A few more transit stats

Passenger trips per hour of transit service, broken out by mode. Source: Federal Transit Agency: 2004 National Transit Summaries and Trends.

A lot of transit enthusiasts howled bitterly about my piece the other day. Some of the accusations were:
  • The Department of Transport transit infrastructure spending numbers I found were off by an order of magnitude, included vehicles, or were otherwise wrong.
  • My statistics were old (around 2000), and it was All Different Now, because ridership was up following the increase in energy costs.
  • It would be Even More All Different after peak oil, when transit would magically become Much More Effective in the US than it seemingly was hitherto.
Well, I found a bunch of great stats from the Federal Transit Agency that allow me to address most of these issues. Here's a quick tour.

These guys see capex on infrastructure as a little less than $10b/yr, and rolling stock as $3-4b/yr. Their numbers are not consistent with the BEA derived numbers the other day, but are in the same ballpark. Transit spending has been increasing sharply.

Capital expenditures on all forms of transit. Source: Federal Transit Agency: 2004 National Transit Summaries and Trends.

This has increased capacity in just about all forms of transit. For example, the total amount of light rail service available roughly doubled in the decade 1995-2004.

Vehicle miles of light rail service. Source: Federal Transit Agency: 2004 National Transit Summaries and Trends.

Bus service went up by a more modest 18% or so.

Vehicle miles of bus service. Source: Federal Transit Agency: 2004 National Transit Summaries and Trends.

Available vanpool service more than tripled:

Vehicle miles of vanpool service. Source: Federal Transit Agency: 2004 National Transit Summaries and Trends.

In response to all this increase in capacity, total passenger trips increased 19% over the decade.

Passenger Trips on All Forms of Transit. Source: Federal Transit Agency: 2004 National Transit Summaries and Trends.

Light rail for example, where available service miles went up 100%, saw an increase in passenger trips of 40%.

Passenger trips on light rail. Source: Federal Transit Agency: 2004 National Transit Summaries and Trends.

So that must mean that ridership is growing slower than the increase in capacity, right? Yup:

Passenger trips per vehicle revenue hour, all forms of transit. Source: Federal Transit Agency: 2004 National Transit Summaries and Trends.

If we breakdown this number (the passenger trips per hour of service provided), we find that the utilization is dropping in most modes. It's particularly serious in light rail, but nothing is doing much better than holding it's own. The best is perhaps heavy rail (but that may be in part because heavy rail service increased less than others - only 20% or so).

Passenger trips per hour of transit service, broken out by mode. Source: Federal Transit Agency: 2004 National Transit Summaries and Trends.

Operating costs per trip started climbing about five or six years ago. I assume this is mainly due to the increase in fuel prices.

Operating expenses (excluding cost of capital) per vehicle revenue hour for all forms of transit. Source: Federal Transit Agency: 2004 National Transit Summaries and Trends.

With costs per hour of service going up, and ridership per hour of service going down, the amount of operating costs covered by fares, never good, is falling:

Recovery ratio of transit (all forms). This is the proportion of operating costs (ex cost of capital) coverered by fares. Source: Federal Transit Agency: 2004 National Transit Summaries and Trends.

Thus the operating subsidy per passenger is going through the roof (and this doesn't include costs of capital):

Operating subsidy per passenger (ex cost of capital). Source: Federal Transit Agency: 2004 National Transit Summaries and Trends.

So, in summary, during this pre-peak run-up in energy prices, we invested more and more heavily in transit. The effect of that was to increase capacity, but lower utilization. Operating expenses increased, and thus the overall financial performance of transit systems degraded significantly, requiring much larger subsidies per passenger (and the number of passengers increased). Overall, we got diminishing returns from this strategy suggesting that the best transit opportunities are already in use, and newer ones are more marginal. Light rail seemed to degrade the worst of any of the modes.

Under the assumption that the post peak-oil period involves still further rises in energy prices, if we invest even more heavily in transit, it would appear to me that we are likely to get even more diminishing returns.

Hi Stuart,

Very interesting graphs, thank you. They in fact spell a bitter tendency because they encompass years of higher energy prices, where people unfortunately didn't abandon private transport for mass transit.

But again you use the single Liberal view point where mass-transit is private and gets subsidies from the State to penetrate the market. And you continue to compare mass-transit to private transportation on that basis - which is ill formulated because it will always cost more to the State. I don't think that Liberalism will ever suit politics towards mass-transit.

Finally you can't possibly assume that in a post peak environment people won't use mass-transit because they aren't using it now. American folk still have wages above most of the countries' folk, and in fact can still afford for expensive private means of transportation. When (if) they loose that capacity they'll have to use mass-transit.

BTW, do you sleep?
Liberalism and conservatism have nothing to do with the success or failure of mass transit. Only people who are blind on both eyes while walking through cities where it actually works will ever raise such claims.

Here is the short list of places where mass transit systems work:

  1. Cities which have too few parking lots to accomodate commuter cars, e.g. New York, London, Tokio. You can drive in but then you have to drive your car back home because there is no place to stop.

  2. Cities which levy 100% luxury tax on cars and gas, e.g. Singapore.

  3. Cities which have too few parking lots and levy a 100% luxury tax on cars and gas. That would be Singapore.

Everywhere else in the world mass transit is an addition to automotive transportation, not its replacement.

It is refreshing to see thought that once in a while someone shows up who just don't get it and still tries to make it a political matter. It is not. Mass transit is simply a matter of urban design. Some urban planners get a fighting chance to implement working mass transit systems and most don't. It does not matter how many billions of dollars get spent on a system. If it is in the wrong place, it will fail.

Now... that does not mean there should be none in places where they fail. There should be. They are simply a public service to those who can not or do not want to keep a car. Availability of cheap public transportation is simply a matter of quality of life. And quality of life, again, is independent of political affiliation. If a town sucks, it sucks for both, liberals and conservatives.


Everywhere else in the world mass transit is an addition to automotive transportation, not its replacement.

Not in Bulgaria and most of Easter Europe AFAIK. Living there gives the distinct feeling that the various forms of mass transit are the primary mode of transportation, while personal cars are considered a luxury (as opposed to necessity) and are accordingly taxed. It is the design of the cities (compact, walkable) which is guiding this policy as well as the cost of energy which has always been high. The mass transit is truly mass - meaning that the majority of the population is using it for most of the trips.

Personally I never missed having a car over there. Having seen both extremes (currently in suburban Atlanta) I can say that if I could choose it will be somewhere in the middle - which is to a great extent achieved in West european countries. I would love if I could take the train to work here and have some free time while travelling instead of being stuck in congestions and risking my life on the highway. I would have still kept the car of course for leisure trips or shopping.

Liberal and Conservative labels do little for examining the  facts in transit. My grandfather was the head of Senior Citizens for Nixon (I Know, a dark family secret) in 1968. He was a friend of President Eisenhower, and spent his carrer as the head of the Lumberman's association.In other words, he was an old fashioned conservative. He rode the bus fro Houston's suburbs downtown every day to work in Houston, and not because wasn't wealthy. During the 1940's in Houston, and nationwide, people's expectations and desires were very different.
My grandparents owned one automobile, as did my parents until the mid 1960s.I rode city busses to schoole until my graduation from High School.
  Cars are fantasticially expensive. Depreciation, fuel, insurance, extra expense for driveways and inside parking, maintaince-I bet if we add it all up then the IRS 0.445 cents per mile deduction looks too low. Thats $5,000 or $6,000 a year for most automobiles and even more for luxery cars like Hummers or $60,000 dually pickups.
  We subsidise automobile with huge streets, highways, bridges, and tax credits on many wastefull automobiles. They make subways look pretty cheap, and thats not counting the CO2 or emmissions problem. Our national security is a farce when we are importing 70% of our energy.
  So be a chump. Label yourself and others Conservative and Liberal rather than take a look at reality-anyone who does this with the peak oil situation is just stupid and gullible. Its an old totalitarian trick-divide and conquer.
I believe Luis was using the term 'liberal' in its historic reference to an economic policy perspective.  Many if not most businesspeople who describe themselves as conservatives adhere to this outlook, while today's 'liberals' are associated often with the idea of the advantages of the welfare state.  John Stuart Mill could provide more insight into the classic liberal view.
For the readers' information, "liberal" in Europe has in the first place the connotation of "free market". Liberal in politics usually means striving for less restrictions both in economic and ethical matters, what would rather be called "libertarian" on the other side of the pond. A liberal over there, center-left in the political spectrum, would be called a socialist or social-democrat or occasionally a christian-democrat in Europe.
My bad, Luis. I guess I'm just overly sensitive from listening to the morons who call themselves consevative and try to tye all kinds of extraneous issues to the original proposition in order to discredit it by association.
  I guess its time to confess to my dark secret, Rove is my ex-husband-in-law. His first ex-wife was my third ex-wife. This is a relationship recognised on the Jerry Springer show and in any trailer park in Texas. It makes me overly sensitive to neocons.  
They used to say the Plantagenet Kings of England were descended from the Devil Himself


William the Conqueror down to Edward III.

Your story is almost as good!  

I remember his ex wife was a Dallas oil heiress, and that brought him to Texas?

The man is the political genius of his age, although the strategy of 'divide and rule' hasn't quite played out: they vacated the political centre and the Democrats have, for the mo' at least, occupied it.

If they hadn't invaded Iraq, then I think things might have come out very differently, so you could say Rove was in part ruined by the materials (ie GWB) that he had at hand, rather than the strategy itself.

the problem with mass transit is a simple one.

The benefits aren't capturable (entirely) by the operator and user.

If you use public transport, I benefit because I can drive to work more quickly, there is less air polllution, etc.

It's an uncaptured positive externality problem.

Interestingly, in the UK, when new roads are proposed, they are allowed to calculate the uncapturable externality (ie benefit to society as a whole) in the cost-benefit equation.

This is not the case when new rail systems are proposed.

In a true Libertarian perspective, ie an economic efficiency one, you would want a system that captured the benefits of taking public transport, for the operators of public transport systems.

Taxes such as the London congestion charge are a first step down this road.

Transit is and has long been the great Rorschach test of peoples attitudes about....well, just about everything.

My father was from a rural area, and hated buses and the very idea of them.  Buses were a sign of poverty.  My mother was raised in the city and was disappointed there were no buses in the country when she married and moved to a small town.  She did not learn to drive until I taught her in her late 30's.  It was one of the happiest days of her life when she got her drivers liscense. :-)

Trains are more interesting.  Older people have a soft spot for trains running out through the country, but are not so keen on commuters or "EL" trains, such as are seen in the city.  And let's not forget the helll that was most high school kids memory of riding a school bus, a moving torture chamber of bullies and jockying for position that often rivaled "The Lord of The Flies."
I got a job and saved money to buy my first car so I could drive to school, a waste of fuel, yes, but a JOYfUL change for me.

What has long been lacking is a very civil and well managed transit system, more in the European mold, that would be more like a cafe than a cattle car.  Some friends of mine and I once did some pretty involved study on the city of Louisville KY, considering a mixed light rail or bus and river ferry system, to carry commuters/shoppers from the prosperous east end of the city down the river to the downtown office and shopping area.  The distance is actually very short, and the river offered free real estate requiring no building of track.  With only a handful of buses and a few nice river ferries, most of the city could have been covered.
There was no real technical barrier to the whole project, but we had to admit it....as long as fuel was cheap, there would be little ridership.  What boosters of mass transit refuse to admit is that most people want a car, for weekends, for "just in case" for social reasons (back to the Rorschach test, try to impress a woman on a date by going by bus or commuter train and you will get a fast lesson in reality) and once a car is purchased, the fuel even at current prices is a very small part of the total cost of ownership, and is miniscule if the car is not driven great distances.  

I have always felt that trolleys and trams succeed in Europe for purely social/psychological differences, and not simply because fuel is expensive there.
Here's a free prediction, probably worth what it cost:  I think privately owned electric cars will have a bigger impact on fuel consumption than mass transit ever will, not because they are demonstrably superior but because they fit the American mindset better.

Roger Conner  known to you as ThatsItImout

I have always felt that trolleys and trams succeed in Europe for purely social/psychological differences

Indeed they did not succeed. They were widely abolished after WW2 throughout Europe. Trams were loud, screeching and badly maintained. Citys didn't want that wiring in the streets any more (that's why Bordeaux/France got a tram with a third (power)rail, that is activated by the weight of the train. This system failed to work at the day of opening, with the french president on board ..).
And, of course, streetcars are obstacles to driving, perhaps the most important reason to get rid of them.

These days a new tram line in Paris/France is opened, with HiTech trains, described as "superbe" by their drivers, very silent, very effective.

France is described as the new tram wonderland.

I mentioned it below, but Karlsruhe has not only managed to hang on to its city wide standard gauge system, but KVV has been expanding its reach and ridership steadily for the last 15 years, with a number of innovations.

Not everyone threw away what they had.

By the way, I think Strasburg/Strasbourg uses the same type of train - they are very quiet, very convenient, but not that well suited for heavy ridership somehow. Just an impression for a couple of years ago.

Not everyone threw away what they had

For very different reasons. If you have experienced a ride on the East-Berlin tram before 1990 (during the GDR era) you will remember one. They couldn't abolish their tram because the socialist subjects had to wait more than 10 years until being rationed a car. So buses and street cars had to do.

In second-biggest western cities such as Nuernberg or Karlsruhe the tram was kind of tolerated for decades, though often majorities of car drivers were hostile to it. The main argument against it was municipal deficits in comparison to poor ridership/comfort and so on, but the real reason is, and was, that they were hindering individual (car) traffic.

The Karlsruhe region is an exception in Germany. Most regions could learn a lot from the planners of the Karlsruhe-Bretten-Pforzheim rail system, which is exemplary in entire Germany.

Unfortunately it is true what someone said: "Car drivers love transit. The others should give up driving and change to transit. The ultima ratio is, and is to stay, "cars first".

Transit will not work as an incentive, it will definetely not get drivers out of their cars; that is wishful thinking. Restrictive measures are necessary (and will not happen since in a democracy those who implement them will be voted out of office).

Our great poet Heinrich Heine knew it - already in 1843 - when he wrote:

"Womit man einlullt, wenn es greint,
Das Volk, den großen Lümmel."

He didn't know cars, of course. But he knew the people.  

Light & other rails are a clear bonus to congestion, but I have not seen convincing improvment on the energy costs of these transport types. Maybe it is time for putting demands to the energy spending of rail transport?  I am sure the learned TOD have experience in this field?

Car pooling with efficient small cars is at least as environmentally friendly as rail.
To illustrate my point. My small 5 year old Italian Fiat Punto not marketed in the US- but the best selling gasoline EU car in the mini segment weighs some 1000 pounds, 80 hp , EURO NCAP 4 star, emits 140 g CO2 per km (even 44.8 mpg average on my 2000 mile summer holiday in France). With 2 passengers well below 50 gram CO2 passenger km. The FIAT car fleet is by the way the lowest consuming car producer in the EU with an average 139 gram CO2/km - see table page 2 here. http://www.transportenvironment.org/docs/Press/2006/2006_10_25_car_brands_co2_en.pdf

Rail emissions:
Now most rail data for emission is in the range of 40 - 70 gram CO2/passenger km. This http://www.transwatch.co.uk/transport-fact-sheet-5a.htm
UK study gives ( table 1 : 14.4 gram times 3.67 = 52.8 g/passenger km )

So my humble conclusion is that Car pooling could be as good as rail - or even better, at least in terms of energy use and emissions and for reducing congestion.  

regards And1

Remember in France nearly all electricity is either nuclear or hydro and hence trams have very low CO2 cost.
They have a lot of nuclear at night, but during the day France imports a lot of power.  Any idea how much of the imports are hydro?
The Swiss (and lesser extent Austrians, Norwegians) have a nice deal with France.  Buy cheap nuke power at night, keep their hydro at a minimum to keep fish alive, and sell expensive hydro at peak.

I see nuke as not more than 50% of US grid under any reasonable scenario.

Best Hopes,


This sounded quite shocking and I decided to take a closer look at the study. Several points:

  1. In UK 43% of passenger rail emissions and whopping 96% of freight rail emissions are due to diesel powered engines. Electrical trains are more CO2 efficient - south east is showing 12.5g/pkm vs 16.9g/pkm for intercity.
  2. The study is comparing real-world data for trains to theoretical data for lorries and cars. For example 10mpg for a coach and 50mpg. for a diesel car are not realistic real-world estimates, IMO. There is technological overhead, stop/go urban driving reduces MPG, etc.
  3. The biggest part of the CO and SOx emissions are from the diesel trains.

Conclusions: instead of moving rail to road transport, better electrify your network and build more carbon-free electricity generation. Increase the passenger rail load factor. Overall the potential for CO2 reductions in rail are an order of magnitude higher than for road transport.
"the potential for CO2 reductions in rail are an order of magnitude higher than for road transport. "

Only if you compare to ICE road transport.  Electric vehicles use less electricity per passenger mile than rail.

You electric car guys are as intriguing as the space traveller crowd.  I wait with enthusiasm for an analysis comparing the economic and environmental costs of an electrified motorized individual transport unit transportation system with an electrified rail mass transit system.  Please include all materials costs, production costs, reproduction costs, operating costs, disposal costs... in your analysis. Don't forget the road infrastructure, public education, licensing, policing and health costs.  Also please don't forget to measure the relative systemic effects, such as land use impacts.
Electric vehicles are a very old technology, older than ICE's.

You're making things much more complicated than they need to be.  For instance, "materials costs, production costs, reproduction costs"  seem to be redundant. "road infrastructure, public education, licensing, policing and health costs. "  are all the same or cheaper for EV's.  Policing???  

EV's cost about $.02 per mile for electricity, versus $.10 for ICE's.  They're simpler, and easier to make than ICE's: simpler transmission (the Tesla has only 2 gears), safer (no more hollywood exploding cars), have lower maintenance costs, negligible pollution at the vehicle and almost zero total pollution if powered by renewables.  The Tesla is so expensive because they're only making about 200/year in the first year.  Still, it's cheaper than comparable ICE sports cars.

The only barrier is batteries.  Batteries are now light & energy dense enough to make the Tesla possible, but they're still too expensive (at $20k for the Tesla battery pack) to compete with really cheap gasoline - an EV would be competitive with gas at around $5-6/gallon.

The Tesla battery pack will likely cost about $.20-30 per mile, depending on miles/year.  ICE's cost an average of $.10 per mile for fuel, and another $.35 per mile for everything else.  All told, EV's probably cost about $.55/mile now, vs ICE's at $.445 (per IRS).

Battery costs are falling relentlessly 7-10% per year and are likely to fall faster with next generation li-ion chemistries with much higher cycle lives.  Gas prices will rise, and the two lines will cross in the next few years.  Of course, if you were to include all of the external costs of oil, EV's would be cheaper right now.

The Tesla is a tiny sports car for car freaks who need a new toy because they got everything else already. It is not a useful vehicle by any means.

Someone make a real utility vehicle for the whole family that can compete with a well designed diesel or hybrid and we talk.  We shall also talk about the extra power plants and power lines that will be needed to recharge that sucker once there are millions around.


Actually, a tiny sports car is the best proof of viability.  A larger vehicle will be much easier to fit batteries into (vehicles become more efficient per pound and cubic foot as they get larger - this is even more true of EV's, where electric motors become more efficient as they get larger, the reverse of ICE's).

No additional power plants and power lines would be needed to recharge: almost all charging could be done offpeak with current infrastructure (see the recent DOE/NREL study for verification), and additional power could come from wind.

Well, you miss the whole point of comparing transportation system to transportation system, including all embedded and operating costs.  And you don't seem to understand that the costs of sytemic affects such as pollution costs associated with different land-use outcomes have to be measured.

Policing.  Transit systems require policing.  So do personal transport systems, including attendance at collisions.  The point is that they need to be compared when preparing the balance sheet.


Download pdf on Rail Transit In America: Comprehensive Evaluation of Benefits.  Chose Executive Summary or Full Report.  Not exactly what you asked for, but close.

Best Hopes,


I don't understand.  Electric vehicles like the Tesla or GM EV-1 look the same & operate the same as internal combustion vehicles.  They don't have new or different land-use dynamics.  They don't need different policing.  Their pollution can only be lower.

What are you concerned about in particular?

Just to clarify, he's not asking for a comparison with ICE's.  To quote, "I wait with enthusiasm for an analysis comparing the economic and environmental costs of an electrified motorized individual transport unit transportation system with an electrified rail mass transit system."

I.e. he's asking for a comparison of EV's with "an electrified rail mass transit system."

Ah.  Yes, I would like that too.

His first line seemed to imply a comparison with space-travel, which would suggest that EV's might be impractical, or inferior to present day vehicles.  Apart from the basic question of battery cost (which I tried to address), I can't see any reason to think that way, so I was puzzled.

I'd be delighted to see a thorough-going comparison of all costs & benefits for EV's and rail.  I think that's the thing that Stuart is trying to start to build.

That only looks at direct energy costs, and that point is quite debateable.

EVs will have (when they arrive) minimal in any indirect energy savinsg by altering the Urban form.  Most Urban Rail energy savings will be via indirect energy savings.



Again, I'm just trying to remind people that "road transport" isn't just Internal Combustion Engine cars, it is also EV's, and EV's are indeed among the solutions to Peak Oil.  

I agree with "silver BB's".  I think rail is great, though more for the indirect, hard to quantify benefits than for the direct costs like energy.

Rely solely on rail?  1) As you note elsewhere, it likely isn't going to get the required commitment, 2) it would be much more expensive than a pluralistic approach, and 3) rail is a centralized, government, long-term capital expenditure kind of thing, and EV's are a consumer-side kind of thing, and the two are going to happen simultaneously.

I think people will get very discouraged if they think rail is the only transportation solution, and I think it's a mistake to say so.

EV's will be much cheaper than scrapping the suburbs: they might cost another $100 per month at most.  That's much cheaper than moving and losing one's home equity, paying much higher housing prices in the city, etc.  I'm willing to pay 3x per sq foot to live in the city, but most aren't.

Not succesful? Trams in the city of Amsterdam are very succesful (nearly always full!)? The point is that you have disadvantage driving with your car enough to stimulate public transport.
Trams in the city of Amsterdam are very succesful

Some months ago I read Amsterdam intends even to use trams for freight transport (and to more and more keep the trucks out of the downtown area). So the Amsterdam tram may also be a hopeful story.

By saying "not succesful" I was trying to describe the past 50 or 60 years in Germany. The tram was abolished in virtually all major cities, as Hamburg or Berlin.

My personal view however, is that trams are by far one of the best means for urban transportation. There are new, modern trams in Nuernberg - really cool. But they are not cool for the average car driver.

I suspect a major factor in the death of trams in Germany was what is known on this side of the Channel as 'urban planning Royal Air Force style'.

Germany had the misfortune to have its major urban centres completely destroyed, at a time when urban planning philosophy was (post war) coming to favour the car.  By contrast Austrian cities were not so badly hit (out of range of Allied bombers) and Dutch et al. cities were not targetted (friendly civilians rather than enemy civilians).

(there were similar effects in a number of British cities from the Luftwaffe bombings.  The cities were rebuilt, but the tram systems never recovered from the damage.)

I suspect a major factor in the death of trams in Germany was what is known on this side of the Channel as 'urban planning Royal Air Force style'.

As of Germany that's not correct - quite the contrary. Since more and more buses were needed during the war the tendency to close down tram lines was significantly slowed! (Shutting down trams had begun long before the war - just at the time when car traffic started growing - or better: sprawling).

Trams were about the first means of transportation to take up service again even in totally destroyed cities as Hamburg or Berlin, and stayed in service until end of the sixties.

Very similar ones in Barcelona:

At first they were lots of incidents with cars, not used to look at the "new kid in town".


But don't you have problems with pedestrians getting hit?  You rarely see tracks level with the platform around here, because Americans, always in a hurry, will try to run across the tracks and beat the train.  (I used to do it myself, until they raised the platform so high you couldn't climb down any more.  Probably on the advice of their lawyers.)


We are in population overshoot anyway right? Lets get the stupid people off the planet 1st.

There are nowhere near as many deaths in Europe from railway and streetcar accidents as there are suicides on US train lines in my personal experience. Somehow people seem to think that public transportation system also serve a one way line to the morgue.
And Cars don't.

I get so mad at people that make a big deal when I plane crashes. I'm always like, "Dude..... Airplanes are the safest thing in the WORLD. Cars.... cars WILL kill you"

Apparently on a per passenger mile basis, aircraft look very safe relative to cars.

But plane journeys tend to be far longer per journey.

On a per journey basis, the death rates are much closer (I don't have the stats to hand).  Planes don't crash that often, but when they do, the effect is gigantic.

Interesting that the number of deaths per aviation incident is steadily rising-- planes are bigger, and when they do crash more people are killed.  This is true even if we strip out 9-11 (the largest aviation incident in history).

Of course, in the USA it's trivially easy to be made a multimillionaire by doing something stupid. Doesn't the New York MTA fork out $10,000,000 a week or something like that to blithering idiots? It's just not as easy anywhere else.
As opposed to being hit by a car?  There are stupid people everywhere.  Some run across the train tracks, some run across a busy street.  You can't stop people from taking stupid risks.  Most likely the expense of raising the tracks is not justified due to a few dumb asses.  
Well, I was certainly one of the questioners, and this certainly answers most of my number based objections.

Your conclusion 'Under the assumption that the post peak-oil period involves still further rises in energy prices, if we invest even more heavily in transit, it would appear to me that we are likely to get even more diminishing returns.' seems well founded at one level, which is the practical/political level typical in America - in fact, it is likely that a declining revenue base will lead to such well-founded arguments being used in discussions concerning how to keep things running along.

The scary thing is that I think Kunstler's point about having reality being the source of America's next discussion about how to live is actually valid.

This is not a discussion of the best forms of transit - I think the 'heavy' and 'light' rail distinction pretty artificial, and that European style systems, which essentially fit into normal traffic as is, are the right thing to consider.

But your work is again disturbing in showing just how deeply the changes must run.

I may add that at least where I live in Germany, the local transit system (KVV) is definitely pulling riders from cars, but the system makes a real effort to provide a service which can compete with a car on a price/time level. Only a few cities in America come close - SF, NYC, Boston to an extent, and in my memory from the mid-80s, Denver - which due to pollution problems, had a fantastic 24 hour bus system, which in no way could have been even 20% covered by fares.

Only a few cities in America come close - SF

I've lived in SF, and silicon valley, and can categorically state that the public transport is an abortion. Just the fact that CalTrain is n blocks south of the centre of the city is a testament to the bad planing.

The commute from SF <-> Palo Alto would try anyone's patience. An equivalent commute in europe / japan...

I lived in San Francisco for several years, and I say their public transportation is excellent. I don't care about CalTrain to San Jose, but an unlimited SF FastPass in $45 per month -- cheaper than auto insurance alone!
Within SF public transportation is quite good, but getting from SF to anywhere in the Bay Area not close to a BART (Bay Area Rapid Transit - heavy rail) Station is a nightmare.

BART is an textbook example of poor transit planning. It uses a 5 ft 6 in rail guage instead of the standard 4 ft 8.5 in, precluding the use of existing right-of-way, motive power, or rolling stock.  The initial computer system designed to run the trains failed miserably.  The system, which started carrying passengers in 1972, failed to connect to SF International Airport (a mere 14 miles from SF) until 2003!  That 14 mile extension cost $1.55 billion to construct. To this day you need to take a shuttle bus to connect from BART to Oakland Airport, and forget about connections to San Jose International Airport.  

I lived in Silicon Valley most of my life before emigrating.  Public transit was a joke, but that was never a better place to ride a bicycle.  Bike lanes are everywhere and the weather is great.

Auckland is even worse than San Jose.  There are all the geographic issues of a Seattle or SF - hills, narrow streets, bridges, no bike lanes.  Plus public transit is limited and riding a bicycle is extremely dangerous.  I avoid Auckland like the plague it is.

Auckland, NZ? Ugliest city in the world. Well... ugliest city I have seen.


I moved to San Jose after my tour with the USN 1971-1974, and worked many years in Cupertino, commuting a helluva lot from Willow Glen.  In 1989 my company (British Telecom) moved to North 1st Street and Trimble, and I moved to a home right off the Blossom Hill light rail station and was able to commute via light rail until 1991 when I moved to London to work for BT.  Commuted every day on the Tube between Marylebone and Moorgate, that was quite an experience!  When I eventually moved to Sacramento, I made sure the home I purchased was close to a light rail station, even though I now work out of my home office.  
The location of the CalTrain station in Palo Alto and Stanford and downtown have been the same for many decades. Walking to the other end of downtown from the CalTrain station is nothing. And there's a free shuttle bus that goes around, sure it's designed to take people to/from Stanford but it covers a bit of downtown too.

The secret is to lose weight, your ankles will lose that "about to break" feeling if you drop 100lbs.

The problem now with that Marylebone to Moorgate line is the delays.  The Metropolitan Line is not bad, but the system as a whole is beset by delays.

And fundamentally over capacity, at least in the core 7.45-9.15 and 6.00-7.00 commuting hours.  Some of the Tube lines are running at more than 100% of theoretical capacity.

I live in SF and most of my clients have been in Silicon Valley.  The only way I found to use the train is to bike to it since both my house and my clients are too far from the station to walk.  I do in fact do that most days, but it's a major hassle in the rain, and always takes twice as long as driving. I justify it by noting that at least I am using all the time productively (either exercising or working on the train).  However, it makes it significantly harder to cope with a busy and complex family/work life, and whenever something unusual happens, I end up driving instead.

My wife takes the kids to school etc.  In almost all cases, even on SF's much better than average public transportation, driving takes half the time of walking/Muni.  Therefore she almost always drives.

This points to my concern about the conclusion of your post..  

 "Under the assumption that the post peak-oil period involves still further rises in energy prices, if we invest even more heavily in transit, it would appear to me that we are likely to get even more diminishing returns."

The investment being made in light rail and other modes of expanded transit represent some respectable long-term planning, and can be quickly thumped by showing their short-term costs.  There are a number of factors that affect how people become regular passengers, including long-standing driving and lifestyle habits, housing choices and urban designs that may barely look at 'distance to stations/stops'. (while in NY, this is in many/most real estate ads as a clear advantage)  Add to that, of course, the reluctance of much of our society to acknowledge our energy precariousness, and it's clear that the investment may not see it's returns quickly, but is as smart an investment as we could be making to prepare for such a likely contingency as a major energy disruption.

  So while your conclusion may be partially correct, in that the investment will still be a heavy burden to bear (and keep increasing) while so many Luxury Hummers are still boppin' along out there, don't you think this is a vital and useful piece of infrastructure to be expanding upon?  Just because they haven't yet given up on the F350's and the mondo-commute is hardly going to convince me that they can't, and that we shouldn't be setting things up for that eventuality just in case.  It may not be possible to make a concrete economic case for it, unless you look Below the bottom line.

Bob Fiske

"don't you think this is a vital and useful piece of infrastructure to be expanding upon?" I expect that the trend to more investment in transit in the United States will continue to grow, and I expect that it will have a negligible impact on our problems. At some point, the financial haemorrage will presumably become enough of an issue to start reversing the trend.
We have a lot of financial Hemmoraging going on, but I don't think this qualifies. Yes, it's expensive.. but it works.  You do have to let it grow. And you have to make sure it's being evaluated in proper balance with the subsidies that it is competing against.

  I think the 'Big Dig' qualifies as a sad example of many wasted tax dollars (subsidising a lot of Drivers), the Medicare non-bargaining policy, the Tax Breaks to XOM last fall, when they were hurting so bad.. the EROEI on the war seems bad, unless you're in Al Quaeda.. - but looking at poor short-term ridership gains as a gauge on transit development seems to be a funny target for cleaning up our budget problems, particularly the ones we can expect to face when people really can't fill their tanks any more.

I agree with this. The oil-gasoline-automobile-highway transport system in the U.S. has massive subsidies and external costs that dwarf transit. In addition, "Overall travel trends indicate little about the cost effectiveness of particular projects and policies" (Todd Litman).
I expect under the current administration that Urban Rail will hardly grow at all.  Just a trickle of federal funding ($1.25 billion/year).  Minimal input, minimal output.  NOT NA SOLUTION* with $1.25 billion/year from the feds.

Best Hopes for spending 1 month of Iraqi spending on new Urban Rail each year,


For $1.25 billion a year you could install 800MW worth of wind turbines a year. That's roughly 200MW average, and that is 4.8 million kWh per day. An EV or plug in hybrid would need roughly 30kWh for a typical commute. In other words... the wind turbines could power 160,000 cars. Does the federal government's investment move 160,000 people more every year? Or 320,000, if an average two people share a ride?
Your handle is InfinitePossibilies, but you are positing an either-or, win-lose hypothetical? How about let's do both -- when full cost-benefit analyses find the investments to be worthwhile.
$1.25 billion for transit systems is not a win... it does nothing substantial, not even for the transit systems already in existence. If you wanted to solve the PO problem with transit systems you would have to spend hundreds of billions... most likely at least as much as has been spent on highways and if I am not mistaken... a lot more. And then you would have to rebuild much of the US so it would look like London, Tokio or Singapore.

How likely is any of that?

I don't mind putting some money into transit. I mind claiming that it is the solution to the PO problem.

Cost-benefit analysis assumes that transit systems and cars serve the same goal. In most instances they don't. Transit systems are mostly a matter of quality of life of a city or a region. Having a car is mostly a matter of quality of life of an individual.

It is therefor a judgement call how much quality of life a city wants to offer to its citizens by building a functional and useful transit system.

PO does not threaten transit systems because cost of energy is not a serious concern. But PO threatens private car ownwership for those who bought the wrong car.

My comments are all fundamentally based on these and similar thoughts which seperate concerns. Public transit takes over the role of cars in very, very few places in the US. New York and Washington, maybe San Francisco and a few others. Everywhere else being on transit sucks. Take it from someone who does not own a car. I couldn't possibly move to L.A. and live the same way I do now. It just does not work.


"I mind claiming that it is the solution to the PO problem."

So who (on The Oil Drum anyhow) is claiming transit is the solution? That's a strawman as far as I can tell. Transit is one silver BB among several others. And it's a necessary BB.

Transit doesn't have to be gold-plated, either. There are European cities with very inexpensive rapid bus systems. Just separate a lane with a continuous wheelstop. Space stops 1/4 mile apart instead of every block. Sell boarding tickets or cards in stores or kiosks, anywhere but from the driver. Voila! You've got yourself a cheap transit line that can get across town as quickly as a car -- or quicker, if traffic is heavy.

About cost-benefit analyses for transit, please see Determining the Value of Public Transit Service.

Lifespan for wind turbines (which I support) is 20 to 25 years.  Lifespan for cars (EVs ?) is rarely 20 years.

I took an 1898 subway to and from the ASPO convention in Boston.  I take the 1834 streetcar line (on 1923/24 rolling stock) pre-K.

Best Hopes for Long Term solutions,


There is an unfunded proposal to bring CalTrain much closer to downtown SF via a 1.3 mile tunnel to the "TransBay Terminal" where a variety of rail and bus lines join for transfers.



Miami is building a Miami Intermodal Center, connected to the airport via people mover, all of the airport auto rental would be located there.  Served by Miami Metro "Subway in the Sky", Amtrak, Tri-Rail (local commuter rail), Greyhound, and local city buses.


So, in summary, during this pre-peak run-up in energy prices, we invested more and more heavily in transit. The effect of that was to increase capacity, but lower utilization. Operating expenses increased, and thus the overall financial performance of transit systems degraded significantly, requiring much larger subsidies per passenger (and the number of passengers increased). Overall, we got diminishing returns from this strategy suggesting that the best transit opportunities are already in use, and newer ones are more marginal. Light rail seemed to degrade the worst of any of the modes.

Under the assumption that the post peak-oil period involves still further rises in energy prices, if we invest even more heavily in transit, it would appear to me that we are likely to get even more diminishing returns.

Huh? I have no idea what you just said. Venezuela will provide you a writing coach. Free of charge. I'm serious. I'll give you my best one. You need help, my friend.

Or perhaps you could try a reading coach? ;-)

In these paragraphs, Stuart concisely summarizes interpretations of all the graphs he presents. That isn't as easy as it sounds.

Stuart says... (blockquotes)

So, in summary, during this pre-peak run-up in energy prices, we invested more and more heavily in transit.

While oil got more expensive, we spent more on transit infrastructure...

The effect of that was to increase capacity, but lower utilization.

So, with that money, we built more available transit seats for people to use. However, the percentage of seats getting used by people went down. (it is also noted later that the total number of seats being used went up, while the percentage of seats being used went down)

Operating expenses increased, and thus the overall financial performance of transit systems degraded significantly, requiring much larger subsidies per passenger (and the number of passengers increased).

The cost to run the transit networks went up, and quite a lot too. The proportional amount the cost to run transit went up was greater than the increase in passenger numbers. That's not good, because it means that the cost per passenger went up. Taxpayers are paying a much greater subsidy per passenger on transit compared to 10 yrs ago, even though ridership is up.

Overall, we got diminishing returns from this strategy suggesting that the best transit opportunities are already in use, and newer ones are more marginal.

Since we spent large amounts of money improving transit infrastructure, but the end result has been an even greater cost per passenger, this suggests that investment in transit is probably a really bad idea, and offers diminishing returns. Looking back, we can see that the most recent spending has resulted in higher overall costs per passenger, therefore we can surmise that the spending that took place prior to 1995 resulted in better costs per passenger than the spending that took place after 1995. The best opportunities to spend money on transit had probably already been used prior to 1995. Spending after 1995 was on projects that on balance weren't really that great.

Light rail seemed to degrade the worst of any of the modes.

Over the period of study, between 1995 and 2004, the cost per passenger for the government to run transit increased on average for all categories, but it increased most for light rail.

Under the assumption that the post peak-oil period involves still further rises in energy prices, if we invest even more heavily in transit, it would appear to me that we are likely to get even more diminishing returns.

If we keep doing this stuff, the end result is more likely than not more of the same.

That means, increased numbers of people using transit, and an increased cost to the taxpayer per person using transit. Put those together and you get a big hole in government spending at a time when presumably, money for projects will be hard to come by.

Hope that helps! Free of charge ;-)

On behalf of all those who, like me, mainly lurk, thank you Stuart for your writing. It is highly valued. When you disappeared some months ago, I very much regretted not having thanked you for your posts.

In response to the points mentioned above:

You don't seperate rolling stock from geological capacity.

Adding rolling stock does not automatically propel people to use them - it has a relatively minor effect once you get past a certain number of trains an hour.  It's pushed upwards by high ridership, it doesn't push ridership upwards.

Adding track and stations, on the other hand, directly adds to ridership in a more than linear fashion - the reach of the network expands, and so more people use it rather than driving, and so it can support more tracks...  TOD takes over near the stations, because the walking distance of a building near a station has expanded from 2 blocks of the building to 2 blocks of any station.

In a post peak or post collapse scenario, the diminishing returns from public transport may simply be due to fewer people needing public transport to travel to work.

In Ottawa, Ontario where I live, we have a well developed public transit system, including light rail. One lane of the main cross town commuter highway has been reserved for public transit.  The bus malls are bright, clean and generally safe, as is the entire system. I have no problem letting my teenagers use the system unaccompanied.

The bridges between Ontario and Quebec connect the greater national capital region and are heavily congested with commuter traffic. However, one lane has been reserved for HOV-3, public transit, and taxis.

In spite of all this, public transit is under utilized. Commuters find themselves parked on the highways and bridges while the public transit lanes remain empty, except the the occassional passing bus.

Maintaining the bridges at HOV-3 instead of HOV-2 is a bit of social engineering which doesn't appear to be working. As well, HOV-3 is not well inforced and there is a lot of defecting.  

Allowing taxis to use the transit lane, to me, sends entirely the wrong message.  If you cannot afford a car, are environmentally concious, or can simply fit public transport into your lifestyle you can move along swiftly.  On the other hand, if you can afford a private rent-a-chauffeur, you can also take advantage of the public transit system.

On another front, there is a general bias against urban sprawl in Canadian cities even though our largest cities suffer from it greatly.  In Ottawa, just outside the core within a 15 minute walk, are beautiful, old residential neighborhoods with expensive, well maintained homes.

However, until recently, the true core of Ottawa has had some poorer, affordable housing that attracted students, welfare recipients, and the homeless who rely on the many shelters.

This is all changing with the aging boomers.  Downtown Ottawa has become "the place to live."  Expensive condominium appartments are springing up everywhere in the heart of the city and in the run up to the recent municipal election, the incumbent alderman for Ottawa Center ran on a platform of moving out the homeless and shelters.  He never quite got around to saying precisely where they were supposed to go.  Perhaps out to the suburbs.

If relying on the charity of wealthy downtown boomers is the "job" of the future, at least these poor souls will have access to an excellent public transportaion system to take them to "work."  That may increase ridership considerably.

I find the idea of HOT (High Occupancy/Toll) lanes attractive.  Instead of a HOV lane, you let anyone go in the lane but they have to pay a toll (electronic transponders or license plate cameras manage the billing).  The toll is posted on signs, varies with congestion, and is set such as to keep the lane always free moving.  A portion of the lane capacity can be reserved for buses, but the rest sold.  Potentially, the toll could also vary based on vehicle fuel capacity or other such things.
My girlfriend gave me the book, "New City," by John Lorinc for my birthday, and in it, he talks about how the demographics of city cores are changing.  Essentially, the idea is that a retired and aging population can't always drive - either for $ or health reasons - and so in order to be well within walking/transit distance of access to doctors and a decent social scene, they've started moving out of the suburbs and into the city cores.  

I've actually noticed this in New Westminster, where I live.  I'd bet the majority of the people here are over 50 or in their 20s.  In that sense, at least, there may be some hope for rejuvenating cities, since both the oldest and youngest (autonomous) generations are starting to return to cities for the accessibility, price (?), and perhaps potential for socializing.

Great work, Stuart.  This is really eye-opening stuff.
Excellent presentation, Stuart. I do have a question however. For you data hounds out there, how much of the construction/maintenance cost of our auto-centric interstate, highway, street systems are funded by gasoline tax revenues which are paid by the users(drivers) of the system? Any additional funds would be like the operating subsidy for transit systems and should be subtracted from same for comparison purposes.
Considering all federal, state and local expenditures on highways, gas taxes cover 35 percent. Sixty-one percent are covered by vehicle taxes, property taxes, income taxes and bonds.  Municipal streets are almost entirely funded by property, sales and in some cases, state income taxes.

That means people who drive less are subsidizing those who drive a lot; those who drive mainly on city streets are subsidizing those who mostly use highways.

See Fueling Transportation Finance: A Primer on the Gas Tax

It is said that road infraestructures can't grow at the same pace as traffic does, so the solution to traffic congestion can't be just more roads.

Does anyone know some references about this?

Anthony Downs:
Traffic: Why It's Getting Worse, What Government Can Do
Why Traffic Congestion Is Here To Stay, And Will Get Worse

The second approach would be building enough additional road capacity to handle all drivers who want to travel in peak hours at the same time without delays. But this "cure" is totally impractical and prohibitively expensive. We would have to widen all expressways and other major commuting roads, demolishing millions of buildings, cutting down trees, and turning much of every region into a giant concrete slab. Then those huge roads would be grossly underutilized in non-commuting hours. There are occasions when adding more roads is a good idea, but no large region can afford enough more roads to eliminate peak-hour congestion.

Another Brookings Institution study:
The Effect of Government Highway Spending on Road Users' Congestion Costs

We find that, on average, one dollar of highway spending in a given year reduces the congestion costs to road users only eleven cents in that year.

Generated Traffic and Induced Travel by Todd Littman:

Urban traffic congestion tends to maintain equilibrium. Congestion reaches a point at which it discourages additional peak-period trips. If road capacity increases, peak-period trips also increase. In the short term this consists primarily of travel diverted from other times, modes, routes and destinations. Over the long run an increasing portion consists of induced vehicle travel, resulting in a total increase in regional VMT. This has several implications for transport planning.

... Ignoring generated traffic results in self-fulfilling "predict and provide" planning. Here is what happens: Planners extrapolate traffic growth rates to predict that congestion will reach "gridlock" unless more capacity is provided. Adding capacity generates traffic, which leads to renewed congestion with higher traffic volumes, and more automobile oriented transport systems and land use patterns. This cycle continues until the costs of increasing road capacity become unacceptable.

Generated traffic does not mean that road capacity projects provide no benefits and should never be implemented. However, current planning practices that ignore some or all generated traffic impacts result in inaccurate forecasts of impacts and benefits. Road projects considered cost effective by conventional models may actually provide little long-term benefit to motorists and make society worse off overall, due to generated traffic. Other strategies may provide greater net benefits when all impacts are considered.

This is what I'm trying to get at. Embrace and accept gridlock. Accept the fact that you can never build/expand/maintain enough highway to avoid gridlock.  Once you have learned to accept gridlock, look at what is a better long term solution. That solution has to be come combination of rail and bus.  To pave over the entire metro area is simply an unacceptable solution although that appears to be one that has been chosen in most metro areas in the United States.

However, if we simply look at the short term return on our transit investments and then compare that to the next phase of fruitless highway expansion and neighborhood destruction, then we make poor investments in the long term. Cities that excel as far as non auto transportation started planning and building decades ago. They are now reaping the benefit.  Would any of these cities like to go back and reverse their investment profile or road vs transit?

Since 95% of Americans get around by car, and congestion causes them major inconvenience, a solution that relies on gridlock is only politically viable until people figure it out. Then they'll be very pissed and throw out of office whoever they think is responsible. Strategies like that are not politically sustainable.
It works in NYC.  :)

The Long Island Expressway, AKA "The World's Longest Parking Lot," was the subject of a planning study awhile back.  It came to the conclusion that even if they double-decked the entire highway and doubled capacity, it would only encourage people who currently take the bus or LIRR to drive, and the congestion would be just as bad as ever.  

The DOT decided it wasn't worth it, and were upfront about it to the people affected.  They understood.  Why spend billions to get the same problem you already have, only uglier?  

New York is a unique place, in so many ways.  The people of Long Island probably don't want the additional development that such an expressway would imply.

If any American city is to try electronic congestion charging, I suspect it will be NYC first.  Or, oddly enough, Los Angeles, which despite stereotypes has a relatively high population density (1) and a history of innovation.

I was struck by the number of SUVs in NYC though.  In Texas, sure, but I would have expected NYC to have the same numbers of SUVs as London (about 1/8).

(1) some distortions there, in that that number is based on the Census Metropolitan Area-- I've heard it said that if you adjust for that, LA doesn't look that dense.  Nonetheless, LA is apparently of above average density amongst American cities.

"policies like that are not politicially sustainable" Sorry, Stuart, people are dumb. Houston, Harris County, Texas has had a policy for years of using the city streets as retention areas to slow down flood waters from our periodic rains to keep them out of the bayous. This results in thousands of drowned automobiles and a whole lot of bitchin'.
  But do they ever vote out County Commissioners? Not in my life time, and I'm 55.
Of course gridlock is "politically sustainable".
There is not a major city on the planet without traffic jams at rush hour. Due to the diminishing returns of paving a city (it pushes uses farther apart, requiring even longer-distance travel across the paved expanses and disadvantages other modes) as the above posts noted, it is impossible to build enough auto capacity to avoid rush-hour traffic jams.
So traffic-jams are not only "politically sustainable" but an inescapable fact of life.

Name one major city that lives without them. A city without gridlock does not exist.

Transit does not serve to prevent gridlock, but only provides an alternative for people who prefer reading a paper on a train to staring at a bumper.

Sorry, I should have been clearer. I agree that the larger the city, the worse the traffic is likely to be and there comes a limit to any practical solution and people learn to accept that (Leanan's example of NYC is a good one). However, I think artificially creating that situation where there's a choice is only going to work if you can mislead people into not realizing what is going on. You risk a backlash.
Part of the future plan for the Denver metro area is to expand U.S. 36 going into Boulder to accomodate what they are say are fugure projections of highway traffic.  Without it, they say, there will be more congestion.

The City of Boulder, which would be the recipient of said traffic, is resisting this approach.  They would prefer no additional lanes with the extra traffic taken care of by light rail and bus.  This political entity is, in essence, saying that they refuse to bow to the paradigm that says we should always acomodate projections of additional traffic with road widening. They are a political entity and will, I belive, be supported by a good majority of the citizens of Boulder.

This approach is not political suicide everywhere.  No doubt there are jurisdictions where this approach would be suicidal. Not here.  

D.C. was often gridlocked when I lived there twenty years ago. My guess is that their mass transit system is successful, in part, because driving is often such a distasteful way to get around, considering the alternatives.


The move to eliminate (dynamite) the elevated section of I-10 from Elysian Fields to Canal Street in New Orleans is still "active".

Hope Springs :-)


Many thanks Laurence!!
I think we should start thinking more about bicycles. They are cheap to make, and everyone can go whereever they want to go. There is no problem with making connections. People may have to adjust to make their trips much shorter, but they can still make trips if they want to.

Transitioning to significant bicycle usage would also help make the big drop in fuel usage that I think we need.

Here in the Atlanta area, I have a difficult time imagining public transit making very big inroads. Everything is spread out, so there is no concentration of traffic going to a single location. I don't see rebuilding infrastructure in concentrated locations at this late date to be a feasible option.

I think you'll find this report interesting: http://parca.samford.edu/roadandbridgeindicatorsweb.htm

I find the section of gas taxes very interesting since it shows that the gas tax in Georgia pays almost none of the cost of roads so most of the money comes from other sources such as sales and property taxes.  It's especially enlightening since the bulk of the population of Georgia believes that gas taxes are too high and that they more than over the cost of roads.  As long as the general public remains willfully ignorant of the actual costs of different transportation modes, things are not going to change.

Very interesting. Is similar information available for other states?

The report also shows that the roads in Georgia are in extremely good condition (Charts 14, 15, 16, and 17). This tends to encourage more road use.

At least for the year shown in the report (2004), GA spending on maintenance and construction was very low (Exhibit 8). This seems to correlate with the low GA gas tax, so less amount available in total to spend. But how can the roads be kept in such good shape, with limited spending? Was 2004 an unusually low spending year?

Thanks! That is a very helpful link...
Hear! Hear! I was pleasantly surprised to read that Baltimore's light rail and metro allow bicycles unless the train is already very crowded.
Yes...but my crystal ball is out of order. How is this more than an empty symbol if it's a riverboat gamble whether you will be let on board? You don't go to the train station because you want to put down chips and roll dice, you go there because you need to get someplace, and you probably need to be there on time.
Agreed. Caltrain has a bike car, but occasionally it's full. It's simply infuriating to have the train one critically needed to get pull out of the station without one.
Ha! I used to take a Ride-On bus to DC Metro in Alexandria, and it seemed that the train was always pulling away just as we were getting off the bus.

The nice part of transit is that you don't have to drive; the bad part is the loss of (percieved) control. Instead of sitting at red lights, you're sitting at the station - waiting for your connection.

If I miss whatever I was headed to, or get dinged for being late for work, that's a reality, not just a perception. I'm usually traveling or commuting in order to meet up with other people, it's not normally a solitary pursuit. And if I can't board, it could be extra hours, not just minutes, until I get where I'm going. After all, there is that row of unstable dominoes called "connections".

Worse yet, if the train I can't board is the last one home at night, then what am I supposed to do? (Just ride the bike home? If that had been practical, what in the world did I need the train for in the first place?)

Read what I wrote. I didn't write "the percieved loss of control," I wrote "the loss of (percieved) control." In your car, even though any number of things can slow or stop your progress, you have a feeling of control.  In the transit system, it is very clear that others have control over your progress.
Fair enough, perhaps, but in the end it's a reality, not just a perception, either way you slice it. And on the whole, I've found public transportation in the USA pretty unreliable. It's probably one of those tail-chasers - if it's unreliable, the "movers and shakers" won't use it, except maybe in Manhattan, and if they don't use something, there's not much impetus to make it usable.
Public transportation is cumbersome and expensive, but finding that they allow bikes on local metro and light rail is a positive sign and greatly expands the range of my biking.
Systems Dynamics theory has a word for this, I'm not sure what it is.

But basically, it's the downward spiral-- happens to any public service.  

You can see assaults like that taking place against public education, mail service, etc.  If you reduce the quality and usership by enough, you get a 'tipping point' and wholesale abandonment and collapse of political support.

Not if the next bus is coming in 10 minutes. In the city I lived in I could always estimate my arrival time with +/- 5min. accuracy. With the traffic here I'm definately hard to beat that. If it is a case of urgency or I'm running late there were always other options at hand - affordable taxis, shuttles etc.

The problem of mass transit in the US is a classical chicken & egg problem. Nobody rides it because its not reliable and nobody invests in it, because nobody rides in it. Therefore unless a major effort is undertaken to invest the huge upfront resources that it requires, I don't see people accepting it, ever. The mistake was made with the urban planning that started some 50 years ago and now I'm afraid is too late to fix it.

My first "real" job was in NYC, and one thing that struck me was that, despite New York's reputation for being full of angry, impatient Type-A people, they were really laid back about punctuality.  Because you never know when the train lines will be delayed or stopped for some reason, or the roads gridlocked due to flooding or a car crash or whatever.  

I was late to my job interview, because I had never ridden a subway before and got on in the wrong direction.  I was half an hour late, and figured I'd lost the job.  But the guy who interviewed me didn't even notice I was late, and hired me on the spot.  I later realized that half an hour late is on-time to the average New Yorker. It's places like Minnesota, where (relatively) empty highways beckon that 10 minutes late is late.

New Yorkers are cool. They'll tell you exactly how they feel and where you stand, none of this passive-aggressive sneaking around. They'll pour their heart out to you, they're probably the most genuine people in the US.

Much has been written about the passive-aggressiveness of the Midwestern personality, and it's worth researching.

Then reflect on the fact the the basic Amerian personality is Midwestern.

Ahhh...all those killings, muggins, rapes and the New Yawkers just stand there or turn and walk away and do absolutely nothing. "Don't wanta get involved." "Not my problem."

Yeah..real people pouring their hearts out.

I can vouch for this. My wife is from the bronx.  There is never any doubt as to what she thinks.  And she cannot stand the passive agressiveness of the midwesterners.
I've always wondered why the really bizarre serial killers are from the midwest.  The coasts have their Son of Sams and Ted Bundys, but it's the midwest where you find the Eddie Geins and Jeffrey Dahmers.  A backlash against all that wholesomeness, maybe? ;)
Sure the wholesomeenss of the New Yawkers.

I have sat in a few bars in NY. You make a simple conversational comment and the guy next to you opens you up like with a canopener. They have compassion a mile wide and a micron deep.

Whatever happened to the old Yankee inguenity? The old "use it up, wear it out, make it last" mentality? I spent a year in upstate, and saw a few old Yankees. Some good folks in Woodstock and Bearsville.

Going to White Plains to our hdqtrs I meet the filth and scum of the universe. Then to Manhattan and sheer terror. I saw a couple standing beside their VW van as it burned to the ground. NO ONE helped them. I saw how the cabbies drove and how if you didnt have the guts to face the crazy traffic down you just sit there forever.

Yeah it was really great. You can have the armpit of the world. I will stay in Kentucky where neighbors help you instead turning their heads and walking away.

Keep repeating your mantra, "I love New Yawk. I love New Yawk."  You know you hate it.

Actually I found New Yorkers very helpful.  As big cities go, I found it to be very friendly.

This is after having visited in the mid 80s, and vowed never to visit again: a scary place 'Escape from New York' was almost a reality.  Of course in the 90s crime and the streets were cleaned up, but there has also been a change of mood.

But after 9-11, everyone says, it has changed-- people wear their hearts more on their sleeves.  I certainly found it so.

London is getting to this point.

Bicycles are, in fact, the only way that we can meaningfully increase traffic thru-put.

Buses have been expanded, to the point they contribute to congestion.  The Tube (subway) is over capacity, and a new £16bn extension is delayed till post the Olympics at least (and maybe never) despite the City (home of 10% of UK GDP) saying that it is vital.

The Congestion Charge has reduced traffic by c. 15%, but by less at peak hours.

Some very interesting stuff there Stuart, but I would be cautious about drawing any firm conclusions at this stage.

One of the key things about public transport is that it has to be pretty good before people start using it to its full capacity. That means that you have to go through an expensive capacity-building phase before you really see any significant take-up. From the capex graph it looks like the ramp-up started around 2000 - perhaps the utilisation will begin to rise? Major cultural shifts don't necessarily happen on the same timescale as the construction projects which enable them.

Another question is how and where that capital expenditure is being spent. Does the US manufacture its own rail infrastructure, or is it imported? Because if it's imported, then the decline of the dollar since 2000 might play a significant role.

Of course, it's also possible that the money was spent on badly-designed boondoogle projects... Or that Americans really are irredeemably wedded to their cars to a much greater extent than we Europeans.

Having lived in both transit poor America and transit rich Europe, I think it's almost certainly because most US development is at too low a density to support transit very well.  However, even in Europe, the trend has been towards driving and away from transit.
I can only agree with this analysis. The US has very few population centers which would be well suited for transit systems. It takes places like London, New York, Singapore or Tokio to make them effective. For most of the US rail systems will never be able to transport significant numbers of people and even bus lines are limited in their reach until the last mile problem can be solved for areas where two-block coverage is impossible.

Bikes are the only form of last-mile transportation that is effective. But they are also completely incompatible with current bus and train designs. That is not only a shame but should also be a call to designers to change railway cars and busses to accomodate more bikes.  

How about segways?  They're a bit expensive now (although not so expensive given the quality), but they would do the job pretty well.
A segway is ten times more expensive than a bike and it maxes out at a rather low speed. I have seen families on bikes with one of the members riding a segway. The result: the bikers had to ride uncomfortably slow to let the segway catch up. The bikers were across intersections in no time while the poor segway was trying to catch up. Personally I would always trade a segway for a really good bike and a vacation on Hawaii.

But it is pretty obvious that current bike designs are not very space-efficient. We need a standard mechanism to fold pedals and handle-bars and make commuter bikes stiff, flat object that can fit into a standardized bike compartment. The current method of hanging them on the bus or somehow jamming them into a heap inside a train car is ridiculous. On Caltrain the conductors are advising people to get folding bikes because those can be taken on any car, not just the bike car. Now... a folding bike is not exactly a very good bike for any purpose. So the choice is between a poor riding experience and not getting on the train...

That's interesting.  It reminds me what I sort've already knew:  segway's are designed for dense urban areas.  They're designed to go where pedestrians go, not where vehicles go.

Bikes are clearly better vehicles on nice, clear, dry roads. On crowded sidewalks, at relatively low near-pedestrian speeds the segway will do better.

Just what we need. More motors zooming by us everywhere. Even small sidewalks won't be safe for walkers. Imagine Segways clogging up the sidewalks of NYC or any city. Only lazy fat ass techno geeks who are adverse to physical exertion could think this stupid invention is cool. There is great invention that goes as fast as a Segway for the fraction of the cost--it's called a bicycle. I vow to kick any Segway driver that zooms by me right in the ass.
Segways are illegal in the large cities in the US too! Old? Semi-handicapped and live in SF? Think a Segway will solve your walking up hills problems? Nope! Illegal! But those battery powered sit-down-and-bump-into-people scooter monstrosities are fine.
Segways are legal almost everywhere, with the notable exceptions of NYC and SF.

I agree that the Segway has been a godsend for a lot of disabled folks.

bikes aren't appropriate for sidewalks - they need a minimum speed to not fall over.  Segways can go at the speed of pedestrians, and not take up more space than a pedestrian.  They're designed to replace cars for trips too long for walking.

You might want to resume your medication....

Could also be an opportunity to see more variations on the Bike, too.  I never rode one of the smaller-wheeled 'fold-ups', but I can see all sorts of possibilities with wheels like the 'Razor Scooters' and on up turning into a folding/packing bike that you can 'George Jetson' and bring into the passenger area with you.

I also think that the "If you build it, they will replan" idea will have more sway in times of tighter LTF supplies.  Businesses and Apartments in the 5-boroughs of NYC have been, for decades now shifting and blossoming in relation to their proximity to transit.  Nothing stands still (unless the subsidy of a cheap energy supply allows it to hover like a Helicopter.. but that get's pricey)  This is an opportunity for the Forward-thinking real gamblers to figure out where the precious transit access lines/points are now sitting undervalued, but could become new focal points of activity and regular foot-traffic.

While I presume that this presentation accurately describes the transit situation in the U.S. as a whole, the interesting information would be revealed if we looked at different cities, with some cities being more successful than others in getting people out of their automobiles and into alternative modes of transportation, including bicyling and walking.

In America, at least, we will never make much progress by simpling tacking on mass transit while still making every effort to accomodating the demands of increased auto traffic. One needs an affordable faster alternative to auto traffic to have a significant impact on people's transportation choices.

I, for example, chose to take a bus to work when I worked in Denver.  The first part of my journey was clearly superior to the automobile as the bus was able to take the HOV lane. In the middle of Denver, however, the bus was forced to share the interstate with all the other autos.  I chose to put up with this unsatisfactory situation anyway because I was so committed to public transportation. I can understand, however, why the vast majority chose the auto as it was a much faster way to get to work.

When I lived in Germany, however, taking transit was  a no brainer.  The bus that took me to the subway was a less than five minute walk.  The subway came by every 3 minutes during rush hour and deposited me a station which was right by my place of employment.

Elsewhere, Alan has reported the massive increases in transit ridership in the D.C. area. I can understand that since that area, where I used to live, has been an auto nightmare for decades.  Rather than try to fix the unfixable, the freeway system, D.C. chose to focus on their mass transportation system, including their fine subway system.

One question would be, how does D.C. stack up against this this overall analysis prepared by Stuart.

Despite these numbers, I think we need to continue to press on with expansion of our transit system.  At the same time, however, we should do nothing to make it easier or even bearable for people to drive their cars in the city. Things will change, but it will take more than higher gas prices.

"Despite these numbers, I think we need to continue to press on with expansion of our transit system"

God forbid the facts should interfere with our theories...

I suggest a very crude rule of thumb.

If you have a large, very densely populated city (on the US scale) that doesn't already have a transit system, or has an incomplete one, it may make sense to add transit to it.  Otherwise, it's a boondoggle.

I would further add that there is an underlying assumption in all of these capital intenseive systems: namely, that the future will be a continuation of today and that people will need transport to jobs at distant locations or to go shopping.  It might make more sense to significantly subsidize PHEVs or PV panels on people's homes with the money.

What has driven me crazy about all discussions about peak energy is that few take the time to consider the longer term consequences of trying to maintain the status quo.  Where's the Precautionary Principle?

Great point.  I would like to think more about localizing production, consumption and jobs as much as possible.  Destroy as much demand for transport as possible then consider how our existing infrastructure can best be used for what little transport we need.  

I know of many people around here who make several trips a day between home and town for errands and shuttling the kids to activities.  Building public transport to support that would be nutty.  

I believe George Monbiot's book "Heat" has some detailed discussion of transport options in the UK.  

While I would agree your numbers largely argue against expansion given the status quo, we should not assume that we will have these same numbers in the future. Part of the key here is that driving needs to be an unattractive alternative to transit. I would agree that if we continue to make driving an attractive option, that there is no sense in spending a a lot of money on transit.  

We need to break this down to see where additional capital expenditures would make sense.  I would agree that your analysis demonstrates that simply building hoping they will come will not work.

To draw that conclusion, you have to look at what they built 20+ years ago, not five or ten.

  Transit alone is not the solution, of course, and should not bear the full burden for 'See, it didn't work'.  An effective long-term plan has to incorporate the ways transit can be a precursor to a shift in community planning.

 Just cause they didn't buy enough Betamax, didn't mean it wasn't a better format.  In that case, it did prove that it wasn't better enough for the market to demand it over VHS.  Question is, will mass-transit provide solutions to energy challenges enough to be worth the heavy setup prices?

  SS seems to suggest that it doesn't, but that's based on New Investments and New Ridership, during a brief, strange period in the US that included $11/oil and two oil wars (at least), a massive Tech Boom for computing, and a rebirth of 'America First'ism .. does the rest of the history of Mass Transit offer any overall benefits that may help reframe this issue?

Stuart has falsely framed the issue, mixing bus with rail in gross #a and then implying that "more rail" is not the issue.  

Of the $10 billion/year, roughly $2 billion in recent years has been for New Starts Urban Rail.

The current administration has heavily pushed BRT (improved bus) and pushed against Urban rail.  I agree with Stuart that more BRT does little good.

IN ALL CASES rail is cheaper to oeprate than buses (even though buses get "free streets").

The operating economics of early light rail (San Diego & Portland as good examples) were better than buses but are far short of the economics today FOR THOSE SAME LINES.

New lines have been added in both cities and the first lines have "matured", altough they are still maturing as new TOD goes in each year.

The economics of maturing systems are interesting.  People get better at what they do the longer they do it (to a limit).  Efficiencies of scale lower costs as new lines are added.  Rail ridership rises year by year on the majority of Urban Rail lines.  As ridership rises, unit costs drop.

Comparing Portland of 1986 to Portland of 2006 would make one think that '86 Portland was a waste of money !  Why didn't you build the much better Portland of 2006 ?

Why. building more '86 Portlands will take you on a downwards spiral of ever increasing costs !  2006 Portlands are "OK" but small & trivial on the national level (Portland OR has meet Kyoto CO2 reductions BTW).  However, diluting them with more '86 Portlands is the path to disaster !

And we spent many billions on buses and more people did not ride more buses, so one must conclude that we should not build Urban rail (implied Urban Rail = Buses, they are all "transit").

Best Hopes for City-by-City case studies, more can be learned that way.

"If you have a large, very densely populated city..."

This is a key point that transportation engineers have analyzed in detail. The denser the development, the more viable the mass transit system. So what is the the minimum density that should be considered? Studies indicate that around 4,200 persons per square mile is an effective minimum. That translates to 6-7 persons per acre, and at 2.5 persons per dwelling, that is 2-3 dwellings per acre.

That's enough for stand-alone garages and swimming pools. At least 25% of the U.S. population (and most likely more than that) lives at these densities or higher.

Assuming that the mass transit system gets these people where these people need to go. If it stops just one mile short... the number of users will drop to close to 0.

I did a rather nasty commute once: five hours total for a distance that would have been less than an hour and a half by car. Of course the problem was that the train pulled into the station the minute the bus left and that the next bus was 29 minutes away. In other words: I had to waste a full hour on waiting and it took the bus longer to get me down the last 7 miles than it had taken the train to get me down the first 30. Had the company been one mile down the road, there would have been no bus at all.

Various municipal systems are implementing vehicle GPS updates sent to web sites, transit stops and cell phones. Those tech advances can help reduce the missed connections and long waits you describe. Fundamentally, though, you're right. Transit lines are going to be more successful where urban form (no matter what its density) is walkable, connected, and in relatively close proximity to transit stops.

One thing that's muddying up these Oil Drum threads on transit is that the discussions cover both current customer preferences and potential post-Peak scenarios. I am interested in the potential for transit post-Peak: What types of transit and other transport solutions can be quickly and cheaply implemented to cope with critical fuel shortages? Whether and how travel behavior will change? Whether and how land use and density will change, and at what rate?

Technically and economically, there is nothing prohibiting rapid change for some of these factors. For instance, density can increase simply by subdividing homes or building accessory units. The more persistent obstacles are cultural and political.

GPS is of no help if they forgot to extend the train line to where it should have gone and now would have to rip up a whole row of city blocks to undo that. The cost for the falling property prices alone would be staggering.

GPS is of no use if the busses are running on half an hour schedule and have rarely more than five passengers. Running them more often doesn't help because the train still doesn't go where people need to go and not everyone is like me and is accepting of a 3.5 hour instead of a 5 hour commute if one pulled all the stops in that system.

This is not about intelligence but about the simple facts of life: you can't cover random traffic flows with public transportation. If the urban area was not planned well, like suburbia, which is not planned at all but just spray painted on a random cheap location next to the highway, no amount of intelligent traffic planning can save the day. A bus works fine over three to five miles distance. Over fifteen it takes longer than anyone can stand... including me.

Transit post peak will have as much potential as before: none. One can't undo a hundred years of missed urban planning and start over. The US is what it is - vast and randomly populated.

"For instance, density can increase simply by subdividing homes or building accessory units."

Good luck subdividing a single family home in suburbia.

Look... some things just don't work. You look at them, shake your head and move on. This is one of those things. Can it be made better? Sure... Can it be made good enough? No, not really. And better is unfortunately not good enough.  

"Look... some things just don't work."

For someone with the handle InfinitePossibilities, you sure have a limited view of what's possible, or how attitudes can change.

In the 1920s, Switzerland made the strategic decision to expand their tram systems (as other nations peaked) and electrify their railroads.

During WW II, the survivied a 6 year complete oil embargo.  At the end, the average Swiss used less oil in a year than the average American uses in a day !  But their society and economy still functioned, and they had a decent quality of life.

It took almost two decades of preparation,

Please note my post elsewhere on this thread on their 1998 rail plans for 2020.

How un-American !

Best Hopes for Long Range Planning !


"God forbid the facts should interfere with our theories..."

There are many problems with your analysis, but the absence of critical information about fixed route transit is the most glaring.  I will mention one: transportation planners today refer to something called 'uplifting' when calculating the costs of installing new fixed route public transit, of which rail is the best example (bus transitways are another).  Uplifting refers to the increased value of land in the vicinity of transit stations.  Why is the land more valuable?  Because ready access to transit increases demand for living/commercial use of the location and allows for intensification.  Municipal taxes rise with the intensified development such as medium-highrise buildings.  This is not theory, but a description of what happens in almost all circumstances.

Intensification takes time, but only incredibly stupid political decisions stand in the way of this market process.
The result is that overtime a 'natural' transit clientele is developed.

The numbers you use indicate an expansion of rail transit capacity, leading to a lower rate of capacity utilisation.

Like too much economic analysis, yours ignores change over time.

Your analysis does not permit you to conclude that building a transit system in places other than very densely populated cities without an existing transit system is a boondoggle.

Maybe his analysis does not permit him to do that but in reality... where is an example of a transit system outside of densely populated areas that actually works for more than a tiny majority of the population? Do you have any examples?
2.5 blocks from my home.  The St. Charles Streetcar Line.

Depends upon you definition of "densely populated" of course.  
Using 80+ year old equipment, it covered (pre-K) 80% of it's operating expenses from fares & ads.

A majority of people used it (still shut down) at least occasionally.  Couple nearby walk to work on nice days to 51 story One Shell Square (about a mile away) and take the streetcar on "bad days".

Houses are a mix of 1,2 & 3 stories, plenty of "green" space.


And now let's hear your plans to build one just like that for the people who commute 60 miles from the Central Valley into San Jose daily. How much ridership do you expect? At what cost?

For the 1000ths time: it is not about recovering operating expenses! For heavens sake... the problem is that rails are one dimensional and commuting is a two dimensional problem...

An Urban Rail SYSTEM is a two dimensional system.

See Miami Plans.


Dark brown lines are post-2016 plans.  I would like to see 100% build-out by 2013/14.

Without additional TOD, 90% will live within 3 miles of a rail station, 50% within 2 miles.  My own SWAG is that 20% will live within 3-4 blocks of a station and 40% of jobs will be that close by 2020/25.

Each station has bus lines servicing local destination.

Best Hopes,


And now let's hear your plans to build one just like that for the people who commute 60 miles from the Central Valley into San Jose daily. How much ridership do you expect?
Post peak oil, such lifestyles will be irrelevant. Substistence living doesn't include 60-mile commutes!
So we're cornucopians now?  Is a 60 mile commute going to be viable in a car with $10/gallon gasoline?  
Ottawa, Ontario.  The bus transitway (separated and separated at grade wherever possible) has engendered huge development at different nodes along the corridor, a process which is ongoing.  Since the inception of the transitway, I believe Ottawa has achieved the highest percentage of commuter traffic using public transit of any city in North America.  Ottawa is not a densely populated city, and was even less so when the transitway was built.

The only thing wrong about the system is the use of buses instead of rail, especially given the higher labour and energy costs of buses.  The system was designed to permit a change to rail at some future date.  

There are other problems with Stuart's take on transit, not the least of which is the muddled thinking on subsidies.  It is important when considering the comparative costs of transportation (or any) systems to think in terms of costs (and of course benefits) to the economy, regional and otherwise. The subsidy to a system operator to meet operating costs that exceed farebox revenue in most cases provide a benefit to the economy.  For this same reason, highways are largely socialized in most countries.

A proper comparison of different transport modes has as is oft noted to consider externalities such as pollution costs. But there are other considerations: what, for example, are the opportunity costs of the dollars exported (from the local/regional economy) to purchase the energy in excess of the energy required for an alternative system? What is the impact in an energy constrained world (supply-led) of effectively unnecessary demand, because transport alternatives exist, on the operations of those providers of goods and services for whom consumption is not elastic.  Ambulance and fire services are charged examples.

The essential problems with Stuart's analysis are, in my opinion, twofold: first, it is static and, second, it assumes a demand-led economy, when we are entering an era wherein the transcendant factor in the economy, energy, or even, energy-matter, is irrevocably limited.


Ottawa WAS the poster child for BRT, until they figured out it wasn't working.  Big dispute whether to build N-S or E-W first, and whether to build light rail in subway down town ("heavy" Light Rail).


Pittsburgh, the US city most devoted to BRT has no plans for more BRT and is currently building light rail to the North Shore, where several potential lines can feed into it.

Best Hopes for Urban Rail,


I'm not a big fan of BRT and was among those lobbying for rail a quarter century ago when Ottawa chose buses over rail.  At the time, the BRT option better fit the funding arrangements offered by the provincial government in terms of the ratio of operating to capital grants, and of course the bus mafia was well represented within the group of municipal planners who provided the local politicians with cost estimates for the two options.  IOvertime it became evident that they had seriously underestimated the cost of BRT, but by then it was too late.  And overtime rail options have only improved.

Nonetheless, it is not correct to say that BRT doesn't work in Ottawa.  It moves a lot of people every day. Which is not to say that any city contemplating fixed route transit today is not better off to choose rail.  

Strawman Alert!
Is anyone proposing adding transit anywhere other than the dense regions of cities?
I have not heard of any plans for light rail in Baggs, Wyoming.
Essentially all major US cities have corridors with sufficient density to support rail transit, yet many of these corridors currently lack transit. Most US cities are adding rail capacity as we speak, even sprawling cities like Phoenix and Dallas-Fort Worth, but they are adding rail along the densest and most heavily travelled corridors. The densifying effect of the Transit-Oriented-Development which follows rail construction is well-documented and visible around every new rail system which I have seen. Over the centuries of useful life offered by rail systems, transit nodes will re-develop to higher densities.

For example the Cinderella City redevelopment along Denver's light rail.
-"A decade ago, the fortunes of this lower-middle-class Denver suburb were inextricably linked to Cinderella City, a mall virtually devoid of retailers, and to the automobile.
Today, where the 1960s mall and its parking lots once sat, Englewood has latched onto a new hope: 55 acres of apartments, stores and offices anchored by a stop on a light-rail line that links Englewood to downtown Denver. The $155 million development features 438 apartments, 350,000 square feet of retail space, including a Wal-Mart, and the town's library and city hall. "Our community has a new core," says Gary Sears, Englewood city manager.
Opened in 2000, CityCenter Englewood is one of a growing number of developments offering a mix of uses that have emerged around mass-transit rail lines, a trend that is particularly notable in the car-friendly West. In Dallas, Mockingbird Station on the city's light-rail system boasts an artsy Angelika Film Center, loft apartments that utilize a 1940s warehouse, and retail and office space. And the Del Mar stop in Pasadena, Calif., along Los Angeles's light rail includes 347 apartments stacked over ground floors intended for shops that will give the project a dense, urban feel amid its low-rise Southern California surroundings when it is completed early next year.
Such projects still have plenty of challenges, one of the reasons the Federal Transit Administration counts only about 100 of them nationwide. Because zoning often favors a single use, such as residential or commercial, developers must work with city officials to change the rules. Land can be difficult to assemble. And upfront costs for developers -- especially with demand to include features such as plazas and parks -- are high with financial returns often slow in coming."_

So, in summary, during this pre-peak run-up in energy prices, we invested more and more heavily in transit. The effect of that was to increase capacity, but lower utilization.

So in the case of mass transit we currently have supply well above consumption - good.

The price spikesfor gasoline the past few years have not been enough to end the Mass Delusion of the Culture of the Automobile.  

It's good to know we at least we have a cushion for when the Post Peak Reality pops that bubble.

"So in the case of mass transit we currently have supply well above consumption - good."

In reality... in my area a real gas shortage would leave a hundred thousand people stranded and the transit systems have the capacity to move a couple thousand more than they actually do. In other words: even if the trains went where people needed to go, only very few would be able to get on the train.

This is a typical case of "Just because you want really badly it does not make it so.".

Thus my call for a "Strategic Railcar Reserve" as Step 5 of my 10% plan.


Even in DC & NYC, all but the Lexington Ave line can accept more rolling stock in an emergency.

In addition plans could be made for expanding Park & Ride lots (DART lots were FULL after Katrina) in a variety of ways, as well as more buses to rail stations).  Blocking off 2 lanes of nearby 4 lane streets and turning them into parking, parking on grass, commandering nearby office & retail parking, etc.

But this takes advance planning.

The slide down the Peak Oil slope will NOT be smooth !

Best Hopes for Rational Planning,


What about

"10% is not an order of magnitude"

do you not understand?

You are a factor of 20 away from covering every commuter and all the systems that actually work are maxed out.

You want to block off two lanes of the nearby highway for parking? Be my guest... the average distance between rail system and highway around here is a mile and a half. I am sure a lot of people will love the opportunity for that walk every morning and every evening. And don't forget to charge them $7 for parking. They will like that even better.


Guys... this is getting ridiculous... you know. PO means you have to give up your freaking SUV and get a smaller car... it does not mean to dig up half of America to install yet another non-functional transit system.

What is there so hard to understand? Life is not "Sim City". You can't just erase the whole game and start over. You have to work with what you got. And what you got are highways.

Not for long.  Highway upkeep is too material and energy intensive.  If you want to develop a system that gives you some longevity, keeping the asphalt fed is not likely to work well for you.  Track takes work, too, but pound for pound it works harder and longer than a highway.

I'd like to think there was a more durable system that we could start resurfacing our roads with.. but in those quantities, what would you propose?   My "Sim-Highway" dream project would be Nano-machines in weather balloons, pulling carbon molecules from the atmosphere and tossing little bundles of 'BuckyBalls' and 'Nano Carbon Tubes' down for new construction.. but I bet it would be hard..

Concrete ties are expected to last 50 years (slightly less for those with heavy coal traffic, longer for "light duty", such as commuter rail).  Ballast basically last forever.  Steel rails depend on weight of rail vs. loads.

Rails used in DC Metro have the better part of a century left.  OTOH, NYC subways have some worn rail that could be replaced next year.  Boston rail is bas in some spots, good in others (spot replacement ?).

Urban Rail can be realistically expected to last the better part of a century to over a century with minimal work (not so true of old wooden ties).

The new Swiss rail (pax + freight) is being designed to last 100 years with minimal maintenance under extremely heavy use.

Best Hopes for Long Lived Investments,


BTW, The under construction Greenbush commuter line south of Boston should be extraordinarily long lived.  Well built, properly banked, cracked granite ballast (the best), concrete ties (except at road crossings) fairly heavy rail (135# from memory), well built and about 30 light commuter rail trains/day (single track mainly, so same rail used coming & going).

High speed switches will wear in decades (3, 4 ?) and wood ties at road crossings was a mistake (IMVHO) but rest should last a LONG time.  Salt is not used on rail lines, so that problem is avoided.

"in my area a real gas shortage would leave a hundred thousand people stranded"

Yep! If we plan not to have a permanent gas shortage, as in post peak, and then we decide not to build the infrastructure to take care of that shortage, then these people will be stranded. In the short to medium term, the required investment might seem to be a very bad allocation of resources. But what of the long term?  

The people in the Denver area decided to tax themselves to provide billions to expand their light rail system.  They will not get their "return" for a couple of decades, if that. Many of them will be long dead before this thing reaches its full flower. Perhaps it is all just a gross waste of capital. Perhaps.  Part of that decision, however, entailed a vision, where people decided that total dependence upon the auto, which has already destroyed much of this area, was simply unacceptable.

One drawback, though, is that they should have embarked on this twenty years ago.  All this investment may be too late for post peak.  If I were in my 20s or 30 or early 40s, however, I know where I would be purchasing real estate.  

They could have kept raising CAFE standards for the past 20 years. If they had... PO would still be a decade away.

Coulda, woulda, shoulda. Great game. And so helpful....


"This is a typical case of "Just because you want really badly it does not make it so.".

It is also a case of one of many "infinite possibilities" not having a high probability.

Stuart, you are presenting a skewed picture by lumping all cities together. Why not compare cities with good transit and do separate stats for cities with bad transit.

S.F. has among nation's highest public transit use among commuters

San Francisco is among the cities with the highest use of public transit, according to a U.S. Census Bureau report released Tuesday. San Francisco tied with Boston, where about 31 percent of workers commute on public transportation.

The only U.S. cities with higher public transit use were New York, with 55 percent, and Washington, D.C., with 37 percent. One-third of the nation's 6.4 million people who travel to work on public transportation use systems in New York, according to the report.

Of the approximately 6.4 million people nationwide who usually travel to work using public transportation, nearly one-third live in New York City, according to a new analysis of American Community Survey data released today by the U.S. Census Bureau.

Chicago and Philadelphia each have 27 percent public transit usage among commuters. Newark, N.J. had 26 percent and Baltimore had 25 percent.

Only one Los Angeles worker in eight uses public transportation, and in Houston, only 6 percent of workers use public transportation.

Overall, 5 percent of the nation's 128.6 million workers use public transportation to get to work.

Other survey highlights:

    * Nationwide, 77 percent of workers drove alone to work, 10 percent carpooled and 2 percent walked.
    * Bus transportation accounted for 55 percent of public transportation use nationally; subway

So the good news is: 5% of the nation's workers used public transportation to get to work.

Can I translate that into bad news for you:

95% of the nation's workers use their cars. Of the 5% which use public transportation, approx. 50% live in areas where the transportation systems are operating close to their peak capacity during commute hours, New York and Washington.

In other words... we could probably transport 8-10% of all commuters on public transportation systems, everyone else, i.e. 90% or more still need to travel by car.

In other words: the energy savings in the best case scenario are marginal.  

5% of 84 million bbl/day=4.2 million bbl/day, not enough but not marginal, its a couple of silver bb's in a shotgun load
4% annual oil shortfall... 3.36 million barrel/day. OK... oilman, what shall we do after the first year and a quarter?
Hello InfinitePossibilities,

...what shall we do after the first year and a quarter?

Kunstler's reality will offer shoe-leather or bicycles.  I suggest to everyone to have a bicycle ready to go.  

Never forget that mass-produced, standardized bicycle tires wear much longer than customized, size-specific footwear.  Compare the price and maximum mileage wear of a couple of bicycles tires to decent quality walking shoes or boots that won't give you blisters.  

PostPeak: shuffling your feet and/or dragging your heels will be seen as a very expensive waste of shoe leather.

Bob Shaw in Phx,Az  Are Humans Smarter than Yeast?

When I was a kid growing up in Los Angeles in the 60s, my family had one car.  When my father used it for work, which was most of the time, my mother, sister, and I used the bus system to get around Los Angeles for shopping, services (dentist, doctor), etc.

Now it is not necessary.  The lower-middle class, which is where I would have placed us at the time, could afford one car.  Now they have multiple cars.  It's an expectation.  Each kids gets a car.

Going back to the buses, giving up the second car, and reserving the only car for absolute necessities, will be unconformtable for most if not all.  Who wants to go backwards?

But the basic idea driving this transporation analysis is that energy will become more dear, more expensive, less driving will take place.  And that choices will have to be made.  Yet I see no discussion of choices.  There is no discussion of the future.  Just an analysis of the past based upon low (and still low) energy prices.

Analysis looking at past transportion usage, during a period of rising but still enormously cheap energy, has little preditive value, in my mind, to what will come.

It's useful to look at.  But I think it tells us little and, other than pulling together some great information on transportation, the presentation does not at all argue or make the case that it has any predictive value.

Past results do not always predict future performance.


As mentioned yesterday.


New Starts Light rail spending is in the range of $2 billion/year (federal & local, I gave the current federal spending yesterday at $1.25 billion/year with 50% match common).

This is the segment that I want to explode with $125 to $150 billion for Phase I.

You and others will note that I have NEVER promoted buses as a solution, other than advocating electric trolley buses to replace FF buses on busy routes.

Buses promote zero changes to the Urban form (TOD)(zero indirect energy savings), have high operating costs and fuel consumption /pax-mile varies from good to terrible.  That is where "Transit" is in the US, with rail as a small afterthought.

In other forum, I have advocated increasing bus replacement cycles to 15 years from 12 years and using the savings for Urban Rail. In other words, shift part fo that $10 billion from bus to rail.

You have caught a labor savings (but electricity consuming) trend among agencies not to decouple trains between rush hours and run full trains sets all day.  Easily reversed if/when the labor rate/electricity cost ratio reverses.

New lines do not have the passenger density of long established line ?

Big deal, I could have told you that !  There is zero TOD for new lines and ridership climbs over time.  WMATA (DC Metro) has been counting on 2% to 3% annual increases for decades. They saw a spike this summer and flat growth since then.


Boston will open a new (non-FTA funded, a rarity) diesel commuter rail line south of town (Greenbush http://www.cbbgreenbush.com/routemap.html I took a tour after ASPO Boston) in a few months.  I forecast that it will have lower than average riderhship (compared to other Boston commuter rail lines) for a decade.  I also forecast that the new line will steal ridership from the line directly to west, and the commuter ferry, thereby lowering your metric of pax/vehicle mile for Boston.

Greenbush was built with future electrification in mind, but several bridges closer to Boston have to be raised for all of the South Boston commuter rail lines to be electrified.

Obviously a waste of money (by your inferred logic) !  Spend money to build and then LOWER average pax-miles/vehicle !  WHAT A WASTE !

The 4 track Lexington Avenue subway in NYC is above nominal capacity at 600,000 to 650,000 pax/day.  Near crush loads at rush hour.  VERY cost effective !  I want to build the 2nd Avenue subway to take the load off of Lexington Ave., and attract pax that will not tolerate such crowding.

Obviously a waste of money (by your inferred logic) !  Spend money to build and then LOWER average pax-miles/vehicle !  WHAT A WASTE !

Other than the stolen pax from the line to the west and the ferry (they will travel to a new station for Park & Ride if it is closer) it will take people who would have otherwise driven into Boston.  And it will add several thousand more daily riders to the "T"s Red, Green, Blue and Orange lines (some will walk from South Station terminus to work).  Oddly, bus transfers may be added later by rerouting existing bus routes but walkups (~15%), bicyclists, Park & Ride and Kiss & Ride will be the source of riders.

The new Urban rail lines have the following in common (A few exceptions).

  1. Lower than than US average passenger density (I prefer pax/mile averaged over all miles, i.e tennysons).

  2. First year ridership is the lowest, going up almost every year after that.

  3. Higher ridership in their first year than the bus service they displaced.

  4. Higher than FTA projected ridership first year

  5. Lower operating costs/pax-mile than the buses they displaced. Thus pax/vehicle ratios 1/3rd or 1/4th of US averages are still more cost effective to operate than buses.

  6. Adding ridership to other rail lines in the system (Greenbush will add subway riders to the "T", but steal riders from commuter line to West).  Thus new Line B added to new Line A will increase cost effectiveness of Line A.  Same is true of Lines C, D, E ...

Systems work better than single lines.  They feed into each other.

  1. Roughly half of new Light Rail Lines are single lines.  Denver (until a few months ago), Houston, Minneapolis, St. Louis (until earlier this year), Salt Lake City (until 2004 ?)  More single lines planned in Seattle, Phoenix, Charlotte,.

  2. Streetcars carry fewer pax/vehicle than do Light Rail, and I think the two are combined in FTA stats.  There several small touristy streetcar lines added in the last decade.  Memphis, Tampa, Dallas, Kenosha.  And the 24 streetcar Canal Streetcar Line in New Orleans.  I suspect some dilution in # due to disproportionate growth in streetcar %.

Did you even read the http://www.vtpi.org papers I pointed towards ?

BTW, I just finished budgetary cost estimates of Elysian Fields & Desire Streetcar Lines using Boh Bros and Public Works #s.  $203 million for 11.2 miles of streetcars, 32 streetcars.

Best Hopes for Proper Metrics being Used,


Trains are like Computers. The more you connect the better they are.

Ever been to Japan? Japan trains are AWESOME. YOU DO NOT NEED A CAR.


I thought the Electic buses in San Franciso were very nice. They allowed connections between major modes of transport( Cal Train, BART, ect) without a massive infrastructure investment.

Lets face it... we need SOME roads. If only for horse drawn wagons.... might as well use them for buses too.

It was amusing when a bus driver took a corner too fast and the bus came off the overhead wire. Apparently the driver was used to that, he just jumped out and reset the connector thingie.


"Japan trains" are fabulous but also expensive - a $127 fare for a 553km or 330-mile trip, about 40 cents/mile. (Which is consistent with or a little on the low side of what I remember, short-notice fares ranging up to 70 cents/mile.) I just can't picture Americans putting out 40 cents/mile for train seats of any description. And in metro Tokyo, the local JR trains were frequent and efficient, but fares ranged up to $5, while even the longest distances were short. People in New York, for example, complain incessantly that their $1.66 effective fare, which can take them 25 miles or more, is outrageous and exploitative. (25 miles would get you halfway to Mt. Fuji - scales of distance are so much smaller in Japan or the tiny countries of tourist-Europe than in the USA.) Of course, maybe that super low fare is part of why the subway is not fabulous.
I was thinking the local and express, not the Shinkensen.

I could get anywhere on the Kanto plains for a few bucks, well worth it.

In another forum, a while ago, I discussed the Salt lake City long term plans (30 years, but they are increasing taxes to build it faster, perhaps in 15) and how they properly matched mode to projected density (after TOD effects in 20 year horizon after opening).

For example, they have Light Rail servicing one side of the airport (from downtown) and streetcar service from suburbs in the other direction.

Expanding upon that, I would propose the following rankings of mode for density (tennysons).

  1. 4 track Rapid Rail (subways, but can be elevated i.e. Miami)  Open in NYC, Los Angeles Red Line should be 4 track down Wilshire (it will not be), CBD section of Miami (it is not), certain sections of DC Metro (most of Red Line, other shorter sections) should have been built a 4 track but they were not.  Lexington Avenue subway in NYC should be carrying 500,000-550,000/day.  With 14 or 16 car platforms, a 4 track subway could pack in 3/4 million/day (US conditions).  Beyond that capacity, build another.  Cars are over 3 m wide.  Typically 3rd rail powered.

  2. 2 track Rapid Rail - As above, but without express service (typically)  Max perhaps 300,000/day.  Rapid Rail is characterized by complete grade seperation and long trains (rarely fewer than 6 at rush hour) and stops every mile (1.6 km) or so.

  3. "Heavy" Light Rail - Grade seperated (subway or elevated) in the CBD and over major highways but at-grade crossings over lower density streets.  Rarely more than 4 cars long at rush hour, stops every mile, 2.65 m wide cars. max capacity about 120,000/day.  Overhead wire.  Dallas would be a good example (except wider cars).

3B) Commuter Trains - On freight railroad type tracks from suburban locations. Trains can be a dozen cars long, stops every 2 to 5 miles (some longer).  Not grade seperated except in high density locations, diesel, 3rd rail or overhead wire. 3+ m wide cars. 1,000/day/line to 500,000/day (vague memory for Long Island RR).

  1. "Light" Light Rail - No grade seperation, 2 to 3 car trains than 2 car trains, 2.65 m wide, stops every mile, overhead wire.  Max capacity ~50,000/day.

  2. Streetcars - Stops every 2-3 blocks, narrow single cars (2.5 m or less), 25 to 40 seats, overhead wire, Canal Streetcar Line was 30,000/day and could max out at 40+ K, but this is beyond most streetcar lines.  Typically max 15 to 20,000/day.

Rail of all types above attract more pax than buses.  Gain is 25% to 100+%.  Rail creates TOD (commuter rail limited TOD effect), buses do not.

  1. BRT - Bus Rapid Transit - Should be (but is not) electric trolley buses.  Basically "Improved bus" with it's own private road for at least part of the route.  Max economic capacity about 12,000/day (at that level streetcars are typically a better choice).

  2. Electric Trolley Buses - Full size electric buses on heavy routes.  Breakeven hard to determine, but routes with 8 to 10,000 pax/day are good candidates at current prices.  ETBs attract 5% to 10% more pax than FF buses due to smooth, quiet ride.

  3. Full Size Diesel (or other FF) Buses - The best solution for routes with enough ridership to fill bus with standees at rush hour.  If service is longer than 1 every 12 minutes, unlikely to be worth electrification at today' prices.

  4. Short Buses are useful to collect pax for neighborhoods for xfer to rail or for "special cases".  All operating costs except fuel are the same as full size buses (they tear roads up far less, but bus wear & tear on roads is not part of transit budget).  As fuel rises in % of total bus operating costs, greater use may be seen of short buses.

  5. Neighborhood Electric Vehicles - Post-Peak Oil, these un-heated/un-air conditioned energy efficient vehicles will work well in walkable neighborhoods with short trips.  Quite suitable for commuting to Park & Ride stations a few miles away, where they will almost double capacity in the lots.  Available today at http://www.gemcars.com  Energy consumption is significantly less than Telsa like EVs.

What I have tried to sketch out is an outline of matching mode to likely demand.  Today, we are "under-building" new Urban Rail due to severe budget constraints (as I said, New Starts Urban Rail is currently getting less than $3 billion/year fed+local funding and was getting $2 billion in recent years).

The under-construction East Side Extension of the Light Rail Gold Line will have ~1.8 mile subway section (2 stations) with the rest of the 6 miles at grade.  It was originally designed as an extension of the Red Line subway.  Ridership would have supported  a subway.  A faster (grade seperated is faster) one seat ride (instead of xfering to the Red Line) would have attracted more pax, but the funding for the better option was not available.  Higher ridership & faster operation > Lower costs of operation & higher capacity post-Peak Oil.

Likewise, St. Louis (from memory) skimped on some start-up stations and they can only handle 3 car trains.  Now the demand is clearly there for 4 car trains, but station expansion is blocked by TOD development.

Washington DC is within sight of ultimate capacity today on several lines due to putting in 2 track subways with 8 car stations in the 1970s/80s.  An additional 10% more then for short sections of 4 track and 10 or 12 car stations, could have doubled ultimate system capacity.  We will need that capacity post-peak Oil.

Stuart has said that I want to spend $1 trillion on "transit" in adapting the US post-Peak Oil and points to past "transit" spending on "infrastructure" (ignoring trillions spent of highway rolling stock) and calling it a waste.  Hirsch & Bezdek want to spend $5+ trillion over 20 years of CTL, oil shale, EOR and improved vehicle fuel economy (more expensive than CTL per Hirsch study).

I refuse to believe that new buses & maintenance/operations/replacement rolling stock are not included in that $10 billion for "infrastructure".  I have some concept of what funds are being spent on, and the sums are just not there w/o adding those items.

I have identified $125-150 billion in "on the shelf" plans where, with "urgent" pressure, actual construction could start in 12 to 36 months (perhaps 42 in a few cases).  Completion would be 2 to 6 years after start (NYC 2nd Ave subway & LA Red Line to the Sea would take longest to build, but have the highest ridership).

I can see a hazy $300 to $500 billion past that in "Phase II".  Lots of low cost streetcars, ETBs, some add-ons (run Miami elevated Rapid Rail into Ft. Lauderdale, run BART around the Bay and to Sacramento, add new lines close to near capacity lines in DC, take part of "Big Dig" to connect North & South stations in Boston), electrify every commuter rail line in the US, do something for Detroit and other overlooked cities w/o plans.

I honestly think that $300 to $500 billion would cover everything that could be reasonably planned and designed (properly) within a 5 or so year period.  1 or 2 new San Jose's would happen, but hopefully controls would limit their #.

Now, once Phase I is built and much of Phase II is open or nearly so (ten + years ahead), demand will grow for more as experience grows and the bite of late Post-Peak Oil sinks in.  A Phase III might run the total up to a trillion, maybe not.  Crystal ball fuzzy !

OTOH, I would expect limited if any expansion of FF bus service (where the bulk of past "infrastructure" spending went).  Buses displaced by new Urban Rail & ETBs would fill demand for new service post-Peak Oil.  Their high costs would limit their appeal to transit agencies. (WMATA in DC covers rapid rail costs by 85% with farebox, soon 95%-100% with fare increases.  Their bus operating expenses are ~40% covered by fares).

Thus I would like to see a 80/20 mix in funding FF bus/rail & ETB go to a 02/98 or 01/99 mix for funding FF Bus/rail & ETB.  Any arguments against the past 80/20 mix in an era of cheap oil are simply invalid against a projected 02/98 mix post-Peak Oil.

Best Hopes,

Alan Drake

How many more people would get moved if ALL these plans were in place? A couple million out of... 100 million? Sounds just like corn ethanol to me... it's a non-starter.
Daily ridership of DC Metro is 564,000 in FY 2006.

Significantly higher oil prices, finishing the Silver Line to Tysons Corner & Dulles, plus the Purple Line in MD and the 40 miles of DC streetcar lines could easily push it to 800,000 with more rolling stock (basically capacity).

Miami's plans are a mirror image of DC (DC original 103 miles, today 106, Miami plans 103 miles of mainly Rapid Rail).  Since Miami is more compact (90% within 3 miles, 50% within 2 miles of a station), 600,000 to 750,000 in a high oil price world seems reasonable but over 1 million could happen (Massive TOD already today, bus feeding rail).

Denver is planning a mix of commuter rail (north) and light rail (south).  117 or 119 miles.  Say 500,000 there.

Salt Lake City with ~90 miles of light rail and streetcar could get a 250,000 weekday riders.  A new city will be built as TOD on an old copper mine.

If Dallas's 2015-2030 plans are built out ASAP, and they have 145 miles of "heavy" light rail, 500,000 is doable "in a high oil cost environment".

30 cities with a starter 1 line light rail (Stuart will not like the cost #s), 35,000 daily riders avg., = over 1 million.  That may be low with "high oil costs".

Red Line subway in LA to the sea, (with Vermont branch) could do 400,000 to 500,000.  Basically capacity for 2 tracks (unless we build 4 tracks for new tunnels).  Add new Light Rail in LA (with downtown connector, 4 tracks subway, 2 surface) "in a high oil cost environment" and 100,000/line is "doable".  Blue Line is 100,000 today.

How many LA light rail lines ? 6 or 8 ?  Lots of plans, lots of potential (they once had 1,000 miles of Urban Rail in LA Basin).  At least 1 million potential in LA basin, bordering on 2 million.

Increase NYC area ridership by 20% is doable, and substantial.

More in Chicago, Boston, Philly with low cost add-ons "in a high oil cost environment".

Not all cities covered above, but a substantial impact.  And that is just for "Phase I" at $125-150 billion.

So now you got 5.25 million out of 100 million. That is 5.25%. PO will reduce oil by 4% a year. In other words... everything you've done so far helps us to get past the first year and a half, maybe.

Any more ideas? How about years 2, 3, 4, 5?

Give it up. These plans look great, on your web site. In reality... none of this will do much in the short run.

What will do something is people getting smaller cars and they will be sharing rides. That alone will beat EVERYTHING you got by half an order of magnitude.

That 5.25 million was just for part of Phase I (not all of it).  Add much more for Phase II.

Add 2 million b/day by electrifying freight railroads and putting tolls on Interstate highways high enough to "Swiss" long distance heavy trucks.

Add another 200,000 b/day with strong pax regional rail (250, perhaps 500 mile) passenger rail system.  A minority of market share, but a savings.

Not 100% of the solution, but a good step towards one !


And then what?  Oil becomes even more scarce and those smaller cars stop running.  Obviously smaller cars, hybrid cars, electric cars, etc, are part of the solution.  But so is increased rail transit.  It's the combination that will get the job done, not one or the other.  I would suggest that smaller cars are also something that simple economics will take care of, without needing that much action on the part of the government, the same is not true for installing rail lines.  Thus it makes sense for cities to follow this path of building and promoting more rail transit.  
I have never been to Japan but I remember hearing it is a VERY densely populated string of islands: 337 people per square km. In comparison the US has 30... some 11 times less. That might explain a few things...


Bada-bing, bada-bing, you get the prize. Not only is the landmass comparable to California, but with four times as many people; most of the population is confined to the narrow coastal plains, which widen out slightly in the Tokyo region.

Metro Tokyo: 25 million people in around 1000 square miles, i.e. nearly the population of California compressed into the acreage of one smallish U.S. county. NYC has nothing on it, we're talking nearly the entire population of New York State and New Jersey combined. It's no wonder the local/metro rail service is comprehensive. It's also impeccably punctual, but not as cheap as in the US.

Oh, and the apartments are tinier than tiny, and stacked high. Now, how many Americans are really looking to live that way? New TOD shows there are some, but really, how many, as a percentage?

So I think, in the midst of all that, you are agreeing with me that the more public funds we invest in installing new rail systems, the higher the operating subsidy per passenger that will be required in the future?

I think the rest of us can draw our own conclusions about whether that is a good use of our tax dollars or not.

Stuart-thanks for the interesting, thought-provoking post. Just one serious que$tion-How much do highway construction and maintainence cost us? How does that compare with the $5,000 0r $6,000 per automobile plus the automotive infrastructure subsidies?  How about adding in the military costs of the war, the equipment for war and the lives from war and pollution and ?My hunch is that public transportation saves a bunch of money. AND ITS MAKES US MORE SECURE!
The important point here is that rail isn't the only alternative.  We can double ICE vehicle efficiency very easily, and double it again by investing perhaps $500 per year per vehicle in PHEV's, and eliminate oil entirely with EV's, which are more energy efficient as rail (excepting TOD effects, of course).  EV's might cost 10% more than your average ICE vehicle now.  Not really a big deal.
I agree that "rail is not the only solution".  We would have to be extraordinarily lucky# if the "electric rail only" solution would be enough.  Any prudent plan needs a large handful of silver BBs !

Best Hopes,


# I do think that under a peculiar set of circumstances "electric rail" alone would be enough.  That would include

  1. An urgent program is started in the 2007 Congress
  2. Peak Exports in 2013
  3. Dollar does not drop quickly or far
  4. Post-peak Oil exports drop 1% to 2% per year for at least the first dozen years.

I will let others judge the probability of all 4.  My SWAG is < 0.1%.
Highway maintenance cost are due to heavy trucks. Get the trucks of the road and highway maintenance will go down.

The war is a consequence of electing an imbecile for president  who is the front for a neo-con ideology that is falling apart like a house of cards. It has nothing to do with transportation cost. Riding the train does not stop idiots from waging war. Never has... just look at WWI pictures with scores of young men riding the train to the front line.

"Get the trucks off the road."
Ok...you say so and your right so remove them. Then go to your supermarket and wonder why the shelves are empty. No tankers taking fuel to anywhere.
A non-solution. Any more bright ideas?

NeoCons. Everyone has the right to vote and vote as they like.Thats who is in office.
Get over it. Thats the way it is for now.
Nothing here so move on.

Trains have NOTHING to do with war. Don't you get it?
Transportation has everything to do with war. Get it?
War has everything to do with trains..yada..yada...

airdale-can be a pretensious a*h*l as well as InfinitePossibilites with his correcto mundo answer for every post that hits the board. Sheeeessshhhh.

Can we call the Nobel Prize committee yet?  

Long-haul transport trucking is wasteful, and unsustainable. Fortunately, there is a move to intermodal, which sees rail increasingly hauling freight for the bulk of its journey.  But trucks are still needed for local distribution.  And for that matter, there is some longhaul for which rail will never be an option.  Damage to roads can be contained by reducing and enforcing weight limitations.  Unfortunately we're going in the wrong direction today on that score.

Vehicles used in construction, especially dump trucks, are responsible for much of the damage to roads, especially local roads.  Here again, weight limits are key.

I expect that new Urban Rail lines will mature, with ridership rising each year, even without Peak Oil.  This process will be speeded up post-Peak Oil.  The % operating subsidy will drop over time for rail operations.  The drops in operating subsidies will be speeded up by adding more Urban Rail lines, thereby increasing long term pax density and improving economics.

Ratonal transit agencies will reduce FF bus service when under cost pressure (if politically allowed).  The costs of bus service are uniformly higher (cannot think of a simgle exception, even San Jose) than rail service.

WMATA (DC Metro & bus) has, in the past, kept rail fares at 80% to 85% of operating costs.  They have experienced decades of 2% to 3% annual rail pax growth and rising commuting share during the cheap oil era.   The same fares (roughly) on buses recovered 40% of operating expenses.

The proposed fare hikes have a large social equity component (middle class/upper class rail vs. lower class bus rider).  I expect at least 95% cost recovery on rail (perhaps turning a profit) being used to subsidize diesel bus operations. All, or almost all, gov't subsidies wil go to poor bus riders.  Because of capacity concerns, rail commuting fares will climb dramatically (max 50%) and off-peak fares will decline, seeking to shift demand.

I think a better solution is to recognize the social/economic/environmental benefits of rail commuting and keep fares at 80% of operating costs.  Increase gov't subsidies for poor bus riders as operating costs increases for FF buses and quickly build out 40 mile streetcar plans in DC, displacing many buses.

Best Hopes for Urban Rail,


The problem is not cost recovery but total ridership. You have to get ten times more people to ride the train than do today. For most of these people there are no and there never will be tracks to their destinations. It is as simple as that. If the train doesn't go to where you have to... you will NEVER take the train.
Electoral results for rail bond issues across the US clearly indicate that the majority have drawn the conclusion that rail investment IS a good use of our tax dollars.

I certainly do not agree that higher rail investment will mean higher operating subsidies per passenger in the future.
This ignores the benefits of complete systems versus isolated lines. It also ignores the effects of real estate redevelopment around transit stops. It ignores the effects of increased gas prices and congestion.
Denver will be building a $4 Billion system, but I can say that the financial impacts on me as a resident of Denver metro will be completely inconsequential. I have already paid $4000 in income tax to finance the Iraq war, and an additional $50 a year in transit sales tax for my family is lost in the noise. The fiscal "pain" and "bleeding" that you are concerned about is not even a hangnail in the real financial world.
Balanced against the 2006 average annual automobile ownership cost of $7384 (not counting all the property tax, sales tax, and income tax that go to subsidize cars), that $50 a year just does not seem very consequential.
Somehow it bothers you, but I and 3 million other people in Denver metro are fine with it.

Tommyvee. Read an interesting and timely article in the Camera today about how Boulder is resisting CDOT's plans to build another lane on U.S. 36.  The City of Boulder refuses to just accept the usual CDOT solution that future additional projected traffic requires an expanded highway.

The also get that if you build more lanes and build light rail at the same time, then you negatively impact light rail usage. They get that light rail and/or bus needs to be seen as preferable to congestion and gridlock.  If you continue to guarantee easy motoring, then you never get people out of their cars.

Boulder gets it.  The traditional approach of expanding highways and eating farmland forever as a solution to our traffic and housing needs is a non starter in an age of peak oil, overpopulation, and global warming.

Yes, Boulder is sort of a fortress against the rest of the front range. And lots of people resent that. But I think Boulder has seen the future that the CDOTers of the world want to impose on them and they don't want to go there.

Having said all that, Boulder has become very unpleasant, traffic wise,over the last several years.  Part of the solution is to get people out of their cars before they enter Boulder. That, needless to say, will be a real challenge. But you clearly don't fix that problem by making 36 hold even more vehicles.

I don't care if there's another one (bus) behind

But you'll only get me on a bus if it goes where I want to go. In the meantime, where's my 4x4?

Under the assumption that the post peak-oil period involves still further rises in energy prices, if we invest even more heavily in transit, it would appear to me that we are likely to get even more diminishing returns.

This conclusions seems to me predicated on the assumption that, as capacity increases, the utilization per capacity will continue to drop.  If the peak is both soon and sharp, it seems likely that that assumption will not hold.  I would posit that, as utilization increases, both cost and energy consumption per passenger mile will drop, since you're still pushing the train cars around, but now they have more people on board.   At that point, we will be glad that we invested what we did, when we did.

Another question: What is "demand response" in the graph you presented?  I'm not sure what kind of system that refers to.

Finally, your thesis in this post and your last one seems to be that radically improving the efficiency for single-passenger vehicles is going to do more, at least in the short term, to reduce our carbon footprint and energy consumption than changes in urban planning or transit deployment.  In general, I agree with you.  But I'm curious to know how you think that Personal Rapid Transit (e.g. Taxi 2000 and ULTra) might impact the picture.  It seems to me that PRT may offer both the efficiency of highly-utilized public transit and the convenience of private transportation.  Of course, it all depends on if, and how, it is deployed.  So any opinions at this point are purely speculative.  But I would be curious to hear yours.

For those interesting in reading a moderately progressive Master transportation Plan in the process of being revised

http://www.co.arlington.va.us/Departments/EnvironmentalServices/dot/planning/mplan/mtp/MTP_Draft.asp x

Moderately progressive for Arlington that is, for much of the rest of the country's suburbs it might be viewed as radical.

Several issues addressed in one comment.

While the numbers are interesting, the analysis is incomplete if you do not look at what is going on in regards to land use. During this time of service expansion and increased expenditures in capital projects, by what factor did the service area expand? Did the population density increase or decrease in an area?

Also, during this time period, what was the increase in communities in the 50,000-200,000 UZA category? This is when federal funds become available for transit capital expenditure, along with federal requirements, and the feds will actually pay a portion of the operating expenses (something that is reduced when a community goes over 200,000).

In the FTA report if you look at the funding sources by UZA, you'll see that the small jurisdictions receive less funds from fare revenues. This could be due to: less riders per vehicle, which can be attributed to a low-density development pattern combined with service that is inconvenient (service that is half-hour at best).

One problem the local (OR) transit districts are facing is not necessarily capital expenditures, it is being able to operate their brand new buses (ie, pay for drivers).

"Operating costs per trip started climbing about five or six years ago. I assume this is mainly due to the increase in fuel prices"

My guess would be the increase is due to labor, not fuel. Increasing the number of service hours will require more vehicles, and thus more drivers. This could be either expanding the service area, or increasing the frequency. Drivers are expensive, as shown in the FTA report.

Also, in the last number of years, many transit districts have seen an increase in demand response transit ridership. These typcially are provided with different vehicles than used for regular service and the number of riders per bus is low. The cost for providing this type of service is relatively expensive for the number of people served.

Finally a comment on yesterdays graph showing "Investment in Transportation Tnfrastructure as Percentage of GDP by Mode". If you read to the end of the BTS article they mention that the railroad figures are for Private Business expenditures. Government investment in rail is shown in the transit number.

"My guess would be the increase is due to labor, not fuel. Increasing the number of service hours will require more vehicles, and thus more drivers. "

The operating cost graph I showed is per vehicle revenue hour (ie per hour that a vehicle was offering service). Thus more vehicles (and thus drivers), or less ridership, could not explain this trend. Either drivers are increasingly getting paid more, vehicles are getting harder to maintain, or fuel costs are higher. My guess is it's mostly the last.

One driver for 1 car LRV "train", 2 car train, 3 car train, 4 car train or 12 car NYC subway train.


Note also various regulatory requirements are causing costs to explode: universal handicapped accessiblity, increasingly lengthy and involved public hearing processes, environmental impact statements, increasingly complex infrastructure and rolling stock engineering standards, increasingly stringent safety standards, terrorism response, insurance, and other factors.
Well, looking at the values for the 2000 and 2004 NTST for the expenses attributed to vehicle operations (form 301 in the 2000 version), the percent increases are:

Fuel & lubricants  45%
Fringe Benefits    41%
Tires & tubes      24.4%
Other salaries     19.8%
Operators salaries 14%
Other materials    11%

Of course, this really needs both the 2000 & 2004 values to be in the same dollars, and then we can look at how these are growing in regards to inflation.

So the increase seems to be attributable to mainly fuel and benefits (most likely the medical insurance that had been increasing >10% per year earlier this century.

I'm confused.  I thought that writers on TheOilDrum understand that fuels for transportation are fast being depleted.  The automobile will no longer be a viable transportion option -- therefore, what help are statistics concerning cars?  The question is, how can we create enough rail systems to avoid total hysteria/collapse when gasoline tops $10 or even $20/gallon?  By the way, not only is the oil peaking, but the dollar is going down too, and when there are 5 dollars to a Euro, the oil states are going to delink the dollar from buying oil, and gasoline will become even more expensive.  I'm guessing that Stuart believes in market solutions, but oil peaking is exogenous to the market, the whole point of looking into the future instead of concentrating on the short-term is to avoid disasters, like most people outside of NYC bankrupting themselves trying to drive.
I'm not strongly ideological, and don't have a rigid bias either to market or government solutions. I'm strongly in favor of solutions that actually will work, rather than ones that will cost enormous amounts of money while making little impact. I'm also strongly in favor of quantitative analysis of what really happens, rather than emotion-based plausible stories about what might happen.
The internal combustion automobile will no longer be a viable transportion option.  Electric vehicles would work just fine.
"The automobile will no longer be a viable transportion option"

That is not correct. Let's assume a 4% drop in oil imports for the US beginning in 2007. All this means is that in 2008 we can either

  1. drive 4% i.e. 600 miles less...
  2. drive 4% i.e. 2.6mph slower...
  3. share a commute ride 4% of the time, i.e. roughly once a month or
  4. drive 4% more efficient cars, i.e. replace 10% of the fleet with with 40% more efficient models
  5. build renewable energy plants replacing 4% fuel a year.

or... do a mixture of all of the above. Obviously 1) will buy us only a year or two, 2) will buy us another 4 years if we go back to 55mph from 65mph, 3) will probably buy us 4-6 years worth of savings and 4) will get us another decade. So putting it all together we have some 19-22 years before 5) will have to kick in big time or else...

18-22 years are a long time to figure it out.

Very nicely put, and that's exactly how I see it going too (except we may get the odd year where the Middle East goes to hell and we suddenly have to save 10-15% in a hurry).
Thanks. I am somewhat of an optimist and actucally hope that we are more likely to have 2-3% of annual import drop. But the more I read TOD, the more I am beginning to believe that 4% is maybe not too far off the mark. My imagination is sufficient to deal with that. 10-15% in one giant step would probably mean a rough year for a lot of people. On the other hand... the world managed to deal with it (and worse) in the 70s... why can't we?

May you live in interesting times...

Yes, but your analysis disregards price, given a 4% drop in imports in one year. What will more likely happen is

  1. Most people will continue to drive "normally", others (the poor) with long commutes will start to get priced out the market. Small businesses that are travel-intensive will get hit hard.

  2. The speed limit will remain the same.

  3. Shared commutes will remain rare since they are so hard to arrange.

  4. The rate of vehicle replacement will change only slightly given the large capital expenditure (for a family).

  5. Renewable energy plants replacing 4%? Coal-to-liquids? Gas-to-Liquids? More ethanol from corn? What about the timing under more realistic (pessimistic) assumptions?

Turnover in the light vehicle fleet is about 8%. I agree that's unlikely to go up (and in a recession will actually go down). However, I think we've seen that $70 oil starts to cause pretty significant changes in the mix of vehicles purchased. If oil went higher (as it would have to to accomodate actual decline in supply rather than just lack of growth in supply), we'd see more radical changes in vehicle mix. Since we already have 40mpg+ vehicles on the market, it's not hard to imagine them having dominant market share over the course of a few years, instead of having an average new fleet fuel economy of 20mpg as we do now. So, over the course of say five years of high prices, we could be getting a 4% wedge from fuel economy alone. And presumably that would motivate the car companies to start figuring out the 80mpg vehicle (which certainly isn't technically infeasible - probably requires customers to accept lighter and less powerful vehicles, which they will do a lot sooner than they'll start walking fifteen miles to the nearest train station).

Not to say this won't be accomplished with significant economic pain (see late 70s/early 80s for details).

I am certainly NOT against better auto & SUV fuel economy.  But is appears sadly lacking as the only response to post_peak Oil.  We would have to be (IMVHO) VERY lucky if that was all that was needed to adapt.

Remember that post-Peak Oil is not a 5 or 10 year event, or even (as you stated) a "generational challenge".  It is a many decade long struggle !

What I advocate is both a medium term (10-12 year) partial solution and a crucial building block to a very long term solution.  

Think 2040 or 2050, convential oil production worldwide is 15 to 20 million b/day, CTL etc. double that.  Every PHEV, 50 mpg turbodiesel, etc. built in 2015 is scrap, gone.  But every Urban Rail line built since 1982 is still open and full every day.

Your approach delays the day of reckoning, but does not avoid it.

Best Hopes for Long Term solutions,


BTW, the Swiss voted 31 billion Swiss francs to improve their already excellent, all (hydro) electric rail system in 1998 (oil $12 ?).  This plan will essentially (among other things) eliminate heavy trucks from Swiss roads by 2020.  Pax service improved as well.

Adjust for inflation and population and this is comparable to the US voting $1 trillion to improve our rail systems over a 20+ year period.

By your logic, a waste of money.  The Swiss children born in 2040 will benefit from this investment for their entire lives.

I can't but agree. Except for the economic pain, maybe. The pain at the pump results from poor choices in the past and is therefor self inflicted. It is up to everyone to avoid it by looking forward and making the right choices the next time they buy a car.
It is up to everyone to avoid it by looking forward and making the right choices the next time they buy a car.

Close, but wrong.

It is up to everyone to decide where to live and work.

I could drive a Hummer and not be overly affected by fuel costs.  With my old diesel, $30/gallon would be quite affordable.

Best Hopes for no more 60 mile commutes !


Your car does not run on price... it runs on gas. Your ICE does not care how much you paid at the pump. Only you do.

  1. Most people in the US are the poor. 30% don't even have health insurance.

  2. You are not required to drive at the speed limit or slightly above. You can drive slower if you want to. Just use the right lane.

  3. The cell phone was invented for a purpose - to communicate everywhere. Brains in conjunction with cell phones can figure out "the complexities" of ride sharing quite well.

  4. Vehicles don't live forever. Who drives a T4? A 1955 Buick? Anyone?

  5. Coal and gas-to-liquids are not renewables. Corn ethanol is a scam. We are talking wind and solar. Maybe some nuclear on top of that.
Vehicles don't live forever. Who drives a T4? A 1955 Buick? Anyone?

1920s Perley Thomas streetcars and 2000s "von Dullens" last forever.  1923 & 1924 built Perley Thomas's are rolling today on Canal Street, and have at least another 75 years left in them.  The 2004 batch of new streetcars have a 500 year design life.

Best Hopes for a Long Life,


Yes, your car does run on price. If you can't afford the gas, your car doesn't run.

Re: #1, yes, that's why I raised the issue. There are lots of poor people and they will be hit hard by a 4% decrease in imports and the concomitant price rises. These poor people are the ones who provide basic services in this society.

Re: #2, that's a behavioral change. I'll start believing in those when I see them.

Re: #3, brains + cell phones = ride sharing???? Sorry if I am a bit skeptical about that.

Re: #4, no one said vehicles run forever. My comment was about the replacement rate. You're not arguing with me, right?

Re: #5, wind & solar to run vehicles? They better be electric vehicles. Lots of investment, technology and social change will be required to make that happen.

Oh, by the way, is that a 1 year 4% decline in imports? Or does that happen every year in your scenario?

Optimism sounds good except when it is confronted with reality.

"Re: #2, that's a behavioral change. I'll start believing in those when I see them. "

Responding to this comment, the following link shows major behavioral change, so maybe you should start believing (unless you think the data is faulty, but my anecdotal experience is that the Skip is indeed full of students).

"Since the inception of the bus pass program in the spring of 1991, student ridership on the RTD system has increased by 500 percent, with annual student bus trips increasing from 300,000 in 1991 to 1.7 million in 1998. "
"Traffic engineering studies for CU-Boulder show that transit use is up from 6.3 percent of trips to and from campus 10 years ago to 10.4 percent now. This is a 67 percent increase, made possible by both growing demand and growing service (supply)."

"The SKIP replaced the RTD 202 and increased frequencies to every six-minutes during the week and increased10-15 minutes during off-peak times. The ridership went from 21,000 as the 202 to 61,500 as the SKIP in the first year. Today, the SKIP moves 65,000 riders annually, and is third highest performing routes in the region."

Much (likely most) driving is done at 55 mph or less today (95% in my case).  Savings from a nationwide, harshly enforced (traffic camera tickets at 61 mph) 55 mph speed limit are, IMHO, unlikely to save much over 4%.

Cars of today are much more aerodynamic than those of 1973, so the lower speed limit savings will also be significantly less.

I like a 50 mph, harshly enforced national speed limit when the House of Saud is replaced by the Islamic Republic of Arabia.

Best Hopes for conservation,


You probably haven't been on our highways, yet. 75 is the average on a good day... until the cops show up and somebody gets a ticket. Then everybody slows down to 65... for ten minutes.


"Cars of today are much more aerodynamic than those of 1973, so the lower speed limit savings will also be significantly less."

That would fall under the category "urban legend", I guess.

Hummers excepted.

The Prius has a Cd of 0.26  The successor to my Mercedes (1986 W124) had the best Cd of any car on the market (then) of 0.30.

From memory, my 1982 240D had an excellent Cd (for the time) Cd 0.38.

1970s GM & Ford cars (neither company owned a wind tunnel then ) had Cds often above 0.50.

Aerodynamics HAVE improved !

Best Hopes for better fuel economy,


Aerodynamics of sedans have improved, but we have a huge number of SUVs and large pickup trucks, not work vehicles, ego-extenders.

Sedans got slipperier starting with the Ford Taurus. It started the "aero" style. The Prius is slipperier than all but the Honda Insight. Tons of Prii around here but still far outnumbered by SUVs.

Apparently the rise of air conditioning has also improved drag coefficient-- people drive with the windows rolled up.

On highway driving, reputedly it offsets the fuel economy disadvantage of running air con (but not the lifecycle disadvantage of the extra weight).

I am reminded the Japanese Prime Minister took to wearing shirtsleeves, to encourage sararymen to follow suit, and thus allow Japanese corporations to save electricity by raising the air temperature.

I don't think he succeeded: culture overcomes good sense, again, I fear.

Cars of today are much more aerodynamic than those of 1973, so the lower speed limit savings will also be significantly less.

Not true.  Some cars are more aerodynamic, it's true, but many are not.  In particular, the explosion of SUVs and trucks has done really bad things to mileage.  They are not aerodynamic at all.  SUVs are too high and boxy, trucks have the empty truck bed doing bad things to airflow.

I do believe that the low-hanging fruit has mostly been picked with regard to conservation.  But the notable exception is driving.  We're driving more powerful cars, and we're driving them faster.  

The problem is that big, boxy SUVs actually start falling the cliff, mileage-wise, at about 40 mph.  It's only cars where it's 55 mph.  Would people accept a 40 mph speed limit?  What about 40 for SUVs and 55 for cars?  


Would be interesting to see "light trucks" (which would include minivans and SUVs, I believe) broken out.

The curves show that 1997 fleet is dramatically more aerodynamic than the 1973 fleet.  Average Cd clearly took a  significant drop.


I was talking about the dropoff at 55 mph.

For fuel efficiency, the last curve is almost ten years old - 1997.  The percentage of SUV sales more than doubled from 1995 to 2004.  More than half of new car sales were trucks, minivans, or SUVs in 2004.  

Mileage was better in the mid-80s than it is now, due to SUVs and the end of the 55 mph speed limit. There is still some low-hanging fruit.

First the scale is wrong for these type comparisons.

A 25 mpg to 15 mpg drop looks "parallel" and thus = to a 30 mpg to 20 mpg drop, when if fact it is steeper. (Other drags are basically proportional to speed once in top gear, aero increases as the square).

Second look at the % drop in the 1997 curve from 55 mph to 65 & 70 mph; quite low !  Only good aerodynamics could do that !

Post-1997, you may be right.  Hummers and supersized brick chaped SUV's appeared.  OTOH, pickups have always been here and their aerodynamics (including the bed) gave only improved.

Best Hopes,



Pickups have always been here, but they were not nearly as popular.  They used to be work vehicles.  Now people are using them for their city commutes.  Once upon a time, you almost never saw a woman driving a pickup.  Now, it's very common.

I'm sure you've seen this story:

Phil Knox's streamlined 1994 Toyota Tacoma pickup saw its fuel economy go from 25 mpg to 32 mpg at 70 mph just by improving its aerodynamics, reducing its Cd from 0.44 to 0.25, the same as the Honda Insight gasoline-electric hybrid.

The good thing is that you don't need to buy a new truck: you can just buy a cover for the bed.  Or make one, as this guy did.

I'm sure those figures above are for passenger cars, not light trucks which is what pickups and SUVs are. And those are on the roads in HUGE numbers, open your eyes next time you're in traffic.

Driving smoothly has a large effect on mileage, it's amazing, I work on driving smoothly and have no problem keeping it above 50MPG in the city in my Prius. I don't mean driving slowly and pissing people off, just thinking ahead more and being smooth, rather than dash-and-screech ADHD style like most people drive. This works in any car - just drive smoothly and thoughtfully.

Now that I'm in a larger place, I have a niche where I can store a bicycle, so I"m on the lookout for a nice used classic. Never forget that the poopiest car (although the Prius is not poopy, I just drive it poopyish most of the time) is a rocketship and a luxury compared to bike or walking (although a bike is often faster in bay area gridlock).

Billions in Highway Taxes Diverted to General Spending.
Only one-third of fees collected by state and local government from motorists goes directly to road construction and maintenance
According to the latest figures from the Federal Highway Administration, motorists gave state and local government $40.3 billion in 2005 for the ability to drive and own a vehicle. Gasoline taxes accounted for $20.5 billion in revenue while registration fees and miscellaneous taxes generated $13.5 billion. State and local toll roads also collected $6.4 billion from motorists.

After accounting for administration and overhead, $28.5 billion remained for all fifty states to spend in 2005. Of this amount, only $13 billion was spent on state and local road construction and maintenance.

A total of $8.9 billion of motorists' money was diverted into unrelated uses. A total of $1.4 billion went to mass transit and $7.5 billion was used for social spending. The remaining amount went to related uses such as paying down transportation debt and funding highway law enforcement.

Source: Federal Highway Administration, 12/12/2006

Highway Traffic Up in 2005
New Data Reveal America's Traffic Congestion Getting Worse

Travel on American highways climbed to an all-time high in 2005, said U.S. Secretary of Transportation Mary Peters. According to the newly released "Highway Statistics 2005," an annual compilation of data reported to the FHWA by all U.S. states and territories, Americans drove nearly three trillion miles on American highways last year. This figure - 2,989,807,000,000 vehicle miles traveled - represents 27.4 billion mile increase over travel in 2004 and nearly 25 percent more than in 1995. [read the full release]
From the press release:
"America is the most mobile nation in history," Federal Highway Administrator J. Richard Capka said, "and, as these new data show, our interstate is every bit the critical infrastructure President Eisenhower foresaw 50 years ago when he created it."

Total expenditures & disbursements: $152,700,000,000

Stuart's conclusion:

So, in summary, during this pre-peak run-up in energy prices, we invested more and more heavily in transit. The effect of that was to increase capacity, but lower utilization....
There are many stories here, given that we are talking about what things cost. However, there is no utilization problem  

Stuart, for me your're whistling a familiar tune.  I live near Phoenix, AZ.  Our city is spread out beyond belief, and most of us commute a fair distance. I don't see any combination of light-rail working for us, although it's upon us.  The city just isn't dense enough.  What would probably work for us is to use the existing infrastructure with smaller, lighter vehicles, like the all electric Tango or ATV's, and to use tighter lanes.  Let the heavy vehicles, from the Titanic sized pickups to 18 wheelers stay in the right lane expecpt to pass, as they do on some roads in France.  
This idea is onto a useful track. Vehicle regulations in most states limit scooters, mopeds, motorized bikes, and neighborhood electric vehicles like the GEM car to streets with a 35 mph speed limit. A useful step toward retrofitting the suburbs for post-Peak will be designate a region-wide network of 35 mph streets so people can actually get outside of their subdivisions with low-speed electric vehicles.
Not to pick on you Laurence, but a lot of thinking about the response to peak oil strikes me as erroneous in the following manner. People look around, say "what uses a lot of oil" and then assume that somehow that is going to have to go.

The entire point of what Campbell/Laharrere/etc have told us is that oil does not run out all at once. We do not have to adapt to oil going away in a very short period. Instead, we have to adapt to a little less every year (which will, as long as we choose to remain a market economy, be strongly signaled to us via prices). So then we have to talk about which parts of our overall economic system can change the fastest and easiest.

I know, from my entrepreneurial experience, that it's very hard to get people to behave differently. Any change or innovation that requires significant behavior change on the part of customers is disfavored. If forced to change, people will pick the adaptation path that allows them to change the least. If that path is enough to solve the problem, then no other paths even get explored much.

Thus faced with a few percent less oil each year, the American system is going to adapt in whatever way is lowest marginal cost over what it was already planning to do, and least behavioral change. Since it's pretty easy to see how car fuel economy can increase a few percent a year, most of us were planning to replace our cars at some point in the next decade anyway, and that's what happened in the 1970s, it's a no-brainer that that's going to be the biggest wedge. If decline rates are low enough for that to suffice, then that's the only thing that will happen to a great degree. Given some supply side response from CTL, tar sands, etc, it's possible. However, if we take global warming seriously (as we in California just said we would), then it's not going to be enough. It also won't be enough if decline rates are high. Then we start to talk about the second wedge. It won't be mass transit, that's for sure.

So when you talk about allowing 35 mile/hour vehicles on highways, there's no way that's going to happen. As a solution, it will be easily beaten out by vehicles that can still do 80mph, but do it on 40mpg (as a Prius easily can today), and then 80mpg, and then plugin hybrids that get 160mpg, and then....

"Instead, we have to adapt to a little less every year"

Yes, that is what the mathematical models predict. The reality could easily be more volatile, with occasional spot shortages due to geopolitical conflict or natural disaster leading to gas lines a la the 1970s.

"Any change or innovation that requires significant behavior change on the part of customers is disfavored."

And I know from my experience that people are sometimes completely fed up with the status quo and are actively looking for a change. You know, there are hundreds, perhaps thousands, of pedestrian and transit oriented developments that started construction in the past decade. Not a single one would have gotten off the ground with the fatalistic attitude you are describing. Every one required proactive risk-taking and a vision of changing status quo expectations.

As I mentioned yesterday, there will be 80 million dwellings built between 2000 and 2040. That will be 50% of the total housing stock at that point. According to the demographers and market analysts, the market will generally not be for large lot, exurban McMansions. It will be for condos, attached houses and small lot houses in mixed use, pedestrian oriented or transit oriented neighborhoods.

So I don't know if transit and walkable neighborhoods will be the second wedge or third wedge or whose wedge is bigger than whose. But I do know that transit and urban form is an significant, necessary and cost-effective part of the solution to peak oil and global warming.

So when you talk about allowing 35 mile/hour vehicles on highways, there's no way that's going to happen.

Subdivisions don't connect directly to freeways; there's almost always a subsidiary network of traffic arterials they connect to. That network would be easiest to retrofit with a separate lane or overall speed limit.

By the way -- sales of low-speed vehicles like scooters and electric bikes are taking off with no coercion or "disfavored behavioral change" involved. In part that's because they're cheap transportation, but also it's because they're more fun.

"According to the demographers and market analysts, the market will generally not be for large lot, exurban McMansions. It will be for condos, attached houses and small lot houses in mixed use, pedestrian oriented or transit oriented neighborhoods."

Source? This sounds like a New Urbanist fantasy rather than something a house builder CEO would say, but by all means prove me wrong...

New urbanist fantasy? Maybe so, if we can change the regulations to allow the the market to work more freely.

Toward a New Metropolis: The Opportunity To Rebuild America

and a related PowerPoint: The Longer View

also, some notes on current market demand.

Ok, so following one of your links, I found a reasonable looking methodology National Association of Home Builders survey here.

Here's one that bears on the question at hand:

When thinking about the factors you took into account before purchasing your
home, what was the most important consideration?


  • 41% Price
  • 39% Location
  • 11% Home amenities
  • 2% Proximity to work
  • 1% Proximity to school
  • 0% Proximity to public transportation
  • 0% Proximity to shopping
  • 5% Other

Notice the "0%" next to proximity to public transportation.  That would sure be reassuring to an investor in a TOD.

Then we've got:

Please rate the importance of the following community amenities that would
seriously influence you to move to a new community, realizing that these features,
in varying degrees, may increase the cost of the home or involve higher
homeowner association fees or local taxes.

  • 44% Highway access
  • 36% Walking/jogging/bike trails
  • 28% Sidewalk on both sides
  • 26% Park area
  • 21% Playgrounds
  • 19% Shops within walking distance
  • 16% Lake
  • 15% Near public transportation
  • 14% Day care center
  • 10% Business center
  • 9% Basketball courts/Soccer field
  • 7% Card-operated gate (no guard)
  • 6% Baseball/softball field
  • 6% Golf course
  • 6% Club house
  • 5% Security guard at gate
  • 4% Tennis courts
  • 3% Equestrian facilities

I think there's more support here for the new urbanist vision.  However, it's interesting that when considering a hypothetical future decision, people give a lot more weight on a long tail of factors that got very little weight last time they actually chose a house.

Then there's this one:

Please rate the importance of these factors if you were to buy a new home today -
with 1 being important and 5 being very important (% who said very or somewhat)


  • 62% Houses spread out
  • 60% Less traffic neighborhood
  • 55% Lower property tax
  • 47% Bigger home
  • 45% Bigger lot
  • 44% Better schools
  • 43% A good neighborhood
  • 40% Less developed area
  • 39% Away from City
  • 35% House with more luxury features
  • 28% Close to work
  • 23% Getting to work more quickly
  • 22% Recreational facilities
  • 13% Closer to public transportation
  • 10% Smaller house
  • 9% Smaller lot

I think you'd have to say here that the "give me a big house out in the sprawl" factors are coming significantly higher than the "give me a nice compact house in a walkable district close to the light rail".


Overall, the survey strikes me as evidence that there is significant market for the new urbanist vision, but there is a larger market for conventional suburbia.

Hello SS,

Thxs for this info.  Shows just how truly strong the 'Iron Triangle' marketing influence is over consumers.  If somehow, Peakoil Outreach could be a big success: it would be fascianting to see how much these lists would change.

I like the obvious delusion in the most desired home features:

62% Houses spread out
60% Less traffic neighborhood
55% Lower property tax

If houses are spread out, it automatically generates more traffic and higher property taxes.

Bob Shaw in Phx,Az  Are Humans Smarter than Yeast?

Here's a direct link to the California Public Policy Institute's survey. It also finds the majority wants large detached homes. On demand for transit in particular, the survey got these results:

31% - Would you choose to live in a high-density neighborhood where it was convenient to use public transit when you travel locally?

66% - Would you choose to live in a low-density neighborhood where you would have to drive your car when you travel locally?

Thirty-one percent market demand is nothing to sneeze at. But again, there is the issue of current attitudes (or to be more exact, attitudes in 2002 when these surveys were taken) and future attitudes over the next several decades. We know there will be large changes in the demographic makeup of the U.S. population and increasing impacts from oil peaking and global warming. That's certain.

In the realm of possibility (rather than certainty) is greater market demand due to examples of good local transit and pedestrian oriented development getting built. I've seen exactly that snowball effect occur time and time again in metropolitan areas around the U.S.

Today there are about 6 million U.S. households living within walking distance (1/2 mile) of rail transit stops. This study, Hidden in Plain Sight projects demand for similar transit access by an additional 14.2 million households by 2025. Add to that the neighborhoods beyond the 1/2 mile radius that can easily commute to rail transit, plus the neighborhoods that can be served by bus, plus the neighborhoods that reduce VMT through mixed use and pedestrian orientation alone, and the overall future impact of urban form becomes significant.

You make good points, Laurence.  A survey reflects the information the surveyed have absorbed.  A sustained price increase for energy and continued economic pressure from the effects of declining eroei will begin to erode faith in the imminent arrival of a messiah (hydrogen, ethanol, electric car), new information will be absorbed, and surveys will generate different results.
Let's take the additional 14 million transit households at face value, and assume that those households all reduce their VMT by 30%, and assume that the current 105 million households continues to grow by 15% per decade over the intervening two decades, so there'll be 138m households in 2025.  Then the total additional impact on VMT in 2025 years will be 14/138*30/100  = 3% of VMT.  On an annual basis then, the impact on oil usage growth will be to a rough approximation (103/100)^(1/20)-1 = 0.14% (neglecting the oil use of the transit entirely).

It's in this sense that I say that land use and transit will be a small wedge.  Under any reasonable assumptions, it cannot make a dramatic difference in the situation.

You are getting beyond my mathematical knowledge here. You've calculated a 3% reduction in total VMT, and as far as I can tell that 3% reduction would be happening every year, year in and year out. I don't understand how you get from that to a 0.14% annual impact; maybe you can explain it better.

As per Arthur Nelson's projections referenced above, there will be 80 million dwellings built between 2000 and 2040 and 60-80% will be apartment, condo, townhouse, or small lot detached. The demand will be for these dwellings to be located in walkable and mixed use neighborhood configurations. So I see the potential total impact of urban form to be greater than that of 14 million new transit households alone.

Let me try to go through it a bit more carefully and clearly, since I see Alan is not following the point either.  It concerns the difference between something that makes a 3% reduction every year (for example) and something that makes a 3% difference at the end of a 20 year process.

Consider If we made a new 3% reduction in VMT every year (by whatever means), and suppose for simplicity, this was the only thing that was happening. The first year, we would be at 97%.  The second year, we would be at 0.97*0.97, or 94.09%.  The third year, we would be at 0.97^3 (97% cubed), or 91.27%.  After 20 years, we would be at 97% raised to the power 20, which is 54.4%.  So our little 3% year after year (what I call a 3% wedge) would overall have made a 45% difference after twenty years.  That's enough to really matter.  That's what I call a big wedge.

On the other hand, to get only a 3% reduction after 20 years, we only need to make an annual reduction of 1- 0.97^(1/20).  That's a 0.152% annual reduction (slightly different than above because I"m working through the 3% reduction case, rather than a 3% increase, and they are almost the same, but not quite).  So if we reduce something by 0.152% year after year after year, it goes as follows:

  • Year 1: 99.848%  (1-0.152%).
  • Year 2: 99.7% (99.48% * 99.48%)
  • Year 3: 99.54% (0.9948^3)
  • ...
  • Year 20: 97% (0.9948^20)

So that's a small wedge.  0.152% reductions year after year, only give us a 3% reduction after 20 years of work on it.  So that kind of reduction does not make a major contribution to solving our problem.

Now, the example above is complicated by the fact that the reductions we are making via the transit households are coming as an offset in a situation that is otherwise growing.  I made the approximations that there were only two things going on: the number of households was increasing at the same rate as it has historically, and that the 14 million households were newly within reach of transit and thus reduced their VMT by 30% each (on average). I assumed that VMT/household was otherwise exactly constant (probably a somewhat optimistic assumption, but it wouldn't make much difference to the analysis of the transit benefit to make a different assumption).

Under those assumptions, without transit, VMT would grow by 15% the first decade, and 15% the second decade, and thus would overall by 1.15^2-1 = 32.25% bigger just because of ongoing increases in the number of households in the United States.  That corresponds to an annual growth rate of 1.3225^1/20 - 1 or 1.4% in the number of households (and thus by assumption in the quantity of VMT).  However, the transit effect, at the end of the twenty years, is to reduce the VMT of 14 million households by 30%.  Since there'll be 138.8 million households by then (32.25% more than the 105 million today), the achieved transit reduction is 30% of 14/138.8 of 2025 VMT miles.  That's 3% of however big VMT will be in 2025 - but that 3% is achieved over twenty years.  So instead of growing by 32.25% to 132.25% of today's VMT, VMT will only grow to 97%*132.25%, or by 28.3%.

To get a 28.3% increase in VMT overall (with the transit reductions) instead of a 32.25% increase (without the transit reductions), we need an annual increase of 1.283^(1/20) -1, or 1.25%, instead of the 1.4% increase that would have obtained without the transit benefits.  The difference is 0.15% a year.  Thus again, we see that this 30% of VMT for 14 million households is a 0.15% wedge.  That's the same answer we got before by ignoring the fact that our transit reduction was modifying an otherwise growing process.  (This is not an accident; the sum of various growth and reduction rates can be treated independently to a decent approximation as long as they aren't too large).

You can wiggle these various assumptions around, but it's only going to make small differences in the answer.  If you want a large transit wedge, then you need a plausible story for how to get ballpark half the US population using transit for ballpark 75%+ of their needs over the course of twenty years.  The mind boggles at how difficult, expensive, and unpopular that would be compared to the alternative of making cars more efficient (given that the transit share is currently 5% and shrinking).

Does that clarify the math?

First, I recognize that most of the oil consumption reductions forced upon us by limited oil availability in the early years of post-Peak Oil/Peak Exports will be meet by improved vehicle fuel economy, reduced VMTs, less winter heating and economic recession/depression.

Second, twenty years is FAR too short of a planning horizon for dealing with post-Peak Oil.  OTOH, my planning horizon gets fuzzy at the 35 to 50 year horizon because I recognize the effects of the "Technology Fairy" can be strong and unpredictable in that time range, the gross effects of GW should be more evident by then, as will the details of oil depletion.

IMHO, 35 years is the MINIMUM time period that societies should use for planning, with 50 years and longer being better horizons.

I will point out that the Swiss voted, in a national plebiscite in 1998, to spend the US equilivent of $1 trillion on improving their rail system.  Project completed in 2020 (if schedule holds) with a major tunnel (about half of total expense) completed in 2017.  There are many "add-ons" in the project that will be evaluated after project completion in 2020 for funding.  One can expect the benefits to maximize some time after completion, and after many 1998 Swiss voters are dead.

Would that American planning horizons were so long !

Third, your steady state 3% compounded is mechanistic and denies the observed reality of major social changes.  I think that we are in the early stages of an "S" curve in reversing the last major change in US urban form. IMVHO, ~1950 to ~1970 was the steep part of the "S" curve.

In 1945/46 returning veterans came home and found that their low down payment, low interest VA loans were of little good in the urban neighborhoods that they grew up in.  There was an urgent need for more housing and developers struggled to adapt.  Over the crucial two decades, new suburban forms were developed (quite different from the earlier "streetcar" suburbs), new commercial zones (regional & strip malls), new transportation modalities (interstates & freeways became universal) and a large minority of Americans moved from urban and rural to suburban.

During those two decades (one could also use 1948-1968), expectations changed, consumer desires changed, what was "normal" changed as new patterns emerged.  It was not a steady rate of change !

I looked for earlier trends in US VMT post-WW II and found only

http://www.bts.gov/publications/journal_of_transportation_and_statistics/volume_08_number_03/html/pa per_03/figure_03_02.html

It clearly shows a 20 year national trend of slowing VMT growth, currently about 1.5% and heading down.  (Add Peak Oil and see it go negative)

You fail to understand my objections to mixing transit and overall adult population growth (closely correlated with households and limited social trends, household growth must stop growing faster than adult population when there is 1 adult/household).  The two trends appear to be largely independent and it is poor analysis to balance one independent trend against another independent trend whilst ignoring other independent trends.  

Will Canadian tar sands production be able to balance growth in US households is as valid a question as your analysis of transit changes vs. growth in US households.  That tar sands do not directly impact VMT and transit does is not validation for your approach.

Since the overarching issues are oil supply and matching oil demand in future decades, I think analysis should be focused on that common focus point (instead of side issues like VMT) with statements like the below (some extrapolated/made up).

In a steady price, steady economic activity environment, US oil consumption increases by 1.2% every year.

Canadian Tar Sands production will be 2.2 to 2.6 million b/day in 2015 and not more than 5 million by 2030.

Price driven fuel economy increases (as opposed to gov't mandated CAFE increases) in the total US fleet are likely to be about 3%/year for the next decade in a high oil price environment.  Politically acceptable CAFE increases can only speed that up to 3.3%/year.

Unemployment increases of +1.0% have correlated with oil consumption decreases of -1.2% measured from the +1.2% mentioned above. (BTW: I think this is the default and will be the largest source of oil conservation).

A massive gov't subsized effort could replace 4% of water heaters with solar water heaters each year after a 4 year ramp up.  This translates into a 0.02% annual reduction in US oil consumption and a 0.5% annual reduction in NG use (direct & indirect for electrical production).

A massive gov't subsidized effort could replace 5% of oil home heating with geothermal heat pumps with 0.3% annual reduction in US oil use.

Transit investments of roughly $350 billion over ten years could reduce US oil consumption (2006) by 4% under a high priced but affordable & available scenario. In an oil supply interruption scenario, this could increase to 7%.  Prolonged and massive economic dislocation (depression) could also increase this percentage to as high as 10% as people and residual economic activity clusters around affordable non-oil transportation.  Electrified rail (Urban & Inter-city freight) has an elasticity of supply that most other mitigation strategies lack.

That same $350 billion transit investment, without further investments, will mature from a 4% to a 6% to 7% savings over a 25 to 30 year horizon.

Best Hopes,


Now I understand. You are talking about the annual rate of reduction, not the percent reduction itself. Thanks for the clarification.

A 30% reduction in VMT is likely an underestimate for TOD. A lot depends on what the reduction is from. If a household would otherwise locate in a dense urban center, 30% might be accurate. Compared the the surburbs, the reduction is larger. For instance, Holtzclaw finds that each doubling of population density is correlates with a 15%-30% reduction in VMT and typical TOD densities can show a 60% reduction. Or, using CO2 emissions as a proxy for gasoline consumption, TODs show 50%-80% reduction.

The VMT reductions associated with urban form occur in a variety of neighborhood types other than TOD. My back of the envelope estimate (hopefully conservative) of 2025 potential, assuming 138 million U.S. households:

  1. 14 million households - new TOD with 60% VMT reduction.
  2. 24 million households - new dwellings in walkable/bus-ready neighborhoods with 30% VMT reduction.
  3. 10 million households - existing dwellings is areas currently with poor/nonexistent bus service, but with sufficient density to be served by upgraded bus service with 30% VMT reduction.

#1 equals a 6% reduction, #2 equals a 5% reduction, #3 equals a 2% reduction. The total might be a 13% reduction.

I agree with Alan that there are various post-peak scenarios possible, ranging from a smooth, incremental powerdown to a situation like Russia in the 1990s. In the case of the former, the U.S. will adapt gradually and current attitudes toward housing and transportation will change slowly if at all. In the case of the latter, change will be disruptive and attitudes will undergo a radical break.

It's in this sense that I say that land use and transit will be a small wedge.  Under any reasonable assumptions, it cannot make a dramatic difference in the situation

Using (IMO) overly conservative assumptions, you show how two unrelated effects (adult population growth and transit use/changes in the urban form) nearly offset each other with transit/urban form change being the larger of the two.  You then reach the strange conclusion that since a positive action is only slightly larger than a negative effect, it is not worth doing.

One could as well use increases in CAFE, (or projected increases in tar sand production) and increased population and say that since the two nearly equal factors (one +, the other -) nearly match, it is not worth doing the positive action.

My estimate is that a massive Urban Rail building program can, over a ten to twelve year period, reduce US oil consumption by roughly 4%.  (Another 6% by moving half of heavy truck freight to electrified rail).  Unstated is that the savings will grow over time past that decade/12 years.


I also point out that is worthwhile promoting transportation bicycling (currently 0.5% of commutes).  I simply do NOT know how much this can be increased to (0.8%, 1.1%, 3% ?) but the cost/benefit seems to clearly be there for whatever energy (& GW & obesity) savings can be generated.

Will 4% savings back loaded to the end of a decade+ be enough all by itself ?  Not by any realistic projection.

And neither will increased transportation bicycling.

Just because a contribution is small in the medium term does NOT mean that it is not worth doing.

Peak Oil is not a short term problem and I recognize it is un-American to do long term planning.  Still, I think it is proper and prudent to take actions "soon" that will help resolve the problems of 2040 and 2050.

Increased CAFE, more carpooling, fewer vacations and all the rest will do less than nothing (i.e. make the situation slightly WORSE) in 2040/50 because they are temporary measures and may delay needed structural reforms.

Electric Urban Rail is a major part, IMO the most important part, of the long term solution.  It can also be a "silver BB" in the medium term.

And electric rail, with proper preparation, can increase capacity by 25% to 100% in a week's time when

... we suddenly have to save 10-15% in a hurry)

post by Stuart Staniford

When we "undershoot" our oil conservation demands, (as we will) Urban Rail & electrified railroads have the elasticity to expand their conservation role.  Much more elastic than most other "Silver BBs".

Best Hopes,

Alan Drake  

You misunderstand the math.  See my detailed response to Laurence above.
There seems to be the concern that we will "overinvest" in mass transit, including urban rail.  My concern is that when people start making the shift to transit in large numbers that the capacity won't be there.  I was in Barcelona recently and, when taking the subway, I was only able to complete half of my trip. For the second leg, I literally could not get in the subway car because it was so crowded.

It seems to me that oil shortages and expense will change everything.  Concluding that additional investments in rail, etc. will be a waste of capital seems to be based upon historical data that will not hold.

To argue that oil shale, oil sands, or CTL will "save" the automobile is to argue that we will go down a passive that will result in massive failure in the sense that we will only accelerate global warming.


The Hiawatha Light Rail line in Minneapolis (opened a couple of years ago) is operating at capacity on weekdays, with growth on weekends only.  (And ridership shrinkage during the winter, one can force fewer bundled up riders aboard each LRV).

I have seen one modest proposal to increase capacity by 1/3 by buying 8 more cars (24 original + 3 more on order) at a total project cost increase of 7% (~2003 cost data vs. 2006 proposed).  A more serious transportation analyst suggested that a fleet of 42 LRVs (24 + 18 more) was needed.

By inference, a 75% capacity increase was needed (this would reduce crush loads under current demand) at a +16% increase in total project costs.

Reality, 3 more LRVs were ordered, 2 to accomadate new ridership from the Northstar commuter rail line, 1 for "surpressed demand".  Since commuter rail xfer ridership is largely "contraflow", the 3 new LRVs will add peak direction capacity.

Those working on the proposed Central Corridor project between Minneapolis & St. Paul complain that FTA pressures and ridership underestimates will reduce the initial fleet by at least ten LRVs (FTA wants to stretch limited fed $ as far as possible).  They hope to bundle a few more LRVs for Hiawatha into the project to accomadate new ridership caused by the addition of the Central Corridor rail line.

Another example is the under construction Gold Line Light Rail extension into Los Angeles.  This should have been a Red Line subway extension, but funding was not there.

Forever more, riders will have to transfer to the Red Line (at Union Station from memory), lowering ridership and wasting time.

Such is reality today.



I agree with most of what you say and below (especially re the predominance of cars and car access in housing choice).

but if oil supply genuinely starts to fall, then prices will move up a lot.  If the supply curve goes vertical (100% price inelastic) then prices will rise correspondingly quickly.

It's really hard to predict what the world will look like, what choices will be made, on $150 or $200/bl oil.  Which is what would swiftly follow a recognition that we had reached 'Peak Oil'.

Much as you say is predictable: sales of diesel cars will go through the roof (40% better fuel economy, before we start).

There may also be changes in how we live and work.  Telecommuting one day a week would save perhaps 10% of the average household gas bill (assuming work driving is 1/2 of all driving).

Why does an electric vehicle have to be low speed? There are no technical reasons for that. Vehicle regulations can be changed in a few months by implementing a new law. Not very hard, really, is it? Unlike building a whole new set of low speed roads... especially in a PO world without asphalt.


two reasons, faster > more drag > more energy/mile traveled.

Also fast drawdown of many battery types damages thme and reduces their life expectancy.


Changing the vehicle regulations works too, if there's a way to mix vehicle types safely. Or maybe people will care less about the safety of mixing vehicle types post-Peak.

The only technical reasons for low speed are range and power. Ninety-nine percent (at a guess) of electric vehicles on the market are low speed.

Then there are economic and social reasons. In some states mopeds, motorized bikes and some scooters don't have the same insurance or licensing requirements as motorcyles. They're treated more like bicycles, and accordingly are restricted to 35 mph streets.

There is a world outside of the United States.  The entire argument is based on United States statistics, with no thought to the question of whether there is a reason why things are different in the United States.

There are plenty of public transit systems in the world that work, that are very pleasant and vastly more efficient than the alternatives, but part of the reason for this is cultural.  The most important question is that of land use and trip demand in the first place.  Most of North America thinks of public transit as a substitute for highway development, rather than as part of the built environment.  Most of North America thinks of shopping as involving a mall with a large parking lot.  If you deal with a highway-development-pattern mentality then you will find public transit REALLY SUCKS.  The positive feedback works to the point that highways become 8-lane or wider monsters yet are still clogged.

Rather than say that investing in public transit is likely to give "even more diminishing returns" the focus should be on altering the built environment to support walking, cycling and mass transit.

Re: carpooling.  Yes, carpooling could "easily" show vast improvements in energy efficiency over single occupancy vehicles.  Yet, it doesn't happen.  Now, why is that?  Let me ask another question.  Which is easier: organizing a group of people with almost identical travel demands who can be ready at a specific time each day, in both directions, or living in a place where you only wait 5 minutes for the next public transit vehicle, where that vehicle has priority, and trips generally only require zero or one transfer?

Car and vanpool are excellent solutions where there is inadequate density, but living somewhere with excellent public transit service is far more convenient.

Now if you want to argue numbers, wrt efficiency:

I'd like to see the carpool that can manage as high as 400-900 passenger-mpg. (Even higher with standees).

Well, it would be about the USA since that footprint is large. One problem I see is that in those places where the transit systems are really comprehensive and quite reliable, such as Tokyo, Amsterdam, and Paris, there are just stacks and stacks of people everywhere - that's what makes the good transit possible - and those parts of life that are not about transportation are not so pleasant unless one is the equivalent of a euro or US$ multimillionaire.

For that reason, good luck on altering the US built environment. Probably few Americans would voluntarily coop themselves up in the typical Dutch or urban Japanese or Parisian apartment or even house. We're the folks who still buy stereo amplifiers that are twice the size of Japanese ones (in all three dimensions) and may be loaded with bricks of concrete or pot-metal to give them heft, and we have room for them. And as for those non-Americans, I'm not 100% sure they're completely in love with wall-to-wall people either. For example, there is the real-estate outfit Hoosiers Co. Ltd. in Tokyo, which I saw advertised on Yamanote line trains. I was told it got its name - a highly obscure and specific word very odd to encounter anywhere abroad - because one of the founders was very, very impressed with the big, comfortable houses available to almost anyone in Indiana.

Good point about the carpooling, since the day and age of the type of factory where everybody arrived at work on the dot and quit on the dot is long gone in the USA. Per a previous thread, carpooling is an excellent way to waste prodigious amounts of time, which is probably why its share is diminishing.

and those parts of life that are not about transportation are not so pleasant unless one is the equivalent of a euro or US$ multimillionaire

How very american. So, if people live without your toys, then they are unhappy with the way they live? Care to ask them about their opinion?

Also 'team players' work back late even if there is really nothing to do. This pissed me off when I had to take a 6pm bus back (actually two buses) to rural areas then do an hour's farmwork in the fading light. Then up at 5am the next day. This is the irony..the workplace was a division of an agricultural department.
It's a clash of different cultures here. 99% of americans would find european way of life unbearable (if they were described how is it like). And would reject it right away. No drive-through coffee shops? Do I really have to walk?

Yet the majority of the americans who have actually done that (living in Europe etc.) have liked it. It is just necessary to accept the different mindset it requires... which unfortunately is too much of a stretch for most people.

"It is just necessary to accept the different mindset it requires... which unfortunately is too much of a stretch for most people."

So much so that they have to make up fantasies about public transit only working in overcrowded large cities, apparently.  You do not need extremely high density to have public transit work.  You don't even have to be in Europe.  Remember North America's "streetcar suburbs"?


I sort of hope this is the last comment.  I've written essays and posted hundereds of times on other forums including this thread to the point where I no longer care to support what I say.  Take it for what it's worth or forget it.  The information is available to anyone who searches for it.

  1.  All of this discussion regarding transportation is an attempt to support the status quo and the status quo is unsustainable.  

  2.  There is a high probability that the US will not only have a serious recession in the coming years but more likely Depression II, the permanent depression.

  3.  There will be sever deprivation and the beginning of a world-wide dieoff, including many in the US.

I was really, really  tempted to add detail and links to the above comments.  It's in my nature as someone trained in the sciences to do so.  But, I really don't debate this stuff much any more.

Todd; a Realist

Yeah, ain't life a bitch.
How much rail can the 11th largest US urban area, with a population of 4.4 million, absorb ?


Please note the Green, Red, Blue & Orange subway lines in the center as well as the commuter rail lines.

1.1 million daily riders.

And the region is far from saturated.  Reasonable enhancement could easily run ridership to 1.25 million.  Post-Peak Oil should also see riderhsip increases.

Best Hopes,


Thought I would throw my two cents in here.  Living in Shanghai, the Chinese government is working to try to stay ahead of peak oil by building 6 additional metro lines and 10 light rail lines in the next twenty years.  In an effort to encouage the use of mass transit in Beijing and take cars off the road the government recently cut the price of mass transit by 50%.
The plans in Shanghai appear to change slightly every year or two.  The last info I had was that eventually Shanghai wanted 17 Urban Rail Lines in toto (built & planned) with about half (9 ?) being metros.

What are the total number of Metro lines and Light Rail lines built + planned ?



For most Americans time trumps everything.  The cliche "time is money" is almost the national motto.  It's hurry, hurry and more hurry.  I don't think they know where they are going, but they have to get there fast.  Then go someplace else really fast. If public transit isn't faster than private cars, it will be little used in the United States and probably little used even if it is faster because of the lack of privacy.

 "A busy man is never wise and a wise man is never busy."  


There's two easy explanations for your spurious correlations.  

First, rail transit usage grows slowly but steadily.  When a line opens, many people do not both live and work near the stations.  Commuting is limited to those lucky enough to find it convenient.  But over time, as more and more people make choices about where they live when they move residences, they have the opportunities to live closer to transit or to move away from it and reduce their housing expenses.  Development around rail stations also takes time.  

So we should always expect established rail lines to have higher usages than new ones.  Since light rail is growing very rapidly, it follows that most of the new lines are not yet near their peak capacity.  Rail is not an instant solution, but it IS a solution.

Secondly, other posters have pointed out the network effects of rail.  Extensions of large systems will attract new riders to the newly opened lines much faster than a system starting from scratch.  Yet over the last ten years many of the new miles of track, and certainly the most expensive segments, have been built in cities that did not have light rail.  With so many new systems starting and needing to build infrastructure from the ground up, or construct main lines through expensive downtown corridors, it's hardly surprising that costs would be rising while ridership is growing slowly.

Yet in nearly every case, ridership for these new systems is FAR above projections, and continues to grow every year.

A better comparison would be to look at year-on-year growth in existing corridors - you'll find that the returns for rail transit are a lot better than you give credit for.

A few complimentary observations.

  1. When a new line is added, ridership (and cost efficiency) of existing lines increases.  This is a major reason why Portland 2006 is a much better system than Portland 1986, and Portland 2016 will be better still (when the Green Line and a commuter rail "feeder" are open and beginning to mature).

  2. The first line usually has teething problems and no efficiency of scale.  A large rail yard and heavy maintenance facility are focused on one "start-up" line.

  3. Workplaces also relocate close to Urban Rail.  Typically, the # of workplaces within easy walking distance of a rail stop exceed the # of residences.

  4. The rates of TOD development vary widely from city to city (One puzzle to me is the Blue Light Rail Line in Los Angeles/Long Beach, ridership 100,000+/day but very limited TOD).  I expect a spike in TOD post-Peak Oil.

Best Hopes,