The Myth That Everyone Naturally Prefers Trains to Buses

When you focus on what really matters—service—much of the difference actually disappears.

Source: CITYLAB – By: Eric Jaffe (Febraury 23rd, 2015)

Take a look at the image above. This rendering represents modern bus rapid transit service. The BRT vehicle travels in its own separate lane, free from the constraints of traffic congestion or traffic lights. The bus is sleek and the shelter is pleasant. If you could see the boarding procedure, too, you’d find that passengers buy their fares ahead of time, enabling them to enter quickly through any door, just as they do on a train.

Now take a look at the image below, which shows a modern light rail service. The scenes are remarkably similar. This train travels in the same dedicated lane and even has the same style. The only real difference you’ll find, if you look very close, is the faint sign of tracks on the ground.


Given what we know from these two pictures alone, there’s no reason to suspect these two rides—modern BRT or modern light rail—would be noticeably different experiences. And yet when transport scholars David Hensher and Corinne Mulley of the University of Sydney Business School showed these images to about 1,370 people in six Australian capital cities, the difference in preference was enormous.

For the study, Hensher and Mulley gave survey respondents the two images above, plus two others whose only difference was older-looking vehicle styles (one bus and one train), and asked them to rank the four images in terms of «which one you would like to travel in most.» They found that 55 percent chose the modern light rail image, and another 18 percent chose the older light rail. Only about 17 percent chose the modern BRT. Just 10 percent chose the classic old bus.

The responses varied slightly among individual cities—Sydney, Melbourne, Canberra, Adelaide, Brisbane, and Perth—but only slightly. A majority of respondents ranked the modern light rail image first in every city except Sydney, where 49 percent gave it the top rating. The modern BRT got more first-place votes than the old bus across the board, but it never eclipsed 20 percent in any city.

What makes those findings more vexing is that every city involved in the survey is familiar with both modes. In other words, lack of awareness of BRT couldn’t explain the preference gap. Even in Brisbane and Adelaide, where BRT is more prevalent than light rail, the train earned top marks: 53 and 56 percent chose it first, respectively, to about 18 percent in each place for modern BRT.


The results suggest, at a very superficial level, «that ‘bus’ has a relatively bad image, and that BRT suffers from its indirect association with bus, with a very high preference for non-bus images,» as Hensher and Mulley put it. So we might be tempted to conclude that people simply like trains more than buses. But as the rest of this study (and others) show, that simplified conclusion would be wrong.

Buses Are Boring — Unless They Run Well

Paraphrasing a former mayor of Los Angeles, Hensher tells CityLab there’s an overwhelming perception «that buses are boring and trains are sexy.» That mindset complicates the discussion of mass transit plans in growing metros: though advanced bus systems can perform as well or better than streetcar or light rail systems for less money, people would rather have trains. In previous work, Hensher has called this emotional preference a «blind commitment»:

The main point is that the enthusiasm (almost blind commitment) for LRT [light rail] has caused many to overlook the potential for more cost-effective bus-based systems and even simpler improvements to bus services that do not require dedicated right of way.

For sure, some people have an almost ideological preference to certain modes of public transportation. But as with so much else we take on faith, that blind commitment breaks down in the face of exposure.

As Hensher and Mulley dug deeper into the data, they found that images alone didn’t tell the whole story. On the contrary, certain ridership factors influenced bus perceptions in a positive way. Respondents who had taken more trips by bus in the past month, for instance, had a higher probability of preferring a bus image to a train image. Ditto for those who’d gotten a seat for the entirety of a recent bus trip. Ditto again for those who’d taken any mode of transit in the past month, compared to those who had not.

The results, conclude the authors, underscore «the importance of exposure and experience in using public transport as conditioning preferences for bus and light rail options.» One reason they suspect Sydney residents gave buses the highest bus rankings, relative to other cities, is that service there has recently improved. In other words, people might indeed have an initial tendency to dislike the bus, but once they get on board and find it’s not so bad, those feelings start to change.

The findings echo a study from 2013 that also showed how transit service can matter more to riders than transit type. Analyzing 44 BRT systems and 57 light rail (and streetcar) systems, Graham Currie and Alexa Delbosc of Monash University in Australia found that the rail systems did, on average, carry more passengers. But once they adjusted for capacity, they found that routes with better service—features like higher frequency and integrated ticketing—»attract more ridership than low-service routes.»

In another recent study, Hensher and Mulley (along with colleague Chinh Ho) showed a head-to-head transit proposal to 1,018 people in eight Australian cities. One proposal was labeled BRT, the other light rail. The construction costs were the same for each system, but many of the 20 service and design features included in the proposals varied, including construction timeline, frequency, travel time, fare, and expected mode-switch from cars.

A University of Sydney Business School survey found a bias against BRT—but that disappeared when researchers adjusted for the service features they deemed irrelevant (above, a sample survey). (Transportation Research Part A: Policy and Practice)

With all 20 features in play—including, critically, the name of the system—people seemed to prefer light rail to BRT. But the researchers also asked respondents to mark which service attributes they deemed irrelevant. When the preferences were tallied again through the lens of what «really mattered» to riders, lo and behold, the bias against buses disappeared. Or, as Hensher and Mulley put it, «the modal image expressed through the name is, on average, of no consequence»:

Hence we might suggest from this evidence that once what really matters (differentially) to each individual is narrowed down, the LRT-BRT distinction blurs into a domain of non-relevance.

Marketing Alone Isn’t Enough

The point is not to pretend that everyone secretly hearts buses. That is not the case. A lot of people really un-heart buses.

Their reasons range from defensible to disturbing. Some rightly equate trains with unfettered movement while seeing bus travel as captive to the same congestion that cars endure, but without the benefits of privacy or trip flexibility. Others suffer from what Hensher calls an «obsession with technology.» Still others—namely, public officials—simply go where the free federal money is. We’ve seen that before with the people mover, and may be seeing it again in places with the modern streetcar.

And, fair or foul, buses carry a social stigma for some people in some places. In 2009, the U.S. Department of Transportation reported results from focus groups held to gauge the public perception of bus travel in Los Angeles. The comments they received about local bus service, in particular, had a pretty consistent (and appalling) theme of «shame»:

  • «I’m ashamed to tell that I am taking buses…In Europe, I wouldn’t. But here, they would think, ‘Did he lose his job?’ «
  • «The shame factor is majorly big.»
  • «I’m just saying that when I was in L.A. and I was in the car and just looking in at the bus…the people getting on….it just seems scary…»

If a shameful image of buses were the whole of the problem, we could probably throw some ad money at it and change the conversation. Indeed, citing this 2009 DOT report, Josh Barro of the Upshot recently argued that cities can save money on rail projects by spending more on bus marketing. But while the right ad campaign could help at the margins, what the research in the preceding section really shows us is that transit service influences transit perceptions—not the other way around.

The Orange Line BRT in Los Angeles has shown that good service will attract new bus riders. (Oran Viriyincy / Flickr)

Barro points to the success of L.A.’s Orange Line, for instance, as evidence that «it is possible to overcome anti-bus bias with the right amenities and marketing.» But in doing so, he mistakes the Orange Line’s integral service improvements, such as high frequencies and dedicated lanes, for amenities at best or marketing ploys at worst, when in fact they represent a fundamentally stronger system. To suggest that reliable service and exclusive lanes are a product of savvy marketing is to suggest that Michael Jordan jumped high because Nike said so.

You can sense the magnitude of the change from buses to BRT in the way L.A. riders speak about the Orange Line, at least as captured in the 2009 DOT report. Riders can’t seem to reconcile that it’s a bus at all; instead, they describe it the same way they’d describe a train. Some actually called it a «train-bus»:

  • «I was informed that what I take is the bus but I don’t consider the Orange Line a bus. I think of it as a train
  • «…my main issue is efficiency, speed…what I associate with that is the Orange Line, the subway system, the railways, any dedicated streets or maybe a dedicated lane.»
  • «At least the Orange Line has its own busway. Nothing but buses. That’s why I like it. And you have the clock thing when the next one’s coming and you feel like it’s a New York subway

So it’s possible that some people just love trains more than buses. But it’s equally likely, in many cases, that people have just used «trains» to mean «good transit» and «buses» to mean «bad transit.» If that’s the case, then marketing better buses as something like trains (or, at least, something other than buses) should weaken this automatic association. But such efforts will fall flat without meaningful investments in well-designed service: dedicated lanes, reliable peak and off-peak service, off-board fare payments, comfortable stations or enhanced shelters, or reconfigured routes, to start the list. A pretty picture alone isn’t enough.