On smartphones in fare collection

From Kytja Weir of the Washington Examiner comes a brief article on smartphones and fare payment. The main thrust is this: “[i]f the U.S. starts to follow the rest of the world, subway riders could soon be using their mobile phones to pay for their transit trips”. The article cites a report by Juniper Research claiming that “500 million people worldwide will be using their cell phones to buy bus and subway tickets by 2015”. I’m not in a position to evaluate the veracity of that claim, but it may nonetheless be useful to look at the technical background. Just how, exactly, will riders pay their transit fare with their cell phones? What are the technologies involved, and who are the players?

The solution of the future is NFC—near-field communication, a technology I’ve mentioned before. An NFC-enabled cell phone is able to not only read ISO 14443 smart cards, but also emulate them—that is, the phone can be used as though it is a smart card. Rather than storing data in a dedicated chip, like a real contactless smart card, when an NFC-enabled cell phone is used to emulate a card, the card data is stored by an application on the phone. In the simplest form, this application just stores the card data between card uses, but more advanced applications can offer a user interface as well. For transit applications, this might mean that an app for emulating a stored-value card would offer the user options to view the value of the emulated card, or even top it up online.

That said, there’s more to mobile payment than emulating existing contactless smart cards (like Oyster, or SmartLink). An NFC-enabled smartphone can also be used to emulate a contactless payment card—a credit or debit card—and I suspect that this is where we will see the majority of applications. Once agencies deploy open payment hardware and start accepting contactless payment cards, it’ll be trivial for them to accept payment from NFC-enabled phones. Card issuers can offer their cardholders an app for their mobile phone—download the app and log in, then use the phone just like a payment card. Of course, this requires the cooperation of the carrier; I doubt that carriers would let card issuers use their network for mobile payment without getting a cut. Carriers may also offer their own solutions, in which transactions made with the phone are billed to the user’s phone bill (like billing for app purchases on the Ovi Store), rather than the user’s credit or debit card. However, the reality is that both card issuers and mobile networks are going to want a cut of NFC transactions—and that’s the approach being taken by the Isis consortium. Isis was founded by AT&T, T-Mobile, and Verizon; they’ve presently partnered with Discover and Barclaycard, but are looking to add additional issuers. Having rolled out an open payment system in 2009, UTA will be Isis’ first deployment in a transit application. There aren’t that many NFC-enabled devices on the market yet (and it’s looking like the iPhone 5 will not include NFC), but it’s possible that by the time the Isis pilot launches in 2012, market penetration will have improved.

There’s also another approach to integrating NFC, which may interest agencies which have already distributed a large number of ISO 14443 cards, and want to add mobile options. Since an NFC-enabled phone can be used as a card reader, you can offer your users an app which in essence turns their phone into a ticket vending machine in their pocket. Users use their phone to buy fare media, then tap their card to their phone to load the purchased product. This app can offer features similar to those offered by FareBot as well, like the ability to view the card’s journey history and current balance.

As I said before, not many people have NFC-enabled phones right now. But there are still things that can be done with a smartphone that doesn’t support NFC. Many transit smart cards (like Oyster and SmartLink, and soon SmarTrip) offer online top-up. An easy, quick fix for offering a limited level of mobile functionality is to deploy a mobile version of the online top-up interface. However, unlike topping up a card with an NFC-enabled phone (as discussed before), using a mobile site for top-up is constrained by the underlying AFC system. It has been my experience that AFC systems which support autoload at the gate have some lag (which could be as much as a day) between the transaction being submitted and the purchased fare products being available at the gate.

There’s a final option, the option taken by airlines, and now Starbucks: 2D barcodes. Any smartphone can display a 2D barcode, which can be scanned by a rider at the faregate. Cubic, for example, has delivered iPhone apps for ticketing on RMV and KVV, both of Germany. Cubic has also marketed a 2D barcode scanner add-on for their Universal Gate. 2D barcodes provide many of the same benefits as NFC, with the important caveat that the communication is unidirectional, from the phone to the faregate. When using a 2D barcode, there’s no way for the faregate to return data to the phone, as is the case with NFC.

Regardless of the solution implemented—and there are a wide variety of solutions available, some dependent on NFC, some not—there’s a lot that can be done with smartphones in transit fare payment. However, these solutions must be implemented with care. With the current trend towards closed app stores, and everyone in the mobile ecosystem wanting their cut of transactions taking place on phones, riders using smartphone apps for ticketing may face steep usage fees or monthly fees for the privilege of using their phone for payment, or flat-out rejection, if their carrier (or the mobile platform) doesn’t approve of a given app. Consider, for example, the plight of AT&T customers who bought a BlackBerry PlayBook, only to find that AT&T has yet to approve the app needed to connect the PlayBook to a BlackBerry smartphone. Not only has AT&T not yet decided if they will approve the app, they’ve yet to decide what (if anything) they’ll charge users for the privilege of tethering their PlayBook. John C. Dvorak, often hyperbolic, posted a strongly-worded article several months ago warning future iPhone users away from NFC-based payment solutions. Dvorak’s warning is a bit strong, but there is a central point to be heeded: NFC as a technology is great, but we must be careful about how those solutions are implemented by carriers, banks, and transit agencies. In addition, users of smartphones which lack an NFC chip need not be left out in the cold; there’s a lot that can be done for them, too.

Update, April 27, 2011: The linked Washington Examiner article wrongly attributes the report to the now-defunct Jupiter Research; it appears to have actually come from Juniper Research. I’ve corrected this post and will pass word to the Washington Examiner.

Update, April 28, 2011: The Washington Examiner post has been updated to reflect the correction noted above.