The bumpy road ahead for Compass

LOST IN TRANSIT | Vancouver Mayor Gregor Robertson tries the Compass card system at Richmond-Brighouse

The Compass card, instituted one year ago, was supposed to give TransLink all the information it needed to make intelligent transit planning decisions. It’s not quite there

It’s been just over a year since we all first saw TransLink’s Compass card. Beta-tested back in 2013, the card had its first partial public rollout in August 2015. As with entrepreneurs at the end of a first fiscal year, the anniversary reminds us of lessons learned and the way year two might be bigger and better.

In the case of the Compass card—the reloadable fare card that works now on all transit in Metro Vancouver—that means thinking about how the card might contribute yet further to increased ridership and lower congestion. In other words, it’s not just a transit issue; it’s also a road-use issue.

Compass BearingsNo reader will have missed that the Compass card had a rocky start. There were lags in uploading funds to the cards. There were half-hour waits on the customer support phone lines. There were complaints of overcharging, though some of those were the result of user unfamiliarity. The system relies on passengers “tapping” the cards on a card reader. But if you “tap in” on the SkyTrain and then fail to “tap out” when leaving, the system defaults to a three-zone charge no matter how short your trip may have been.

In fairness, transit smartcards are one of the thorniest system challenges of our day. Toronto’s Presto card was 10 years in development and design. Calgary ended up cancelling the Connect card after six years of trying, and ripping card readers out of the buses in which they’d been installed. Transportation economist Robin Lindsey of the Sauder School of Business at UBC thinks this is endemic to complex systems involving multiple transit authorities. The worst example he remembers is a proposal in Los Angeles from the 1980s that a transit expert recently told him had still not been successfully rolled out. By that standard, we haven’t had such a bad first year.

Can we make it better? Undoubtedly. And here, it’s interesting to consider a system feature TransLink originally wanted but which got left behind: a tap-out feature on buses. While you tap in and tap out on the SkyTrain and the SeaBus, telling the system exactly where and when you travelled, on buses we presently only tap in with Compass. Since the system therefore doesn’t collect point-to-point passenger travel data for bus trips, TransLink was forced to scrap the three-zone system for buses in favour of a single zone.

Setting aside the implied revenue loss, the bigger liability in not having that tap-out feature is the loss of what Lindsey refers to as “trip differentiation”—a crucial tool in the management of transit and road systems in reducing congestion. Trip differentiation with Compass would allow the system to read each ride distinctly, drawing on point-to-point data collected by the card. With that data, the system could theoretically price every passenger trip to reflect the cost of that specific trip based on duration, route, time of day, direction travelled and user profile. This is what economists refer to as “marginal cost pricing” and, in an ultra-rational frame of mind, it’s the holy grail of road and transit pricing. People travelling farther pay more to reflect their greater use of system resources (fuel, bus operator hours). People travelling at busy times or in busy corridors pay more to reflect the higher cost their travel imposes on other users. And given smartcards can be linked to individual users, a system with this capability could also tailor discounts for seniors or students on particular routes or travelling at certain times. Lindsey notes that he is participating in the current TransLink fare review, where many of these issues are being discussed.

Critics of price differentiation point out studies that show that people prefer knowing with certainty the cost of their travel in advance. Dynamic pricing could threaten that until a user becomes familiar with how different routes and times will affect what they pay. In cities like Vancouver, meanwhile, living far out of the centre is a function of real estate prices for many people; charging suburbanites premiums to come downtown, says Lindsey, can be seen as “adding insult to injury.” But distance isn’t the only way such a system could differentiate. Along the busy Broadway B-Line to UBC, there are 4,000 daily instances where travellers have to wait for another bus when the one in-station is full. If a system objective is to get UBC-bound commuters across town—and have them choose transit over their car—TransLink could manipulate peak pricing along that route while exempting passengers riding all the way to the university.

Granular trip data—obtained by tap-out technology on buses, or by road transponders, or other technologies—could be a crucial part in deploying dynamic pricing to manipulate system usage across Metro Vancouver. But only if TransLink has the data telling it where each passenger is getting off.