With official ridership stats flat or falling over the past few years, and the annual pressure to raise fares to balance the budget, the issue of fare evasion comes up regularly as an untapped revenue source. This became a particular concern with the move to all-door loading on, primarily, the streetcar network where the absence of a fare check at vehicle entry gives more scope for evasion than on buses or in subway stations.

Toronto’s Auditor General (AG) has issued a report and a video on this topic. They will be discussed at the TTC’s Audit & Risk Management Committee meeting on Tuesday, February 26, and at the full Board’s meeting on Wednesday, February 27.

The political context of fare management comes in on a few counts, and should be remembered when reading about dubious decisions and practices as flagged in this report.

• As the TTC shifted to larger vehicles, primarily on the streetcar system, an important goal was to increase the ratio of riders to operators. However, as all-door boarding and Proof-of-Payment (PoP) became more common, the need to validate fare payments went up. The politicians who control TTC funding at the Board and Council levels have a fetish for “head count” where limiting the growth in staff, or better still reducing their numbers, takes precedence. The result was that the number of Fare Inspectors did not keep pace with the growth in PoP.
• Presto was forced on the TTC by Queen’s Park under threat of losing subsidies for other programs. There is a strong imperative to report only “good news” about Presto both at Metrolinx and at the TTC for fear of embarrassing those responsible at both the political and staff levels for this system. Getting the system implemented took precedence over having a fare system that worked.
• Historically the TTC has claimed that fare evasion on its system amounts to about 2% of trips. With fare revenue for 2019 budgeted at $1.2 billion, this would represent a loss of about $24 million in revenue. If the actual evasion rate is higher, assumptions built into the PoP and Presto rollouts especially about the scale of enforcement required, are no longer valid.

Through all of this, there are many examples of poor co-ordination between Metrolinx/Presto and the TTC, of poorly thought-out implementations of procedure and of operational practices that simply do not achieve the best possible results. There is plenty of “blame” to go around, but a fundamental problem is that the system “must work” for managerial and political credibility.

The AG conducted a six-week review of actual conditions on the subway, streetcar and bus networks in November-December 2018 and found that the actual evasion rate was substantially higher, especially on the streetcar system.

The dollar values shown here are built up from mode-specific evasion rates and the level of ridership on each mode.

Problems with Presto contributed about 5% to the $64.1 million total in lost revenue, but this does not include issues with fare gates or TTC practices regarding “crash gates” in stations which allow fast entry for riders with media that can be checked visually. The proportion of such riders has dropped substantially with the end of Metropasses, and will fall again when tokens and tickets are discontinued later in 2019.

The report contains 27 recommendations all of which have been accepted by TTC management. The challenge will be to see how they are implemented.

Summary

The Auditor General’s findings fall into broad groups:

• The challenges of self-service fares where entrances are not always checked
• Presto equipment reliability and performance
• The ratio of fare inspection staff to the number of passengers
• Deployment issues for fare inspectors

A related issue is that the way the TTC estimates ridership might not accurately reflect conditions in the field. The reported drop in “ridership” in the past few years could lie as much in the methodology of counting multi-trip (pass) usage and shifts from old-style passes to Presto as in a real loss of riders and system demand. Moreover, a weakening in the rate of growth is clear going back longer than Presto has been available on the TTC, or Proof-of-Payment was in widespread use.

Boardings vs Ridership

An important distinction between “ridership” and “boardings” has come up recently in TTC reports, and this is echoed by the AG’s report. When the TTC cites a “ridership” figure, this is actually a value calculated from revenue, not a direct count of passengers. On the other hand, counts by Service Planning are based on a sample of real observations in the field and data from automatic passenger counters on surface vehicles, where they exist (primarily on buses).

• A “ride” is one passenger paying one fare for a journey based on the TTC’s transfer rules. In the case of Metropass holders, travel was converted to a trip count based on user diaries. With the conversion to Presto, the trip count comes from the Presto data. (Ridership associated with free trips, mainly children, is estimated from overall data.)
• A “boarding” is one passenger getting on one route. For this purpose, the TTC considers the subway to be a single entity and does not count transfers between the lines as a new boarding. Trips can consist of a single boarding, or a string of boardings that would cost one fare under the transfer rules. With the move to the 2-hour transfer, the definition of when one “trip” ends and another begins has changed. This will affect comparative reporting in years before and after the switch.

Although the revenue-driven ridership number is falling, the boardings are stable or rising. This suggests that riders pay less per boarding than they used to. This could be due to various factors including:

• Overstatement of “ridership” based on an outdated trips/month value assigned to Metropass holders.
• A shift of Metropass users to pay-as-you-play single fares through Presto.
• Increased fare evasion affecting revenue but not the counts of actual on-board travel.

As a statistical problem, the uncertainty about Metropass “ridership” values goes away with the move to Presto where the actual usage will be tracked. A problem does remain that this will report boardings where taps are required (or at least desired), but will not pick up transfers between routes in paid areas. There will still have to be a “fudge factor” for those with passes on their Presto cards.

The shift away from passes to single fares is driven by two factors. First, for many riders a pass was a matter of convenience even though it might be barely a break-even cost compared to single fares. Second, the introduction of the two-hour transfer allows more riding per fare, and reduces the number of individual fares charged to riders who make multiple short trips.

Riders who drop out of the pool of Metropass users are likely those with below-average trips/month, but for the purpose of “ridership” estimates, they count as if their previous usage was at the average level. For example, the break-even trip count for an adult pass versus tokens is about 49 trips ($146.25 pass / $3 token). However, the average Metropass holder takes about 74 trips/month. If a rider changes to paying as they go, 25 trips that they never actually made disappear from the TTC’s counts. This would be compounded by the two-hour fare which could reduce their fare payment below 49 depending on travel patterns. When trip monitoring catches up with an “average” profile of remaining pass users, will this be corrected, and there is a lag between the official average and actual values.

The Long Term Decline in Ridership Growth

Even during the period when annual TTC ridership continued to rise (red line below), the rate of the rise declined over several years (blue line). The TTC celebrated the increases while ignoring the warning signal that each year represented less growth, proportionately, than the years before. In other words, this change in “ridership” has been underway for over half a decade, not just for a few years. The rate of growth peaked in 2011 at 4.8%, but the rate fell every year thereafter until by 2017 it went negative. This trend is not a new phenomenon. [Ridership stats from 2017 Annual Report for years 2008-2017, and from CEO’s report for 2018]

The Shift from Metropasses to Presto

The AG’s study was conducted while the old Metropass swipe cards were still in use through November and December 2018. Pass holders represent the majority of fares paid (as “trips”) on the TTC although not the majority of customers. In other words, passholders take a disproportionately high number of “trips” compared to their actual numbers.

The uptake of Presto by TTC riders overall went from 45.5% in December 2018 to 77% in January 2019. Assuming that the change was mainly due to former Metropass users, the number of trips (and taps) would have changed more than the “adoption rate” which is based on customers, not rides taken.

The AG’s video presentation includes a clip of riders boarding via doors on a Flexity streetcar, and the majority of them do not tap on. This is no surprise considering that, at the time, many trips would still be by Metropass users who did not, indeed could not, tap onto vehicles. This video is the one disappointing part of the AG’s report because it gives the impression of widespread evasion which is not supported by their own on-vehicle fare inspections. Sadly, there are members of the TTC Board and Council who will cite this misleading video without understanding why it is meaningless in the context of the discussion.

Another factor in the shift is that strictly speaking a Presto user with a pass on their card does not have to tap on to a vehicle because there is no barrier to open. The tap is something the TTC wants for statistical purposes, and it also serves as a signal to other riders (and any watching fare inspector) that, yes, this rider does have a valid fare. Under crowded conditions, however, that tap may be ignored.

Failing to tap makes fare inspection harder because the hand-held Presto readers do not check for a pass first, but rather present recent tap-on history. The Fare Inspector must drill down to find the pass rather than having it shown automatically. This is the direct opposite of the design of Presto card readers in vehicles and fare gates where the first thing they look for is a pass, then for a currently active fare (within the two hour limit), and finally for a balance sufficient to pay a new fare on the card. (I have simplified this a bit by omission of transfers between TTC and other operators such as GO.)

Fare Evasion Statistics

Over a six-week period, the AG’s team together with TTC Fare Inspectors checked over 24,000 passengers using both uniformed and plain clothes staff, and both off and on board vehicles. A further 4,626 entries via fare gates were reviewed with video footage.

As noted in the table above, two factors were not included in the estimate of lost revenue:

• Functionality of fare gates: Out of service gates allowing passengers to enter stations without tapping.
• Unattended “crash gate” operations: Gates locked open for quick handling of Metropasses, transfers and cash fares, but unattended during a staff break.

Moreover, fare checks were not attempted on board vehicles during peak periods because of problems with crowding and the inability of Fare Inspectors to move through vehicles.

When the fare evasion rates shown above for each mode are multiplied by the number of riders, this produces an estimate of revenue loss. In this table, the value used for the average fare is the system-wide allowing for the mix of full adult fares and concessions (seniors and students). The calculation assumes that each group evades payment at the same rate.

The AG recommends that the TTC embark on a campaign to show riders the importance of paying their fares, but with a strange marriage of two ideas in one paragraph.

Passengers need to be made aware of the impact of fare evasion on TTC. Just as shoplifting affects the ability of retail stores to keep their prices low, fare evasion affects transit agencies in a similar way. Passengers should also receive more education on how the PRESTO card payment process works, the Proof-of-Payment system, and the consequences of a $235 ticket if found to be evading fare. Passengers should also be made aware of the City of Toronto’s Fair Pass program designed to assist eligible adult residents receiving Ontario Disability Support Program or Ontario Works financial assistance. [p. 14]

It is not clear what the Fair Pass program has to do with urging riders to pay beyond an implication that somehow this program would be undercut by the TTC’s lack of revenue. In fact, the main problem with the Fair Pass lies with the City’s slow implementation pace, the small number of eligible riders, and the problem that for many who are eligible, the pass is overpriced compared to their actual transit usage. This is an odd bit of political spin in the middle of what should be a technical report.

An obvious thing the TTC does require, and this is the AG’s first recommendation, is to decide just what is an acceptable rate of evasion. This is an important step – don’t just define a target, but recognize that there will never be zero evasion, merely an acceptable level that is part of the cost of doing business. By comparison, the City is content to forego considerable revenue, waste road capacity and tolerate unsafe conditions through its failure to enforce traffic laws. If this were audited, the percentages would probably not look good compared to TTC fare evasion.

In a separate recommendation (3.b), the AG urges that ongoing statistics be published rather than kept secret so that the achievement of goals (or not) is clear to all. This brings us to two problems:

• As the AG notes, past estimates of evasion suffer from flawed methodology (see below) causing the rate to be under-reported.
• A common problem with the TTC generally is that the stats don’t always tell the full story, but might be crafted to show management in the best possible light. This is already an issue with service quality numbers as I have written elsewhere.

The AG explains how past reports misrepresented the fare evasion rates, even though these reports are cited in estimating lost revenue.

Although the 2011 study continued to be used as the source of TTC’s publicly reported system-wide fare evasion rate of 2 per cent (and $20.5 million estimated total passenger revenue loss), the audit scope did not include buses and only included one streetcar route. There were also much higher evasion rates for certain areas examined, including 13.8 per cent fare evasion for discounted Metropasses, 5.4 per cent on streetcars, and 5 per cent for invalid transfers. However, these were not mentioned when the 2 per cent was publicly reported. [p. 16]

…

The 2014 Internal Audit study was specific to the one Proof-of-Payment streetcar route which was found to have a five per cent fare evasion rate and 23 per cent for passengers entering the rear doors without Proof-of-Payment. The results from this report were not presented to the Audit Committee or the Board.

TTC engaged an external consulting company to conduct a fare evasion study in the first half of 2016. Their overall fare evasion rate was measured at 4.4 per cent (bus 4.89 per cent, subway 4.32 per cent, and streetcar 2.85 per cent).

However, senior management had concerns about the consultant’s methodology, and did not accept the rates of this study and therefore did not report them publicly.[p. 17]

A particularly damning observation shows that the reported evasion rates from the TTC Fare Inspectors’ own reports were understated.

In 2018, TTC staff reported 1.8 per cent as its streetcar fare evasion rate based on its 2017 Transit Fare Inspectors’ inspection results. However, in conducting our audit work, we noted that this rate could be inadvertently understated.

For TTC to calculate the fare evasion rate, Fare Inspectors need to record the number of passengers inspected on streetcars as the denominator. During our audit observations on streetcars, we noted that instead of recording the actual number of passengers inspected, most Fare Inspectors were recording an estimated number of passengers on the streetcar.

There were significant differences in our results for the number of passengers inspected compared to the Fare Inspectors’ results. In some cases, the Fare Inspectors’ estimate was almost twice our inspected numbers. The impact is that the denominator for the fare evasion rate calculation from the fare inspection program was erroneously inflated, resulting in an understated fare evasion rate.

When asked why they recorded an estimated number of passengers, Fare Inspectors explained that they have an unwritten target of 500 inspections per 12-hour shift, and that it can be difficult to achieve this target at times during on-board inspections for various reasons (e.g. number of tickets issued, slowness of PRESTO hand-held devices combined with increased adoption of PRESTO card, travel time, inability to board congested streetcars). There was some apprehension of possibly being disciplined for not achieving this target.

Throughout our audit, we had many interactions with multiple Fare Inspectors. We observed that this practice of estimation appeared to be an ingrained, common practice. This calls for better training to ensure all Fare Inspectors understand the appropriate data to be collected, how the data is used, and how their work is important for the TTC and passengers.

Equally important is for management staff to develop realistic and clear performance targets for Fare Inspectors, conduct ongoing monitoring of work performed, as well as undertake regular reviews of the data collected to ensure they are accurate and complete. [pp. 17-18]

This is a classic example of management gaming the system while also creating targets that staff cannot achieve. The reported numbers “confirm” the evasion rate that is routinely reported to the Board.

Needless to say, the AG recommends that both the statistics and the targets for inspection rates be realistic and accurate.

Evasion Rates and Passes

A useful statistic missing from all reports on fare evasion is to distinguish between riders who pay by single fares and those who travel on some form of pass. If one has a pass, there is no incentive to cheat.

We know already that over half of all TTC trips are taken using some form of pass as the fare, and this implies that the evasion rate among riders for whom each trip has an incremental cost is probably well above the numbers cited as system-side averages. This begs the question of whether evasion rates would go down if more riders had a fixed cost per week or month based on a pass or a capped travel fee.

Capping as a replacement for the Weekly Pass is expected to come into effect later in 2019.

Streetcars

The AG reported a fare evasion rate for streetcar riders of 15.2%, and cautioned that this could be low as her team was unable to check during peak periods.

We were not able to find industry benchmarking standards specific to streetcars, as TTC is one of the few remaining transit agencies that has streetcars. Most other transit agencies are now using light rail transit instead. However, the rate is in the double digits, which is high. [p. 20]

This is an odd remark considering that streetcars and light rail transit are the same vehicles, and only the implementation differs with the latter generally being on a more restricted right-of-way. The real difference, I believe, is that the AG refers to on board vs prepaid fare collection. This will be an issue when Metrolinx begins “LRT” operation that has “open” stations just like streetcar islands now on Spadina or St. Clair, and it would apply to any “BRT” (bus rapid transit) operations with all-door loading from the sidewalk or from stations in reserved lanes as on new facilities in the 905.

The real distinction is that the moment one moves to all-door access without a prepaid area such as a subway station, there is an opportunity for fare evasion. Bus stats look good only because most routes in Toronto still use the conventional model of entering at the front where the operator can check for fare payment.

The AG tested evasion rates with both uniformed and plain clothes inspectors. The rates were quite different at 9.49% and 15.24% respectively. Note that this only covers onboard fare inspection, not the results from checks as riders left vehicles. To no great surprise:

… when passengers saw uniformed Inspectors on board, a number of them proceeded to pay fares or stopped boarding the streetcars. [p. 20]

Rates also differed by vehicle type on the streetcar routes and were highest with the new Flexitys (18.6%), lower with older streetcars (7.6%) and lower still with buses operating on streetcar routes (5.9%).

The majority of fare evasion modes on streetcars was concentrated in three categories:

• Did not pay / no proof of payment: 46%
• Did not tap Presto card: 31%
• Invalid pass: 16%

Buses

Although fare inspections are not carried out now on bus routes because they are not part of the Proof-of-Payment system, the AG’s team and TTC Fare Inspectors conducted a review of 26 different routes. The evasion rates measured fell between zero and 22.2% with an average of 5.1%. On the articulated buses, the average rate was 6.6%, as compared to a 4.4% average on standard buses. Results for individual routes are not included in the report.

There is a fundamental difference to streetcars in that operators check most riders who are boarding while on new streetcars, the operators do not interact with passengers much less check their fares.

Because buses do not operate under PoP rules, it is “legal” for a rider to be on a bus with no proof of payment (a valid Presto tap, a pass, a fare receipt or a transfer). If the TTC moves to a higher proportion of bus routes using unattended all-door loading, it is likely that evasion stats for this mode will rise.

Subways

A major problem for the TTC has been the conversion of automatic entrances from high-gate turnstiles to the standard low-gate versions. This was done to make the entrances accessible. In the process, the TTC created an environment where fare evasion is much simpler than in the past. 42 of the TTC’s subway stations have 56 automatic entrances between them.

The AG reviewed video footage of four entrances (Victoria Park, Sherbourne, Ossington and Spadina) from 6:00 to 11:00 am and 9:00 pm to 1:30 am for one day, and found many examples of illegal entry. Half of the fare evaders “tailgated” another passenger through the gate without tapping, and the highest incidence of evasion occurred in the morning from 7:00 to 8:00 am.

There are no statistics available from the fare machines themselves:

TTC’s system has the capacity to record illegal entry data through the sensors attached to each fare gate. We had initially planned to analyze this set of data to determine the prevalence of illegal entries at all TTC subway stations. However, two years into the implementation of fare gates, staff advised that they are still in the user acceptance testing phase due to the design of the sensors, and the data was not ready for analysis. Staff also need to ensure the completeness and reliability of the data in generating future reports. [p. 27]

The AG recommends various measures including a possible return to high gate turnstiles or closing the entrances, but more generally better monitoring of the automatic entrances. Whether these are practical remains to be seen.

Metrolinx Equipment Reliability

The contract between Metrolinx and the TTC sets out in excruciating detail the requirements for the type of services Presto is supposed to provide to the TTC. Buried in this are answers to questions such as why there are only two classes of fare readout (regular and concession), and what type of fare structures might be supported. There is supposed to be an Appendix containing a “Service Level Agreement” that would set out factors such as reliability and maintenance response times. However, the AG reports:

The Service Level Agreement between TTC and Metrolinx had still not been finalized by the end of 2018, six years after the initial Master agreement was signed. [p. 50]

I asked Metrolinx for the SLA, and it will be interesting to see how they respond.

Single ride vending machines (SRVMs) have two problems. One is that they are slow and a bit cantankerous, just the sort of behaviour one wants in a device that will be disproportionately used by infrequent riders. Moreover, as the AG reports, the credit card function often did not work. This capability was removed on December 3, 2018 and SRVM reliability immediately went up, but they are still troublesome.

During our 80 hours of streetcar observations (over two weeks) in November 2018 for on-board inspections, we noted 40 passengers on 12 streetcars who were not able to pay because the vending machines were not working. Among the 40 passengers:

• 26 attempted to pay their fare but were unable to do so because the vending machines were not working. These 26 instances were not included in our fare evasion rate calculation.
• 14 did not attempt to pay until asked by the Transit Fare Inspectors; however, when they went to pay, they found the machines not working. [p. 29]

These observations were made while the credit card function was still active, if that word can be used to describe something that rarely worked.

During off-boarding inspections of streetcars at four stations, among 170 passengers who could not provide the required Proof-of-Payment, 71 complained about broken vending machines. [p. 30]

…

Many of the Fare Inspectors raised concerns about the unreliability of the Metrolinx equipment and that it can also impact fare evasion and their fare inspection. They stated that many regular passengers are aware of the machines always having issues, and some use it to their advantage to evade fare payment. [p. 30]

…

Many PRESTO card readers were out of service during our audit observations in November and December 2018. Although at least one other PRESTO card reader on board was functional the majority of these times, passengers were not always able to reach the other reader due to congestion or did not choose to go to the other reader to tap. [p. 31]

…

Metrolinx is also responsible for the maintenance and repair of the PRESTO card readers. Under the agreement between TTC and Metrolinx, there is a target functionality rate of 99.99 per cent for the PRESTO card readers. [p. 31]

A major problem is the divided responsibility between Metrolinx and TTC for servicing equipment. In the case of Presto readers, Metrolinx does the first level work, but reports from Local 113 of the Amalgamated Transit Workers (the TTC’s primary union) reveal that there can be a substantial lag between reporting of failed devices and their repair. It is possible for vehicles to be held out of service because there is no working reader on board. As for Metrolinx, the stats reported to their board are based on the premise that if there is a reader somewhere on a vehicle that works, this constitutes “availability” even though riders might be physically unable to reach the working device.

Fare Gates

Fare gates are a TTC responsibility for initial support, but dispatching repair teams is an issue:

Since installation, TTC has been accountable for its own first-line maintenance and repair of the TTC fare gates. Metrolinx would be responsible for any issues in regard to backend servers or software.

During our audit observation, we noted many instances of malfunctioning TTC fare gates that were stuck in an open position. Among the 15 subway stations we visited, in 14 of them within one to two hours of observation we noted multiple instances of fare gates not operating. In total, we noted over 40 instances of malfunctioning
fare gates during 22 hours of subway observations. In many of these cases, the fare gates would remain open and then ‘self-close’ shortly after. But at other times, they did not and could stay open for long periods. [p. 33]

…

In addition, at an automatic entrance, we noted an accessibility TTC fare gate that was stuck open half way […]. TTC staff advised that when a fare gate fails and is stuck open, there is no automatic message to fare gate maintenance staff to let them know that the gate went out of service. We were also informed that the only way fare gate maintenance staff know that the gates are out of service is if TTC staff notify the maintenance staff by creating a ticket in their maintenance system. Unless TTC station staff regularly check and report on TTC fare gates at automatic entrances, any malfunctioning fare gates at those entrances will be stuck open for potentially a long period of time, allowing passengers to freely pass through the gate. [p. 34]

The AG will conduct a further review of Metrolinx equipment and Fare Gates in phase 2 of the audit.

Crash gates

At high-volume locations, the TTC opens a fare gate and provides an operator with a farebox to quickly check riders with passes and transfers, as well as to collect single fares. With the move to Presto for Metropasses, the volume through a crash gate line will diminish and they will not be required. However, there were operational issues with crash gate practices.

Based on our discussion with crash gate staff, they are instructed by TTC to lock their fare box during their scheduled 15-minute break, but they are unable to close the crash gate. Only the fare collector in the booth can close the gate using the computer system located inside the booth. [p. 34]

…

We were advised by TTC staff that not all fare collectors know how to use the computer system to close the gate, so this may require additional staff training. [p. 35]

Unattended crash gates were a source of fare evasion as observed by the AG’s team, although is unclear how many who walked through the open gate were used to doing so when it was attended and had valid fares (e.g. an old-style Metropass).

Presto Cards for Children

The Auditor General has flagged extensive problems with the availability of children’s Presto cards that allow free travel on the TTC.

The number of cards in circulation has ballooned, and this was supported by an ad campaign urging people to get these cards.

According to TTC staff, 12,584 unique PRESTO cards with child concession were used for 867,238 rides on the TTC from January 1 to October 31, 2018, compared to only 3,962 cards for 162,231 rides for all of 2017.

TTC’s Transit Fare Inspectors have found that the number of passengers who fraudulently use Child PRESTO cards has increased since the advertisement campaign. The total related charges and cautions […] have increased from nine in 2017 to 80 from January 1 to October 31, 2018. [p. 37]

During the AG’s team survey, 22 bus, 2 streetcar, 56 subway child cards were used by riders who were not children, but none was presented by an actual child.

It is important to note that during our six weeks of audit observation work on all three modes of transit and covering many different times of the day on TTC, we did not come across ANY children aged 12 and under using the Child PRESTO cards. We saw parents letting their children through the TTC fare gates and children walking onto the bus and streetcar for free, which is fine with the current fare policy.

This raises a question of whether the reported number of Child PRESTO taps, just over one million rides in 2018, were truly used by children, and what percentage could be passengers fraudulently using the cards. Based on our observation results, it is likely that a large percentage of the Child PRESTO taps are fraudulent and the annual revenue leak for TTC could be in the millions. In addition, given the increasing number of Child PRESTO cards, combined with an increasing adoption rate for PRESTO cards on TTC, the annual revenue loss from fraudulent use of Child cards could rise even further. [p. 38]

TTC ridership numbers include an estimate for children. However, not clear whether this is based on historical patterns if Presto “child” taps are affecting this number. There is a further problem across the system with “children” who are really students. This is a wider issue than just the Presto cards.

Other transit agencies using PRESTO also have Child PRESTO cards but none seem to have reported issues similar to TTC. This is likely because of TTC’s fare policy allowing children aged 12 and under to ride for free. The other agencies have a fare policy for children aged five and under to ride for free, so presumably a child would not be travelling independently in the other jurisdictions. Also, the other transit agencies appear to have in place ways for their bus drivers to easily see when a Child PRESTO card is being used. Another major difference is that TTC is the only agency (other than GO Transit) in the GTHA operating a rail system where passengers can enter the system without any interaction with staff. [p. 39]

The cards for children are identical to regular Presto cards, and so an operator or Fare Inspector cannot make a simple visual check for their use by riders who are not children.

We were informed that TTC staff have attempted to negotiate with Metrolinx to provide visual distinction on the Child PRESTO cards, but this was rejected by Metrolinx citing additional inventory costs, according to TTC staff. Metrolinx staff advised that the card is meant to be used for several years and that they don’t want to limit the ability of passengers to have the concession type changed, e.g. student to adult, without purchasing a new card to do so. [p. 40]

When a Child PRESTO card is used on TTC, it flashes yellow on the PRESTO card readers – the same as with other concession cards such as students and seniors. This makes it impossible for bus and streetcar operators to identify the inappropriate use of the Child cards, as the PRESTO card readers only have one colour (yellow) and sound when any type of concession other than adult is tapped. It would also be more efficient for Fare Inspectors if there were a distinct light and sound for Child PRESTO cards, as they could focus their efforts on catching fraudulent use of these cards, instead of needing to check all concession types with a yellow light.

In other transit agencies that use PRESTO, such as York region and Mississauga, their bus operators have a PRESTO tap monitoring device enabling the bus operators to see the specific type of concession being tapped.

Without the proper monitoring device in place, TTC bus drivers cannot see whether the PRESTO cards being tapped are adult, student, post-secondary student, senior, or child. According to TTC staff, the lack of this monitoring device was due to a TTC decision in the system chosen for TTC buses and not due to a limitation with PRESTO. [p. 41]

It is likely that when the TTC’s Presto program began, they did not want to attempt to integrate a readout of Presto card types with the antique hardware and software of the old “CIS” consoles on vehicles. However, this should have been integrated into the new “Vision” consoles. This only makes sense on buses where front door entry streams most riders past the operator who can validate the type of concession fare a rider is using. In any event, a problem will remain that a substantial portion of all riders will enter the system at a gate or reader that is not monitored, and the free travel available with “Child” Presto cards will continue to be a problem.

With “fare integration” a key desire in provincial planning, it is pathetic to see how inter-agency bungling limits the ability to police the abuse of Presto cards. Presto cards with a Child concession coded are issued through Shoppers Drug Mart. No proof is required that the buyer actually has a child, and of course there is no photo id on the card. The AG found that cards are offered online as a way to obtain free transit. However, when a rider is found to be using such a card, there is no mechanism for TTC Fare Inspectors to seize it because the card belongs to Metrolinx.

When a TTC Fare Inspector identifies a passenger fraudulently using a Child PRESTO card, the Fare Inspector can issue a ticket of $235 on the spot. However, a TTC Fare Inspector does not have the authority to seize the Child card used fraudulently, as TTC would normally do with a fraudulent TTC Metropass, since the PRESTO card is the property of Metrolinx. The TTC Fare Inspector is to email TTC management for the card to be blocked. TTC staff then request on the PRESTO website for Metrolinx to deactivate the card. However, TTC does not receive a report from Metrolinx confirming the card has been deactivated and it is possible the individual may still be using the Child PRESTO card fraudulently. [p. 43]

…

We found a large number of fraudulently used Child PRESTO cards during our observation period and there are numerous serious control weaknesses. In our view, the Child PRESTO cards should be temporarily suspended until appropriate controls are put in place by the TTC. [p. 45]

The AG recommends:

13. The Board request the Chief Executive Officer, Toronto Transit Commission, to re-assess whether there is a critical need to issue Child PRESTO cards, balancing provision of good customer service with the risk of fraudulent use of the Child Cards.

14. The Board request the Chief Executive Officer, Toronto Transit Commission, to NOT distribute the Toronto Transit Commission’s promotional Child PRESTO cards until appropriate controls are in place. [p. 39]

Fare Inspection Program

Fare Inspectors cannot enforce fare rules as they are not Enforcement Officers with the power of Special Constables. The history of this lies both in a desire to limit the number of higher-cost Enforcement Officers, and to allow Fare Inspectors to present a less-authoritarian image to riders. This also ties in with concerns that employees with Special Constable status might abuse their power based on past incidents. However, the result is a cadre of Fare Inspectors who can check fares, but are powerless to deal with passengers who simply ignore them.

During our audit period, we observed that if passengers had not paid the appropriate fare, Fare Inspectors used their judgement on whether to issue a ticket, written warning, or verbal warning. When passengers cooperated by providing their identification and contact information they received a ticket. When passengers just walked away or were aggressive, they did not receive a ticket. There is no repercussion – we saw many evaders simply walk away when asked for proof of payment.

Many passengers appeared to know that if they walked away from a Fare Inspector, there was nothing the Fare Inspector could do about it. In one instance, we observed that a passenger became aggressive and the Inspector de-escalated appropriately, but could not issue the ticket, even though it was obvious that the passenger evaded fare. This raises the question of whether TTC’s fare enforcement is fair and effective.

…

The presence of Transit Enforcement Officers helps to minimize the number of walkaways and address the safety risk. This would be particularly important when TTC expands its inspection program to subways and buses, where the safety risk can sometimes be higher, based on our observations. [p. 47]

In past discussions, TTC Board members vacillated between a hard-line demand that staff do everything in their power to collect fares, and a desire to treat riders without confrontation, in part for staff safety and in part to avoid the political fallout of an assault/arrest situation. The message to management and staff was to collect every nickel possible, but don’t hassle people, except when the goal is to keep the budget under control.

Fare Inspectors are not provided with the best of equipment to do their job, and the fault here lies with Presto.

Fare Inspectors use a PRESTO hand-held device to inspect fare payments made with PRESTO cards. The hand-held device is used to check the concession type (e.g. student, senior, or child) and the transaction history for the last 10 transactions on the PRESTO card. The only way for Fare Inspectors to inspect a PRESTO card payment is with these hand-held devices, which are supplied and maintained by Metrolinx.

Both the speed and the reliability of this device is important for Fare Inspectors to be able to do their jobs effectively and efficiently. This tool has become even more critical as the rate of adoption of PRESTO continues to increase significantly on TTC.

During our audit, we noted that the speed of the devices was very slow and many Fare Inspectors commented on their frustration with the slow speed. Compared to Metrolinx’s devices which perform at 40+ inspections per minute for GO Transit inspections, the devices used by TTC are much slower at about 15-20 inspections per minute.

…

We also observed that TTC’s hand-held devices were frequently out of service. Several times during our observations, a Fare Inspector’s device would crash. Sometimes the Fare Inspectors simply needed to re-boot the device, but often the device did not re-start and the Fare Inspectors had to call the supervisor to deliver a replacement unit, or return to the reporting location to obtain another device. This could cause significant interruptions to fare inspection and lower the efficiency.

There was also one day during our audit that ALL of TTC’s PRESTO hand-held devices crashed, but all Inspectors were still expected to carry on and perform their inspections that day. [p. 49]

Many Presto card checks are done off of vehicles typically where there is a large volume of transfer passengers between a streetcar line and the subway. The physical layout varies so that in some locations passengers a funneled through a path ensuring that they all pass Fare Inspectors, while in others passengers emerge from all doors of a streetcar onto an open platform. There are rarely enough Fare Inspectors present to monitor all doors for the latter cases. In any event, if an Inspector has to step away from checking passengers to issuing a warning or a ticket, this further reduces the actual inspection count. One could argue, however, that the sight of someone getting a ticket will instill in other riders the need to have a valid PoP on hand whether they are actually inspected or not.

Any attempt to constrain exit from streetcars to a few doors where there are Fare Inspectors would present a considerable delay and confusion on board. In any event, the new streetcars do not have controls which allow selective door opening.

Concluding Thoughts

Fare inspection is not simple nor is it cheap. The TTC has moved to a station, vehicle and fare media design that requires inspection at a rate sufficient to deter freeloaders. This does not mean everyone will be caught, but that evasion is kept down to an acceptable level, whatever that might be.

The lack of a clear set of agreed service levels between the TTC and Presto must be corrected, and the TTC must be able to provide front-line support for Presto devices especially if Presto itself refuses to provide this on a timely basis.

The inability of TTC Fare Inspectors to seize fraudulently-used Presto cards must end. Riders who know that cheating could cost them their card might think twice about evading fares.

There are functionality problems with Presto obvious to anyone who uses the system although, to be fair, the situation has improved over the past year. Any “next generation” of Presto must be designed to work in the real world with the wide array of usage patterns riders on a big system like the TTC will present.

The Auditor General has shown that there is a substantial group of problems with issues of implementation, of procedure, of management and of ongoing support. If what we are looking for is “blame”, there is plenty to spread around. Finger-pointing is a favourite exercise especially when multiple organizations and departments are involved, but that is a useless exercise. Riders simply want a fare system that works.

Any scheme that is implemented, or modifications to current practices, must acknowledge the real world in which fare collection takes place including very large volumes of riders most of whom pay their fares properly.