volume and quickly. I would have added the train doesn't stop at your factory, but then the factories are now in Asia, and so the relevant concept is a rail interface at sea ports.) One big part of the dollar efficiency has been as you observe an artificially low (in US$ terms) oil price.

Of course that is not all there is to mass transit. More later.


At November 20, 2004 at 6:30 PM, Anonymous Anonymous said...

How stupid can I be??? I was always looking at it from a personal/business/green/recyclying "cost savings" point of view, and not from a price of gas point!

Yes! let's follow up this discussion!!! CM and Oldman.

While you are at it, consider a tunnel system for all the utilities lines under/beside the train lines. This would make it more economical for competition in the utilities industries. This would lead to the ideal of a grid system/layout. Squarish in design, with the main trunk branching out/in to smaller lines.

Jim Coomes

At November 20, 2004 at 7:44 PM, Blogger Oldman said...

I would argue that for short distances commercial shipping and distribution is more cost and time effective if done with conventional trucking systems. I'm only talking about replacing long haul trucking with a national series of hub/spoke rail distribution centers. You want to ship something from Missouri to Florida? Haul it over to the nearest rail distirbution center, see it get picked up in Miami, and then a truck there will deliver it to a company. Right now what happens is that except for industrial grade large commodity lots most of the time, companies will ship on trucks the whole distance.

At November 20, 2004 at 7:50 PM, Anonymous Anonymous said...

I'm reminded of an architect I read about that designed a college campus. Instead of laying paths out from building to building,she planted grass, and waited until students had trodden the path down before pouring any cement for sidewalks.

What I also find interesting was the wide range of approaches to public transportation that emerged around the same time at the end of the 19th century after the first few waves of immigration: street-cars, subways, trolleys, cable cars, etc. Cities chose forms of transportation based on their own needs and the experiences of other cities. Hopefully we will see the same process at work during the crunch, and I like the approach you seem to suggest of focusing on the transition at this point. Urban planning will be a growth industry, though hopefully the focus will be more on understanding the emergent flow across areas and make incremental adjustments than centralized planning to meet poorly understood needs.

Transition into the crunch will lead to centralization into the core, as you suggest, and along with the need to make sure these existing centers are densely connected it is also worthwhile to encourage a bit of decentralization and help as new centers away from the existing cores.

The focus of your question is on the movement of people to centers of production. I think it will be equally important to think about distribution of goods in the same terms. A store like Wal-mart, for example, could transition into warehouses slightly distant from the centers, that gather orders from an area and delivers goods. While some may spend more time going to work as the result of shared transit, the amount of time running errands shopping could drop dramatically for most.

For those who cannot 'telecommute', the crunch economy might also lead to longer and more flexible work hours, to minimize public transportation bottlenecks and number of passengers daily. People will no doubt use human-powered transportation more, and may rely more on local resources and neighbors to meet needs. It may be unduly optimistic, but I hope after the crunch we find people to be more fit, with more free time out of work, and more connections to friends and neighbors.

a Fool

At November 20, 2004 at 8:16 PM, Anonymous Anonymous said...

Oldman: cost effectiveness of rail transport.

There are certain parameters that play into this which are unknowns to me, like: the road system is maintained largely by the taxpayer, whereas the rail system & infrastructure maintenance cost is charged directly to the user on effectively a mileage basis; rail worker unions and grown bureaucracies may have arranged for living wages and certain overhead costs (not sure about trucker unions and truckers' working conditions -- sleeping on a mat in the back of the driver's cabin?); rail transport is more schedule constrained and less scalable (try to put more trains/cars on a well-used rail grid); trains have fewer alternate routes in case of disruptions; etc.

Some of these are direct costs, others go under the heading "flexibility". It's like companies that have a choice between employees, temp staffing, and contract workers.

To a large extent, flexibility is a euphemism for unwillingness to plan and manage proactively. If I were to use public transit to work (not that this is feasible in my situation) instead of driving my car, or use transit to shop, my choices will be greatly constrained, most activities will take substantially longer because of delays (getting to the bus stop, transitioning between lines, ...), and I will have to start planning my daily activities at least a week in advance, and will have to let go of some things altogether because I don't have enough time. Not that this would be necessarily bad, but it will be a fundamental change of lifestyle. I have been living in such an environment, so I think I know what I'm talking about.


At November 20, 2004 at 8:27 PM, Blogger J Thomas said...

The time to start thinking about this stuff was 40 years ago, and there were people thinking about it. There's a big body of work to build on.

The time to start doing it was roughly the beginning of Carter's second term. We didn't do it. I think it's too late to do it before the crunch. It may be too late to do it at all, the economy needs to recover first. You don't build giant mass transit systems with an austerity economy.

One obvious alternative is to move the urban poor out of the slums and into suburban shanty-towns ala Rio, and move the employed suburbans into the slums. The ones with a lot of status -- higher managers and such -- can try staying home and telecommuting.

I heard about the beginning of that from some talk-radio guy in maryland. He was pretty bemused by it. They had found a hidden spot and built shoddy multi-room houses. They'd tapped into the water system and had running water, but no sewer. Oil stoves. Whole families living there with their kids not just a few guys with sleeping bags. The authorities bulldozed the whole thing, of course. It was before its time.

Suppose we did try doing effective mass transit. I note that the DC Metro is having some trouble now despite a whole lot of funding. Services are down and rates are up. They're talking about a "death spiral" where reduced usage requires higher prices right down to complete failure. Of course this is political. As the driving gets worse they can afford to raise rates, and somebody gets to skim the profits, and the federal government etc can't let it fail so they'll likely add subsidies, etc. They are of course adding stops but it costs a lot of money and takes a long time.

At November 20, 2004 at 8:36 PM, Anonymous Anonymous said...

Fool: Yes, people would be more fit. (I was more fit when using public transit and walking to shop. You try to carry a few pounds of merchandise forth and back for up to a mile a few times per week. I was probably working off the beer that I consumed while getting it and returning the bottles.)

But please forget your notion of more spare time. Au contraire, your commute will take you *significantly* longer, and your work hours will not be cut. On the other hand, you can read or play the gameboy while on the train/bus, which you can't in the car. You will also spend more time on behalf of those activities where you now drive, but you may be able to cut the gym if you are using one.

On balance, you may gain more "idle" time when you are in transit that you will quickly figure out how to enjoy (if your concept of enjoying yourself strongly includes reading, listening to music, or doing anything that has to do with thinkning and making notes; the latter applies to students or researchers as well), and by cutting a lot of activities that become infeasible your total stress level may go down.

The idea of telecommuting I believe is an artifact of .com times, and is available only to a select group of workers. (And is probably mostly used for working from home on evenings and weekends in addition to office hours, but here I'm speculating.) But who knows, maybe it sees a revival when transport costs become high enough. On the other hand, I know from my own experience that while you can do routine interaction over email, telephone, etc., you need physical presence for most meetings, in particular highly interactive planning or technical sessions in an R&D setting. Also forget video conferencing, it's but a crutch put on the telephone, and showing slides or scribbled notes does not add that much, as it is not really very interactive.


At November 20, 2004 at 9:00 PM, Anonymous Anonymous said...

cm: Commuting will take much more time, and may eat into time spent shopping and 'running around' to use the modern term. The commute burden will be worse in cities that don't have the beginnings of a public transport system already. Those of us lucky to be a few blocks from a subway or light rail station that is convenient to work already will have to deal with crushing crowds, so the gameboy idea will be out! For most, the connection time will be outrageous, and lead to pressure to relocate.

Oldman has suggested a 'grapes of wrath' worst-case for an economic crunch, which I associate with huge journeys of transient farmers from unproductive agricultural areas. What is the analogue this time around? The cities on the coasts are the best prepared to make the public transportation extensions and may adapt quickly once the distribution networks adjust. Factories and farmlands elsewhere will also be magnets for population growth (those still economically viable of course).

I'm curious also how sprawling cities like Houston or Dallas evolve. I think in the long term, the fragmentation pressure is great enought that you will have the 'collection of villages' effect of 19th century Tokyo, London, etc.


At November 20, 2004 at 9:16 PM, Anonymous Anonymous said...

cm: Telecommuting is a place-holder indication of the kind of thinking about work decentralization that may have to take place. The technology is inadequate and indeed an artifact of the dot-com era, but I have no doubt it will get a second look.

Organizationally, I'm thinking of what comes next as 'hot seating' on a metropolitan scale. A large financial services firm, for example, may decentralize operations into a dozen or so much smaller offices scattered within a 20 mile radius, and adjust divisional structures as needed. Many workers may adjust responsibilities to work closer to home. Other workers may report to different offices depending on the day of the week. For services, this may work, for production of goods, this solves nothing. Factory workers may have the longest commutes, though if they have been forced to the new periphery newer factories would be constructed nearby.


At November 21, 2004 at 1:25 AM, Anonymous Anonymous said...

Fool: I used to live in a city that grew in a time when access to cars was not ubiquitous, and as a consequence had developed a quite comprehensive public transit network.

But even so, it was very apparent that there is a fundamental problem with any public transit -- it doesn't really scale into a large area. Your typical European city/metropolitan area has grown outwards from the initial settlement, and its major traffic routes have ring and spoke structures, with the spokes roughly converging in downtown. More often than not, major public transit routes are laid out similarly. As a consequence, Getting from point A to point B generally means getting to a stop at the A-nearest spoke, making it to the ring, taking the ring, getting off at the B-nearest spoke intersection, and taking that spoke. (Add shuttles and additional bus lines as needed.) Of course this is a somewhat extreme scenario, but in large cities it can happen. Usually you also have various specialized no-change routes criss-crossing the area, typically buses, as you cannot have rails everywhere.

As the city expands outwards, the distance between the spokes will increase, area coverage will become more difficult to achieve, and traveling between points by changing lines N times will become untenable. This is where public transit breaks down -- a "modular" system of lines will "suffocate" in the central region, and at the major intersection points. The inward intersecting stations where people change lines cannot be substantially expanded, as the city has already grown around them in earlier days, and they simply cannot physically handle the volume of passengers that have to change lines. You don't want so many people on the platforms that those at the fringe fall on the rails.

Same with buses -- how many people can you have waiting at a bus stop, crossing streets walking from one bus stop to another, or crowding themselves onto a bus depot platform? How many bus depots can you build and maintain?

As the city grows and central real estate becomes more expensive, companies relocate outwards, forcing more outer-to-outer-area traffic patterns, etc. pp. (And traffic patterns are constantly changing to begin with, which shortens public transit planning horizons and makes it more difficult to install large fixed cost facilities like rail routes that also take long to build.)

None of these are new insights. So I have enumerated problems and no solutions so far, but what say you?


At November 21, 2004 at 9:18 AM, Blogger Oldman said...


I want you to think of a Curvilinear track layout system with hubs and tangentially convergent intersections. Think French Curves.

At November 21, 2004 at 12:21 PM, Anonymous Anonymous said...

Oldman: A quick Google search on "curvilinear" yielded too much noise. But I found out what kind of thing you call "french curves".

Anyway, I think I'm getting the idea. I suppose what you mean is covering "new" area with linear segments, and connecting them with curve segements that fork out of the "trunk" lines. Like for example, when you have an existing radial/spoke system, filling the empty space between outwards extending spokes with cut-off spokes that are connected by curve lines to the old spokes?

Well, I am not saying the problem cannot be technically or logistically solved (and what I meant to say in my previous post and failed to express concisely was that it is difficult to upgrade/extend existing historical transit systems with relatively little cost & effort -- what I really meant by "scaling", because then you run into logistical and physical limitations).

But I'm afraid the actual problem is the difficulty of transitioning incrementally into a public transit system, both from an economic and psychological side. First, there is little low-hanging fruit, and small & cheap efforts will not buy you much. Any new construction in a populated area with existing infrastructure leads to disruptions, as roads must be narrowed or closed, property must be bought off and bulldozed, and the construction itself creates noise and pollution nuisances (have you lived close to a major construction? I have, if you leave the window open, your place will be filled with dust and sand quicker than you can clean).

However the largest problem I see in the various disruptions to individual lifestyles and patterns of business that people will fiercely resist, which will express itself in voting patterns and the resulting disincentives for politicians. Many people, and indeed most of society, have constructed their lives and organizational structures around cars. There is the issue of individual lifestyle, but also goods distribution as you point out, and work commutes. As long as the current arrangement appears to work, nobody will want to touch it and induce upheaval. When the resource crisis arrives, the willingness will grow, but it may be too late, as building the new system depends on the same type of resources.

I think we should focus on the latter, social, aspects. With the technical stuff I'm less concerned. But I don't have easy solutions. Anybody?

At November 21, 2004 at 2:28 PM, Blogger J Thomas said...

How about this: When you want to go somewhere at some specific time (say, from your home to a distant bus stop or subway station or whatever) you post it on the net. You can offer a price that you're ready to pay or not. You then get bids from people who offer to take you there, and if you didn't post a maximum price they'll tell you their offers. You pick one.

If you have a regular schedule you might agree to a long-term contract.

Anybody who has transportation can get into the deal, whether they drive an SUV that can hold 8 passengers or a smaller car that holds 4 passengers. They charge enough to get by, you pay what you have to. You're safest with a full car which is how they make the most money. There would be a certain number of kidnappings but on the whole the system might work.

No investment needed except the website and database. Anybody who looks at the data and figures he can buy a minibus and get a profitable route can go do it. Any big busline who looks at the data and figures a better way to run their routes can do that.

All it takes is enough of a felt need that a whole lot of people use the service, plus programming the service, and funding for the website.

If the boom hadn't already collapsed, that website would probably be up already.

At November 21, 2004 at 7:11 PM, Blogger Oldman said...

Yes, you bring up a good point CM about already established cities. I think there's only one solution. Elevated tracks.

At November 21, 2004 at 9:21 PM, Anonymous Anonymous said...

J Thomas: No offense meant (as always), but I'm not buying it, and frankly consider it excess belief in computer technology. Computers can assist us, but won't save us. And I say this as a computer professional who does software development for a living. Computers are overrated, though important.

The problem is not in matching ride-seekers to ride-providers, but in that people are reluctant to accept the restrictions that come with it. Both the ride-seeker and ride-provider have to commit to a certain schedule, and worse, have to negotiate their daily constraints ("can we go 1/2 hour earlier/later, I have a short-notice meeting/leave earlier/etc."). Which BTW also requires them to be able to communicate on short notice.

My spouse used to carpool with some guy working close to her place, but it didn't really work out because of aforementioned problems, and they broke it off (the version I heard is she dumped *him*, but then he was the driver, and I suspect he was not opposed). Now I'm carpooling with her, which means going a bit out of my way.

When going for lunch, I either go to the corporate cafeteria which sucks overall, or I have to find somebody for company, and then the argument becomes who drives. Absent a car, I would be stuck with the cafeteria. And once the cafeteria operator knows this, will their price/performance ratio go up or down? Same with stores and other businesses.


At November 21, 2004 at 9:28 PM, Anonymous Anonymous said...

Oldman: Even elevated tracks need space to put the pillars, and take up space in the air. And you need a route regardless. Whether your route is on the ground or in the air, nearly the same houses have to be bulldozed. And they probably create even worse to protect against noise problems. (I'm of course biased by what I saw in Europe, where new train routes literally have to go between existing houses, at maybe 10-20 meters from your window in some cases.) And most of the routes I'm talking about are elevataed to begin with, as you cannot sacrifice the existing road(s) beneath.

BTW, the post @ 11/21/2004 12:21:38 PM is also mine and I forgot to sign it.


At November 21, 2004 at 10:54 PM, Blogger Oldman said...


You are right and that these are not easy problems to solve, nor trivial ones, however look at the alternatives. Yes there is a balancing feedback about the maximum profitable density about mass transit clusters ... but the alternative is individual internal combustion and the compaction/density issues are worse and not better. Yes you don't have any easy way to fit elevated tracks, but the alternative is sacrificing even more ground space or digging underground and that becomes phenomenally more intractable (see Big Dig project on google).

We're not talking about an ideal situation here. We're talking about the "only" solution. Because if you don't adopt it, you will have huge unemployment, huge homelessness, out of control urban densities, etc.

Energy-distribution issues will force better efficiency measures. Mass transportation is the only available way to address the energy-distribution equation. Not an ideal solution, the only available solution. If people could have flying cars or just teleport or if we had next generation cities maybe it wouldn't be so pressing.

Many solutions in life are forced upon us by circumstance and not desire.

For the past 20 years I have been against mass transit. When I began looking at automobile technologies and alternative energy, I realized that the effective commuter range would be cut in half. Cut in half CM. Do you know what that will do to city infrastructure?

There's a reason why cities sprawled. Cities sprawled because we took cheap energy costs and traded it for a horizontal urban infrastructure and direct transfer production shipping. Cities in the 19th century were built around railroad distribution, because they didn't have the automobile technology. Living in a 19th century city sucked.

Unless we want our lifestyle and living standard to fall back to Victorian England we have to consider a way to balance the energy-distribution question differently.

Hybrids, eletric cars, hydrogen, fuel cell, doesn't matter. If you look at the break "even" point of energy-distribution to commute-shipping it halves roughly. If we half the effective radius of most consumer travel, the current system will experience unacceptable social shocks.

At November 22, 2004 at 12:44 AM, Anonymous Anonymous said...

Oldman: I think we are both on the same page, and I'm not disagreeing with any of the points of your latest post. But I'm not exactly sure what you are arguing. First you say that oil consumption has to be reduced, then you seem to say how this would impact life styles/social workings. While not technically contradictory, I'm not getting what you try to say.

Based on what I said earlier, I don't believe public transit is a comprehensive solution. Maybe a good start could be where public transit can serve all those people for whom it works (which may mean either effectively the same commute time, or longer commute time with time to read or otherwise enjoy or stimulate themselves intellectually, etc.), and load is taken off the roads.

I have good confidence that this can be managed technically and logistically, once the public transit system is there, and the system is constructed such that it can respond to changes in traffic patterns (or else you will get back to individual traffic). How to get it in place and sustain it I don't know. The big problem is the financial, and thus political, part. Political will aside, in the steady state the convenience and cost of all transport modes has to be in balance, where convenience/cost includes the impact on the lifestyle. I cannot see, for example, how having an extra hour of commute when using public transit can be tolerated by most individuals. I was in approximately that situation -- my public transit commute was 50+ minutes one way with 1 or 2 changes of train depending which initial train I catch, and approx. 15-20 min of that was wait times on mostly open platforms being exposed to wind, rain, and snow (not pretty) -- that is, the platforms had roofs, but it's at times difficult for all waiting passengers to fit underneath. Often enough I had to miss trains because I could not physically get on! Once I had a car, the time was cut to 25+ min. Obviously my public transit route was not very optimal, but so it will happen for a number of people. I basically had to take a radial line to the ring, take the ring for a while, and get off on a different radial line outwards. But I grant you that the cost of driving the car was higher, even in the marginal case (i.e. you have the car's fixed cost anyway, and your commute cost is just marginal mile cost), but not enough to go back to spending 50 min more daily. The job paid above average for the region so that I didn't attempt to look for alternatives (in the area, anyway).

So aside from my good salary, what would have convinced me to take public transit? High enough fuel costs -- but then ticket fees would have matched, convenience of rides (likely to get a seat instead of being wedged between people with bad breath -- no joke, and having to work your way to the door one station ahead), better connections with shorter waits, accomodating office hours (not squeezing 20% overtime as it was) -- I don't really know.

Still no solutions, but let's keep discussing. Maybe we come up with something. I suspect that as opposed to many others here I have actually lived with public transit, and a not too shabby one at that given the ridership volume. Why don't others share their experience? I would be interested.


At November 22, 2004 at 1:31 AM, Blogger Oldman said...

What you're talking about CM is the marginal cost of resources vs. the marginal cost increase from opportunity cost. Both are part of a supply-demand curve. There was sufficient supply of cheap commuting train in terms of resource, but the increase in time meant a lower demand because of opportunity cost from time lost.

Obviously any economically feasible mass transit system has to solve the opportunity cost issue of time. I suggest however we are having difficulties communicating because I am suggesting that we inherently build in volume and frequency of demand into mass transit supply.

For mass transit to work there must be for a peak demand a peak supply, meaning if there are a lot of people who want to get from point A to point B at time X then there must be sufficient cars there scheduled at that time so they suffer only transportation time lapse and not queing time which indicates a shortage situ.

The analog to the students on the grass is that you watch where people want to go from point A to point B most and at what time, and you build around that with shortest route, most cars, least number of interruptions.

Most mass transit is based on the idea of peak time scheduling but geographic homogeneity ... to cover the most area. I want a least time solution not a uniform geographic access solution.

If you are having trouble understanding what I say please wait and I will propose something more detailed later on but I must emphasize it is very different from the spoke/ring model you speak of.

At November 22, 2004 at 2:15 AM, Anonymous Anonymous said...

Oldman: I cannot parse this: "... because I am suggesting that we inherently build in volume and frequency of demand into mass transit supply."

Otherwise I do understand what you are saying, at least in the abstract. And I was not proposing the spoke/ring model, just pointing out that this is what many European (not sure about US) transit systems look like, and their problems. It may not be very relevant to the discussion, but my hope was that by illustrating the problems I am by analogy pointing out similar problems that other transit systems may have.

The problem with the valid point about peak supply that you bring up is that peak supply requires redundant infrastructure (rails, platforms, cars, fee collection stations, ...) that will only be used during peak times, but has to be maintained all times (and charged into ticket prices or whoever funds the system). You can crank up train frequency on a single rail, but only so much, and it doesn't give you the desired peak to non-peak ratio. As an example, the off-peak train frequency where I lived was 1 in 20 min per line, but several lines went through one rail on "shared" routes. When cranking it up to 5 min per line (factor 4), then with 2 lines you get one train every 2.5 min, which is already borderline impossible given boarding/unboarding times, and how much train frequency the rail metal can handle. Then you can vary car count per train (which they also did), giving you another 2x factor or so in terms of capacity, but not at a 2x marginal cost reduction (the drivers still get their full pay, and energy use does not scale linearly with car count). Finally you can build more rails and platforms, or distribute load between stations by having lines not cross at one, but several stations, or have express trains that don't stop at every station, in which case you definitely will need bypass rails because you cannot speed across platforms, less so when another train is stopped there, and you don't want to run into harder to schedule general trains (boarding times!) between platforms.

Last post today, I have to go to bed, and I'm looking forward to your next one.


At November 22, 2004 at 10:54 AM, Blogger J Thomas said...

Anonymous, I wasn't suggesting a comprehensive solution, but a do-able solution. And I didn't mean to suggest car pools so much as taxicab pools. Get one super-dispatcher for everybody who wants to use the system. Anybody who wants to make money can try to participate and over time develops a track record. Established bus lines can look at where the high-value unmet traffic needs are, and look for ways to skim some of that. Likewise for other mass transit.

Say some people get electric cars whose charge comes from coal-fired generators. Even with limitations in the cars they might find ways to fit into the system and make a profit. But efficient cars that get used twice a day for a 20-minute commute are for rich people. When we have a lot of unemployment, why own a car that you drive to work and pay for 9-12 hours parking so you can drive it home again? Somebody who works cheap could be driving that car all day.

If people find it inconvenient to use the internet to access the thing, someone might set up a telephone system; call their 900 number and they'll do the access for you. Or some other payment system. We might easily get a 20-30% improvement for a modest cost, and the data might help the planning for a better system.

I don't think this is an optimal solution. It looks to me like the *default* solution. It's what we'll get if we can't manage anything better.

At November 22, 2004 at 11:32 AM, Anonymous Anonymous said...

J Thomas: You must be referring to my post. I did not mean to say that your way of operating a "mass cab transit" service and matching riders wouldn't work at a technical/logistical level.

My problem is that I question whether such a system can be flexible and reliable enough to serve customers' (perceived) needs, and economical. For example, you will run into the issue of peak demand, e.g. during commute and lunch hours. This means you need a lot of drivers, most of whom make essentially all their money in the peak hours, and charge accordingly. Then in order to be economical, riders have to find company for sharing, which introduces constraints on flexibility. Around peak times, it may be difficult to get rides, as most drivers are booked. (No matter how the free marketeers will say this can be mediated by rate structures.)

However we may be talking about completely different contexts -- I'm trying to explain why people would never transition to such a system *now*, you perhaps mean why such a system can work *after* the resource crisis has kicked in.


At November 22, 2004 at 12:31 PM, Blogger J Thomas said...

It's expensive to arrange supply for peak demand, you wind up with a whole lot of unused capacity other times.

I'd suggest an economic approach. Raise the fee during peak demand high enough that the system isn't strained. For a railroad system you'd have premier cars for extra cost where everybody gets a seat, and maybe low-class cars where they fold up the seats so they can get more people on. Use some of the profits from the peak-demand times to expand capacity. So people who ride at the peak times are investing in peak capacity.

To the extent this is an issue it encourages businesses to do flex-time. Hire the staff or part of it to work 6-3 or 10-7 and get the benefit of avoiding much of the rush hour. And of course more people using the system an hour or two before or after the rush encourages the system to run more cars then and reduce waiting times.

About routes, this is always a political problem. Property valies near a rail line are reduced, but property values near a stop are improved. So the real estate guys who run city governments fight each other very hard to decide where the lines can go and where the stops can be. It's a mess and you can't get around it unless you can eliminate that democratic process.

When I think about how to design a modern passenger system I wonder whether the UPS system would work. Start with a spoke system. Have a bunch of cars and assign a destination to each car. At each stop the train keeps going and the cars for that destination go onto a siding and slow down to stop. AFter the car fills with new passengers it gets pushed back up to speed to link with the next train coming back.

So people don't have to stop at other people's stations. You don't get the energy cost of stopping the whole train at each station and then starting it up again. If the engineering challenges work out then it's faster and cheaper. More chance for a serious accident, of course.

It's a waste of time to go a long way down one spoke just to go a long way back up the next spoke. You can deal with that however you want. You could have one set of spokes that curve clockwise and another set that curve counterclockwise with a station wherever they cross. Then instead of going all the way to the center to switch, you could switch wherever your clockwise hub meets the counterclockwise hub you want.

But I have the strong impression that mass transit systems generally start with one line, designed to maximise income. They extend that line more or less linearly until they see that it would be more rewarding to start a second intersecting line, and so on. The actual paths vary according to some combination of expected profitability for the mass transit and profitability for the real estate interests. It would be very interesting to collect a whole lot of money to build a transit system all at once, according to some overall plan.


At November 22, 2004 at 3:13 PM, Anonymous Anonymous said...

J Thomas: As long as you are not the one who has to work from 6 or until 7 (and with commute has to get up at 5 or come home at 8), this should be fine. The "overlap" under your proposal would be 5 hours, with an assumed 1 hour lunch timeout 2x2 hours. That does not leave a lot of time for meetings or individual helping each other out interactively, that are inevitable in many environments. When push comes to shove with the overlap requirements, I suspect the "inertia landscape" is such that early-shifters will be asked to stay in the office rather than calling late-shifters in early (which is infeasible on demand anyway). Also you will have to arrange the same extended work schedules for all other businesses and institutions, i.e. if you are supposed to be in the office at 6, you bring your kids to daycare/school at 5? And with vacations and sicknesses, some people will have to switch shifts, which is not very healthy plus it destroys all their schedules. In short, I have serious doubts.

Well, obviously I'm much better with pointing out problems than proposing solutions, but I'm already through the exercise of trying to avoid rush hour by going to work 1-1.5 hours earlier. That worked quite well including the getting up around 6. The part that didn't work very well was leaving earlier accordingly, and my final resolution was to go after the rush hour, not before. I would have preferred otherwise though, as I would rather work in the morning and have spare time in the evening.


At November 22, 2004 at 9:31 PM, Blogger J Thomas said...

cm, yes, I think the carefully-dispatched gas-guzzling approach might work after we get a crisis with no good solution.

As for the flex-time idea, yes, the support industries would have to extend hours some, which balances against reduced peak usage. And while there would be various problems for employees who have to do it, they won't complain too much when they feel lucky to have a job at all.

At January 3, 2006 at 1:23 PM, Blogger James Baker said...

Hi Blogger!I like your blog! Keep up the
good work, you are providing a great resource on the Internet here!
If you have a moment, please take a look at my site:
find properties make money
It pretty much covers find properties make money related issues.
Best regards!

At January 8, 2006 at 3:09 PM, Blogger Investment Center said...

I just came across your blog and wanted to drop you a note telling you how impressed I was with the information you have posted here.
I also have a web site & blog about intl investing
so I know what I'm talking about when I say your site is top-notch! Keep up the great work!

At January 13, 2006 at 8:52 PM, Anonymous cheap flights said...

Hi, I was recently surfing, checking out the latest information on airline tickets and I came accross your site. While airline tickets was not what I was searching for, I found your site to be an interesting one.

Hi, maybe you can drop by my site one of those days airline tickets

At January 30, 2006 at 9:03 AM, Blogger Paul Adams said...

Congratulations Friend for your excellent blog on used uk car prices directory!Keep up the good work!
If you have a moment, please visit my site:
used uk car prices directory
I send you my warm regards and wish you continued success.
Have a nice day! :-)

At March 14, 2006 at 5:22 AM, Anonymous Anonymous said...

Unbelievable! Thanks.
Was searching for traffic winner.
traffic winner


Post a Comment

<< Home