The Washington, DC region is great >> and it can be greater.


Lessons from a South American Bus Rapid Transit system

Curitiba, a city of 1.8 million people, is the capital of the state of Paraná in Brazil. In late December, I visited Curitiba at my own expense and was briefed on its Bus Rapid Transit (BRT) system, one of the world's first and most highly-regarded. Because Montgomery County is studying BRT as an option for its residents, I wanted to find out how it is working in other communities. I found many positive aspects and some less positive.

Councilmember George Leventhal enters a bi-articulated express bus in Curitiba.

My sincere thanks to Silvia Mara Dos Santos Ramos of URBS (Urbanizacao de Curitiba, S/A), the city's transit agency, and André Vinicius Marchezetti of transportation consulting firm Logitrans, who were the guides for my visit. Officials at EMBARQ, a project of the World Resources Institute, and its Brazilian Center for Sustainable Transport were also extremely helpful in helping me make contact with the guides in Curitiba.

Positive aspects of the system

Curitiba's system is respected around the world and has inspired other BRT systems in Latin American cities including Bogota, Mexico City and Guatemala City, as well as U.S. cities including Cleveland, Pittsburgh, and Los Angeles. Click here to see a good overview of the system in Portuguese, but with easy-to-understand graphics. Click here to see a map of bus routes.

The genius of the city's plan is that on primary streets, the center two lanes of traffic have been reserved for buses. This enables the buses to avoid automobile traffic and move smoothly, with minimal interruption. "Tube" stations are placed every 500 meters along primary routes. A city law states that no resident may live more than 500 meters from a bus stop, so the primary and collector routes cover the entire residential area.

Inside a tube station. All photos from the author.

On the main bus line, the 351 tube stations are sleek and modern. Passengers pay when they enter the station, avoiding a delay when entering the bus. The buses are long, with several doors allowing passengers to enter and the bus doors correspond with several doors at each station, enabling speedy entrance to and egress from the buses. Buses on primary express routes are "bi-articulated," containing three cars with accordion-style dividers that enable them to navigate tight turns quickly and efficiently.

This video demonstrates a biarticulated bus navigating a tight turn.

Within the stations, bus stops are identified with a clear map similar to the Metro route map. A tourist like myself can easily navigate most of the city by following the bus route on a map. Unfortunately, because the buses are used on multiple routes and because the bus network covers so much territory, complete routes are not shown inside every bus.

A single fare of R$2.20 Brazilian Reals (US$1.22) enables a rider to take a single bus as far as it travels, even from Curitiba's furthest suburbs into the city center. Fare revenue accounts for 100 percent of Curitiba's operating budget. (Its capital budget—the cost of building the system—came from international investment including the World Bank and French Development Agency.)

Property values adjacent to the bus line have shown consistent increases when new express routes are constructed, increasing tax revenues to local governments. Buses utilize 20% Biodiesel to minimize emissions of carbon and other pollutants.

My son Daniel Leventhal rides the bus.

Less positive aspects of the system

Because the current system dates to 1974, the 29 primary bus terminals that facilitate transfers between collector and primary bus routes are aging. The two I visited were not especially attractive, and abundantly covered with graffiti.

Ms. Dos Santos told me that "the rich do not use the system much." Those who have access to the comfort and privacy of their automobiles opt out of using the BRT system. In general, I did not find the BRT experience comparable to fixed-rail systems in terms of comfort. Curitiba does not have a system like Metrorail, although there are plans to construct a rail line in time for the 2014 World Cup.

Taken from inside an express bus, this photo shows the division of the BRT lanes from automobile lanes.

Lessons for Montgomery County, Maryland

According to Ms. Dos Santos, the rate of automobile ownership in Curitiba is 22%, while transit usage is 40% and the balance of commuters travel by motorcycle or motor scooter, or walk. This helps to explain why ridership of the BRT system is so high and why farebox revenues cover operating costs.

By contrast, in Montgomery County, 66% of workers drive to work alone while only 14.9% commute by rail or bus, according to the 2008 American Community Survey.

Another important difference is the price of gasoline, which is substantially more expensive in Brazil (R$2.49 Brazilian Reals per liter, or approximately U.S. $5.20 per gallon, while I was there) than in the U.S., making automobile travel substantially less affordable for working Brazilians.

The critical question is whether increased frequency, speed and convenience would persuade enough Montgomery County residents to ride the bus to make the system financially sustainable. Even in Curitiba, where per capita income was $17,977 Brazilian Reals (U.S. $10,005) in 2006, upscale people do not ride the bus. And there are far more upscale people in Montgomery County, where per capita income is U.S. $35,684.

The county's current Ride-On bus system generates only approximately 15% of its revenue from fares, with the remainder subsidized from general tax revenues. The current budget crisis has highlighted the Ride-On system's significant expense to taxpayers, although the County Council has so far resisted cuts in Ride-On service proposed by the County Executive. How much more can we afford to expand the bus system even if the current 85% subsidy from all taxpayers decreased to a subsidy of 80%, 75% or lower?

Also relevant is that the county's bus storage and maintenance facility is at its absolute limit of available space, and a new storage and maintenance facility proposed for Clarksburg has been delayed because of concerns over runoff into Ten Mile Creek, so there is currently no place to store additional buses in the county.

An express bus departs from tube stations.

Curitiba has a long history of transit service, dating back to horse-drawn trolleys in 1887. Planning for 20 primary express bus routes began in 1966 and the system opened in 1974. For more than 35 years, primary roads have been designed to accommodate the BRT system. Municipalities in the Curitiba region served by URBS delegate management of the transit system to URBS, whose board members are appointed by the Mayor of Curitiba.

In Montgomery County, however, major roads would need to be redesigned to accommodate bus rapid transit. The state of Maryland's recent reconstruction of Route 29 represented a critical missed opportunity to develop express bus lanes in the middle.

Our county's primary roads (Route 29, New Hampshire Avenue, Georgia Avenue, Connecticut Avenue, Routes 28 & 198, Veirs Mill Road, University Boulevard, Wisconsin Avenue/Rockville Pike, and I-270) are owned, designed and maintained by the State Highway Administration. It would require close coordination between the county, the state and possibly the Washington Metropolitan Area Transit Authority (WMATA) would be necessary to accomplish the BRT vision. Ridership will be a very important criterion for winning federal funding, but ridership is the key question regarding whether the system can succeed here.

Accommodating through automobile traffic on primary roads will be another key issue. Because the buses travel on a dedicated route, automobiles may not cross primary streets at every intersection. In Montgomery County, this will affect many neighborhood streets whose residents are accustomed to being able to cross major streets or make left and right turns to exit their neighborhoods—all of which could be restricted with rapid bus routes down the center of major streets.

My visit to Curitiba was a great experience and there is no question that I would love to see similar technology employed in Montgomery County. In the near term, I will advocate for BRT on Veirs Mill Road, which along with University Boulevard has just received a federal grant for bus transit improvements, and Georgia Avenue. I am also optimistic about prospects for BRT on the Inter County Connector and the proposal for BRT on Rockville Pike contained in the White Flint Sector Plan, now pending before the County Council.

On the other hand, my visit to Curitiba did not persuade me that BRT compares favorably to fixed-rail systems as an effective inducement for riders to leave their automobiles at home. I will continue to advocate strongly for light rail on the Purple Line and, while I understand that cost factors may ultimately persuade Governor O'Malley to select BRT for the Corridor Cities Transitway, my preference remains light rail for that system as well.

George Leventhal is an at-large member of the Montgomery County Council. He serves on the Transportation, Infrastructure, Energy and Environment Committee.


Add a comment »

Another important difference is the price of gasoline, which is substantially more expensive in Brazil (R$2.49 Brazilian Reals per liter, or approximately U.S. $5.20 per gallon, while I was there) than in the U.S., making automobile travel substantially less affordable for working Brazilians.

To me, this is the biggest factor as to why people ride transit in other countries. Driving is just too cheap an alternative in the US.

by MLD on Feb 26, 2010 10:50 am • linkreport

Thank you for the well-written piece, Councilmember Leventhal.

On the other hand, my visit to Curitiba did not persuade me that BRT compares favorably to fixed-rail systems as an effective inducement for riders to leave their automobiles at home.

Your conclusion is well written. Buses and trains are complements for each other, not substitues. They are different mobility and planning tools.

I look forward to your continued advocacy for the Purple Line and CCT.

by Cavan on Feb 26, 2010 10:51 am • linkreport

Sounds like a very interesting trip, and thanks for writing it up for GGW.

I've been trying to understand the scale and right-of-way width needed for a high throughput bus system such as Curitiba. Your first picture of a busway, the one from inside the bus, shows perhaps a 25-30 foot ROW and bus stops without a separate passing lane, through good timing of the shot, some cross traffic just down the road. The second picture shows what is perhaps a 75 foot ROW, with a passing lane outside the lane used to service the station.

Where in the context of the route structure are these two locations? I'd guess the first is at the outskirts and the second is closer in, perhaps served by multiple lines. How prevalent are the two types of configurations? How much of the system is 4 lane, versus two lane?

My tentative conclusions are that this sort of system is an excellent way to repurpose existing large rights-of-way--like a highway--to serve more people. (I think this is what was largely done in Bogota, but I'm not entirely sure.) But it also seems like the scale needed to achieve high throughput is much too massive to be compatible with the sort of lively, dense, walkable neighborhood that, e.g., this site tends to favor. (Or, for that matter, any place where cross traffic can't be heavily interrupted.)

That is, a system like this could be a way to make the most of the (unfortunately overbuilt) road infrastructure already in place in Montgomery county, it would not be a good option to increase commuter capacity into downtown DC.

by thm on Feb 26, 2010 11:05 am • linkreport

Curitiba has done much more for their transportation beyond the BRT system. In conjunction with the BRT lanes, Curitiba utilizes a Trinary road system. Like a larger implementation of K Street, Curitiba's BRT lanes share the street with local streets that allow cars to reach destinations, shops, restaurants, etc. On either side of these local/BRT streets are express roads or "Rapidas". Theses are wide, one-way express streets meant to facilitate rapid vehicle movement across the city. Then, closer to the destination, you'd switch to the local street in-between.

It's an interesting approach to the city's traffic problem, but one that required a massive overhaul (not very likely in the U.S.)

Photo below is one of the Rapidas streets:


by kidincredible on Feb 26, 2010 11:07 am • linkreport

Also, it should be noted that Curitiba gets a lot of press for their BRT system, the city under mayor Jaime Lerner has done much to advance the concept of Urban Design.

Some small examples:

1) Portions of the city were very flood prone. Poor residents lived in shanty towns in these floodplains. Whenever it rained too much, their homes were destroyed, possessions swept away and so on. The City arranged "buyouts" to have these poor residents move out of the flood prone areas and into better homes (still low-income housing, but better than the shacks they were living in). The floodplains were then made into city parks so when the rains came, the parks could flood, and no one's homes would be swept away. The money for this relocation was paid back by the taxes on the property surrounding the new parks when property values went up.

2) Unfortunately, the city was still relatively poor and parks can be costly to maintain. Particularly, mowing the grass can be expensive. So the city purchased sheep. Sheep. That roam the parks, feeding on the grass and keeping the parks looking cleaner.

3) Complementing the bus system and the park cleanup, the city constructed a recycling facility. Materials from this facility went into construction, and other recycling efforts. The city arranged to provide bus tokens to citizens for each X pounds of recyclables they brought. This allowed poorer citizens to help take better care of their neighborhoods, provided a way for breadwinners to reach jobs with "free" tokens, and gave mothers a way to help out the family while at home.

All of this comes from memory from watching this film a couple years ago:

It was a fascinating film to watch and I highly recommend it for anyone interested in Urban Planning. (I'm actually going to buy it today after reading this article and remembering all about Curitiba)

by kidincredible on Feb 26, 2010 11:24 am • linkreport

IMHO, price is only a small factor in getting "rich" folks into transit. The main factor is convenience. People who don't use transit are intimidated by hard-to-find schedules and unknown routes. They'd rather spend 15 minutes searching for an expensive parking spot than wait 5 minutes for a (cheap) bus.

So what you need to do is make buses route and schedules clear and reliable. An easy start can be made by putting clear bus schedules and maps at every bus stop. The map should not only show the route of the bus(es) of that stop, but include an overview of nearby bus lines and connections.

Furthermore, bus lanes and queue jumps should be made as visible as possible so that car drivers get confronted as much as possible with passing buses. Drivers tend to ignore buses, but are not able to do so if line 3B passes them three times on their way to work.

Buses should also look good. If someone is passed three times by an overcrowded, loud and dirty bus, they're not going to be felt left out. However, if a car driver is passed three times by a clean bus with everybody nicely seated and a big ad on the side saying "6 times an hour from A to B in 15 minutes reading your newspaper", they will feel left out.

It's a mind game. Too bad WMATA leadership has no mind, or it out of their minds.

by Jasper on Feb 26, 2010 11:51 am • linkreport

I just had an "argument" with a student at U Delaware (where I will be speaking to the urban studies and civil engineering depts.) about BRT because a number of faculty there do a lot of writing on the issue.

Generally, I agree with Councilman Leventhal. BUT we are both missing the point, or not clarifying the issue properly.

Yes, BRT works great in places where you can create dedicated busways and where people are so poor that they are transit dependent, and where they are willing to tolerate crush loads double that of people in North America, i.e., places like Curitiba, Bogota, Johannesburg.

BRT in North America, especially in the U.S., doesn't work well because we can't get the ridership (compared to fixed rail) or the loads per vehicle, because people have other choices.

In the north american situation, fixed rail works better than BRT.

HOWEVER, I had to concede to my colleague at U Delaware that in the context of planning a transit network for a metropolitan area BRT as a mode needs to be considered legitimately, as a potential component of a robust network, depending on the LOQ and LOS you want to provide, and the various land use and population conditions, which vary across the service area.

In areas with the right population density and tight coupling of origins and destinations, fixed rail usually is the best option.

I can't say exactly where BRT would work well generally, or in the DC area specifically, but I am sure some routes in some places might work. In most places not. Because for BRT to be R you need density. And if you have density you're not likely to have enough roadway to be able to provide dedicated roadways for the BRT service. Without dedicated roadways it can't be R. And if it isn't R, given that people aren't likely to ride it anyway (becuase it is bus not rail) then what's the point. (Subways deal with this by being underground. And light rail by having signal priority.)

The reality is that BRT was pushed by the FTA because they didn't want to fund fixed rail. But the North American context is so different from where BRT succeeds. And our friends at the orgs. that tout BRT in other countries (i.e., WRI, ITDP, etc.) aren't very good at delineating the different contexts that we have to work with _here_, which is all that matters.

I suppose the MetroRapid service in LA is good, but that's because of the population in LA. Most other metropolitan areas in the U.S. aren't likely to be able to generate the equivalent amount of ridership.

PGH has about the best BRT service in the U.S., with dedicated roadways, and it's been in revenue service for more than 2 decades. Few people ride it though, even though it works well. Now, PGH is a shrinking region and the CBD is no longer the dominant destination for work trips. Still, I think it shows how despite getting everything to work right, bus service of any type is still a hard sell.

by Richard Layman on Feb 26, 2010 1:39 pm • linkreport

and sorry for not saying "thank you for writing the article, and for writing it so well."

by Richard Layman on Feb 26, 2010 1:44 pm • linkreport

Richard, is it really necessary for successful BRT that riders "tolerate crush loads double that of people in North America"? Couldn't you support the same service frequency with half the density, half the number of riders and double the fare, given the higher average incomes here?

And yes, great article!

by Erica on Feb 26, 2010 4:50 pm • linkreport

I lived in Curitiba for 6 years. I would have recommended some great restaurants, had I known that you were going.

@thm - The picture of the very wide ROW looks to be the newest BRT line (linha verde). They built it in the very wide median of an old highway, so there was lots of space. There are roads on both sides of the tube because the biarticulated buses load on the right side, and the silver express busses load on the left. That means cross platform transfers. Terminals also have both types of buses, but are actual buildings.

Part of the beauty of the system, is that the lines are color coded based on purpose as seen here

So big red = BRT running in dedicated lanes.
Regular sized silver buses run in mixed traffic, but only on high capacity streets, they also load in tubes.
Green = cross city service along regular streets
orange - Service BRT terminals
Yellow = regular bus lines

There are also some buses that only service hospitals and a microbus for downtown.

Another important thing to note is the city zoning. If you look at curitiba from afar, you see lines of high rises....high rises are only allowed next to the BRT routes.

Here is one of the silver buses loading on the left (end of this particular line)

High density

Just a few blocks from the BRT line, single floor buildings. The closer to the BRT line, the taller the buildings

by J on Feb 26, 2010 6:23 pm • linkreport

Why the trip to Curitiba to look at a BRT when most of the BRT lines in the U.S. are a failure, not because of lack of ridership but because of lack of capacity. You can only run so many busses without impacting cross traffic and this is certainly the case in Los Angeles. Many European cities have high capacity tram or “Mini Metro” lines such as Budapest (lines 4&6) that run on reserved right of way in the center of the street with much greater capacity, faster operating speeds, a much more pleasant ride, the right of way can be narrower and more attractive with greenery and landscaping with no need for pavement and they are much less expensive to operate per passenger mile. So why are BRT’s even being considered when LRT lines can do everything a bus can do and for less money and better? Before any final decision is made as to BRT or LRT a trip to Europe needs to be made to see just how well a LRT can work in the center of a major highway on privet right of way and its many advantages over any kind of a BRT.

by Interurbans on Feb 26, 2010 11:17 pm • linkreport

"Another important difference is the price of gasoline, which is substantially more expensive in Brazil (R$2.49 Brazilian Reals per liter, or approximately U.S. $5.20 per gallon, while I was there) than in the U.S., making automobile travel substantially less affordable for working Brazilians."

Gas is expensive in Brazil because it is highly taxed. Brazil produces Ethanol from sugar cane, and has for many years. Most all cars sold in Brazil can run on gas or any proportion of gas and ethanol. The price of ethanol is about 1/2 that as gas, so even though it has less energy than gasoline, the price is about on par with gas in the US.

On the original topic, BRT might work in a few places in the US, but on the whole better B's could be made by having more expresses and skip-stops so that there is better R. Even without throwing out the BRT acronym, there is a heck of lot of improvement that could be made to good old-fashioned B.

by spookiness on Feb 27, 2010 2:44 am • linkreport

I wanted to mention that the "Metro Rapid" buses in Los Angeles are not "BRT" in the same sense as in Curitiba or Bogota. These buses have limited stops (every 1/2 mile), improved bus stops in some places, and high frequencies (every 10 minutes most of the day). Most street with a Rapid also have a local bus with stops every 2 blocks.

Los Angeles also has the "Orange Line" rapid bus, which has an exclusive right-of-way (used to be a rail line), special stations every 1/2 to 1 mile and extra-long buses with more doors. It manages to average 18 mph (including stops and street lights), about the same as some slow rail lines, but significantly slower than our subway (30 mph average, including stops). It is also over-crowded at peak periods despite buses every 5 minutes or less, due to lower capacity than light rail.

by Joseph E on Feb 28, 2010 3:18 am • linkreport

The article is knaive-ly written at best. You (and readers) should have read the 2007 article in NY Time ( which contains 5x the amount of information. One of the real keys to the rapid bus system is the stations, which expedite boarding (on/off) and take fares like a train station. Another major key is the complete integration of the rapid bus system into a wider city planning, which most cites refuse to do beyond the next election.

by Marc on Feb 28, 2010 12:18 pm • linkreport

the "negative" aspects are that it is "not especially attractive" (because of graffiti) and "the rich" don't use it. that basically means there is nothing wrong with it. brazil has levels of inequality and poverty much greater than in the US, so who cares about the rich - they're riding around in helicopters and hummer limousines. the point is that it's efficient. it would indeed mean major road re-designs if implemented in the US, but if the cost-benefit analysis yields the conclusion that the redesign would be worth it, it should definitely be done.

by andrew on Mar 1, 2010 2:46 am • linkreport

Thank you for taking time out of your day to expose us to your experience, Councilmember Leventhal. I truly appreciate it.

Although I live in Washington, your experience illuminates the vast difference between the reliable bus service you expereienced in Brazil with the daily frustration we experience with Metrobus here in DC.

I have long advocated buses, but I'm afraid they are not or have not been as "sexy" as rail. But for me, it's really the way to go.

by Jazzy on Mar 2, 2010 8:01 am • linkreport

Congratulations Mr. Leventhal for genuinely making an effort to explore BRT. I only wish leaders that in other cities across the world (particularly North America) would make such an effort.

I was one of those people who truly doubted the potential of BRT until I saw how well a fully executed and properly developed BRT works. I remain astounded when I encounter consultants who have yet to visit the exceptional systems in Latin America (Colombia in particular now has 4 systems running).

During one of my trips I was astonished to discover that a delegation from South Africa had visited no less than 6 different BRT and rail system to understand how they operated before moving forward with the BRT plans in Johannesburg and Capetown.

So hats off - great article. I would encourage others to do the same. Cali in Colombia is smaller than Bogota, and Periera in Colombia is only 1/2 million. Both also use BRTs very effectively.

or this - start @ 0:50

by Adriana on Dec 15, 2010 10:08 am • linkreport

Add a Comment

Name: (will be displayed on the comments page)

Email: (must be your real address, but will be kept private)

URL: (optional, will be displayed)

You can use some HTML, like <blockquote>quoting another comment</blockquote>, <i>italics</i>, and <a href="http://url_here">hyperlinks</a>. More here.

Your comment:

By submitting a comment, you agree to abide by our comment policy.
Notify me of followup comments via email. (You can also subscribe without commenting.)
Save my name and email address on this computer so I don't have to enter it next time, and so I don't have to answer the anti-spam map challenge question in the future.


Support Us