TTC’s 2017 Operating Budget: More Creative Accounting (Updated)

The TTC’s 2017 Operating Budget will be discussed at a special Board meeting on Monday, November 21, 2016. When work began on this round, the TTC stared at a $231 million hole in its potential 2017 funding, and it was apparent that the Mayor’s request for a 2.6% cut in subsidy was small change beside the TTC’s much larger problems.

Updated November 17, 2016 at 6:40 pm: Responses from the TTC clarifying the treatment of externally recovered costs have been added to this article.

The Budget Report is now public, and initial media comment suggested that the TTC had wrestled that huge potential deficit to the ground. However, a lot of that is spin and accounting trickery, not a real reduction in the TTC’s needs.

Still facing a gap of $61 million, TTC management list many unpalatable ways that operating costs could be trimmed. In effect, the message to Council, and especially to Mayor Tory is this: being a “transit mayor” costs money, and it’s time to pay up.

This article looks at the Operating Budget, the one that provides service and handles day-to-day maintenance activities. In a separate article, I will review the Capital Budget.

Understanding The Budget Mechanics

The TTC (and all City agencies and departments) begin work on their next year budget midway through the year. The 2017 budget has been “in the works” for months and in many ways is based on 2015 rather than 2016 results because the year was barely half-over when the 2017 budget cycle started. This can lead to problems when the “current” year does not turn out as expected as happened in 2016.

An important first step in looking at TTC budget numbers is to recognize that any year-to-year comparisons are relative to the 2016 budget, not to the probable actual results for the year. This has some important effects:

  • The fare revenue projection for 2017 is based on a lower projected ridership than was used, but not actually achieved, in 2016. Therefore, revenues go “down” in this budget (without allowing for effects of a fare increase) simply to get the estimated ridership back to a reasonable level. 2016 was described as a “stretch target” for ridership, and the budget elastic didn’t stretch as far as hoped.
  • Some 2016 costs are coming in under budget, notably for employee benefits but also for diesel fuel. These are savings in actual spending in 2016, but they also show up as “reductions” in 2017 when they are rolled into the budget. It is important to distinguish between reductions in costs that affect actual spending in 2017 as opposed to simply being a lower budget number.
  • The 2015 budget included a “capital from current” item for the purchase of new buses. This is not an “operating cost” in the traditional sense, but it avoided putting the item on the City’s Capital Budget. For this reason, 2015 is an odd year in any stats unless the capital portion is factored out (notably from the claimed subsidy per rider). There is no such payment in 2016, and so a direct comparison with 2017 is possible without adjustments, at least until some of the TTC’s new budget tricks are factored in. For 2017, the TTC proposes to shift some operating costs onto the Capital budget and, as a result items that were billed to “operations” and counted as part of the rider subsidy in past years would disappear. The City Manager’s Office does not concur with this tactic. The point here is that year-to-year budget and subsidy comparisons cannot always be made without adjusting the figures to a common basis.

Another important factor is that in the total numbers reported in the media, the “conventional” and Wheel-Trans (WT) budget numbers are often conflated. When demand for WT is growing quickly, as is now the case with improved eligibility criteria from Queen’s Park, costs for this service grow proportionately. This cannot be wished away by budget hawks who care only to limit tax increases, but worse it can create a situation where the conventional system is pillaged for dollars to handle the WT demand. That is not the sort of conflict we should be seeing in budget debates, but it is inevitable when the extra cost of WT for 2017 is almost equal to the revenue from a 1% property tax hike.

Finally, there is a distinction between true savings that represent lasting reductions in expenses, and one-time benefits from procedural changes or special accounting provisions. The impression can be given that a budget is in much better shape than reality by giving the impression that a large initial deficit can be whittled down.

Getting From $231 Million to $61 Million

The budget gap was “closed” by a number of measures, some of which are simply savings against the 2016 budget that are carried forward into 2017. On an “actual spending” basis, the savings have already happened. They are not the result of new, recent actions by TTC management.


As I reported previously when reviewing at the Budget Committee discussions, the reduced healthcare costs were actually achieved as a saving in 2016 against that year’s budget. They are a reduction in 2017 only on a budget-to-budget basis because the $10.3 million was part of the 2016 budget as a starting point.

The recommended savings in the current report are:

  • Reduced PRESTO fees due to the delay in rolling out Metropass support. With less of the TTC’s revenue flowing through PRESTO, the cost of serving this does not have to be included in the budget. Note that this treats PRESTO costs as a net addition and does not include an offsetting saving in handling costs for Metropasses likely because the two modes would co-exist during a transition period.
  • As in past years, energy costs are expected to be lower than the previous year’s budget level. The saving shown here is $12 million, but $5 million of that is already projected to be saved in 2016 actual results as per the October 2016 CEO’s Report, p. 50.
  • Capitalization of City construction impacts. When the city tears up a street (for example the Queen Street watermain project now underway), the TTC incurs extra cost to divert and supplement service. This has been borne out of the operating budget but without an explicit chargeback to the capital project for which this should rightly be a cost. This appears to be a new practice for 2017 and it is unclear whether the City and its agencies have agreed to pay these charges. (See update below)
  • Elsewhere in the budget, there is a section on cost recoveries from Metrolinx for its construction project effects on service. This would be done on a cost recovery basis as with City construction, but it is unclear why one of these has been included as a “saving” but not the other. (See update below)
  • Delays in Bus Reliability Centred Maintenance. TTC management had proposed a higher fleet spare ratio and maintenance practices to pro-actively get ahead of failures rather than doing predictable repair work after a vehicle had already passed the likely failure time and possibly actually broken down in service. In the Capital Budget, the TTC is proposing a very large order of buses to replace the Hybrid fleet before its due date, and this would reduce maintenance needs by substantially lowering the average age of the fleet. It would also place a large chunk of the fleet under warranty effectively transferring operating maintenance costs to the Capital budget. This tactic has a downside as the TTC has seen in the past when a homogeneous fleet reaches its non-warranty period, and later the higher cost of maintenance (and staffing requirement) of older vehicles. This is a time bomb built into the budget even though it could “solve” a short term problem.
  • Reduced contracted services. It is unclear what this refers to, and I have sought details from the TTC. The largest “contract service” the TTC has is the provision of service in York Region, but this is done on a full cost recovery basis. If YRT takes over a service the net change to TTC’s budget should be zero. (See update below)

Update: The TTC has clarified the handling of externally funded costs in an email from Vince Rodo via Brad Ross.

With respect to City construction projects:

It has been a long standing practice for the TTC to charge the incremental cost of service to TTC capital projects.  For example, when we tear up streetcar track, we run replacement bus service during the construction period.  We charge the difference between the regular streetcar service and the bus service to that project as part of the cost incurred because of the project.  In the past, we have not done that for City of Toronto construction projects.  The city has agreed that the practice for TTC projects can be used for city projects too.  So for example, if the city were closing an intersection for work and we had to re-route service round that, we will now be able to calculate that extra cost and bill it to the city, who will charge it to that project.  Since these costs had to be covered by the TTC operating budget in the past, they have been included in the TTC operating budget.  From now on, they will not show in our expenditures because there is no net cost to the TTC.

With respect to Metrolinx cost recoveries:

The Crosstown Master Agreement calls for Metrolinx to reimburse the TTC for incremental operating costs associated with the impact of Eglinton Crosstown construction on TTC service.  For 2015, we billed them and they reimbursed us for $5.2 million.  I don’t have a final figure for 2016, but I suspect it will be in that range.  That is not included in our budgets because there is no net cost to the TTC.  For 2017, it is similarly not included in our budgets because once again there will be no net cost to the TTC. So the treatment of this and the city construction above will be completely consistent on a go-forward basis: no net cost included in the TTC budget.  We flagged this in the 2017 budget report because: (i) the quantum is increasing substantially in concert with the ramping up of Crosstown construction, (ii) it stays high for at least the next 5 years and (iii) because we are hiring 169 TTC employees to provide this augmentation of our service.  If it were the same range as 2056 and 2016, we may not have highlighted it.

With respect to York Region cost recoveries:

The service the TTC operates in York Region under contract to YRT is included in the budget as both an expense and a revenue and that has been the case for decades.  It is the service they request us to operate on their behalf.

With respect to contracted services:

The “reduced contracted service” has nothing to do with any of the items mentioned above.  It has to do with reduced purchases of services such as IT, human resources, IT licenses, cell phones, etc.  No impact on service on the street.  It all about saving money everywhere we can.

The different treatment appears to arise from whether there is a net delta in the budget (i.e. a new condition, and therefore reported as a budget-to-budget change) or a continuation of an existing practice where costs and expenses always balance out. [End of update]

Four additional sources of revenue or savings are proposed:

  • A draw of $14.4 million from the Transit Stabilization Reserve. This money comes from “surpluses” (planned subsidies that were not needed in good years) that have been banked against lean years of which 2017 is most definitely one. This is not a “saving”, but rather one time revenue that can offset the budgetary pressure for 2017 only. The underlying costs will not disappear, and they will show up as part of the 2018 pressure.
  • A fare increase of 10 cents on the adult token rate, pro-rated through other types of fare (the details are discussed later in this article). The added revenue is net of the anticipated loss of riders. If pushback from the increase is less than expected, then the TTC could do better with new riders than planned, but many other factors will affect riders’ decisions about staying with transit.
  • The TTC proposes that the cost of new batteries for Hybrid buses of $8.5 million be transferred to the Capital budget. This is an interesting accounting debate because parts replaced during a major bus overhaul (typically at mid-life) are paid for from capital, while parts replaced in normal day-t0-day maintenance count against operations. However, the lifespan of these batteries is short enough that capitalization is an odd treatment.
  • Some TTC capital assets are not subsidized by the City in part because their lifespan is too short, and in some cases this is likely a holdover from the days of provincial subsidy when certain items were excluded from that funding. As an accounting mechanism, the TTC funds these purchases out of working capital, and recovers the money as a depreciation charge against the operating budget over their lifespan. The TTC proposes that this amount be treated by the City as a capital cost thereby shifting $29 million out of the operating budget.

The City’s response to the final two items is not exactly welcoming:

Both of these items were reviewed with City staff and not supported because they reflect a shift from the operating budget to capital, requiring City capital funding. Staff suggest these items be given further consideration by the City as they might help address the operating pressures while immunizing customers from service adjustments or further fare increases, to the extent possible. [pp. 5-6]

Fare Increase Options

TTC management recommend a ten cent increase in the adult token fare from $2.90 to $3.00 with proportional changes in all other fares except cash which would stay at $3.25. This is expected to bring in $27 million in new revenue, net of the loss of 1.2 million riders and the PRESTO fees that are a percentage of the revenue stream.


There is no discussion of the issue of special fares as a social benefit, and that issue will get tangled up in proposals to deal with the (at least) $61 million remaining gap between projected revenues and expenditures.

One set of proposals involves larger fare increases. These would inevitably trigger higher ridership losses than the proposed ten cent level, but this is really uncharted territory for the TTC. Moreover, there is the question of perceived value, and whether riders feel they would actually be ripped off by a fare increase if service did not materially, and fairly quickly, change for the better. Paying more for what many perceive to be inferior and declining service is no recipe for retaining customers.


In addition to fare increases, there is the question of the many forms of discount fare. The table below shows the estimated “cost” to the TTC of these fares. This is the sort of issue where fare:ridership elasticity can get very tricky depending on the nature of each market and the extent to which elimination or reduction of a discount is considered to be “unfair”. A related problem is the TTC’s perennial treatment of these concession fares as a cost, as if giving people cheaper rides drives up the cost of service, or represents revenue that might be available if only we could get everyone to pay a higher fare. This is directly contrary to many City objectives to reduce barriers to travel among many groups of citizens, and the recognition that mobility has a value.


Note: The 8 million lost rides by children does not incur a cost because they are travelling free today. However, it is an indication of how much more riding is done today by children, and the degree to which this unexpected bump in one rider class is masking declines in others, notably adults.

Service Cuts

Another way to trim the TTC’s budget gap is to roll back service improvements, possibly even to levels below those imposed by the Ford/Stintz regime. This would be a bitter pill for the “Transit Mayor” to swallow, and in the midst of such grandiose spending plans as we see on major capital projects, the idea that these services would not be funded should be deeply embarrassing.


This table shows quite clearly that the usual poster children for “waste” in the transit budget will not yield a great deal of savings, notably the “low ridership, high subsidy routes” and the night services. The big money is to be found is reducing Service Standards, and rolling back both the 10 minute network and the full service 19-hours-a-day policy.

Here are the current loading standards for TTC vehicles. Note that these are based  on averages over a peak hour and some vehicles will have more passengers while others are half-empty either due to bunching, or because they are short-turned and are of little use to many riders. The TTC only reports averages, not max-min values nor standard deviations.


As things stand, the service budget for 2017 includes:

  • A 0.4% increase over the budget for 2016 for the target ridership level of 545 million, but no provision for ridership growth.
  • Annualization of improvements made in 2016 including express buses and earlier Sunday service.
  • Restoration of full streetcar operation on all routes including the conversion of 511 Bathurst and 514 Cherry to Flexity operation.
  • Opening of the Spadina extension (TYSSE) in December 2017 and concurrent reductions in contracted service for York Region.
  • Provision of new bus service to the Renforth Gateway.

Major Changes in Expenses

Several cost areas will contribute to the increase expense budget for 2017:

  • Collective bargaining agreement increases: $24.3m
  • PRESTO commissions and new faregate maintenance: $14.5m. This cost is almost entirely due to PRESTO fees because maintenance costs on the new faregates are largely offset by savings in maintenance on the old ones. At this point there is no offsetting saving shown for staff reductions due to PRESTO, but a discussion of this comes up later in the report.
  • TYSSE opening: $6m for 2017, projected at $30m for years following. Note that the extra cost of operating the extension is equivalent to revenue from a 1% property tax increase.
  • Cap & Trade: A $5.2m additional cost in fuel. According to the report “This is expected to increase the TTC’s diesel costs by 4.7 cents per litre and its natural gas costs by 3.3 cents per cubic meter.” The TTC has not produced consolidated figures showing the combined effect of price changes in fuel, the benefits of hedging, changes in consumption, and the new tax.
  • Hybrid bus battery modules: $8.5m. This is an operating cost that the TTC seeks to transfer to the Capital Budget as described earlier.
  • Accident claims: $6.2m. Actuarial evaluation of existing claims indicates a need to increase the provision for settlement. TTC self-insures except for disaster coverage.
  • Traction power and utilities: $5.5m. Again, it is clear that “lower energy costs” of $12m cited as savings earlier in the report have offsets elsewhere in the budget.

A full list with explanations is in Appendix D of the report.

Workforce Effects

A perennial issue at City budget time is the matter of “headcount”, to the point that some Councillors fetishize this to the exclusion of any other measure of a budget. Transit service, of course, requires people to operate and maintain it. If you buy a bus or a subway train, someone has to drive it, someone has to maintain it, and in the case of a subway, someone has to maintain the infrastructure the train runs on. The TTC projects changes in their workforce for 2017, but the big increases come in two areas: operators needed to provide service that will be paid for by others (Metrolinx) resulting from their construction projects, and additional staff needed to operate and maintain the subway extension net of savings on the surface network.


Of the 210 staff covered by 3rd party payments, 169 are for service operation in the Eglinton corridor where the TTC expects to receive $72.5 million from Metrolinx from 2017 to 2021. 36 are to provide frontline PRESTO support for which Metrolinx is responsible, but the TTC is acting as their agent. The remaining 5 are for vehicle repairs that are charged to others.

The staffing required to open a subway extension is considerable as shown in this breakdown for the TYSSE. Note that the saving in bus operators is considerably lower than the extra staff needed to operate and maintain the subway.


As for the PRESTO rollout, the lion’s share of savings from elimination of Station Collectors will be offset by the new staffing model for subway stations. Something that the TTC has never made clear is the degree to which the value of the Collectors was included in the fare collection costs that PRESTO is supposed to offset. Whether the “evolution” of stations will “meet and exceed customer expectations” is difficult to say considering that we still do not know the actual function and service quality the new Customer Service Reps are expected to provide.


A complete description of the purpose of all workforce changes is in Appendix G of the report.


The Wheel-Trans budget is a major source of cost pressure for the City of Toronto. Demand is rising quickly due both to demographic changes and increased eligibility mandated by Queen’s Park. It should be noted that the TTC (and other municipalities) receive no provincial subsidy for their accessible services. Indeed, the cost of running WT in 2017 will be about 90% of the value of the operating and capital gas tax contribution Ontario makes to Toronto. This is not to suggest that increased WT service is bad, indeed it is long overdue, but to show the relative level of provincial support for transit generally against the cost of providing just this component.

Wheel-Trans expects to carry 28% more trips in 2017 than in 2016 which itself is expected to be 14% over 2015. WT has almost no revenues (fares cover a trivial amount of the total cost), and this rate of increase means a big jump in subsidy requirements from the City. The extra demand is projected to add $24m to the WT budget offset by only $1.5m in fares.

There is a very small increase in WT workforce because the additional trips will substantially be handled through contract services, not the TTC’s own fleet. Indeed, the TTC projects a reduction of trips carried on WT vehicles as trips shift to other modes and as the new “Family of Services” program diverts more trips to being partially served by the conventional system (i.e. WT handles the “last mile” portion of a trip between a subway station and the rider’s origin and/or destination).

How successful the TTC will be in shifting its WT demand between various types of service remains to be seen. This will involve not just a more complex booking system, but also the ability to ensure that connections between legs of mixed-mode trips actually work as planned.

Accessible transit is a fast-growing part of Toronto’s network, and City Council should ensure that it can be properly funded without endangering the base system used by all riders, including ambulatory WT passengers who can, in part, ride conventional transit if it is “there” and not crowded or erratic beyond their endurance.

Ridership Growth Strategy

At Budget Committee meetings, there has been talk of a new “Ridership Growth Strategy” to improve the TTC’s attractiveness and to return to an era of steady growth on the system. A report on this is supposed to be coming before the TTC Board early in 2017, although the rather grim situation painted by the budget report suggests this will be a wish list for the future. One might joke that it’s just the sort of thing someone might run on for re-election in 2018, if only there were a sense that there would be political support to pay for it.

Just keeping the TTC at the level it is now at in service and fares will be a huge political struggle with a Mayor and his supporters on Council who cannot get past their promise to limit tax increases and fund any growth or improvement from those mythical “efficiencies” we have heard about for years.

Anyone who looks through the list of “savings” in the TTC budget will realize that little of the reduction from that original $231 million gap for 2017 and the now-proposed $61m number is due to  management actually squeezing blood (or possibly gravy) out of a stone. Some is the luck of changing costs, some is a matter of accounting, and some is wishful thinking that the City will take on more costs without actually treating them as part of the “operating subsidy”.

The shell game continues.

16 thoughts on “TTC’s 2017 Operating Budget: More Creative Accounting (Updated)

  1. Steve – one of the budget items that you list is “provision of new bus service to the Renforth Gateway”. Any idea what’s planned?

    Steve: Nope. It’s a passing reference in the text with no details.


  2. Hope the TTC better not roll back the service improvements to what it is before May 2015 or else it will be a disaster! I was happy with those service improvements the TTC made last year! They really should be trustworthy to not change it again!

    Liked by 1 person

  3. ttcawesomefanner said: “Hope the TTC better not roll back the service improvements to what it is before May 2015…”

    If they do, I think that John Tory should hold multiple press conferences to ensure that people associate him with the changes. I suggest one the day of the budget decision, another when the schedules come out, and a third one the day the cuts come into effect.

    I haven’t seen the “Transit Mayor’s Podium” out in a while; it looks like it could be time.


  4. TTC wants to increase fares again. There was a time when I used to pay my fare honestly but when oil prices went up, TTC increased prices and when oil prices went down, TTC increased prices and when ridership went up, TTC increased prices as prices increase with demand and now when the ridership is down, TTC is still increasing prices even though economics suggests that prices should go down with decreased demand – this is more than just creative accounting, this is creative economics as well. The bottom line is that TTC will increase prices no matter what and while I love your blog Steve, I am proud to admit that I don’t love the TTC and I am one of the most dishonest TTC users ever but at least I am honest about being dishonest. The fact that the TTC has not been able to prevent me from using their crappy service for free speaks volumes on their competence. Someone more competent like Steve should become the head of the TTC’s enforcement activities so that the TTC can have better luck in cracking down against people like me. The TTC does succeed in catching small fish every now and then but what about serial fare evaders like me who has not paid a cent in so many years? I do honestly pay GO, UPX, MiWay, Brampton Transit, YRT/VIVA, DRT, etc not because I can’t manage to travel for free (been there, done that) but because there is vastly better service than TTC can ever hope to provide.

    Steve: No, you are just justifying stealing free rides with a bunch of economic sophistry.


  5. In this year’s round of Transit Fantasy Budget, I’d raise PRESTO/Token fares in the fashion of TTC to $3.25, cash fares to $4.00, and bring the Metropass multiplier down by however much the raise in cash would give (Net ~$43.9M, ~-12.2M riders). I would harmonize cash fares (Net $6M, -2M riders*). I’d cut the MDP/VIP discounts (Net $16.1M, -1.4M riders). No Stabilization Draw (-$14.4M).

    This leaves council with the ability to either axe the Batteries & Depreciation shift to Capital and increase the supplement by $38.7M or to allow them to shift and either have a $1.2M increase in the budget or a small draw from the stabilization reserve.

    *The table for revenue/impact of harmonized cash fares doesn’t make sense. With the 10/20/25 cent fares all being harmonized from $0.05 more than the PRESTO to $3.25. Shouldn’t this table be the other way around? Prices rising from $2.00 to $3.25 nets $3M with a loss of 1M riders, but prices rising from $2.25 TO $3.25 nets $6M with a loss of 2M riders?!

    Steve: Yes, that table didn’t make sense to me either.


  6. Really the TTC needs to reinstate the child fare 5-12 so that the TTC can boost more money for improvements. Elimination of child fare hurts the TTC budget.


  7. So do the pricing tables mean that the Metropass will be around in 2017 — possibly even including MDP and VIP plans — unlike the prognostications of TTC brass when Presto was launched?

    Of course it’s possible that whoever did up the tables wasn’t privy to Higher Strategic Decisions, and just went with what has been done before, i.e. including Metropasses in any fare increase scenario.

    Steve: The TTC has always intended to keep the pass for at least a six month overlap with the introduction of Presto, and they are not yet even at a point where Presto can handle the volume of transactions these passes represent. With that sorted out, there would still be some equivalent of a monthly pass (Presto supports this for other systems today) and it needs to be priced regardless of what we call it.


  8. Steve writes:

    “Wheel-Trans – The Wheel-Trans budget is a major source of cost pressure for the City of Toronto. Demand is rising quickly due both to demographic changes and INCREASED ELIGIBILITY MANDATED BY QUEEN’S PARK [my emphasis]. It should be noted that the TTC (and other municipalities) receive no provincial subsidy for their accessible services. Indeed, the cost of running WT in 2017 will be about 90% of the value of the operating and capital gas tax contribution Ontario makes to Toronto.”

    This is absolutely ludicrous and reminds me of Mike Harris downloading all sorts of costs that had been previously funded by the Province onto cities and towns so Queen’s Park’s budget would look rosy (although the ink would have been black and not rosy red for a deficit). It sucks for the Liberals that the pinch point regarding the aging population in Ontario seems to have been starting while they’ve been in power but based on StatsCan and Ontario Ministry of Finance numbers, there will be an up-to-fourfold increase in the population of 70+ -year-old residents in the period to 2041 (See 2nd table here).

    This means – very clearly – that in larger centres that offer WheelTrans or similar services, especially if low-floor buses or streetcars are not in the cards for a while, the demand for this service will most certainly go up. And this doesn’t account for access for current physically disabled patrons (not age-related) or those requiring specialized mobility services who are not able to afford retrofitted personal vehicles to get around.

    It is incumbent on the Province that, if they are mandating through law a more equal access to transit for all citizens – and I am not disagreeing with that at all – that they cannot also directly impose the associated higher costs on the budgets of those communities without some sort of financial assistance. I would call it a case of “Mobility Noblesse Oblige” (MNO).

    However, given how the Province has been, with regards to funding “regular” public transit – and this is operating costs, not capital costs – those who need “specialized” services aren’t even going to get on the bus, never mind what the law says. And I do also blame the City Councillors for this travesty in Toronto, because – Bombardier Flexity streetcar production/delivery problems notwithstanding – they continue insisting on building castles in the sky – with the blessing of the Province, for the political outcomes – and somehow just shrugging off the increasing frustration of transit riders at the increasing costs to them for the privilege of using “The Better Way(TM).”


  9. I’m surprised there is still no Presto monthly pass…even if it was only available in a limited quantity via their website or something…if they are having backend capacity issues it makes sense to ramp up rather than go “all in” one month…

    The board should suggest a 0% increase in fares, contingent on getting the funding from council, otherwise a 15 cent increase…this puts the ball in council’s court…better to hold council’s feet to the flames as far as funding goes…


  10. The TTC adamantly refuses to introduce a 2 hour timed transfer because of perceived revenue loss. This ignores potential ridership increase. Yet, at the same time they justify free unaccompanied children and the significant revenue loss to get them accustomed to using transit when they grow up. Yeah, right. Crappy service will really encourage children to grow up being transit users. NOT!

    While I am at it, is there any adjustment in these budget figures for revenue loss due to implementation of Presto fare gates? For MONTHS people just walk in past the fare collector and ignore paying, yet there is no enforcement people to stop this. Dufferin Station alone lost a fortune.

    Steve: This has never been mentioned. To me, the implementation of those gates has been very poorly handled. They were “sold” as an integral part of Presto, but were only needed (a) to make automatic entrances accessible, and (b) to provide for future fare by distance which has never been approved by TTC or City Council.


  11. Raymond: While I am at it, is there any adjustment in these budget figures for revenue loss due to implementation of Presto fare gates? For MONTHS people just walk in past the fare collector and ignore paying, yet there is no enforcement people to stop this. Dufferin Station alone lost a fortune.

    Trust me, it’s happening all over and not just Dufferin. North York Centre took 6 months to install Presto fare gates during which it was free entry for all even as sometimes as many as 8 customer service officers stood by to give people directions + one or two collectors and sometimes even supervisors. Even now when the new gates are finally there, people just walk in for free and nobody cares as the collectors, etc are getting their salaries anyhow.

    This I overheard on a TTC bus: the on-duty bus driver talking to another bus driver standing near the driver’s seat (while the bus was on its way to Scarborough Town Centre) and making his case that bus drivers should get paid more than streetcar drivers as streetcars he claimed are easier to drive and streetcar drivers have less responsibility as they don’t have to collect fares and collecting fares also results in at least one bus driver being assaulted every single day and so bus drivers with more responsibilities and dangers than streetcar drivers should get paid more than streetcar drivers. I never thought of it that way but it makes sense though not the best thing to do talking while driving a bus.

    Steve: I think one common thread in all of this is that Presto and the new hardware are perceived as (a) a badly implemented system and (b) not a TTC project. There is no buy-in when staff see something badly handled like this.

    As for bus and streetcar pay rates, I could point out that streetcar drivers work in much worse traffic (that lasts for more hours a week) and are responsible for many more riders. Maybe we should pay subway operators a fortune, particularly for one person crews! Utter nonsense.


  12. John Tory needs to stop being a hypocrite: he said “there are pressures on The TTC budget” but then feels all budgets need to cut 2.6%. If he wants to actual do some good for transit grow a set and bring the car registration tax back or a gridlock congestion charge. Enough of treating TTC users like second class citizens especially when the effects costs us approximately $6 billion if not more in lost productivity, unacceptable. Tired of this crap by the last two mayors of Toronto 😦


  13. Two inspectors boarded my eastbound 501 at the first stop out of Long Branch loop. Out of about ten or twelve passengers, three were dinged.

    One was a guy who (to my eyes) looked like a habitual fare-dodger. He turned belligerent when challenged.

    The two others were possibly students and had Presto cards that they hadn’t bothered to tap. I don’t know if they were playing the Presto lotto or just hadn’t thought to tap (they boarded through the rear door at Long Branch loop, probably coming from Miway or a GO train). All three were taken off the streetcar a couple of stops later.

    Unlike transfers or Metropasses, it takes a while to check a Presto card. I have a hard time imagining how POP checks would be done on a full vehicle once everyone is on Presto. In the meantime, people board through the rear doors, possibly with valid transfers or passes….or possibly with Presto cards that they just forget to tap.


  14. When I was in France a couple of years ago they would have herds of scruffy looking characters who would get on a vehicle en masse, block access to the fare machines and check everyone’s fare for validity. They also checked anyone who got off. They would have upwards of 12 people who would suddenly hang badges around their necks and start checking. Anyone with an invalid ticket was removed. The whole process took about 1 minute.

    All the fare inspectors I saw were in plain clothes and looked scruffy. They never more uniforms but just hung a badge around their neck after they boarded or flashed the wallet.


  15. Ed said: “I have a hard time imagining how POP checks would be done on a full vehicle once everyone is on Presto.”

    The simplest solution with Presto is to have the fare inspectors force a tap on with some sort of “punishment charge” for those who forgot to tap and then go after those with insufficient funds on their cards.


  16. Brian said:

    John Tory needs to stop being a hypocrite: He said “there are pressures on The TTC budget” but then feels all budgets need to cut 2.6%. If he wants to actually do some good for transit, grow a set and bring the car registration tax back or a gridlock congestion charge. Enough of treating TTC users like second class citizens especially when the effects costs us approximately $6 billion if not more in lost productivity, unacceptable. Tired of this crap by the last two mayors of Toronto😦

    Speaking of second class Citizens…

    First off I agree with your operational concerns and am very serious about my next statement. The most realistic way to rectify yours concerns going forward is to build and the SSE and Eglinton East LRT within the next 10 years (as well as the other current projects). At that point will be enough pressure from all sides of the City to improve TTC operations. Although that would likely mean atleast 4 more years of Tory. Personally I’ll take this slippery tongue over extreme polarization of City’s geographically biased Left party and subsequent polarizing politicians like the Fords any day.

    As of right now we have large areas of this City who have little reason to share your concerns. It’s just the reality. As much as the current Politics are farce (lust like they always are) I do see a brighter future for this City “if” we can raise the appropriate taxes to fund the projects on the table with minimal delay and the usual divisive opposition.


Comments are closed.