Request for Information: Canadian Ridership (Updated)

Some colleagues and I have lately had occasion to compare public transit performance in young metro areas of 1-3 million, across North America and Australasia, and we’ve been turning up draft charts like this one (click to enlarge):

Canada miracle
The x-axis is population, the y-axis is public transit trips per capita, per annum.  The red triangles are the Australian cities: Adelaide, Perth, Brisbane, Melbourne, and Sydney.  The purple circles are US cities: Honolulu, Portland, and Seattle.  The blue dots are Wellington and Auckland in New Zealand.

But we’re especially struck by the green dots, which represent Canada.  The four dots are Ottawa, Edmonton, Calgary and Vancouver, and all are at the top of this heap despite relatively low populations.  Even tiny Edmonton and Calgary are outperforming all the other cities studied, including Sydney and Melbourne.

I’ve been asked to write a few paragraphs about why Canada’s cities do so much better than very similar cities in the US or Australia/NZ.  I have my own speculations, but anyone with references or insights on this, please comment!

UPDATE:  The consultant in charge of this project has reviewed the comments so far (Oct 8 0900 Sydney time) and observes:

Some rapid comments, having skimmed through the blog responses:

  • Area boundaries. Always a problem. We have tried to cover the whole metro service areas in each case.  For SYD, this has meant including Newcastle and Wollongong etc, given the CityRail services extend to these areas. Similarly we have included Gold Coast and Sunshine Coast in the Brisbane figures.
  • ** Linked v unlinked v return trips.  I doubt if this is a problem—we have aimed for unlinked trips (boardings) in all cases.
  • ** Densities may well be one of the explanators of the differences, and we have estimates of these for the  North American metro areas (but not the Aus/NZ areas. I am reluctant to pursue these in more detail—as you well know, this is a minefield.

The whole topic would be good for a PhD—why has no done it already??

Good question!  More on this topic here!

55 Responses to Request for Information: Canadian Ridership (Updated)

  1. Brent Palmer October 7, 2010 at 3:29 am #

    The population figure for Brisbane is almost 3 million. Does that include the Gold and Sunshine Coasts, i.e. the conurbation of Southeast Queensland?

  2. Jarrett at October 7, 2010 at 3:30 am #

    Brent.  Yes.  J

  3. Warren Rempel October 7, 2010 at 3:36 am #

    I have a couple of thoughts, completely speculative, and off the top of my head:
    Calgary and Edmonton are both in the oil revenue-rich province of Alberta and it may simply be a matter of them having more money at their disposal. The province established the Heritage Fund years ago, which sets aside oil tax revenues for expenditures on large public projects, research, etc. The cities are also quite competitive, so if one gets rapid transit, the other is close behind.
    Ottawa is the nation’s capital, so prestige could be a factor here.
    And finally, Vancouver is just a very progressive city… It’s doing almost everything right.

  4. Alon Levy October 7, 2010 at 5:29 am #

    The numbers in Calgary seem wrong to me. Calgary Transit claims much lower ridership numbers, 88 trips per capita.
    Conversely, the Sydney numbers seem too low. We know from other sources what all those cities’ transit mode shares are, and Sydney’s is the highest on the list, at 25%. The next highest, Ottawa, has a 21% share.
    Are you sure the comparison is apples-to-apples, rather than linked trips in Sydney and unlinked trips in Calgary?

  5. Jarrett October 7, 2010 at 5:39 am #

    @Warren. Thanks! Those were exactly the speculations that I had, but I am hoping to write something a little more factual.

  6. Matt October 7, 2010 at 5:59 am #

    I’ve made a chart of some data published by Paul Mees (available at which shows transit mode share, density and population for some US, Canadian and Australian cities.
    Comparison of the mode share with your ridership data is interesting. While it looks like there are some differences between the metro area definitions, you can see that:
    – Melbourne has a lower mode share but higher boardings than Sydney
    – Adelaide has about the same mode share as Edmonton but much lower boardings
    When you look at the density vs city population map, Canadian cities are typically higher density at all city sizes than their Australian and US counterparts. Perhaps this plays a part.

  7. Paul C. October 7, 2010 at 6:02 am #

    There was some recent discussion on this in Ottawa lately in the context of a mayoral candidate’s claims re: BRT v. LRT ridership impacts. A journalist did some digging and posted his results here:
    His numbers would put Canadian cities far closer to what you have found for U.S. and Australian samples.

  8. Jonathon October 7, 2010 at 6:20 am #

    These numbers depend largely on inclusion of the car dependent suburbs. This journalist pegs Toronto at 76 per capita, however recent statistics would suggest that the actual city of Toronto (not Vaughan or Mississauga or Brampton) gets closer to 185 trips per capita.

  9. John October 7, 2010 at 6:29 am #

    It must be the good weather in Canada.
    Seriously, I think Alon is probably on to something. There’s a problem with either linked vs. unlinked trips or round trips (aka: return trips) versus single direction trips.

  10. Geodarcy October 7, 2010 at 6:55 am #

    The numbers for Edmonton and Calgary do seem out of line with other values I’ve heard though Calgary does will with its LRT system because a majority of the jobs in the city are centred in the downtown core and parking is very expensive. Its whole LRT system currently consists of three lines radiating out from the core.
    If these numbers are true, I would suggest that weather, daylight hours, immigration and maybe poor alternatives. While commuting in Calgary, I wouldn’t ride my bike the 14 km between home and work in the winter because of snow and having to ride both directions in the dark between mid-October and the end of February. Also, anyone who doesn’t live very close to the recreational bike path network, myself included, must deal with riding in unfriendly traffic. My wife immigrated from China five years ago and still doesn’t have a driver’s licence. I know a number of other immigrants who cannot or will not drive.

  11. Ed O October 7, 2010 at 7:20 am #

    Jarrett, Warren, – but Portland is progressive and has invested in LRT as well. There may be more to the success of the Canadian cities than the factors you are speculating about.
    In addition to re-checking the PT trips/capita numbers for the Canadian cities, check also whether it’s reasonable to use the population of Sydney’s Greater Metropolitan Region as a basis for calculating PT usage for comparison purposes – it’s a huge area including people in the Illawarra and Lower Hunter, some more than 160km from the CBD.

  12. David M October 7, 2010 at 7:41 am #

    Hi jarret
    Some thoughts to explore:
    1. Canadian cities are all operated by a single transit authority and all modes of transit are integrated. This leads to services supporting each other
    2. Edmonton and Calgary are contained planned cities – their densities are generally, but when taken the entire metro area,a re higher than most US cities.
    3. U-pass program and other pass programs. Most students in all the cities have access to transit as part of their university costs. Many big employers have pass programs.
    4. None of the Canadian cities have major freeways into the city centre – but have worked on providing a quality transit link into the core making transit a much better and faster option.
    5. Mentality – there is just a stronger use of transit in our mentality than elsewhere. Using transit in Canada is not a stigma but the opposite, a sign of being progressive.
    6. Value added – many transit systems offer value added for passes – such as weekend family free with pass holder.
    7. Strong nodes of employment – Edmonton has the downtown and Universities; Calgary has a large concentration in the city centre, Greater Vancouver has the city centre, Broadway, Metrotown and other town centres; Ottawa has a large government population working downtown.
    8. Taxpayer support in Canada – leads to a lot of political debate, discussion about funding and I think awareness of the system and its benefits
    9. Relatively quick to address service concerns – (i.e. Vancouver just extended rush hour service on skytrain by 15 minutes to accommodate cyclist of all things)
    10 Practical approach to transit planning – know not going to get everybody out of their cars, so accommodate that with free or inexpensive park and ride lots at stations, or at points where express buses run direct into the city centre.
    There’s probably more…

  13. Tom West October 7, 2010 at 7:43 am #

    What’s the source for the Canadian numbers? CUTA/ACTU is the best bet, but make sure the methodology behind their numebrs (especially teh definition of “population”) is consident with other countries.
    I got the following (transit trips/service area population/trips per capita):
    * Calgary: 94,892,400/1,042,892/90.9
    * Edmonton: 66,092,169/752,412/87.9
    * Metro Vancouver (Translink): 178,803,205/2,271,224/78.2
    * Toronto (TTC): 466,700,000/2,508,281/186.1
    * Ottawa: 93,934,710/898,150/104.5
    * Montreal: 382,521,000/1,877,693/203.7 (!)
    Toronto’s figures excludes rides on GO Transit (the inter-regioanl transit provider) (which would add another 40-50m trips), but also excludes people outside of Toronto proper from its service area population, who certainly ride TTC in great numbers. If you were do calculate “rides on TTC and GO Tranit by Torontonions”/”Toronto’s population”, you’d get a very differnt number, probably lower.
    Similarly, Edmonton/Ottawa’s ridership include rides by people from St. Albert/Gatineau respectively, but the service area population doesn’t.
    I’ve no idea why Montreal’s is so high…
    Conclusion: the transit systems of many large Canadian cities have larger numebr of riders from satillite cities, who may have transit system of their own. Thus “rides per capita” is hard to compute for a citiy – you must do it for a self-contained region (transit-wise). Calgary and Vancouver’s numbers do that.

  14. Kyle October 7, 2010 at 7:45 am #

    Apta confirms the ridership rate in Calgary : page 33.
    As a former Calgarian I would cite the following for ridership:
    *consistently pushing system expansion
    *a decision years ago to limit the number of traffic lanes into downtown Calgary
    *a cap on total number of parking spots in the commercial core
    *very long term planning, for example the line currently under construction first had land reserved for the ROW in early 1970s
    *the availability of park and ride spots for around 10% of users helps people ease into transit use
    *mandatory transit passes for all post secondary students
    *a dense city by North American standards – it might be really suburban but few neighborhoods were built at a density of less than 6 units per acre, with more recent neighborhoods hitting 12 upa
    *a right sized road network – roads in Calgary don’t trick people by having an excess of capacity far from the urban core, they back up everywhere
    I guess a lot of these issues come from one thing: Calgarian’s are pretty cheap, and they elect politicians to limit property taxes. With little federal or provincial (state) assistance in building infrastructure choices had to be made very judiciously on where to spend money. Transit just moves a lot more people for the dollar.

  15. David M October 7, 2010 at 7:46 am #

    Another to look at:
    1. age of populations – Edmonton and Calgary have lower age due to the oil industry. Lower age usually means more trnasient populations, less settled and maybe more prone to using transit.
    2. In canada, the average city sees 50% of its residents renting as opposed to ownership. I think for many reason, some tax, home ownership is much higher in the US and maybe Australia. Renting doesn’t mean low income, but usually means less disposable income, less likely to own a car.
    3. Which brings me to the last one – car ownership – do Canadian’s own as many cars as US and Australia? We have a lot but is is still less than the average?

  16. David Reevely October 7, 2010 at 7:48 am #

    Hi, I’m the Canadian journalist whose blog is linked to above.
    Do NOT count on my Toronto number. It’s definitely wrong — it only counts transit ridership on the City of Toronto (that is, downtown) transit system, but it incorporates all the populations of Toronto’s surrounding suburbs. They all have their own transit systems, too.
    For the purposes of my post, I just had to establish that Toronto’s transit ridership rates are substantially higher than Ottawa’s. The figure I calculated does that, without bothering to consider factors that would push the rate even higher.

  17. Tom West October 7, 2010 at 7:50 am #

    David M: “1. Canadian cities are all operated by a single transit authority and all modes of transit are integrated. This leads to services supporting each other”
    Umm…no. See Edmonton, Ottawa, the entire Greater Toronto Area, and Montreal.
    That said I agree with David’s other points… expecially the lack of stigma compared to the USA, lack of freeways into the downtown core, and the UPass scheme.
    Overall, I think the bigegst thing is effective planning by cities as they grow. Most large Canadian cities realised in the 60s/70s that they couldn’t support future growth through roadways alone, so opted for gorwth based around transit usage.
    Calgary is one the best performing by any measure… they planned the routes of their LRT system in the early 80s, and all future growth was based around the LRT being built someday, so room for tracks and stations and approraite zoning was always included in new devlopment.

  18. David Reevely October 7, 2010 at 7:51 am #

    And one more thing: my figures are based on reported rides taken so far in 2010, which means they only cover up to the end of the second quarter. Only about half the annual figures, that means.

  19. David in Ottawa October 7, 2010 at 7:53 am #

    These numbers are out to lunch.
    Ottawa is around the 120 trips/capita mark, Calgary around 90. Vancouver is slightly less than that, Edmonton lower still.
    Fwiw, Ottawa and Calgary both have in place suburban planning policies requiring that new suburbs be designed to facilitate feeder bus service (<400 m to a bus stop is typical). While both cities "sprawl" their sprawl isn't as bad as US sprawl in that there's a lack of uncontrolled leapfrog-type development and fairly strong planning oversight to keep it that way (Ottawa had a regional government, Calgary annexed mercilessly). This ensured that providing transit coverage was more economical, especially when combined with the "built in" transit routes of new suburbs. Both cities clamped down on downtown parking and both cities have limited freeway access into their downtowns. Both cities have large educational institutions on their respective rapid transit systems (three in Ottawa, two in Calgary) that provide good all day ridership.
    So there's plenty there.
    Culturally, transit in Canada never quite took on the deprecated status that it got in many American cities, and I think there was a greater understanding that bigger freeways meant that neighbourhoods would have to be destroyed. Transit was widely seen as early as the 1970s as a better solution to mobility issues than freeway building. A more civic or public-oriented cultural ethos may well be at play in Canada as well in that taking transit is seen to be a good thing to do.

  20. Kyle October 7, 2010 at 8:02 am #

    [2. In canada, the average city sees 50% of its residents renting as opposed to ownership. I think for many reason, some tax, home ownership is much higher in the US and maybe Australia. Renting doesn’t mean low income, but usually means less disposable income, less likely to own a car.]
    In 2006 in Alberta 73.1% owned the dwelling they lived in –
    [3. Which brings me to the last one – car ownership – do Canadian’s own as many cars as US and Australia? We have a lot but is is still less than the average?]
    Canadians own more cars per capita than Americans or Australians

  21. Robert in Calgary October 7, 2010 at 8:12 am #

    Time for just a quick post…..
    LRT in Alberta has nothing to do with the Heritage Fund.
    1. Calgary has a very strong downtown. Transit has a high market share.
    2. We do have high parking prices.
    (might be the highest in Canada and one of the highest in North America, based on recent surveys released.)
    3. We did the opposite of Edmonton. Started on the surface downtown and spent more money expanding the system. More legs sooner boosted the ridership.
    4. Calgary officials say a key success is the bus system feeding into the stations.
    Here’s two links for you, the first is entire page listing.
    The second is the “First 25 years” presentation made in St. Louis in 2006

  22. Matt G. October 7, 2010 at 8:15 am #

    A lot of Canadian municipalities have gone through amalgamations and therefore the tax/voter base includes more of the surrounding suburbs than seems to be common in the US. But that’s just my impression overall.
    The only hard data I have is that the city of Ottawa (where I lived for a while) includes most of it’s suburbs.

  23. Robert in Calgary October 7, 2010 at 8:28 am #

    And a bit more,
    Calgary actually runs LRT as rapid transit.
    From the start, peak hour frequency has always been been in the five or six minute range.
    Our three car trains are packed. We’re (very) slowly moving towards running four car trains.
    Route 201, the combined South/NW lines, has major shopping centres near stations. Chinook Centre is one of the top 3 in Canada.
    The NW line has the University and SAIT (Tech Institute)

  24. David M October 7, 2010 at 9:19 am #

    Thanks – times have changed over the the years with respect to home ownership. However, according to Statistics Canada, in the Edmonton CMA, 30% are rented. Of all the housing types, 57% are single family dwellings.
    In Edmonton proper, Stats Can say that 37% rent and 50% of housing stock is single family dwellings.
    That last number is probably more important as I bet in the US and Australia, SFDs will be a higher proportion for similar size cities.
    Anybody else?

  25. Mike JA October 7, 2010 at 9:26 am #

    Bear in mind that the backbone of Calgary’s transit system was built for the 1988 Olympics, not in response to the massive oil revenues which have characterized the Albertan economy since the late 1990s. It seems simplistic to conclude that oil bought it.
    A few additional comments from a Torontonian:
    – Vancouver’s urban structure lends itself to transit, insofar as there’s a clearly-defined downtown, and the inner city and inner suburbs are all mid-density. You don’t hit conventional cul-de-sacs-and-McMansions suburbs until you’re well out of the city proper.
    – Toronto has maintained much the same transit structure as it has had for about a hundred years now. Lines have been converted from bus to streetcar to trolley bus to subway to whatever else, but there have been very few major grid-defining shakeups, and this has allowed development to follow transit rather than vice-versa. (For example, if you stand at an intersection like St. Clair and Yonge, you find 20-storey buildings in a place which, by all rights, ought to have none–but because this intersection has had a subway stop for the last 40 years in addition to a well-served streetcar line in place for nearly a century, development has followed.)
    – There is precisely one major expressway into downtown Toronto, which does not enter the downtown itself, and is constantly on the cusp of being demolished. The only way to get down there is on a bike, on foot, or by transit.
    – Montreal’s subway is pretty sweet.
    – As a result of the weather, downtown Toronto has an underground city connecting every major building between Yonge and University. Some people are able to do all their shopping, working and living for weeks at a time without going above-ground. Other cities (Montreal in particular) have similar types of development.
    – People in Toronto seem to like subways, to a point where we demand they be built in even developmentally-inappropriate contexts. (See: Rob Ford.)

  26. David Arthur October 7, 2010 at 9:38 am #

    I’d definitely agree with the emphasis on Calgary’s core as a cause of ridership. The C-Train has its faults, but it’s really good at getting people into the surprisingly dense city centre.
    Vancouver, in addition to the excellent SkyTrain, has the advantage of one of the most logically arranged bus systems I’ve seen. It’s still not quite as simple as a good rail network, but it’s a lot easier to work out what buses go where than it is in many cities. Frequencies also tend to be good.
    There also doesn’t seem to be the same cultural prejudice against buses in Vancouver that exists elsewhere – this may be connected at least in part to the retention of the trolleybus network. (People in Vancouver have even been known to complain when their bus is replaced with a train.)

  27. David in Ottawa October 7, 2010 at 9:39 am #

    Can you tell us how the ridership numbers you’ve posted were derived, particularly for the Canadian cities?
    Canadian agencies’ and cities’ own reporting is coming up with numbers considerably lower than in your chart.
    John | 10/08/2010 at 00:29 noted there may be an issue with linked/unlinked trips. It’s my understanding that Canadian statistics (e.g. the 120 per capita for Ottawa, 90 for Calgary) are linked trips.
    If you’re using unlinked trips for all cities then it really does mean that the Canadian cities are outliers.

  28. Jake Wegmann October 7, 2010 at 9:47 am #

    One point about Calgary, which has already been made, but I think it bears more emphasis: it has exceptionally centralized employment by North American standards. Its downtown is comparable to Denver’s in terms of jobs, in a metro area that is one third the size. While that leads to enormous inefficiencies in the the LRT system, in terms of massive flows in and out of the downtown at the beginning and end of the day, it does make it an easy place to serve via transit. Especially in light of the aforementioned parking restrictions. Somehow Calgary has managed to keep white collar jobs from sprawling out of its CBD.
    I think this is an underreported story — people in Calgary often don’t even seem to realize how unusual of a situation it is.
    Is there a city on the continent with more centralized employment? New York is the only other place that I could think of that is comparable.
    In a lot of ways, Calgary seems to function more like an early 20th century North American city than a typical polycentric 21st century one.

  29. Michael Druker October 7, 2010 at 10:24 am #

    This has been mentioned, but I think an important reason is highways into the center of the city. Portland may have light rail, but it also has a ring of expressways around the CBD. If you look at American and Canadian cities on Google Maps, the difference in grade-separated highways into the center of the downtown is consistent and striking.

  30. ant6n October 7, 2010 at 10:28 am #

    Calgary comes up a lot because of their really integrated transit/city development. At the same time they focused more on getting transit out there rather than building expensive subways just in downtown:
    At the same time they focus on feeder buses to lrt stations; try to keep bus/lrt transfer penalties low (i.e. the stops should be as close as possible) and keep park+ride rates low.
    Also, according to their deisgn guide, they build lrt stations and bus stops to be within 400m of people, they try to create street grids that guide people better towards stops than the average suburb, they view transit stops as little centers in the city with a focus on community activity and density pockets.
    One note on Montreal: the city proper has 1.8M, the metropolitan area 3.5M people, because a lot of the low dense suburban neighborhoods demerged a couple of years back. one would have to be careful which living population and which riding population is picked to calculate the ratio

  31. Agustin October 7, 2010 at 11:08 am #

    I am surprised that the Canadian cities are up so high (with the possible exception of Vancouver).
    However, as others have said, Calgary does have really high LRT ridership during rush hour. I think this is in large part because of the extremely high parking rates downtown, and because a very large proportion of the population works downtown but hardly anyone lives there.
    I can’t speak for Edmonton because I’ve never lived there nor used the transit system there.
    I have used Ottawa’s transit system a few times and it’s gotten me around the city quite well. Reports from my sister-in-law are that busses are often full, especially in the winter.
    Vancouver’s transit system is fantastic and very highly used. Its modal share may be lower than in other similar cities because people tend to live closer to work and so walking has a fairly high modal share. (And cycling, increasingly so.)
    The U-Pass for students (in use at least in Calgary and Vancouver, and I believe Ottawa as well) seems to really increase ridership.
    When I visited Portland I was surprised to see how few people used the busses there. I saw a lot of people on bikes, but not that many on busses.
    Just one man’s observations… no data here, but hopefully I’ve said something useful!

  32. Robert in Calgary October 7, 2010 at 11:16 am #

    Jarrett, I’ll add some more.
    First, the Olympics were a bonus for building NW LRT, nothing more.
    An anecdote from today. Midday frequency is 10 minutes. I boarded the CTrain at 42nd Avenue station heading downtown at 10:10am. The middle car of the three car train. All seats were full, with about a dozen standing. After two more stations, 18 were standing.
    At the first station downtown, we gained more people and had about 36 standing. It’s the middle of the morning with 10 minute frequency and we have a “rush hour” crowd on the train.
    People outside Calgary underestimate how many people are traveling though downtown.
    Calgary has key gaps in the road network. This encourages congestion. Many newer communities have only one, two or three entrances pouring out traffic onto a limited road network. Hello congestion!
    By Dec. 2012, Route 202, West/NE LRT will be about 26km long.
    By 2014 route 201 South/NW will be about 34km long.
    Here’s the route animation of West LRT.
    West LRT is about 8km long. Only 30% is at ground level. Still an “easy” route in Calgary terms. 6 level crossings and LRT will have automatic priority at 5 of them. The one exception is 11th Street downtown.
    A lot of folks want to start construction on SE LRT as soon as West LRT is completed. The city has an “approved” route for North Central LRT, BUT, it runs in the Nose Creek Valley where no one lives. Opposition to this approved route is growing. We want some sort of Centre Street route.
    Calgary’s LRT success is why I argue that Skytrain is the better choice for Metro Vancouver. Metro Vancouver doesn’t have enough “easy” LRT routes to handle the crowds they would have with their higher densities.
    Double the density, more than double the demand.

  33. David M October 7, 2010 at 11:41 am #

    Some other possible reasons to think about:
    1. Weather – snow and cold conditions make driving a pain. ridership on transit is usually higher on bad weather days.
    2. Price of gasoline – has always been consistently higher in Canada than in the US mainly due to taxes on the fuel.
    3. Insurance costs – insurance can be expensive but I’m not sure if it is more or less than in the US
    4. Social verse individual paradigm – Canadian society has largely grown up on the principle of putting society ahead of individual needs, hence the greater acceptance of higher taxes and social programs. The US by contrast is largely based on individual need over social needs. This could translate into a social conscience that using transit is good for society. In fact, in my office, it’s almost embarrassing to admit you drove to work by yourself – walking, cycling or bus are popular.

  34. ChrisC October 7, 2010 at 12:15 pm #

    Can you share the source of your data? Every other source I’ve seen shows Vancouver’s per capita ridership slightly higher than Calgary’s, and substantially higher than Edmonton’s. And the commute mode share stats from the 2006 census do this as well.
    As for Canada having higher per capita ridership, we have much better levels of service. You’d have known this had you taken even a brief glimpse at transit schedules and service frequency charts. It’s no rocket science.

  35. NCarlson October 7, 2010 at 12:16 pm #

    I can’t speak too much for Edmonton either, but (without looking for the numbers) their LRT ridership is quite remarkable on a per mile basis as well, especially given the NW line is largely through declining industrial areas. It can’t hurt that this line is (or was until the SLRT opened anyway) built as more of a subway than typical LRT, and even the southern line is to very high standards, much like Calgary’s. It’s also another city with limited highway accessibility, and some real bottlenecks around the river valley, that the LRT nicely bypasses.

  36. Danny October 7, 2010 at 1:30 pm #

    You should notice that ridership has seen significant but very steady growth in all of the above mentioned Canadian cities. I interpret that to mean that ridership is being driven by slowly changing land use.
    My perception is that in Canada there are fewer euclidian zoning laws, and the few that they do have are a) much easier to appeal, and b) more likely to be influenced by significant transit investment.
    In the US (can’t speak for NZ or Australia), it seems like transit planners and land-use planners are almost at war with each other.

  37. D October 7, 2010 at 1:35 pm #

    Interesting pattern; you should do a study on Canadian readership, too, judging by the volume of comments.
    I’ve lived in Calgary, Montréal, and Vancouver, and would agree with most of what’s been written above.
    I’ll add that:
    Canadian cities have relatively little ‘super sprawl’ where things are very very low density; I’m thinking of the areas south of Minneapolis, or some of the outskirts of Denver, where land uses are incredibly spread out.
    As pointed out, freeway systems don’t generally go right into the downtown (other than Montréal and Toronto); they’re also much less extensive across the board, save perhaps for suburban Toronto and Montréal. I’d heard somewhere that Québec City actually has the highest kilometres of freeway per capita, but am not sure if this is true. Oh, and if anyone tells you that it’s because Canadian planners/politicians had the ‘courage’ to say no to the freeway, that’s really not the whole truth; we only discovered that courage after we realized that we didn’t have the money or technical expertise of our American counterparts.
    Calgary’s LRT system is indeed what your commenters say it is; it was extended (not built) for the 1988 Olympics and, sorry to burst your bubble ‘ant6n’, that design guide is largely meaningless, though this may (hopefully) change over the next decade, particularly with the West LRT.
    Jarrett’s already waxed lyrical about the virtues of Vancouver’s SkyTrain and bus system; these also bump the numbers up, as they really are pretty good. Again, it’s also just difficult to drive in Vancouver, as there are few freeways and the arterials really aren’t up to the standards you’d find in the US, or even in Calgary, for that matter.
    It’d be great if you could add Montréal into this analysis; but I can see that it’d be difficult to figure everything out with the commuter trains (AMT), on island transport (STM), and all the off-island transit agencies. Statistically, Calgary is a lot ‘cleaner’, while in terms of transit agency coverage, so is Vancouver.
    Anyways, great discussion; hope you look into this some more; it would be interesting to see similar charts for walking/cycling…

  38. ant6n October 7, 2010 at 2:07 pm #

    Yeah well, Calgary’s doc’s do sound to good to be true. But they are probably trying; it would be nice to see how well all this is actually implemented.
    Regarding Montréal, the origin destination study done by the AMT probably provides enough stats:

  39. Jarrett October 7, 2010 at 3:09 pm #


  40. Peter from Calgary October 7, 2010 at 3:44 pm #

    Calgary’s amazingly high ridership is due to two things:
    1. It has one of the highest downtown job concentrations in the world. The city is a one industry town (head offices for oil and gas)and every energy company wants to be in the downtown. This makes the density destination of the transit trip incredibly high. In 2008, the City reported something in the league of 297K LRT boardings per day. And that is for a city of 1 million!
    2. Parking policy. The City of Calgary has been restricting parking in the downtown for a few decades now and downtown parking rates are higher than everywhere in North America save for Manhattan. The cost of parking really drives transit ridership.

  41. jack horner October 7, 2010 at 4:06 pm #

    1. As discussed elsewhere, in any such report please please please be clear about how you are defining the unit of study, here called ‘metropolitan area’. Certainly in Australia, and I guess often elsewhere, political boundaries are pretty much irrelevant to transit planning.**
    2. I was under the impression that Sydney’s per person transit use is somewhat better than Melbourne’s. Including Newcastle (pop 400,000, 160km away) and Wollongong (pop 200,000, 80km away), which are both well separated from Sydney geographically, probably pulls the average figure down as Newcastle and Wollongong have much lower per person transit use than Sydney proper.
    ** ie irrelevant to competent metropolitan transport planning. Of course they may still be of keen interest to politicians whose main interest is avoiding responsibility.

  42. Daniel October 7, 2010 at 4:18 pm #

    Lots of great comments here, all very interesting.
    I would just note that Australian cities’ frequent networks are not particularly well-developed, eg check these maps from Melbourne’s 15 minute services at different times of day (these are from 2006, and a handful of new frequent routes have been added since).
    Apart from the outer edges of the maps, most of those suburbs have been populated for decades, and many are car-dependent.

  43. Brisbane October 7, 2010 at 4:20 pm #

    Propensity for transit ridership, it must be genetic!

  44. Alon Levy October 7, 2010 at 5:36 pm #

    Unlinked trips can be a real problem to compute, and to use. For example, for a subway, the headline figure is always linked; unlinked numbers can be produced (APTA certainly tries), but mean very little, because transfers are within the fare control system, and often there’s no way to know whether people are transferring or not.
    A high ratio of unlinked to linked trips probably means people are transferring from one mode to another. For example, in Calgary people transfer from buses to the C-Train, and in Vancouver they transfer from buses to Skytrain.
    The most apples-to-apples number is trip-to-work mode share. For Canada, the data is here. For Australia, I believe the numbers are Sydney 25% and Melbourne 18%; I posted a link in a comment a few months back, after Googling, but don’t remember where it is right now.

  45. Corey Burger October 7, 2010 at 10:41 pm #

    U-passes in Vancouver: I don’t know about the other cities in Canada, but only UBC, SFU, and a few smaller colleges have the U-pass, which leaves out a lot of students, so there is likely more growth in the transit ridership that is sitting in the table.

  46. ChrisC October 7, 2010 at 11:20 pm #

    >>Unlinked trips can be a real problem to compute, and to use. For example, for a subway, the headline figure is always linked; unlinked numbers can be produced (APTA certainly tries), but mean very little, because transfers are within the fare control system, and often there’s no way to know whether people are transferring or not.<< There's no way to know how many people board vehicles even if they aren't transferring either. For example, each morning and evening when I hop aboard transit, I shot my monthly pass to the driver, and he just lets me on. Same with the majority of commuters on the line. I do not swipe my card through a reader, nor does the driver hit some button to count the number of passengers. Everything is an estimate.

  47. Paul C, Vancouver October 8, 2010 at 2:25 am #

    I do believe the buses in Vancouver have sensors at the doors to count passangers as they load and unload. Or there are plans to install them. Once the smart cards come in. There should be a more accurate idea of when and where people are loading and unloading.
    As for reasons on why transit ridership is higher in Canada. I think most ideas have been suggest.
    Things like not building a full extensive freeway system. Density wise the cities tend to be more compact. Higher fuel costs. In the case of Vancouver the much higher housing costs I feel have had an impact. In a lot of cases people have had to decide on giving up 1 or more cars to afford a place to live.
    Although I also feel Vancouver is put lower due to its higher population downtown. Where a lot of people don’t really have to take transit and can just walk or cycle.
    It would be interesting to compare the cities with cycling and walking added as well.

  48. K October 8, 2010 at 3:50 am #

    Something I’ve noticed in my reading about Vancouver is that they have planned very well. And they have a history of following through on plans. Some of their land use plans are pretty much transport plans and vice-versa. I think a strong recognition of the link between the two has influenced them greatly.

  49. Michael Lewyn October 8, 2010 at 2:45 pm #

    I have some statistics on Canadian cities (comparing them with American cities) in this work in progress:
    (see pp. 12-17)

  50. Jonathon October 9, 2010 at 6:47 pm #

    Toronto’s transit system is remarkable for the amount of frequent off-peak service it provides. I can catch a streetcar on Spadina Avenue every 1-2 minutes every day of the week. Buses come frequently (5-10 minutes) every day on many major arteries, and almost every grid street (approx. every 2 km) in the city can count on no less than a bus every 12-15 minutes, often better. Also, a lot of Canadian universities have parking policies to actively discourage driving even though their campuses may not be downtown.

  51. Matt Fisher October 11, 2010 at 4:49 am #

    In Ottawa, Carleton (where I’m a student now) currently has the U-Pass. Yes, Ottawa has the highest per capita ridership in the image shown, and it is no doubt attributable to the success of the Transitway, but it is also based on that Ottawa has a high transit mode share (20% of the population).
    I can’t believe there are people in Ottawa who say, “We don’t have the population for LRT”. We do! Calgary and Edmonton had 500,000 or less at the time LRT was being built, but Ottawa (including Gatineau, where a busway similar to the Transitway is currently being built) had more than 700,000.
    Indeed, BRT proponents like John Bonsall will never admit they were wrong about BRT. They will never apologize for what they did. They will defend what Ottawa has done to the world, saying, “Look at Ottawa’s high ridership. It’s the result of the Transitway. That’s why your stupid rail city doesn’t have a mode share as big as our bus city right here in Ottawa. See, Ottawa is a green city and yours isn’t.” That’s what I don’t like about BRT, this whole sort of exceptionalism. But I agree that the Transitway is a success.

  52. David Ngo October 11, 2010 at 11:19 am #

    Canadian transit ridersihp numbers on here look correct if they are unlinked trips.
    The official numbers reported by Canadian transit agencies are linked trips, while American report ridership as unlinked trips. I’m guessing the chart is using unliked trips and that would be why the official numbers don’t match the numbers on the chart.
    So, yes, Calgary does have around 90 riders per capita per annum by Canadian standards. But by American standards, Calgary actually has around 140 riders per capita per annum.

  53. Jarrett at October 11, 2010 at 1:53 pm #

    @all.  Yes, we are clear that all the numbers here are unlinked trips. 

  54. Scott H January 17, 2011 at 7:27 am #

    Calgary Transit’s linked and unlinked trips for 2009:
    linked: 94.15 million trips
    unlinked: 148.88 million trips
    Calgary 2009 Census Population: 1.032 million
    linked trips per capita: 91.2
    unlinked trips per capita: 144.3

  55. Nicolas Derome October 21, 2012 at 5:41 pm #

    Montreal has a lot of high density neighbourhoods for its size. The percent of Montrealers living in high density census tracts is greater than any North American city but New York.
    I believe this is mostly due to the fact that Montreal was considerably poorer during the mid-20th century than other American cities, so it started building typical low density suburbia later than other cities. By the time that did start happening, Montreal’s growth was slower since Toronto had overtaken it as Canada’s primate city.
    Boroughs of Montreal like Saint Leonard, Saint Michel, Lasalle and Montreal Nord are all very dense with apartments and attached homes. When these Montreal boroughs were being built, most other cities were building bungalow suburbs. A little later, in the 60s/70s, Toronto started building towers in the park, but they were more in the middle of nowhere compared to Montreal’s dense mid 20th century areas.