Getting smart about smart cards

SEPTA is preparing to bring Philadelphia into the 21^st century with new payment technology and leave tokens behind.  During my fellowship I’ve had the opportunity to meet with some of people behind both the successful implementations in Taipei and Adelaide and the struggling implementation in Melbourne and use the system in Hong Kong.  

In Melbourne the implementation of the Myki (myki) system has been arduous with costs exceeding initial estimate by AUS$500 million, though opposition politicans are fond of using a different baseline and citing AUS$1 billion in overruns.  Australia often uses special purpose entities for the implementation of large projects and created a Transport Ticketing Authority (TTA) to implement myki. The TTA has its own executive team and board of directors.  The system is also far behind the initial schedule set for implementation and experienced variety of deployment hiccups. The project has been the subject of numerous inquiries by various government departments (though nothing criminal). It’s my opinion that governance was not necessarily the issue here as special purpose entities have been successful for smart cards and other projects. After siting down with some of those who've witnessed the issues first hand, here are some lessons learned from Melbourne:

Recognize that new payment technologies are complex IT projects - Public transit agencies typical build and project managing infrastructure projects and IT is a different beast. An ombud’s review of the Melbourne implementation sited research done at Harvard Business School which finds IT projects are more likely than other costly projects to not simply have cost overruns, but to have cost explosions, with final costs being several hundred percent more than initial estimates. Any who has lived through the implementation of any sophisticated it systems has probably witnessed this themselves. 

Give the project the time it needs – Hong Kong implemented its system over five years. London implemented the oyster card over nearly 10 years.  No system is the same, every new payment system is essentially custom built, so doing it right takes time. As Secretary Betts noted in recent testimony, “You do not want people implementing new ticketing systems to be leisurely because it burns money as time passes, but the two-year time frame was always going to be unrealistic.”  The time frame that Melbourne set for itself put managers and politicians in a corner in to meet public expectations, which resulted in an implementation before the system was ready and a raft of initial problems.

Recognize that these projects have a higher degree of policy risk than other projects - While it is true that smart card systems can accommodate any kind of fare structure, changes cost money to implement.  Having a clear view upfront is important.

Contracts should not be solely outcome based - Contracts for smart card technologies need to include technical requirements. It's fasionable to talk about outcome based performance measures, but the Melbourne experience reminds us that with too much focus on outcomes it is hard for the either the public sector or the private sector to be able to adequately assess the projects risks and costs.  

If you are interested in getting into the details here are some links to reports:

Audits Committee Parliamentary Inquiry – the testimony of transportation secretary Jim Betts is a good read (if you like reading this sort of thing).

Ombuds Report on IT projects including myki

Auditor General Report on fare evasion

 

Adelaide

Adelaide launched to the general public their smart card in just the last few weeks seemingly on-time and budget.  The good news for Philly is that Adelaide and SEPTA are using the same firm ACS for smart card technology. All the officials I met with had praise for the professionalism of the team. When we discussed factors critical to the success of the implementation to items kept coming up communications and testing. The Adelaide system was tested over 12 months by 10,000 customers. They startedg with a group of employees, then moving to a select group of commuters, before releasing a few thousand cards "into the wild." Minister for Transport Services Chole Fox told me, “we must of launched this six times already, but it’s not just because we politicians like launching things. Testing the system to get it right was important.  The launches were also opportunities to educate the public and get them excited about having their own MetroCard.” Customer up take of smart card technology is critical for systems to experience the benefits of smart cards.  Marketing for the system is everywhere, including a small army of customer service agents promoting the system at transit hubs, libraries and malls to help people get their cards.

Taipei

Taipei’s Easy Card has been in place for more than a decade and shows the potential of smart card technology. Like Melbourne and the MRT system, Easy Card is its own corporation. Founded in the year 2000, the company has a capitalization of about US$25 million. Government shareholders include the Taipei City Government and the Taipei Rapid Transit Corporation own about 40 percent of the company with private corporations including: the Fubon Bank, 12 bus companies in Taipei City and New Taipei City, Mitac Inc., Cathay United Bank, Taishin Bank, China Trust Commercial Bank, Mercuries Data Systems, China Engineering Consultants, Sanmen Technology and other companies owning the remaining 60 percent.

Over the past twelve years the Easy Card Corporation has expanded opportunities to use the card. The EasyCard’s are now accepted at for small purchases at convenience stores; registered cards now function as library cards and for using the Taipei bike share system. The card is also accepted for Taiwan High Speed Rail Trips, Maokong Gondola service, Taipei Zoo admission, bus services and river boats.

During a lunch with former Easy Card Chairman Sean Lien (2009 Eisenhower Fellow) he cited the 2009 strategic alliance with 7-11 and the marketing of cards as a fashion item as key achievements. The strategic alliance with 7-11 brought Easy Card payments into the thousands of stores across Taipei. With each transaction, Easy Card earns revenue.   

Each red dot is a 7-11View Larger Map

Easy Card has issued more than 30 million cards since its launch. The company regularly releases new cards with latest pop culture icons, hot brands and new initiatives on them.  During visits with the DOT, MRT and U Bike (bike share) offices, everyone offered me their custom branded card.

Some key features of successful smart card systems

Widely available – Make the card available for sale at convenient outlets over the long term (e.g. supermarkets, convenience stores, banks, newsstands etc).  In the short term, make them at unique high volume outlets like malls and stadiums.

Easy recharge and auto recharge – Offer an auto recharge option and make sure to offer recharge on line, at stations, on vehicles and at outlets where cards are available for sale.

Lost card value recovery – A great incentive to get a card is that if you register your card you can get back the value if it is lost or stolen. Most cities offer this service if you have registered your card with the agency.

Guaranteed lowest fare – Guaranteeing passengers that they always pay the lowest fare (e.g. once you hit the cost of a weekly ticket your travel is free for the rest of the week) makes operators uncomfortable, but customers shouldn’t pay an "idiot’s tax" as one executive told me. The lowest fare guarentee  encourages more trips when people know that they will always pay the lowest fare.

A tourist and occasional user friendly system – Adelaide is maintaining a single ticket system. Taipei has single use tickets that come in the form of a plastic token. Hong Kong has specially branded tourists smart cards. Making the system easy to use for tourists will be critical in a City like Philadelphia. That means multilingual information and sales available at hotels and tourist sites.

Slow phase out of the old system – It is important to allow time for the new system to stabilize.  There will be problems with the new system.  An agency needs time to get the cards in the hands of customers, while some will say its too difficult of expensive to run two systems doing it for 18 months to two years makes sense.  

 
A final thought
There is the potential that SEPTA may leap over some of these issues, particularly the tourist and occasional user issue, by accepting credit cards at the fare gate. Even with this option, I think Philadelphian’s will prefer to use a SEPTA card (to be named).  With a few millions cards out there, it is interesting to consider the potential of spinning off the fare payment technology divisions of US transit systems to provide them the flexibility to generate additional revenues.  With the agency as the major shareholder the profits would then be poured back into the transit system.