On Wednesday, March 21, Toronto Council will consider a report recommending that the Sheppard rapid transit line be built as an LRT from Don Mills Station east, initially, to Morningside. This is the same scheme that was on the table in the Metrolinx 5-in-10 plan, and approval of this recommendation will more or less put Transit City back on track where it was before the election of Mayor Ford.
There is a main report and many background documents, including an alternative subway proposal, what might be called the “Chong Dissent” from the otherwise pro-LRT conclusions of the panel. This article provides a summary of material from many sources. For the definitive word, please refer to the originals as I am not going to attempt to cover every detail here.
As a general observation, the materials present a review of the situation in considerably more detail than we see for many transit planning decisions, notably those surrounding recent budget debates. With luck, and with a less transit-hostile TTC board, we might see the same level of interest turned to basic questions like “where’s my bus and why can’t I get on when it shows up”.
The pro-subway folks claim that the report is biased, that it is hogwash, and advance their own dissent purporting to show the superiority of a subway option. The misinformation and factual errors in this dissent are disconcerting, putting it mildly, considering that billions in provincial spending and the future development of our transit network might have depended on such twaddle. I will turn to this in detail later in a future article.
The panel’s report evaluated three options for the Sheppard corridor:
- Option A: LRT from Don Mills to Morningside
- Option B: Subway from Don Mills to Scarborough Town Centre
- Option C: Subway from Don Mills to Victoria Park, and LRT from there to Morningside
This evaluation used input from the TTC, Metrolinx, City Planning, Toronto Transit Infrastructure Ltd. (Dr. Chong’s home pro tem), and material from other sources. Each of the options was rated against various criteria to develop a score showing its relative benefit on each point, and the scores were summed to produce an overall ranking. This sort of scheme is always open to misuse both by biased scoring and by inappropriate weighting of components. I will discuss each of the component scores as I come to them in the report.
The panel recommends Option A, but also urges the City Manager to develop a communication plan about the significance of transit, and of the Sheppard corridor in particular, in Toronto. The panel also urges Queen’s Park to accelerate work on the Metrolinx Investment Strategy which will be critical for funding projects beyond the 5-in-10 plan.
The panel also wants to see the City Manager bring forward a comprehensive transit plan that would be integrated with the Official Plan. The lack of such a plan goes back to a period when the TTC jealously guarded its role in transit, and the current OP does not contain a full transit component. Oddly enough, this situation prevented a subway-oriented plan, then the TTC’s pride and joy, from acquiring the imprint of Official Plan status. Indeed, the OP contains a few well-hidden references to LRT (including an illustration of an LRT right-of-way at a redeveloped Eglinton East and Kingston Road). This is a chance to finally bring Toronto’s plans in line with each other after many decades of drift.
The financial context of the deliberations was quite simple: Queen’s Park is prepared to pay up to $8.4-billion (2010) for a set of rapid transit lines in Toronto, and Ottawa has a further $333-million on the table. To date, there is no municipal commitment to supplement this spending. Queen’s Park wants whatever is built to make sense in a regional context and, by implication, not simply be a vanity project to suit political or personal aims. If the City wants a plan different from the one Metrolinx was working on, then it must absorb the cost of any work that cannot be used as part of the revised project. Still unclear is the question of who will pay for the work on the Eglinton all-underground option given that Council never approved this scheme, and Queen’s Park foolishly went ahead on the strength of a deal Mayor Ford could not deliver. Queen’s Park is also not prepared to tolerate any further delays.
Much of the $8.4b is already spoken for on projects that have Council’s blessing: the Eglinton-Crosstown LRT (including the SRT conversion and extension) and the Finch LRT. All that remains is to decide what should be done on Sheppard.
Option A is a 13km LRT line starting at Don Mills Station and running east to Morningside with a GO connection at Agincourt. The underpass for this is already under construction and would have been required for GO whether or not the LRT was built. Possible add-ons include an extension east on Sheppard to Meadowvale and/or south via Morningside to the University of Toronto Scarborough Campus (UTSC). Presuming that the SRT is extended up to or beyond Sheppard, there would be a link between the two routes. This option has 25 stations and a capital cost of $1b.
Option B is an 8km subway extension from Don Mills to Scarborough Town Centre. The line leaves Sheppard Avenue east of Kennedy to swerve south and into STC. To maintain a connection with the GO service, Agincourt Station would be shifted south. There would be a station at Progress between Kennedy North and STC, but all riders from eastern and northern Scarborough would access rapid transit via bus feeders. This option has 7 stations and a capital cost of $2.7-3.7b depending on some underlying assumptions in the design.
Option C is also 13km long, but runs as a subway extension to Victoria Park where there would be a link with an LRT line east to Morningside. The link with GO at Agincourt would be the same as with Option A. This option does not include a major bus terminal at Victoria Park, only basic facilities for an LRT interchange, nor does it assume the need for additional subway train storage. Most of the routes now serving Don Mills Station would remain there to avoid the extra cost of duplicating that terminal a few km to the east. This option has 26 stations (Victoria Park counts as 2) and a capital cost of $1.5-1.8b. About $900m of this is attributable to the subway extension.
City Planning and the Official Plan
[This section originated with the City Planning department and statements in it do not necessarily reflect the views of the panel.]
The planning department argues that several criteria should apply to evaluation of the options:
- Access and mobility
- Community impact
Several of these have multiple dimensions. For example, “community impact” can be seen from the viewpoint of disruption of what is there now, as well as from the outlook that a new transit line will stimulate positive changes. “Cost” involves not only the actual cost of a line in the Sheppard corridor, but the alternate spending elsewhere that might be precluded for a more expensive option.
Potential growth in the Sheppard corridor would be higher with a subway line, although there are two important caveats. First, much of the growth would be concentrated at the west side of Scarborough where the subway actually provides service, and there is no guarantee that the level of growth would actually occur given competing sites elsewhere in the GTA. Second, the subway projections run out to 2061 while the LRT-scenario projections go only to 2031.
Although the 2061 numbers are higher, they are not generally proportionately higher given the longer period involved. For example (see Table 2 on Page 16), population growth to 2061 is projected as 24k on a base of 43.9k (55%), but employment growth is only 10k on a base of 28.9k (33%) in the area between the 404 and Agincourt. Population and employment growth is much more robust in the Progress-STC corridor for the subway option, and this implies that the modellers foresee very substantial redevelopment of these lands. Oddly enough, the larger effect of a “Sheppard subway” is to stimulate growth around STC, not along Sheppard itself.
(Table 2 contains long footnotes qualifying the values shown there, and I urge readers to digest this information before using the numbers in their raw form.)
Because the major source of projected growth is in the STC corridor, this causes major differences in effect of various options. A line staying entirely on Sheppard will benefit riders north of the 401, but it will do nothing to stimulate the redevelopment of vacant and light-industrial lands around STC. Is the function of a Sheppard subway to serve Sheppard, or to facilitate the long-held dream of a dense Town Centre and the development aims of current landowners?
Ridership forecasts for the 2061 model are not available because this would be the complex product of actual development patterns over half a century and the evolution of the transit and road network in the area. For example, if a frequent and financially attractive service were provided on both GO’s Uxbridge line and on the CPR corridor through Agincourt to North Pickering, a great deal of traffic that might otherwise be modelled onto a local transit system of any flavour might shift to the regional commuter rail system. The absence of such foresight was a major flaw in the transit plans of the 1980s which loaded all population and riding growth onto the TTC system at a time when it was clear GO would evolve into a much more important part of the overall network.
In the 2008 EA for the Sheppard LRT (Table 3, Page 17), the TTC projected that an LRT from Don Mills to Morningside would yield a peak hour/direction demand of 3k in 2031 westbound approaching the Consumers Road area, while a subway to STC would yield 4.2k. Demands at Yonge on the existing subway would rise from 4.5k to 6k (LRT) and 7.8k (subway). These forecasts were based on the much smaller actual growth achieved in the Sheppard corridor as compared with the very optimistic values used in the 1992 subway EA.
The projected ridership, as we shall see in more detail later, is also related to the actual travel pattern of people living in Scarborough. Contrary to opinion that sees most of Scarborough desperate for a subway to take residents anywhere else, there is a very large amount of local traffic within Scarborough in both the east-west and north-south directions. Neither of these is particularly well-served by a subway line whose primary benefit is to enhance the development potential of lands around STC.
On page 18, planning staff state:
In the end, City Planning staff concluded that in the absence of reliable long-term ridership forecasts, support for a subway at this juncture would be based on a long-term city-building vision.
They go on to cite a considerable number of preconditions, and conclude:
Unless these conditions can be met, a subway is not warranted, and the LRT would be a viable option to meet transit needs in the corridor over the next 20 – 30 years, and may be sufficient beyond that. City staff is concerned, however, that the LRT would under-perform as a City-building option if it doesn’t link to the Scarborough Centre.
Again we see that the goal of “city building” focuses on STC, not on Sheppard itself. Moreover, it is unclear how attempts to force STC to become a major development node would fare in a regional context over the long haul given the singular failure of such efforts here and at North York Centre for past decades. Indeed, the very incentives needed to seduce developers away from their preferred sites may work contrary to other financial tools which seek to capture added value from the new development.
When we look at the actual travel patterns in Scarborough (for example on page 47 of the report), we see that one third of all trips in the planning district north of the 401 are local to that area, and a further 25% flow north into Markham and south across the 401. The remainder goes to downtown, midtown and various parts of North York. The subway option does not address this distribution of trips well in part because the line ends at STC and in part because it does little to serve local demand. (This topic is explored in background papers to which I will return in a future post.)
[This section originated with the City Finance department and statements in it do not necessarily reflect the views of the panel.]
This is a long section of the report examining various options for financing transit expansion and, specifically, the cost of the subway options beyond the funding already committed by Queen’s Park and Ottawa. Three main sets of funding streams are examined for their potential and for the complexity of their implementation, if any.
- Conventional funding which requires no legislative changes (grants from senior governments, municipal debt)
- Private sector funding through an investment tied to future revenue growth triggered by transit construction (provincial approvals required)
- New funding sources (e.g. taxes, most of which require provincial approval)
Table 4 on Page 20 summarizes the cost of the options and the sources of funding currently available. Note that this table includes projects already approved by Council (Eglinton-Crosstown LRT, Scarborough RT conversion and possible extension, Finch LRT).
- For the all-LRT plan, there is no requirement for a City capital contribution.
- For the Sheppard subway option, there is a shortfall of $1.7-2.7b that would fall to the City, possibly offset by $0.5b saved in terminating the SRT at McCowan rather than Sheppard.
- For the subway-LRT option, there is a shortfall of $0.5-0.8b that would fall to the City.
In all cases, the estimated cost of the Eglinton line may rise depending on options selected west of Black Creek and the cost of widening Eglinton in Scarborough to retain the HOV lane that would have been traded off to the LRT right-of-way. Known “sunk costs” for work on options that would not be pursued in a subway or hybrid option amount to $45.9m and these have not been included in the City’s share.
In a conventional funding scenario, the City would debenture any costs not covered by subsidies already committed to the project. Allowing for a likely staging of construction and timing of the borrowing, this leads to a total tax increase over a seven-year construction period of 4.2-6.5% for the subway option. Translated to annual increases, this would require an additional 0.6-0.9% property tax hike for seven years to fund the debt for the subway option. The hybrid option requires much less City financing with annual tax increases of 0.2-0.3%.
For the subway option, the difference lies in the two estimates for the subway’s construction. As I have already written, there is good reason to doubt that the (lower) Chong estimate for the subway (mistakenly called the “Metrolinx” estimate) was prepared on a consistent basis with the TTC’s estimate. Much of the difference is a question of underlying assumptions.
Council imposes on itself a debt ceiling so that the percentage of tax revenue devoted to debt service does not exceed 15%. Obviously, if taxes go up, the total borrowing can go up too, but there is a limit to how much additional property tax is politically tolerable, and whether all of the headroom available should be devoted to a Sheppard project.
In the private sector funding scenario, the City would commit future revenues from subway-related developments to pay down private sector investment in the line. This is little different from borrowing the money on the open market and hoping that revenue growth will offset the extra cost, but there is the precept that a private investor might have the incentive to build cheaper (presuming they were actually the builder, not just a bondholder) than might otherwise occur. There are three sources of funding that could be used:
- Tax Increment Financing (TIF)
- Development Charges (DCs)
- Sale of municipal properties in the corridor
TIF depends on the assumption that new development will actually occur along a subway line, and as we already know, most of this is projected for the lands around STC, not on Sheppard. With all those new buildings will come incremental tax revenue that would pay down the debt for the subway construction. However, there is no guarantee that the development will occur up front, if at all, and the City could be left responsible for debts whose TIF-based revenue might not show up for decades.
DCs assess a premium onto a new development (typically a per-suite surcharge for condos). Developers don’t like them because the cost is passed on to the buyer who may choose to look elsewhere rather than buying a condo on Progress Avenue.
Municipal property sales are one-time revenue sources, and the City may or may not have land available in the corridor and marketable for sale.
KPMG looked at the TIF option for the Sheppard subway, but the scope of their TIF Zone (the area subject to dedication of incremental taxes) stretched along the entire Sheppard and Eglinton-Crosstown-SRT corridors for the next 50 years. This would yield $5.3b in new taxes, but much of this money would go to pay down the Sheppard debt, not to build or maintain other infrastructure or to provide municipal services to the growing population and employment centres in these corridors. It is hard to credit the taking of substantial funds for one comparatively small project (Sheppard) from such a wide swath of the City over such a long time.
DCs could raise $2.2b over 50 years based on roughly $2k extra per 2-bedroom condo, but it is unclear what the scope of this charge would be — how far afield would the affected properties lie, and how much future revenue would this sequester for the Sheppard financing. Getting $2b at $2k a crack is going to take a lot of condos.
In practice, the amount of money that could be actually raised in the short term (5 years) would be about $651m. This number is much lower than figures cited above because it takes into account what would likely happen, the risks to investors and the fact that current investors would be betting on a future, uncertain return.
Available municipal property in the Sheppard and Eglinton-Crosstown-SRT corridors is valued at $227m. Again, this figure is inflated by the inclusion of areas that have nothing to do with Sheppard in fundraising for that project, and the value of the properties — that might be needed to fund other civic projects — would be lost to pay for one subway.
Even with all of these revenue streams included, there remains a 26% shortfall in the City’s ability to finance its share of the Sheppard subway. This number is based on the lower “Metrolinx” estimate for the subway’s cost, and any increase beyond that figure would add to the shortfall in available City funding.
A further problem lies in the diversion of tax revenue that would otherwise flow to the general coffers, and the distortion of development patterns through incentives to build along the subway corridor. If development along Sheppard and Eglinton replaces development elsewhere, this means that tax revenue that would otherwise come to the City is bundled into the TIF/DC net. For example, a condo might not be built in the waterfront, but on Sheppard, and its taxes would be earmarked for the subway debt. This reduces the actual net revenue from TIF/DC sources to 59% of projected levels.
None of the additional servicing costs (roads, utilities, etc) caused by the new development would be offset by new taxes as these would be dedicated to paying for the subway.
There is a potential effect on the City’s credit rating, and hence on its cost of borrowing, because any subway-related debt is, eventually, backstopped by the City. If the expected revenues do not materialize, the City will be on the hook to pay the costs through other revenue streams.
The potential revenue from land sales is considerably smaller than might be thought because the most valuable parcels have already been earmarked for sale through Build Toronto, the City’s real estate development agency. This revenue has already been “counted” in future budgets as anticipated dividends from Build Toronto. It cannot be counted again as subway-specific financing.
Finally, Alternative Funding Streams (e.g. new taxes) could provide revenue to fund the City’s share of the subway project. Table 8 on Page 29 details these, and by far the single largest in the list is a toll on vehicle kilometres travelled (VKT). It is unclear whether it would be practical to implement all of the options cited here (for example, a toll scheme might run headlong into problems co-existing with a general tax on kilometres travelled). Large amounts of money are available over a 50-year timeframe if only Toronto (and Queen’s Park) has the will to levy the new taxes. Once again, however, one must ask whether the political effort to implement any of them could be saleable for only one project — the Sheppard subway — which is not high on most travellers’ priority lists.
A final consideration for any new revenue tools is that, to date, this entire discussion has focussed on one comparatively small expansion of the subway network, and proposed taxes would take money from a large part, in some cases all, of Toronto to finance that one project. There are many others, and these too need to be financed including:
- Pearson extension of Eglinton line ($1b)
- Finch West extension to Yonge ($.5b)
- Downtown Relief Line East ($3b) and West ($2.9b)
- Yonge subway to Richmond Hill ($3.1b)
- Don Mills LRT ($1.8b)
- Jane LRT ($1.5b)
- Scarborough-Malvern LRT (1.4b)
- Waterfront LRT (west) ($.5b)
This list does not include the Waterfront east projects, nor a number of GO Transit improvements. Although these may have funding in part or whole from Queen’s Park or Ottawa, the same set of revenue tools now eyed for Sheppard are in the list for the Metrolinx “Investment Strategy”. There is only so much money to go around, and we have to spend whatever we raise wisely.
Evaluation of the Options
At this point, I am going to close off part I of this article. Due to website interruptions today and other business, I have not had time to complete as much as I would prefer, but will return to the main report and its conclusions soon.