May 19, 2008

Scale, cost and impact of a public transportation overhaul

Many environmentalists love to point to public transportation as the blessed technology and policy which would reduce oil consumption and dependence. I agree that increasing public transportation can be helpful, but how helpful and how long would it take to have what level of impact ?

In the big energy picture, the USA uses 40% of its oil for cars. So more mass transit will not help with oil used for factories, home heating, trains, freight trucks and other uses.

I will show that going from 3-5% public transportation up to 20-30% will still leave 70-80% driving cars. It will cost 1% of GDP for 50 years. So 40% of oil for cars goes to 30%. US 20.8 million barrels of oil per day goes to 18.5-20 million barrels of oil per day. After decades of a massive public transportation system overhaul which would be politically difficult [perhaps impossible] to enact in many cities.

Which is why the nextbigfuture energy plan has a different focus.

Book on Australia and New Zealand public transportation It is from the early, mid-90s but there has been no significant change in the situation.

Public transportation trips in Auckland over time

Comparison of some other cities over times
Perth -7.1% per capita ridership since 1981 versus 2005.
Portland and Brisbane are among the cities with more public transport per capita since 1981.

Perth and Adelaide and Auckland all are among the cities with less mass transit per person. How much would it cost and how long would it take to get up to the higher levels. Add 60km of rail per person. In each low transit city.

Comparing transit and commuter trips for USA, Australia, Canada, europe and Asia.

Transportation in Australian and New Zealand cities

Stronger versus weaker rail cities in Australia and New Zealand

2005 view of public transportation in many cities

Public transportation for cities in 2001-2005.

So copy Vienna and 50 years of 1% of city GDP (or more for a shorter period of time, this would be trillions of dollars on a global scale.) on public rail plus all decisions needed to transform city planning and layouts for higher public transportation usage to get closer to the higher commuter usage rates. Note: the highest usage rates are 20-30% in Europe. Maybe some highly dense asian cities are doing better. The Philippines with a lot of jeepnees. (Local made jeep/buses where people with less money ride at to 20 to a vehicle, some hanging on the outside. Two jeepnees crashing together can result in 30-50 deaths.)

So up from 3-5% up to 20-30%. That still leaves 70-80% driving cars.

So 40% of oil for cars goes to 30%. US 20.8 million barrels of oil per day goes to 18.5-20 million barrels of oil per day. After decades of a massive public transportation system overhaul.


James said...


Its generally reasonable to assume that public transportation use decreases over time with large amounts of land available (such as in America or Australia).

Group transportation methods save money and energy at the expense of privacy and convenience and time. As societies get richer, which free societies do over time, they trade off money for convenience in greater amounts.

In very dense environments, public transportation may always save time which means it will continue to be used by most people.

As sort of an aside - You say that the US has oil consumption of 22 million barrels per day.

I've read other people say that the US uses 20 million barrels per day.

According to US energy department statistics - found here: - the US imports about 13 million barrels per day (for all uses), produces another 5 million barrels per day, and then exports about 1.2 million barrels per day (mostly of refined products).

Doesn't that mean we burn about 17 million barrels (18 in the summer) per day, not 22, not 20?

James Becker

bw said...

You are right James 20.8 million bpd was the peak. (so not 22 million bpd.)

However, you are confused about the oil production. The USA has 8.5 million bpd of total oil production. The 5 million bpd is crude oil. Thus 12-12.5 million bpd of net imports.

Link for EIA for all countries

Gregor J. Rothfuss said...

Have you factored in how high fuel prices lead to deserted suburbs, thereby increasing density? Suburbs, the slums of the future.

Once most people live in urban environments again, I bet the equations look different.

bw said...

there will be changes to suburbs and cities.

Vancouver/LA and other places are looking at rezoning to allow for taller buildings in suburbs and other areas. Plus suburbs will have areas rezoned for more offices and mini-ciy centers.

any new plan has to have profit in it for developers to make money changing what exists.

Some places may get abandoned. Detroit may not recover if the car companies do not recover.

But LA with a lot of population growth will be able to adapt and adjust and re-invent itself.

the equations however still do not change much. Because building a lot of public transit and reworking infrastructure and city layouts takes a long time (decades) and a lot of money.

What is the fastest rebuilding of a city that you have seen.
Building a new city can be done faster than rebuilding an old one.
the fastest that I know of is Shenzhen and the other chinese cities that have sprung up in the Chinese economic boom. But I do not think western cities will follow the more lax building standards.

Although it can be nearly as fast if one were to bomb/destroy the old one (ie. Berlin and other German cities after WW2).

Jeff said...

unimodal (skytran) seems like it solves the time and some of the convenience issues of existing forms of public transportation. When I've talked to people at work, time is the number one reason why public transport is unacceptable. However we now have the technology to create much faster than automobile public transport. I would also be interested in seeing Curitiba added to the graphs you've produced.

Richard Kulisz said...

> I will show that going from 3-5% public transportation up to 20-30% will still leave 70-80% driving cars. It will cost 1% of GDP for 50 years. So 40% of oil for cars goes to 30%. US 20.8 million barrels of oil per day goes to 18.5-20 million barrels of oil per day.

So what? 2 million barrels per day (and rail transit usage is always underestimated) is 700 million barrels per year which at 100$ per barrel is 70 billion per year. You say it would cost 1% of GDP yet that's 110 billion. So already 2/3 of the costs are recouped just in saved fuel. Nevermind reduced costs in traffic accidents, mortality, and air pollution.

But nevermind that, do delete my comment.

bw said...

Richard, your history as an internet troll is well documented online but occasionally you can contribute to a conversation. So I published the prior comment since it is not rude and has a contribution. (not necessarily fully correct but a contribution to the discussion)

Now to answer your post.

Yes, increasing transit has some value. However, the payoff that you are talking about does not start for many decades. Plus it requires generations of politicians (local and regional) to work together for decades. Many of the public transit/public works projects end up failing [see Big Dig] and some of the places that spend quite a bit of money do not achieve the highest levels of ridership of the best cities.

There is also the comparison of trillions on mass transit projects versus other kinds of technology.

Optimally rail mass transit can provide the equivalent of 500mpg.
However, new electric vehicles can achieve comparable mileage equivalents. (Aptera electric vehicles)

Robotically driven vehicles or assisted (enhanced cruise control) driving could allow platooning of vehicles. Formation driving where gaps of one vehicle length can be safely maintained. This can provide 30-50% reduction in drag and increased fuel savings.

Sometimes it does not make sense to have a one size fits all policy and expect that people in LA will become like people in Vienna in regards to adoption of transit. So the need to consider what people in a locality will be willing to do and to influence where possible but to also accept what is not changing and optimize it.

Make the each item in the buffet of transportation choices acceptable or more tolerable.