Updated May 26, 2014: Philip Webb caught two typos in numbers in this article which have been corrected. These changes do not affect the overall argument.
The madness that passes for political policy in Toronto continues in the provincial election with a proposal that a Tory administration under Tim Hudak would transfer control of the rapid transit system to GO Transit as a regional asset. The conventional wisdom is that the subway on its own would be “profitable”, and that Toronto would be stiffed for the money-losing surface network.
Quite bluntly, any claim that the subway makes a profit and could be uploaded at no net cost to Queen’s Park is pure bunk, and it says something about the quality of Hudak’s advisors that they don’t seem to know this (among many other fiscal facts of life). Just like the operation of a house or a car, two things many voters must deal with day-to-day, there are two budgets:
- Operating: Here we have the bills that roll in regularly such as taxes, utilities, insurance. Unless we are renting out our homes or vehicles, there is no offsetting revenue, but in the case of the subway, there are fares and other much smaller sources of income.
- Capital: Now and then, major expenses come along such as a new roof or foundation repairs, a new furnace or other appliances, fixing the plumbing and electrics, building that nice new patio you always wanted. These don’t happen often, and the expense covers an asset that should last decades, but some level of capital spending is unavoidable.
I have omitted mortgage costs here because they do not have a direct equivalent in transit budgets where the cost of borrowed money is not visible. If this were included, then capital-intensive modes like the subway would have a higher operating cost with the debt service charges included.
Capital Costs for the TTC
The TTC has a 10-year rolling budget projection for capital costs (see 2014-2023 Capital Budget). The appendices contain a detailed breakdown by area and project with a summary in Appendix A and the details in Appendix B. Those who have a real appetite for mind-numbing detail can browse the two volumes of the “Blue Books”, all 1,800 pages or so, that are available for review on request, but by personal visit, not online.
This table subdivides the areas within the 10-year budget by mode. The allocation to each mode has been done as follows:
- Where an item is clearly for a single mode (such as “purchase buses”), the cost is allocated 100% to that mode.
- For a few large items that contain many projects (notably “Buildings & Structures”), I delved into the line items to determine how they should be allocated. The percentage split reflects that breakdown (details not included here).
- For other lines, I allocated costs based on my understanding of the relative portion of the asset related to each mode. Some will quibble with the values I chose, but it is important to note that these are not major budget lines, and shuffling the percentage splits will not have a big effect on the totals by mode.
The TTC’s capital plans (including the so-called “below the line” projects for which they do not yet have funding), total about $8.6-billion over the 2014-2023 period. This breaks down as:
The streetcar amount and percentage are high because this part of the capital plan includes the complete fleet replacement and the initial stages of expansion not to mention the construction of a new maintenance facility. Spending on these budget lines will not continue at the same level into future decades.
Similarly, some subway costs such as the new signal system and fleet replacement do not recur at the same level decade-by-decade.
That said, it is clear that over half of the TTC’s capital budget, with no provision for system expansion, goes to the subway network at an annual cost of about $450-million. If Queen’s Park takes over the subway, they must also take over this capital expense.
The only long-term capital funding from Queen’s Park today is the gas tax revenue at an annual rate of about $71m (Toronto receives more, but directs the rest to the operating budget). Ottawa kicks in about $154m in gas taxes bringing the total from this source to $225m. However, only about half of that can reasonably be “uploaded” to Queen’s Park as part of a subway takeover, and so there is only about $113m in gas tax that should shift to the province as part of the transfer.
This would leave a net additional cost to Queen’s Park of $337m per year for capital maintenance.
The Operating Budget
The TTC’s Operating Budget covers all modes and does not break out the subway system as an explicit cost centre, although this breakdown has appeared in the past. On a broad scale, the budget looks like this:
Although the TTC does not publish subway and RT operating costs, we can get a sense of their magnitude by working backwards from the costs assigned to the surface routes.
Taking the reported daily cost and factoring by 300 to get an annual estimate (this treats weekends as the equivalent of one weekday), we can figure out the “leftover” costs that must be due to the subway. Note that 2012 costs are used here because these are the most recently published on a route-by-route basis.
|Daily ($-million)||Annual ($-billion)|
|Bus System Cost (2012)||2.9|
|Streetcar System Cost (2012)||0.6|
|Surface System Cost (2012)||3.5||1.050|
|Inflation to 2014 @ 2%||1.090|
|Total Budget (2014)||1.600|
|Net Subway Cost||0.510|
(If someone at the TTC wants to challenge this estimate or correct it, I would be more than happy to see numbers that, for some reason, disappeared from the route-by-route allocations years ago.)
A common refrain is that there are millions to be saved by conversion of the subway to automatic operation, or even to unattended trains. I wrote about this in my critique of the Neptis Metrolinx review. The total number of subway operators is only about 605 out of a 10,700-strong workforce (2012 data). Some savings would be possible, but these would be offset by the cost of roving passenger assistance agents and security staff.
(The subway has a large fixed cost associated with the operation and maintenance of its complex infrastructure and stations that is independent of actual usage or level of service.)
When we combine the operating and capital maintenance costs together, we get:
|Total Annual Cost||0.935|
|Current Capital Subsidies||0.225|
|Subway Subsidy Share (50%)||0.113|
|Net Cost of Uploaded Subway||0.822|
However, the total revenue obtained by the TTC is only $1.17b, and roughly half of that “belongs” to the surface network. (There are actually more riders on the surface network than on the subway although many trips involve both networks.) If the subway actually is “owed” half of the revenue, $585m, this would roughly balance the annual operating cost with little left over to fund capital maintenance (or to pay a “dividend”).
Cooking the Books
The subway can be made to appear profitable simply by changing the revenue model, notably by splitting it off as a separate fare zone. Even if a dual fare were granted to riders who start out on a TTC bus or streetcar (rather like the co-fare provided to GO riders who start their trips on a local bus system), the effect would be to increase fares for all riders whose travel uses both modes.
We can have a robust debate about whether the subway should be priced this way, and given the projected operating losses of extensions into York Region, the idea of a subway surcharge is superficially attractive. However, this goes directly against the design responsible for the TTC’s success as the city grew — providing a single fare ride with free transfers between routes — and also risks penalizing those riders who cannot organize their lives to avoid the subway system.
A long-delayed policy discussion at the TTC should examine the practicality and effects of moving to a new fare system such as zones or a time-based model. It does not matter which one we choose or how we tweak the scheme, some riders will pay more and some will pay less. After all, the talk of “regional integration” is at its heart a desire by riders who must cross the 905-boundary for a lower combined fare on their journeys. “Integration” is not simply a matter of having one card that would pay for all travel but with an unchanged tariff.
The effect of rebalancing fares can be offset either by raising subsidies, or by accepting that some riders will pay more. This discussion would be central to uploading the subway to Queen’s Park because there must be some way to decide how fares are calculated and revenues are split between the systems.
Another crucial issue is that of service standards. We have seen this battle in Toronto. Some believe that transit should be available at reasonable crowding levels and good service frequency throughout the day. Others hold that service should only be operated when it does not lose too much money, and that vehicles should be stuffed full as the standard and target for service levels.
Nobody ever talks about the subway service operated with nearly-empty trains to the ends of the system until 2:00am every day. Would GO Transit provide such generous service on its rail network, and would the TTC’s frequent off-peak subway service be a victim of uploading? A Hudak government can hardly be expected to run “gravy trains” out into the suburbs for only a handful of riders when a bus would do just as well. That is, after all, the GO Transit model.
Whenever there is talk about “integration” or “uploading”, we never see a full explanation of just what is proposed. This can leave voters and transit riders hoping for their ideal world such as a single TTC fare for trips well beyond the 416 into downtown or a reduced cost of trips including GO and other systems. What they actually get may be a surprise that pandering politicians would rather not discuss, presuming they even understand that issue.