A question often arises about just how Metropass riders use their passes. How many trips do they really take? How much of a “deal” are they getting compared to those who pay by tokens, tickets and cash?
The TTC conducts a rolling survey of passholders on a weekly basis with about 30 riders who keep track of where they travelled. It is a new group every week, and so over the course of a year, the TTC will have about 1,500 separate surveys.
The information recorded by riders is converted back into a trip count (allowing for “normal” TTC transfer rules) to arrive at a trips/week value for each person surveyed. With a small sample set, the values bounce around a lot, but aggregated over time, they can give an idea of what Metropass usage actually looks like. The data is used to calibrate the conversion factor from pass sales to “rides” in the TTC’s regular reports of “ridership”.
With over half of all “rides” now taken with passes, this conversion factor is important, and a small change in the multiplier used can have a big effect on the calculated ridership. Moreover, if Metropass sales fall, the presumed “loss” of rides is at the average for the whole group even though it is more likely that the lost customers will be relatively low users of passes.
Wondering about just what the numbers looked like, I asked the TTC for statistics from their weekly diary surveys spanning January 2015 to June 2016. The raw data are from the TTC, for which much thanks, but I have consolidated and reformatted them for this article. The presentations and interpretation are my own.
The overall numbers for the 18 months are shown in the table below.
This table groups the data by the number of trips reported in the week.
About two thirds of the diaries report between 10 and 19 trips a week, and the overall average is 16.28. Note that the “trips” values shown here are actually calculated from the individual values (i.e. number of diaries times number of trips).
Another way to look at this is to plot the percentage of diaries reporting individual numbers of trips.
The mean and median values are shown in the next chart.
There is a very slight upward trend in both values over the 18 months, but not much. Although there are week-by-week variations, the long-term numbers are consistent. This suggests that those riders who have stayed with Metropasses are using them at roughly the same rate.
Another factor could be the loss of low-frequency passholders back to single fares because they don’t consider a pass worth the expense. Removing them from the overall pool will increase the “average” usage of those who remain even though their travel patterns might not have changed at all. The actual proportions have changed a bit over the 18 months, but no group has declined precipitously relative to the total.
The next charts present the diary and trip counts in a way to show the relative size of each group (based on trips/week) and of the trips they generate. Ten riders at 15 trips/month generate more trips than ten riders at 5/month, and so the frequent riders produce proportionally more trips than their actual numbers. (The underlying data are the same as in the chart above, but formatted differently to illustrate each aspect of their behaviour.)
The first chart shows the proportion of diaries each week falling in various bands of usage. Of the diaries, about half fall at or below the 10-to-14 trip group, while the other half are above. The proportions vary each week, and effects such as holiday weekends show up as a drop in the more frequent riders.
The second chart shows the proportion of trips represented by these riders. The break between the 10-14 trip group (yellow) now lies at roughly the 35% line showing the relatively lower contribution of the less frequent riders to trip counts. About 40% of the rides are due to frequent riders who take 20 or more trips/week (purple). (Click for larger versions.)
An important caveat here is that these data are not subdivided by type of pass, and even within a single pass type (e.g. “adult”) the effective cost to the rider (and hence average fare) will vary depending on whether they receive a discounted pass and how much tax benefit they obtain for buying transit fares in bulk.
Pass Pricing and Trip Multiples
Adult passes sell for $141.50 per month at full price, or $129.75 on the subscription “Monthly Discount Plan” (“MDP”). With the token fare at $2.90, these prices correspond to multiples of 48.8 and 44.7 respectively.
Passes for Seniors and Students sell for $112.00 and $102.75 (MDP). With the ticket fare of $1.95, the multiples are 57.4 and 52.7.
The Post Secondary pass sells for the same rate as a Senior’s pass, but these riders would otherwise pay full adult fare. Therefore, their multiple is 38.6.
The non-refundable federal tax credit reduces all of these numbers by 15%, although it is not a benefit a rider “sees” until long after a pass has been bought and used. There is no benefit to someone who has no taxable income.
An average trips/week of 16.28 translates to an annual rate of almost 850 trips, or a monthly value of 70.7. This number is considerably higher than even the most expensive fare multiple for passes. However, it is important to remember that many so-called “trips” would probably not be taken if they required separate fares because they are short, convenience trips, or because they form part of a “trip chain” that would be uneconomic without a pass to eliminate transfer issues.
The idea that somehow the “extra” trips beyond the break even point represent “lost revenue” is fundamental to arguments by those searching for any way to reduce TTC subsidy requirements. However, an increase in the marginal cost of those trips would drive many of them away from the system, or would simply result in increased “fare evasion”.
As a simple example, yesterday was a “four trip” day for me, but three of those trips logically constituted a “trip chain” with stop-offs between each leg that would normally trigger the need for a new fare payment. This brings us to the basic question of whether the transit system should be hitting up riders for a new fare as often as possible, or if it should be making travel as simple as possible. Motorists, by contrast, do not face a large marginal cost for multi-stage trips, beyond the agony of finding free parking, and will commonly arrange their travels to avoid high cost stopovers where parking is limited and/or expensive.
If the idea of transit is to provide an alternative to cars, then pricing its use in a way to even further reduce the perceived convenience runs counter to the claims many politicians make for having transit in the first place.
Re-Pricing or Replacing the Metropass
The trip diary data covered in this article do not include a fare change for the Metropass, and the pricing multiples actually fell in 2016 because token and ticket rates went up while pass prices were frozen. This freeze was a direct response to experience a year earlier when pass sales dropped after the price went up.
It is worth noting that the actual usage of passes, on average, has stayed almost unchanged since January 2015. The number of passes sold may rise and fall, but average usage stays the same. This is not surprising because once a rider has broken the link between paying at each step of a trip and simply using transit as a convenience, then the amount of use is more a question of service quality and availability, and the rider’s day-to-day needs for travel.
During the 18 months covered here, there were 7 diaries returned with zero trips reported, probably because the users were not in Toronto, or otherwise not using the transit system. There was also one diary reporting 81 trips in the space of one week, or more than ten a day. The important point is that this is an outlier, and a fare structure should not be torn apart because a tiny fraction of riders get a very good deal out of passes. The bulk of riders, and those for whom any pricing change will have the most effect (both on them and on TTC revenues) lie in that central block of the trip distribution chart above.
Where the challenge lies is with riders who sense that a pass may not be worth the investment either in absolute terms, or because they don’t want to spend all of their “TTC” budget in a single payment. The option to not buy a pass is not available to MDP subscribers, and so the reference point would mainly be full fare adults for whom the break-even is about 49 fares/month or just over 10/week. Based on the diary numbers, this is not a large proportion of the total, but neither is it trivial. Repricing the pass to lift the multiple would take more riders into that gray area where they would think twice about getting a pass.
Increasing the multiple would shift more riders below the break-even line, and would increase their mental debate about whether convenience outweighed cost. That decision will change with a move to Presto when the convenience factor (use any turnstile, board at any door) of passes will vanish. Indeed, some of the low-end pass users may drop back to paying single fares because they can do so much more easily than today. Oddly enough, this will show up as a “ridership loss” for the TTC because passes are counted at the average usage level. Someone might take only 10 trips/week with their pass, but if they moved back to single fares, they would count as a “loss” of 16 pass-based trips, or a net loss of 6.
Conversely, the convenience of a self-reloading Presto card could encourage them to take more single rides much as a prepaid coffee card ups the sale of lattes and double-doubles.
The TTC already has plans to replace the Day Pass with a daily cap on fare costs through Presto. The machinery to do this already exists within Presto software, but it has not yet been turned on. A similar scheme could be used for monthly ridership much as GO Transit now caps charges so that rides beyond 40 in a month are free. Either of these caps will fundamentally change the way that people “access” pass-like pricing because riders would not have to decide in advance whether they should pay for a pass.
Another factor that would bear on this is the concept of the “two hour transfer” already in place in GTA transit systems. Rather than enforcing the strict rules about stop-overs and doubling back on trips that have been a TTC hallmark for its entire existence, a “fare” would be valid for a fixed time period after the first “tap on” entering the system.
The TTC has been mulling fare strategy for the past few years, but the two hour transfer was among the casualties of the 2016 budget because “we can’t afford it”. The anticipated lost revenue is about $20 million, although there is reason to believe that this estimate is too high by a considerable margin. In an era when the TTC faces subsidy cuts, the last thing anyone would propose is something that will “lose” money even though this might address other transit issues such as simplicity of fare collection and making transit more cost-effective for users whose riding does not fall into the classic two-trips-a-day commute pattern.
Over half of all trips taken on the TTC are now paid for with some form of pass. The idea that fare debates turn on the cost of a token is becoming almost as quaint as the token itself when the real issue is how riders will pay for transit service in bulk.
Any debate about the future of fares in Toronto must recognize that passes are the way people prefer to travel, and price the service on that basis rather than agonizing over the fictional “lost” revenue that might be regained if only those pesky riders paid their fair share.