Researchers Refute Higher Density=Better Transit Principle

Prevailing wisdom is that transit mode and frequency of service is dependent on residential density, which leaves low density, outer suburbs in a lurch, instilling an auto-dependent lifestyle. Not so, says Australian researcher and author Paul Mees.

"Residents in the outer suburbs should not have to wait for higher housing densities before getting better public transport, according to research that could defuse one of the most bitter controversies in urban planning."

Writing in the journal Australian Planner, Dr. John Stone, of the University of Melbourne, and Dr. Paul Mees, of Royal Melbourne Institute of Technology University present an alternative to what many view as Smart Growth 101: The key to better transit is to increase density and mixed uses.

"The keys to increasing public transport use in outer suburbs are more frequent buses, running at least every 10-15 minutes, and not just in peak hour; better co-ordination with rail services; more convenient transfers; and fares that allow free transfers between modes."

''Alternatives to the car will need to be effective at existing urban residential densities''.

Thanks to Allen Tacy

Full Story: Transport study derails thinking on outer suburbs



Transit does need density to be efficient

Thist "study" looks at city density, not transit corridor density, and declares that the lack of sufficient correlation between city density and transit usage means that there is no connection.

In other words, Los Angeles, which is overall fairly dense, doesn't have much transit usage, and that means density doesn't matter. The fact that Los Angeles destroyed their transit system, built lots of highways, and is only now building a transit network is irrelevant.

The giveaway is the conclusion that suburbs "need" a lot of transit service, with no mention of cost, much less of cost per rider served. Transit needs density at stops, and more generally, along corridors, if we want to serve the most people at the lowest cost.

Low Density Means Low Energy Efficiency for Buses

"The keys to increasing public transport use in outer suburbs are more frequent buses, running at least every 10-15 minutes, and not just in peak hour; better co-ordination with rail services; more convenient transfers; and fares that allow free transfers between modes."

If buses in low-density suburbs run every 10 or 15 minutes in off-peak periods, how many people will be in each bus? Few enough people that it will not only be very expensive but also very wasteful of energy, probably more wasteful than if those people drove.

More frequent bus service in low density suburbs will obviously increase ridership, as the article claims - but at a huge cost in money and energy.

Charles Siegel

The Planetizen Interpretation of the Australian Article

I must say that I eagerly awaited this issue of Australian Planner so I could criticise the authors of the article. However, I did not quite take away the same interpretation of the authors' research that the person who posted the article on Planetizen did. I will criticize the article for (a) poorly documenting its assumptions and (b) mixing local and regional scales for purposes of the evaluation. I'd suggest that the person who posted the Planetizen version of the article go back and reread the article (as I will do). I might also suggest that the researchers be more diligent in census-based research (consistency is key, especially when comparing data among several countries where there are many differences in data collection, methodology and even questions).


Follow up... Having read the article more thoroughly, I must say that I'm quite surprised that this article survived its peer review. My apologies to those who posted the news link on Planetizen, as the article does accurately interpret the information presented in the article. My first read of the article didn't include a look at the graphs and charts.

The article states that planners claim density needs to be increased across a whole city or region before significant improvement in PT patronage can be reached. I've never heard a planner say this, but I'm sure some planner out there might have made the claim (I have heard a planner say that density isn't needed to make transit sustainable - and I disputed it - so maybe this is effectively the same statement).

Obviously, transit service is both delivered and utilized at a localized level, and its productivity at a regional level is effectively the sum of the parts of the transit network (which typically doesn't cover an entire Census-defined region or even an entire metropolitan planning region). Density is a critical factor in transit patronage with the dual caveat that density take place along transit corridors and that transit corridors serve density.

I don’t disagree that networked transit planning can yield higher ridership (the subject of another discussion); this seems to be the central point of the article in spite of its title, but I don’t think the article actually makes its point with any of the data analysis. The article finds "serious prospects" that transit (PT) may become the dominant motorised travel mode in Australian cities with no basis in fact or research; this subjective notion seems to be the only statement linking the content of the article to its title.

The major flaw in the report appears to be in the mixing of scales (e.g. city versus region versus areas actually served by transit), as well as the mixing of data points (as well as irrelevant information). For example, the article mentions Canada Line ridership during the 2010 Winter Olympics which is irrelevant to a discussion around residential densities and the corresponding transit patronage, since many of the Canada Line users were not Metro Vancouver residents. The article also characterizes the Canada Line as light rail before making an “apples to oranges” comparison by comparing SkyTrain with Brisbane’s QR (a regional commuter system).

As for the entire SkyTrain system, one only needs to look at station area development to find the ridership explanation (station area densities are washed out in regional data comparisons). The author goes on to make value-loaded statements about “democratically sanctioned urban consolidation,” implying that everyone hates density (but apparently loves transit), as well implying that there is some relationship between whether a neighbourhood along a transit corridor "voluntarily" accepts intensification and a statistical analysis of whether urban density has a relationship to PT patronage. I'm not sure why such statements were in this article at all.

The article refers to “recent changes in reporting methods by US statistical agencies” that allow these comparisons with Australia before citing earlier literature (no citations are given for actual data). I’d be curious to know which agencies have changed which statistical measures, for reporting purposes. Since the most recent US Census data is from 2000 (Canada 2006), with 2010 US Census data only beginning to arrive in early 2011 after the article was published, this statement seems implausible. Add to that the a fairly long history of consistent reporting of transit statistics by the FTA, I am simply unable to identify which US statistical agency has changed their data reporting methodologies “recently.”

As far as density goes, Figures 5 and 6 in the article based on “urbanised hectares” but there is no definition of how these are derived (yet Los Angeles, Toronto and San Francisco are all remarkably identified as having roughly the same residential density). The labels and place names in the article and charts correspond to cities (that is, the names of specific municipalities) and not to the regions defined by the Census as CMSAs (CMAs in Canada), PMSAs or by regional planning regions established in urbanised regions for planning purposes. It is unclear whether the researchers separated urban from non-urban using the Census designation by Census Tract, or whether these are the population densities at the municipal or county levels. The data appears used in this article appears to be sourced from mixed geographies (cities, Census regions, planning regions, and/or possibly even counties could have been used, and research is quite unclear without proper citations).

Toward the end of the report, the authors claim that a single agency is responsible for transit network planning in Toronto, which implies that the authors only considered the City of Toronto (and not the GTA or GTHA). Obviously, this ignores GO Transit, and the suburban transit providers, as well as the interesting divisions of planning and asset ownership under the Metrolinx framework since 2006.

I was hard-pressed to find a "learning opportunity" in reading this article.


Further follow up. I had an email exchange with one of the authors. It appears that the author is making an indirect criticism of intensification strategy in Section 2 of the Melbourne 2030 Plan. It seems that the author dislikes urban density (or perhaps urban change/transformation). I'm sending in a letter to the editor regarding PIA's peer review and publication standards.

Irvin Dawid's picture

Australian Transit & Density research

I'll just say kudos to Keith for the above analysis of the article - the study appeared to me to be if you do your transit well, you can overcome lower density disadvantages.

I will just offer that anecdotally it appears that in low-density, particularly affluent areas where most households have multi-vehicles, I think the only transit that appears viable are school buses where they are mandated.

I think taxis make the best transportation option here - can they be called transit? In fact - more anecdotes, I think for commuter rail stations in affluent suburbs, taxis really do make sense - of course, this means having to phone for a taxi in the morning - but I suppose that if done regularly, a good relationship good develop between patron and taxi provider....
Irvin Dawid, Palo Alto, CA

Prepare for the AICP* Exam

Join the thousands of students who have utilized the Planetizen AICP* Exam Preparation Class to prepare for the American Planning Association's AICP* exam.
Starting at $245

Essential Readings in Urban Planning

Planning on taking the AICP* Exam? Register for Planetizen's AICP * Exam Preparation Course to save $25.
T-shirt with map of Chicago

Show your city pride

Men's Ultrasoft CityFabric© tees. Six cities available.
Planetizen Courses image ad

Looking for the perfect last-minute gift?

A one-year subscription to Planetizen Courses is the gift that will continue giving all year long.