Australia ranks second for traffic congestion

ScreenHunter_277 Nov. 14 12.57

By Leith van Onselen

Demographia’s Wendall Cox has produced an interesting report in New Geography drawing on satellite navigation data from Tom Tom (called the Tom Tom Traffic Index) to determine which regions have the worst traffic congestion. Under the Tom Tom Traffic Index methodology:

…travel times during non-congested periods (free flow) [are compared] with travel times in peak hours. The difference is expressed as a percentage increase in travel time. We take into account local roads, arterials and highways. All data is based on actual GPS based measurements and for each city the sample size is expressed in total number of measured kilometres for the period.

According to Cox, New Zealand has the worst traffic congestion in the five regions surveyed, with Australia coming in second. The United States, by contrast, suffers from the least congestion, with cities operating liberal land supply/planning policies fairing the best:

Tom Tom produced data for 122 urban areas in the higher income United States, Western Europe, Canada, Australia and New Zealand. This included nearly all urban areas with more than 1,000,000 population, and some smaller. It might be expected that the “sprawl” of US urban areas, and their virtual universality of automobile ownership, as well as the paucity of transit ridership in most metropolitan areas would set the US to to the nether world of worst traffic congestion. This is not so, and not by a long shot.

1. New Zealand: The trophy goes to, of all places, New Zealand (Figure 1).  The average excess time spent in traffic in the three urban areas of New Zealand rated by Tom Tom was 31.3%. This means that the average trip that would take 30 minutes without congestion would take, on average, approximately 40 minutes in the three urban areas of New Zealand. This is stunning. New Zealand’s urban areas are very small. The largest, Auckland, has a population of approximately 1.3 million, which would rank it no higher than 25th in Western Europe, 35th in the United States and 4th in Canada and Australia. Christchurch and Wellington are among the smallest urban areas (less than 500,000 population) covered in the Tom Tom Traffic Index, but manage to rank among the 20 most congested (Figure 2). Christchurch has no freeways and Wellington’s are not long.

ScreenHunter_278 Nov. 14 13.49

2. Australia: Second place is claimed by Australia. The average trip takes 27.5 percent longer in Australia because of traffic congestion. All five of Australia’s metropolitan areas with more than 1,000,000 population are among the 20 most congested urban areas in the higher income world. In the case of four urban areas (Sydney, Brisbane, Perth and Adelaide), every larger US urban area has less traffic congestion. Melbourne is the exception, but is still “punching well above its weight,” with worse traffic congestion than larger Chicago, Dallas-Fort Worth, Houston, Toronto, Philadelphia, Miami, Atlanta, Washington, Riverside-San Bernardino and Boston.

3. Canada: Canada is the third most congested, with an excess travel time of 24.8 percent. Vancouver ranks as the third most congested urban area (36 percent excess travel time) in the higher income world, and has displaced Los Angeles as suffering the worst traffic congestion in North America. This is a notable accomplishment, since Los Angeles has more than five times the population, is more dense and only one-third as many of its commuters use transit to get to work. None of the other five largest urban areas in Canada (Toronto, Montréal, Ottawa, Edmonton and Calgary) is rated among the 20 most congested in the higher income world (Figure 3). Toronto is tied for 6th worst in North America with Washington (DC-VA-MD) and San Jose (Figure 4).

4. Western Europe: Fourth position in the congestion sweepstakes is occupied by Western Europe, where the excess travel time averages 22.2 percent. Marseille (France) and Palermo (Italy) are tied with the worst traffic congestion in the higher income world, with excess travel times of 40 percent. Excluding Christchurch and Wellington, Marseille and Palermo are among the smallest urban areas among the most congested 20, though their large and dense historic cores complicate travel patterns. Rome, Paris, Stockholm and Rome, all with strong transit commute shares, are tied with Vancouver for the third worst traffic congestion (36 percent excess travel time). Other Western European entries to the most congested 20 rankings are London, Nice and Lyon in France and Stuttgart, Hamburg and Berlin in Germany. Western Europe contributes only 11 of its 54 rated urban areas to the most congested 20 list (the most 20 most congested list includes 24 urban areas because of a five way tie for 19th)…

5. United States: The United States is the least congested in these rankings with an excess travel time of 18.3 percent. Even after losing its top North American ranking to Vancouver, Los Angeles continues to be the most congested urban area in the United States, with an excess travel time of 35 percent. San Francisco (32 percent), Seattle, and much smaller Honolulu (tied at 28 percent) are also in the most congested 20. Only four of the 53 rated US urban areas is in the most congested 20.

ScreenHunter_280 Nov. 14 13.50

While the results might seem counter-intuitive, they are not. The fact remains that urban consolidation policies can slow travel times, since they 1) pack more people into a given area, increasing congestion in the process; and 2) tend to favour public transport over roads, despite most people preferring (and using) the latter option.

[email protected]


Leith van Onselen
Latest posts by Leith van Onselen (see all)


  1. Ummm…. Christchurch has had its motorways and most major roads knackered by a rather large set of earthquakes. And , hey, 10 minutes extra during working hours compared to after hours? (which is about what is is from the outer suburbs of Christchurch to the City centre) I’ll take that, given that when I lived in London there were days it took me an hour just to get out of my suburb!

    • Janet … congestion certainly has been a problem here in tiny Christchurch (population circa 370,000) for years prior to the earthquakes.

      We should be scoring at the same level as small mid North American cities.

      Excessive numbers of overpaid bureaucrats certainly do put a dampener on the infrastructure spend.

      While of course the Annual Demographia Surveys ( ) rate housing affordability, in essence they are measuring the functionality of local government.

      Poor housing performance and poor transport performance seem to go hand in hand.

      Hugh Pavletich

      • Amen….!

        I have been starting to notice, and evidently Wendell Cox has too, that these planned growth constrained cities have the worst traffic congestion as well as the least affordable housing.

        Which makes a mockery of the whole point of constraining urban growth in the first place.

  2. Wow, a problem here: “Melbourne is the exception, but is still “punching well above its weight,” with worse traffic congestion than larger Chicago, Dallas-Fort Worth, Houston, Toronto, Philadelphia, Miami, Atlanta, Washington, Riverside-San Bernardino and Boston”

    Boston, to pick one, is nowhere NEAR bigger than Melb. They’d have to be referring to the Greater Boston Area which is about the same pop as Melb, but 50% bigger in area.

    Likewise, Chicago, Toronto etc etc – highly doubt these CITIES have pops of 4M+.

    • “Boston, to pick one, is nowhere NEAR bigger than Melb. They’d have to be referring to the Greater Boston Area which is about the same pop as Melb, but 50% bigger in area.”

      Demographia measures the continuous urban area, not the city limits. This is the correct approach.

      Boston’s metropolitan statistical area is 4.6 million people, whereas its urban area is 4.2 million. Greater Melbourne is about 4.2 million, which is on par with Boston.

      Chicago (9.5 million metro) and Greater Toronto (6 million) are bigger than Melbourne, population wise.

      • Yes, what this means though (supporting your argument) is that urban areas which are spread out tend to have less congestion. This is pretty obvious.

        But the way it is explained by TomTom would lead one to conclude that Melb is REALLY BAD because it has worse congestion than much smaller CITIES. This would not be a correct conclusion IMO.

        For example, I think Boston CITY has a population of less than 1M.

        Generally speaking, the US is far more spread out than Aus. Aus would be far better-served to follow that model than to attempt to cram more poeple into less area.

        But Melb generally, being quite spread out, is not too bad and should not be compared to Urban Areas which have been designed/evolved to be far more spread out. The idea that Melb traffic is even close to as bad as London, for example, is a sick joke.

      • To add to this, the Boston MSA includes Providence, RI which is 50 miles away. Should we therefore include Geelong and other similar places in Melbourne?

        • Greater Melbourne includes towns outside the boundary, according to the ABS:

          “As GCCSAs are designed to represent a socio-economic definition of each of the eight State and Territory capital cities, this means the greater capital city boundary includes people who regularly socialise, shop or work within the city, but live in the small towns and rural areas surrounding the city. It does not define the built up edge of the city.”

          As regards to Boston, its Urban Area is smaller than the MSA at 4.2 million, so it is comparable to Melbourne.

      • Not convinced, I feel the spread of population is far different US to Aus. However, this supports your argument.

        I just feel the TomTom analysis is open to significant misinterpretation.

      • Davel: Round the London proper – especially west end/ embankment! – of course but it took 1 hour to get from St Kilda Rd near Domain to Flinders street!! That is at least as bad as London traffic man.

        Edit: Trams don’t help in the inner ring

      • If you simply consider the average speed at peak time for the total trip from airport to CBD, I think you will find Aussie and NZ cities are actually as bad as Mumbai, Mexico City and Sao Paulo, as well as London. I recall some time statistics in Ed Glaeser’s book, that he expressed horror over, but of course he is an American comparing these developing megalopolises with US cities. My reaction at the time of reading it was, “but that is not a lot different to the speeds we do in Australasian cities in similar circumstances”…..

      • Migtronix, obviously there are issues. But lots of people regularly commute from my neck of the woods in SE Melb to the CBD, say 12-15kms, every day in the rush hour in less than 30 mins. In London, I lived in Putney and to get to the office (Blackfriars, edge of the City) in taxi would usually take 1.5 hours, even if you left at 6.30am. And it was only 10ks or so, and that’s not even a particularly bad route. I also played football all over London and we used an assumption of 10mphs when planning when to leave (admittedly this was midday Sat). Apparently this is the average speed of traffic in London, at ANY TIME.

        Melb comes nowhere near that. Sydney might.

    • Yes to two of those, add Lagos and Nairobi to the nasties list 🙂 We should aim to build up our cities like the Asian model, that will fix congestio for sure 🙂

    • Sao Paulo!!! Oh my f god!! London doesn’t even come close, seriously! Lived in Bernado do Campo for 4 month, lucky Sao Paolo has good weather for cycling 🙂

    • Lived in Bernado do Campo for 4 months, thank god Sao Paulo has good weather for cycling it was insane 🙂

      Lisbon is pretty bad too, narrow streets and trams

    • Exactly….!

      Growth containment = unaffordable housing and worse traffic congestion.

      I also suspect it = greater local fiscal strain. I suspect all the shallow assumptions on which growth containment policies are based, are flat wrong.

  3. 30%? Gee that has averaged out nicely, more like 130% for western Melbourne.

    After 6pm to the city I can be door to door from home (Truganina) in 25mins.

    8am to the city you would be looking at well over an hour easily. It would take 25 mins just to get on the freeway!

    • You obviously didnt try and get to the CDB from Point Cook, the city with one road in and one road out and woe betide you if there is an accident. My wife took me to Laverton railway station on several occasion and it must have taken around 40 minutes each time.

  4. It is well worth following the link to the original Wendell Cox article and reading the whole thing.

    Not only does constraining urban growth CAUSE increased traffic congestion, diverting road spending to public transport ALSO CAUSES increased traffic congestion.

    Each increase in urban density adds far more car drivers to the given area than it induces mode shift to public transport. The resulting increase in traffic congestion is always of greater cost than the benefit of a small increase in public transport mode share. Congestion delay tends to increase exponentially because the change from free-flow to “stop-start” conditions represents a drastic reduction in road throughput even as more cars are trying to use it.

    Linear reductions in car use and a majority mode share to public transport only tends to occur at population densities 20 times what Australian cities are.

    Another inconvenient reality: in the long term, every person km travelled on roads comes at a public subsidy cost of around 1 to 2 cents. There is hardly a public transport system in the world that would come near this. The best patronised routes in Australia would cost ten times this – even Manhattan’s subway subsidy cost is around 5 to 6 cents per person km travelled.

    So every dollar of public spending diverted enables approximately one tenth of the actual travel, which is certainly not a formula for local prosperity.

    Roads, once built, cost almost nothing and it is almost all “payback” on the capital investment from then on. But public transport’s operating costs and frequent capital replacement costs are ongoing.

  5. We can still make it to #1. Let’s just keep increasing population without improving public transport.

    Also let the countries above us come and work in Australia without the need of a working visa… oh hang on.

    • You’re missing the point. Public transport, and urban growth containment to attempt to force more use of public transport, is a surefire route to failure.

      In contrast, just letting urban areas spread and providing a proper road network, actually works. The USA has some incredibly fast-growing cities – like 20% plus per decade – with less traffic congestion than Aussie cities – and in every case they just “sprawl” and build roads. There is nothing wrong with their local fiscal situation either.

      It is taking academia an awfully long time to work out what is intuitively self-evident to quite a few ordinary people – such as the voters in US cities that mistrust the smart-growthers and train-set promoters. They are well and truly having the last laugh now.

      • I think you are missing my sarcasm. It is quite obvious the concept of a CBD along with increasing population and an outdated public transport system are the cause of both the traffic chaos and over priced properties in an around the CBD.

  6. For anyone interested, here’s an interesting extract from a book written about the cities that we live in. The extract makes for a good follow-up read to UE’s article:

    “There is a clear connection between social deficit and the shape of cities. A Swedish study found that people who endure more than a 45-minute commute were 40% more likely to divorce. People who live in monofunctional, car‑dependent neighbourhoods outside urban centres are much less trusting of other people than people who live in walkable neighbourhoods where housing is mixed with shops, services and places to work.

    A couple of University of Zurich economists, Bruno Frey and Alois Stutzer, compared German commuters’ estimation of the time it took them to get to work with their answers to the standard wellbeing question, “How satisfied are you with your life, all things considered?”

    Their finding was seemingly straightforward: the longer the drive, the less happy people were. Before you dismiss this as numbingly obvious, keep in mind that they were testing not for drive satisfaction, but for life satisfaction. People were choosing commutes that made their entire lives worse. Stutzer and Frey found that a person with a one-hour commute has to earn 40% more money to be as satisfied with life as someone who walks to the office. On the other hand, for a single person, exchanging a long commute for a short walk to work has the same effect on happiness as finding a new love.”

    • “…..People were choosing commutes that made their entire lives worse…..”

      This is missing the point. The number one reason is not bad choices, it is lack of choice due to the prices of the options.

      Cities with dispersed employment (the norm these days) AND a low house price median multiple, give everyone the maximum choice of locating close to whatever job they might get, and in the kind of house they prefer.

      On the subject of the social cohesion of cities relating to density, the following 2 papers apparently show that “social sustainability” is NOT higher in denser communities (i.e. the smart growth advocates are wrong about happy neighbours and community spirit and conviviality and all that).

      “Social sustainability and urban form: evidence from five British cities”
      Glen Bramley, Nicola Dempsey, Sinead Power, Caroline Brown, David Watkins

      “Urban form and social sustainability: the role of density and housing type”
      Glen Bramley, Sinéad Power

      Then there is the following two goodies:

      “Social Interaction and Urban Sprawl”
      Jan K. Brueckner and Ann G. Largey (2006)

      Various authors, most notably Putnam (2000), have argued that low-density living reduces social interaction, and this argument has been used to buttress criticisms of urban sprawl……
      …….The key element in this argument is a positive link between social interaction and neighbourhood density, and our paper tests empirically for such a link. The results are unfavorable: whether the focus is friendship-oriented social interaction or measures of group involvement, the empirical results show a negative, rather than positive, effect of density on interaction.
      The paper’s findings therefore imply that social-interaction effects cannot be credibly included in the panoply of criticisms directed toward urban sprawl. In fact, the results suggest an opposite line of argument. With a negative effect of density on interaction, individual space consumption would tend to be too low rather than too high, tending to make cities inefficiently compact, as explained in section 2. Thus, the empirical results suggest that social-interaction effects may counteract, rather than exacerbate, the well-recognized forces (such as unpriced traffic congestion) that cause cities to overexpand.”

      Joshua D. Gottlieb and Edward L. Glaeser. “Urban Resurgence and the Consumer City” (2006)

      From Section IV: Cities, Sprawl and Social Capital:
      “……..while density is correlated with consumer amenities, we show that it is not correlated with social capital and that there is no evidence that sprawl has hurt civic engagement……
      “…….The problem with Putnam’s logic (“Bowling Alone”, 2000) is that sprawl is not responsible for longer commutes and that, in fact, commutes are actually shorter in low-density metropolitan areas. Lower density, sprawling areas are not associated with longer commutes, rather they are associated with shorter commutes. As such, it shouldn’t surprise us that these places have more social capital. The longer commutes in dense places discussed in section II should deter social engagement. While high-density living is certainly a plus for many forms of social connection, civic associations do not seem to thrive in higher density areas. Within metropolitan areas, there is no connection between central city residence and most forms of social capital. Across metropolitan areas, density is associated with less, not more social capital, perhaps in part because density is associated with longer commutes. Sprawl may have negative consequences along other dimensions, but it cannot be credited with killing social capital…..”

      Robert Fishman, in “Megalopolis Unbound” argues (correctly in my opinion) that modern social networks are “a la carte” and based around the motor vehicle and communications. Out of every household, there will be members networking with other sports players, with other musicians, with other fishermen, with fellow professionals, with workmates, with church members, with chess players, with schoolmates, with other parents of same-age children, with debating societies, with Kiwanis-type organisations, with political party members, etc etc etc. The “a la carte” urbanism of automobile dependent society is far richer than its critics looking on from afar can discern.

      High urban density is marked just as much by inter-neighbour conflict as by positive interaction. Well separated private properties enable every household to pursue its own interests without giving its neighbours cause for objection. Patrick Troy has some choice comments about all this in his book “The Perils of Urban Consolidation”.

  7. There are high density cities with excellent public transport where congestion is not too bad. Try Hong Kong or Singapore and some European cities. The problem is that Australia never really commits to setting up quality public transport, because it would be expensive to do. So we end up with networks that don’t link up, run late and are pretty much unusable. There’s a reason people don’t use public transport much in Australia.

    We’ve also spent the better part of a century convincing people that a large house in the suburbs and a big car is an indication of a good life. Tho expectations will take a long time to change. In the meantime, everyone is going to keep trying to drive their SUV from their McMansion to work while ranting about the money being wasted on bus lanes and cycle ways.

    • As I said above:

      Each increase in urban density adds far more car drivers to the given area than it induces mode shift to public transport. The resulting increase in traffic congestion is always of greater cost than the benefit of a small increase in public transport mode share.

      Congestion delay tends to increase exponentially because the change from free-flow to “stop-start” conditions represents a drastic reduction in road throughput even as more cars are trying to use it. Linear reductions in car use and a majority mode share to public transport only tends to occur at population densities 20 times what Australian cities are.

      This is why HK and Singapore “work” at all, but most Australians would not like to live in the “housing” there, or at the prices involved.

      There is no European city with public transport that is as fiscally sustainable as HK and Singapore. My criticism above, of the subsidy cost per person km, applies to European cities only slightly less than it applies to Australian ones. Here are the conclusions of a 2008 European Parliament Conference on Transport Policy entitled “Liberating or Strangling Europe’s Potential” – MEP’s Ari Vatanen and Malcolm Harbour being the authors.

      “90% of travel in the EU is by car”. “Transport modes are not simply
      interchangeable”. “Public Transport operates effectively within
      specific niches”. “In the great majority of cases, travel by road
      cannot be made any other way”. “The smooth running of modern economies
      relies on road transport. Cars play a large role in economic productivity and the enlargement of markets”. “The high costs of public transport subsidies weighs heavily on Europe’s economy”. “The “external costs” (air pollution, etc) of vehicle use is covered many times over by the net taxation revenues specifically levied on road users”. “Since 1985, emissions levels of each new vehicle coming to market have been reduced by a factor of at least 10, and even though traffic volumes have increased, air quality in Europe’s cities is improving spectacularly”. “Investments in Rail would take 10,000 years to recoup in terms of reduced CO2 emissions”.

    • “….We’ve also spent the better part of a century convincing people that a large house in the suburbs and a big car is an indication of a good life….”

      HUH? People don’t choose this for themselves because they genuinely prefer it?

      Is there any country in the world where “development” and people rising out of poverty into the middle class, has not involved those people seeking more living space? According to Shlomo Angel et al in “Making Room for a Planet Full of Cities”, the path of urban evolution is fairly consistent. Wendell Cox’s ongoing series of articles on “The Evolving Urban Form” shows this happening pretty much everywhere in developing economies.

      Europe’s cities are usually misjudged by people who visit as tourists and as business visitors to the CBD. A look on Google Earth will reveal “automobile dependent” suburbia differing only slightly in scale from our own. In fact every city in France outside Paris, is less dense than Los Angeles.

      High urban density continuing to exist in the present day in the developed world, is dependent on two factors. Firstly, the city has to predate the automobile. Secondly, the city has to have evolved economically in such a way that its primary source of income is high-density, high-income employment sectors like banking, finance, bureaucracy, law, accounting, advertising, media etc.

      This has to be only a minority of cities. These employment sectors are wealth-consuming and wealth-transferring ones, and a sustainable modern economy needs its Houstons and Silicon Valleys and Wolfsburgs and Ruhr Valleys to create the systemic wealth in the first place.

      • Europe has a lot of urban sprawl, to be use, but public transport is much better even in outer areas. Australian public transport is poor in comparison.

        People’s housing preferences aren’t innate. Residential construction and the automobile industry have focused their marketing on selling the suburban dream. I can’ t really think of many ads about the freedom of the bus or the individual expression that comes with taking the train, but I’ve seen ads like that for cars. Ads for dream homes usually show a house in the suburbs.

        High density cities are often pre-automobile, while a lot of low density places have been built since the personal car became available.

    • Hk and Singapore dont have bad congestion? Never heard that one before.

      High density = shit commuting experience, thats just how it is. Name a Euro city with over 2M people that is good in this respect.

  8. Thanks Phil. In the end, my point is that HK and Singapore work, but we both seem to agree that most Australians wouldn’t want to live in that sort of housing. It is also abundantly clear that Australia could provide far better public transport and that there are situations where this would be beneficial, like in the inner city, which will almost always be high density. Public transport could be a viable and efficient option for people within 15km of the CBD. But cars will almost always work for longer trips, unless you are flitting from inner City Sydney to inner city Melbourne.

    Statistics on the cost/person/km are fine as a starting point, but actual decisions need to be made based on more specific factors. In Sydney, it could be a case of upgrading train lines v the cost of constructing road tunnels and acquiring land.

    • There is an Aussie Uni Professor by the name of Paul Mees who is both anti-rail and anti-car; he says the public transport solution has to involve more flexibility and lower subsidy costs. This means abandoning “radial” routes in favour of “grid” routes with smaller vehicles.

      The principle he is observing is sound.

      Radial rail based public transport systems really should be paid into more highly by the sectors that benefit from them, as these, both the riders and the property owners, are already among the most well-off in the economy. This is pretty much a principle for public transport everywhere.

      If we want to help provide mobility for poorer people especially, we should completely deregulate and de-licence “public transport” including taxis, and just give poor people “mobility vouchers” to spend with whoever provides them the mobility. If it is their next-door-neighbour giving them a lift, what’s the problem with that?

      The empty seats going everywhere all day in cars, are a massively cost-beneficial solution to the problem of the mobility of “carless” people – who are only a minority anyway. THIS makes interesting reading:

      In the USA at least, car use is only a few percent below average among the lowest quartile, and their cost of automobility is spectacularly below average. This is probably because the average is dragged up by depreciation on new and expensive cars, whereas the poor buy already-depreciated but still reliable used cars. Anthony Downs pointed out in “Still Stuck in Traffic” (2004) that one of the most significant contributors to increased average VMT per capita, is the explosion in automobility among the poor.

      Lower average VMT per capita in a nation is often explainable by lower car use by the poor, and by women, and by the elderly. These groups would not exactly be thrilled to learn that some advocates are saying that giving them this mobility was a mistake.

      • I am both anti-train, anti-bus and anti-car and agree with in deregulating the taxi industry. Likewise, I agree there are other potential causes for long commutes, like stamp duty which discourages people selling their homes and moving closer to their workplace.

        But if it was about economic benefits, then we would constrain new road development and plan cities around cycling and walking. Unlike public transport and motorised transport, cycling and walking provide an economic return, not a drain on the economy.

        Think about how much we spend on purchasing, maintaining, insuring and fuelling our motorised vehicles? Compared to the positive benefits of active transport. It isn’t merely about density (high or low), but simply having an appropriate amount of employment within reasonable distances of housing and policies that make it easy for people to move around if they need to.

        Unfortunately, our cities are spread out so while cycling is viable for a significant minority, walking often isn’t. To make matters worse our cycling infrastructure is often either nonexistent or dangerous and far from international best practises. Our cities are explicitly built around cars, and motorised transport is heavily prioritised over cycling and walking.

        Likewise, given that our public transport is of such poor quality, it is no big surprise that most Australians continue to think that cars are the only way to get around. I suspect Australians feel that way because they aren’t given a real choice.