Friday, March 21, 2014

OPAL CULT - SYDNEY'S $1BN TRANSIT SMARTCARD


Opal smart card at Circular Quay Sydney
Opal staff hand out fliers at Circular Quay- note the interchange signage in the background (DRC)
“Cheaper fares with Opal!” say the more enthusiastic promotions people, echoing the text of the flyers they hand out at Sydney’s Circular Quay commuter wharves. It is an ironic choice of location. Circular Quay is Sydney’s only tri-mode bus, ferry and rail transit interchange.  Here more than anywhere else, commuters are going about a modal interchange, in many cases to the third mode of their journey. A feeder bus, a ferry and a train is not a rare itinerary.

It is these interchanging passengers who are to endure the greatest fare increases under Opal. It is little wonder that the Opal team have had to ramp-up the propaganda; as commuters come to the realisation that the new smart-card penalises them for their modal interchange, they are less willing to abandon their old zone-based multimodal mag-stripe tickets.

General comparisons between the old multi-modal tickets and Opal are difficult, because of the sheer diversity of multi-modal journeys. It is with this problem in mind that Jani Patokallio developed OpalOrNot.com in his spare time. A complete web-app with the sole purpose of cutting through mind-numbing complexity of calculating the comparative prices of Sydney transit fare options.

OpalOrNot.com screenshot for a typical two mode commute (15/3/2014)
Patokallio observes that for many multi-modal journeys, costs will significantly increase. This is the most recent change in a ten-year history of Sydney's transit pricing increasingly favouring long single-mode journeys, at the expense of short multi-modal journeys.

While the press-releases of governments of both persuasions have been laden with pro-integration rhetoric (and have even restructured and rebranded the transit instrumentalities to appear integrated), fare structures increasingly impede integration.

What Was the Point of Opal?


The story of Opal is one borne out of a political response to the previous Government's abortive attempts to implement T-Card integrated ticketing. The broader narrative around the need for a new "integrated ticketing" smartcard system has its origin in a 1996 major Inquiry by the newly-formed Independent Pricing and Regulatory Tribunal (IPART). IPART observed that integrated ticketing could improve the flexibility, convenience and accessibility of transit  [5] and that a smartcard system could enable the delivery of integrated ticketing.

The then Government's response was included in its 1999 'Action for Transport' plan, a central pillar of which was to develop an integrated ticketing system, expanding a combined intermodal fare structure beyond the then existing ones in inner-Sydney to the privately operated areas in the outer West and Illawarra. Between 1999 and 2010, consecutive Labor Governments grappled with implementing the technology to deliver integrated ticketing, culminating in the collapse of the T-Card integrated ticketing project, and the near bankruptcy of Australian technology company ERG. In April 2010, after the failure of T-Card, the then Labor Government awarded the contract for a new smartcard system to the Pearl Consortium.

Amidst the mire of T-Card's challenges, the original intention of improving the "flexibility, convenience and accessibility" of transit was somehow lost. As early as the year 2000, Australian transport researcher Paul Mees was sceptical of the New South Wales Government's characterisation of its integrated ticketing project:

“In Sydney there is still no discussion of a genuinely multi-modal fare system. Instead what is proposed is an electronic ‘smart-card’ which makes it easier for passengers to pay a separate fare for each mode used…” [4]

In 2003 a Ministerial Inquiry into Sydney's transport, led by the chairman of IPART responsible for Sydney's transit fares,  redefined Integrated Ticketing as "The use of a single stored value card to purchase travel (referred to as the ‘smart card’)" [6]. This belief that "integrated ticketing" was nothing more that a unified contactless charge card spread throughout the transport administration, and the entire T-Card project was effectively redefined as one with the purpose of producing an electronic purse for transit, rather than an integrated multi-modal pricing regime.

When the new Liberal State Government was swept to power in 2011, it inherited the smartcard contract awarded by the previous government. The new government nonetheless had a very clear mandate and motivation to get a system in place that lived-up to the ticketing systems that Sydneysiders often see interstate and overseas (with the thought "why can't we do that in Sydney?). In September 2011, the press release came unveiling the new ticketing project:

"MEDIA RELEASE
OPAL A GEM FOR PUBLIC TRANSPORT
London has the Oyster, Hong Kong has the Octopus, and Sydney will have the Opal.
Transport Minister Gladys Berejiklian today announced the Australian national
gemstone would find its way into the pocket of public transport customers in the form of the new electronic ticket."

As with the wayward T-Card after about 2003, the Opal project sought to imitate international examples like London's Oyster. But with the original purpose of integrated multi-modal fares lost amidst its highly political rebirth, it has struggled to get beyond replicating superficial form.


Superficial Form: Solutionism and Cargo Cults


In his recent book “To Save Everything Click Here”, Evgeny Morozov wrote on the problems of what he calls “solutionism” and “internet centrism”. An aspect of his argument is that the internet has formed a reality distortion field, whereby things that the technology of the internet can enable are assumed to be inherently desirable.

“Ironically, in addition to sanctioning the hideous ideology of ‘solutionism’, with its never ending quest for solutions to often non-existent problems, Internet-centrism is beginning to block our ability to think of effective technological solutions to problems that do exist. Solutions are not addressed based on their merits but rather on how well they sit with the idea of [The Internet]” [1]

It struck me while reading Morozov that the phenomenon he described was applicable to other technologies with a political dimension: that any technological centrism can block our ability to think of effective solutions to actual problems. It also reminded me of a related concept described by Nobel-winning physicist Richard Feynman in his 1974 Caltech Address.

“In the South Seas there is a cargo cult of people. During the war they saw airplanes with lots of good materials, and they want the same thing to happen now. So they've arranged to make things like runways, to put fires along the sides of the runways, to make a wooden hut for a man to sit in, with two wooden pieces on his head to headphones and bars of bamboo sticking out like antennas--he's the controller--and they wait for the airplanes to land. They're doing everything right. The form is perfect. It looks exactly the way it looked before. But it doesn't work. No airplanes land. So I call these things cargo cult science, because they follow all the apparent precepts and forms of scientific investigation, but they're missing something essential, because the planes don't land.” [2]

All aboard the Opal plane: A Melanesian "cargo cult" with their aeroplane, but alas, no cargo
Both of these concepts describe the outcome of a mental shortcut: the recognition of, and focus on, a superficial form. In Morzov’s description of “Internet Centrism” the logical error is that because something has a superficial form that aligns with the superficial form of other good things enabled by the same technology, it too must be good. In Feynman’s description of “Cargo Cult Science” the logical error is the belief that replicating the superficial form of complex system produces the same outcome as the complex system.

A focus on technology and infrastructure (and its appearance to the electorate) rather than an operational outcome is not an unusual problem in transit. Paul Mees touched on the dynamic in the field of transit in his thesis “A Very Public Solution”, observing that, in order to improve transit:

“Many of the preoccupations and conventional wisdoms of traditional public transport planning will require reconsideration or rejection if real change is to take place. Infrastructure and technology fetishism will have to be abandoned, although getting technological decisions correct remains important.” [3]

Sydney’s new Opal card has fallen victim to phenomena described by Morozov and Feynman. The outcome is over $1bn being spent on technology that does not solve a problem, but rather replicates the superficial form of successful revenue collection systems like London's Oyster. Opal has the contactless smartcards, the backend data warehouse, all the visible physical form of systems like Oyster, yet is missing something essential- the fundamental mechanics of an integrated fare system.

Integrated Fare Systems


The two dominating considerations in commuter travel are how long it takes, and how much it costs. These are the major elements of what transport planners call Generalised Cost, the value which travellers compare across the various travel options they can use.

These are linked, by virtue of the different attributes of different modes of transportation. A short, low patronage, local trip is best served by a car or bus. If you want to reliably and quickly move 30,000 people per hour in a single lane over medium to long distance you'll need a train. If you want to go over water without a bridge, you'll need a ferry. And the fastest journey for most people would involve a combination of these. But if you apply a cost to each of these component trips independently, you very quickly destroy the Generalised Cost value proposition of the door-to-door journey.

For this reason defining pricing of urban transport around accessibility, rather than around individual trips on individual modes, is necessary to provide the equitable Generalised Cost and price signalling supporting the free interchange of passengers between modes of different scale. Integrated ticketing aggregates the network access or flag fall component of cost across a product called mobility rather than a product called a ride on a bus.

Integrated ticketing is of particular importance for transportation modes that do not work very well in isolation. Ferries are the supreme example, being captive to water, there are very few journeys they can serve door-to-door. Being blessed but divided by a wide harbour, Sydney has few harbour transit crossings other than ferries; only two significant rail crossings (the North Shore line and the Main North), compared to say London's 23 rail crossings over or under the Thames.

Accordingly there has been in Sydney a long history, though a piecemeal and declining one, of integration. From the 1930s, there was collaboration between different operators, both private and government, to provide multi-modal services. Often, as with the Balmain to City commute, a faster journey and better use of capital (keeping trams out of slow-moving City congestion) could be achieved by the Government Tramways collaborating with a private ferry operator, offering combined integrated tickets at the same price as single-mode (tram only) tickets.

Sydney multi-modal tram and ferry ticket from 1945
A Multi-modal tram and ferry fare from the late 1940s (DRC)


Many cities have now embraced generalised system-wide integrated ticketing based on time and zones, including London's Oyster, with its underlying zone TravelCard system. (I explained the key aspects of London's time and zone system which underpin Oyster in this post on Opal and Oyster).

There was a long (30 year) gestation for the concept of integrated zone fares in Brisbane leading up to 2003, which concluded in ultimately successful lobbying by groups such as The Brisbane Institute and Queensland Conservation Council. Transport planner Eric Manners was a key author behind community-interest lobbying such as this one from SmogBusters. In it Manners identifies a succinct recommendation from consultants and transit revenue experts Wilbur Smith & Associates which is as relevant to Sydney today as when published in Brisbane in 1970:
"The efficient and economical operation of public transport requires full coordination of services, the elimination of wasteful duplication and the use of each type of transport vehicle and system to meet the particular needs it can best serve"
US Transit planner Jarrett Walker described a similar challenge in specific relation to Sydney in 2010:

"My basic view of connection penalties has always been this:  The connection is not an added convenience for which the customer should pay extra.  It's an inconvenience required by the geometry of any efficient, legible, and frequent transit system.  Transit systems must therefore do what they can to minimize this inconvenience and encourage people to make connections.  Connection fare penalties are exactly contrary to this imperative."

Few have explained the importance and rationale of integrated transit pricing as clearly as Dr F Pampel, Director of the Hamburg Transport Authority (HVV). The HVV was the pioneering example of an authority overseeing an integrated approach to multi-modal operations across more than ten operators, most of which were private enterprises. They pioneered multi-modal integration, inspiring similar celebrated transit systems like Zurich's ZVV. In 1967 Pampel presented a paper to the Union Internationale des Transports Publics (UITP):

"The essential prerequisite for the desired integration of transport in the Hamburg area was a passenger fare system with uniform rules of price calculation and tariff provisions for all the services in the Community... 
When the fare system was worked out, great importance was attached to identical prices for broken and continuous journeys of the same length in view of the large number of broken connections resulting from the transport network structure with the rapid transit systems as the main carriers and the buses as feeders and distributors. The joint fare system allows passengers to change vehicles as required. All fare barriers existing between different operating branches and undertakings have been eliminated. All services in the Community are offered to the passenger under a unified fare system." [7]


In some cities the principles set out by Pampel have only been achieved with arrival of newer generation technology. It was indeed the Oyster Prestige smartcard in London that enabled an electronic purse with an upper limiting zone fare defined by the TravelCard zone system. It was the GoCard smartcard that enabled the implementation of Brisbane's Translink zone system. It was the Clipper Card that enabled San Francisco's MUNI to role together time-based multi-modal fares with the Bay Area Rapid Transit's fare system (previously a confusing paper-based transfer system).

It is perhaps for this reason that there is a Post hoc ergo propter hoc logical error contributing to the cargo cult mentality; that the smartcard itself has become the object of desire, rather than the fundamental and unfortunately prosaic pricing structure and revenue-division framework at its foundation. As understandable as it is, it is an error that needs be corrected if Sydney's transit system is ever to become an integrated one.

Actions and Words

In the introduction I included a screenshot from OpalOrNot.com, showing a fairly typical combined ferry- rail journey increasing $827 or 36% from $2293 to $3120 for a year of travel. This in itself is a damning contradiction of the "Opal Fares Cheaper" press releases.

Ministerial Press Release "Opal Fares Cheaper" 12/12/2013


But if we look over a longer period we see an even greater discrepancy between single-mode and multi-modal journeys, particularly those involving a ferry.

Above: Per Cent increase of Sydney transit weekly fares, single mode and multi mode over ten years to 2014, without Opal and with Opal (full backing sheet, inputs and sources are available in this spreadsheet)

There are two important things revealed in the above graph:
  • Long single-mode journeys have increased at (or at less than) inflation for the last ten years, ferry fares have increased about 60% (double CPI) and multi-modal fare price has more than doubled in ten years, with a 107% price increase (more than triple CPI)
  • Single mode journeys enjoy a moderate price decrease under Opal, multi-modal fares suffer a significant price increase
It is also interesting to note that ferry fare have massively outstripped rail fare increases, despite ferries consistently maintaining a higher cost recovery [8]. This graph is not comprehensive. There are some cases that are better, and some far worse. But it is illustrative of a fundamental disconnect between what policy allegedly aspires to, and what is actually happening. The bureaucratic response is to downplay the number of people affected, according to Sydney Morning Herald journalist Jake Saulwick:

"Transport for NSW says that its analysis shows that just 7 per cent of people would pay more under the Opal, with another 7 per cent potentially paying more. Those paying more would typically be people who buy monthly, quarterly or yearly public transport tickets.
But the department has consistently refused to release the full analysis upon which these figures are based."

Whether it just effects 7% of people now is hardly the point. Given the massive price differential that has established between mutli-mode and single-mode tickets over the last 20 years, it is no surprise that multi-modal fares have declined from over a quarter of total passenger revenue in the late 1980s [9]  to "7% of people" today. That does not tell us how many people are not using transit, or are using a fare-minimised route rather than time-minimised or "optimal use of infrastructure" combination of buses, trains, ferries and trams.

Not only does $1bn worth of smartcard technology not support multi-modal pricing, it makes multi-modal journeys considerably more expensive than single-mode journeys. In this central functional respect, Opal is the polar opposite of integrated zone-based ticketing systems like London's Opal, and even Brisbane's GoCard and Perth's Transperth tickets.


The outcome


No doubt the focus will stay on the expensive technology, because like the fires on the runway, the superficial appearance is what gets noticed by the casual observer. There will be self-congratulatory back slapping and perhaps even more "smart infrastructure" awards acclaiming the success of the rapid Opal smartcard rollout. And from a hardware rollout point of view, it has been a remarkable turnaround. The success of this technology project- of quickly and successfully setting the fires along the runway, and donning the coconut headphones in the bamboo control tower, is immense.  But Opal delivers none of the central benefit of integrated ticketing systems after which it is superficially modelled, at a greater cost.

In fairness, Opal can be credited with two achievements: replacing life expired-technology (like on-board bus magnetic stripe systems) and improving revenue tracking to non-gated rail stations. Whether the two practical outcomes warrant over $1bn expenditure on smartcard technology is highly doubtful, particularly as smart-card systems are now being superseded by low-capital mobile technology such as that recently rolled out on Boston's MBTA.

Price signals blocking multi-modal transit are subtle and sinister in their effects. By way of just one example: commuters stop catching the bus to the train station, they drive instead. The car park fills up, the road becomes congested. Patronage of the feeder bus drops. Bus revenue falls. The operator reduces service levels. The bus becomes less attractive to the few remaining people using it because of its infrequency. The Municipal Council is motivated into building a multi-million dollar car park or a bigger car park.

As Transport for New South Wales pushes for more multi-modal infrastructure, such as Sydney's currently under-construction Eastern Suburbs Light Rail, more modal interchange will be necessitated, and cracks in the "one journey on one mode policy" will propagate. Whether there will be the economic impetus (of equitable cost-recovery) or political impetus (of disgruntled commuters), it is conceivable that these kinds of ad-hoc kludges will converge to a geographic zone-based cap.

If the current course of Opal is maintained (replacing multi-modal periodicals altogether with capped trip-based fares), ferries will be particularly hit. Cross-elasticity of demand will draw passengers to cheaper more time-consuming capital and labour-wasting options, such as sitting in buses jammed water-side roads like Spit Road and New South Head Road. And the empty seats will continue back-and-forth on the harbour irrespective.

There is still an opportunity for recovering value from Opal's costly infrastructure. A zone-based system (like Sydney's former TravelPass) could be restored as the capping mechanism for Opal, as TravelCard is for Oyster.  But it requires both leadership and understanding of the issue at hand, of which there has been little evidence to date.



The original version of this post incorrectly indicated that the Opal project was awarded by the current New South Wales Government. In fact the Pearl Consortium which delivered what was subsequently named Opal was awarded the contract by the previous Government. This has been corrected.

[1] p75, To Save Everything Click Here, Morozov E, 2013, Allen Lane/ Pearson, London
[2] Cargo Cult Science, Richard Feynman, 1974 address to Caltech
[3] p289, A Very Public Solution, Mees P, 2000, Melbourne University Press, Melbourne
[4] p288, A Very Public Solution, Mees P, 2000, Melbourne University Press, Melbourne
[5] IPART, March 1996, An Inquiry into Pricing of Public Passenger Transport Services, Fare Structures for Public Transport, Transport Interim Report No 4, p27
[6] p107, Ministerial Inquiry into Sustainable Transport in New South Wales, 2003
[7] p11, The Hamburg Transport Community, an example of coordination and integration in public transport, Dr.-Ing. F. Pampel, Presented before the UITP, 1969
[8] p38, NSW Auditor-General's Report to Parliament- Volume Eight 2013- Transport Overview, 2013, Sydney
[9] p12, Urban Transit Authority Annual Report 1987-1988

10 comments :

  1. The contract with the Perl consortium that is delivering Opal was signed by the previous administration in it's dying days, not the current administration. The current administration named it and have managed to mostly keep their hands off and let the contractors get on with the job, but it's certainly NOT their initiative.

    The 'simple' fix to the Opal multi-modal issue would be to follow the Singapore model of applying a 'transfer rebate' to passengers who change modes. Each mode can have it's own distance based fare scale, but the passenger making an intermodal journey isn't slugged with multiple 'flag falls' like in Sydney.

    ReplyDelete
    Replies
    1. Hi Matthew, thanks that it is indeed correct that the Pearl Consortium was awarded the contract by the previous Government. I have updated the post accordingly. However, it was clear that the current Government had a mandate to fix the errors of the preceding one. And you can see from the press releases I quote that the new government took complete ownership of the project, and never identified the contract as an impediment to integrated transport. Three years have passed, which is ample time to have developed a new fare system- or at least apply the existing one including multi-modal periodicals.

      In relation to your second point, I don't know Singapore's model, but it sounds like what you describe could be an option. I would be reserved about describing anything as a simple fix- there are pros and cons to each of the solutions. The examples I use are based on my own experience (both as a user and in a professional sense), and time/ zone based systems have been well proved and validated in those examples. There could be other options.

      Delete
  2. Why not mention the Melbourne multi-modal ticketing system? The basic system long predates smartcards, and even magnetic stripe tickets. A limited multimodal system started in 1981, with a system based on neighborhoods dating from 1983. A three zone system was introduced in 1989 - very similar to today's ticketing system though using punched, scratched, or stamped paper tickets. Magnetic stripe technology ousted this paper based system around 1997/8, and we've just lived through the transition to a smart card system. However, it is important to note that the last two didn't change the basic multi-modal system.

    ReplyDelete
    Replies
    1. Hi Andrew, there are plenty in the world I could have mentioned. I focussed on London- because that's what Opal is often compared to. I focusses on Hamburg- because it was the first example of generalised multi-modal fares. Brisbane and Perth were both late but successful adopters of multimodal fares. For that matter, Sydney had a 9 zone system established in 1984, entirely on (non-mag) cardboard. You'll find in that paper of Eric Manners I cite he refers to Melbourne extensively as an example for Brisbane.

      Delete
  3. This comment has been removed by the author.

    ReplyDelete
  4. I'm not in favour of 'zone' based systems, they tend to over price short trips, under price long ones and cause all sorts of messiness on zone boundaries. Zone systems existed originally only to reduce the number of ticket types needed. It was a solution to the technology of the time, printed ticket stock. You didn't want too many variations.

    We have the technology now to charge by distance and each trip start and end being tracked.


    The system as it was when I was last in SG had different distance charges for Metro and buses, and not only that different MRT lines had different charges!. Buses are different depending on if they were airconditioned or not!
    This web site appears to list their tariffs
    http://www.ptc.gov.sg/FactsAndFigures/fares_new.htm
    It appears the aircon vs non aircon bus fare difference is disappearing in about 2 weeks!.


    But the system is relatively simple - you get charged a distance fare on each mode. To remove the mode change penalty, when you make a transfer, the system gives you a 'transfer rebate' that nullifies the 'flag fall' the tap on to the new vehicle charged you.

    Buses are generally cheaper than the trains, but not by much, the time saving makes it worth changing to a train despite it being a little more expensive. But because you don't get slugged for a 2nd 'flag fall' like Sydney, it's certainly worth paying a little more for the faster journey.

    ReplyDelete
    Replies
    1. Matthew, I can assure that zone fares needn't under price short journeys, that is a matter of design choice. Thanks for the link- an interesting case study. All I can say is you may be right, but in most of the world's best multi-modal systems, geographic zones and time are the two dimensions of charging.
      There are complexities relating to linked and chained trips, as in how you define the end of one journey and the start of the next. You can come up with obscure rules (like Sydney) or you can just say: area for A for time B. But hey, there are many potentially better solutions when your pricing structure is as bad as Sydney's

      Delete
  5. This is an excellent post, but I think it is sad that it needs to be written at all. It is difficult to understand why the concept of an integrated fare structure is not already an accepted part of the transport policy environment in NSW, when it so widely embraced elsewhere. And the issue doesn't stop with fares. There does not appear to be a good understanding here of European approaches to network design (Taktfahrplan), infrastructure planning or an awareness of the patronage growth and improved farebox recovery achieved where these strategies have been adopted.

    ReplyDelete
  6. excellent article, a couple of observations:

    1) Zones are designed for the subsidisation of voting constituents who live far away. It's political pandering to not do $/km but $ vs the approach to postcode 2000. For someone that actually lives *in* Sydney, I think the first way to get a fairer fare is to start reducing some of the subsidies for people that live far away. I can understand for housing reasons, etc, but it makes no sense from a transport point of view. (and maintains the view that Sydney must somehow only sprawl from pc 2000)

    2) So in defense of Opal, what I need to point out is: while Opal doesn't solve the multi-modal problem (essentially a transport, policy problem), it does solve a few other major problems with ticketing. It will make it much quicker to board buses, for starts. At peak hour, an extra 3-5 seconds per passenger is an absolute lifetime in the. Furthermore, if we had the T-card 15 years ago, that's 15 years of a mountain of paper and trees that would have been saved. Also, I get ticket failure fairly regularly - for my 6 months of travel last year, I had 3 ticket failures, all had to be mailed back to RMS.

    Anyway, having come from Singapore and having used Octopus in HK for the past 25 years, I have to say that the Opal is a welcome change. It may not solve all the problems, but it's a first step to solving some of the issues for Sydney.

    ReplyDelete
  7. Thank you for this most informative post. I agree with the points you made with about the T-card, especially this: 'Amidst the mire of T-Card's challenges, the original intention of improving the "flexibility, convenience and accessibility" of transit was somehow lost.' I remember being allocated a T-card back in highschool, and at the time believed it to be a wonderful step forward in terms of public transport within Sydney. In retrospect, I can see all the flaws with the system, and hope that Opal will be learning more from these mistakes. The biggest flaw, like you mentioned, is flaw with commuter transfers. This needs to be fixed. But regardless of that, I don't believe our transport system is as bad as everyone makes it out to me. Behind, in terms of the rest of the world - yes, but not horrible.

    Joy @ Love Your Trains

    ReplyDelete

Note: Only a member of this blog may post a comment.