The Sheppard LRT Report (Part II)

In the previous article, I reviewed the three main options under study for Sheppard East as well as the comments of the City Planning and Finance departments on various related issues.

In this article, I turn to the Expert Panel’s evaluation of the options, their scoring system, and the question of bias in the process.

The analysis and scoring begins on page 39 of the Expert Panel Report.  The panel chose three broad areas for analysis, and subdivided each of these into three subcategories.

  • Funding & Economic Development
  • Transit Service
  • Sustainability and Social Impact

In each of the 9 subcategories, the highest possible score is 5 points for an overall raw total of 45.  However, the weights assigned to each group are different with Funding & Economic Development getting a weight of 3x, Transit Service 2x, and Sustainability and Social Impact 1.5x. Once the weights are applied, the total potential score is 95 points.  These values are normalized up to a “perfect” score of 100.

Table 15 on page 41 summarizes these scores.  In order that readers can see how the weights affect the outcome, I have recast these data to show the buildup of the weighted scores to a 100-scale.

Funding & Economic Development

Economic Development

The subway option will require “significant concentrated employment growth at North York, and Scarborough Centres, Consumers Business Park and Agincourt Secondary Plan Area beyond the 2031 forecast levels”.  Whether this growth will actually materialize is a matter of conjecture especially given the past three decades’ history of low growth.

Any new transit service will produce an economic uplift.  This has two components:  the scope of the effect (how far from a corridor the uplift is felt depending on mode) and the scale (how much a given mode adds).  These values are summarized in Table 16 on Page 42.  An important issue here is that the LRT is much longer than the subway option.  Therefore, its potential effect is greater even though both the scope and scale are smaller than with a subway line.

One effect buried under the covers is that “economic development” in the Benefits Case Analysis methodology used by the Metrolinx contractor, SDG, is that the cost of building a line counts as “economic development”.  In other words, if we spend $2-billion, this will produce twice the effect of spending $1b.  The problem is that a BCA looks at a single line, not at the best way to spend a fixed pool of capital.  If we have $8.4b overall, a combined view of the situation would yield the same spinoff benefits from construction.  However, the methodology skews this to favour subways which are more expensive than anything else.

The raw scores for Economic Development for each option are 3.71, 4.14 and 3.57.  It is impossible to determine the contribution of each element to these scores, but the fact that the hybrid ranks lowest suggests that the greater development benefit of a subway to Victoria Park is outweighed by the loss of the “economic impact” of a cheaper construction project.

Cost Effectiveness and Fiscal Sustainability

This item is built from three components:  cost per new rider, attracting new riders, and equity in transit investment.

Although the subway option generates more rides, the cost is much higher and so the cost per new rider is worst for this option.  On the remaining components, the scoring looks at the effect of spending on the Finch LRT and benefits in that corridor.  The total new ridership for Sheppard and Finch together would be higher than for a Sheppard subway alone.  While this is interesting and a useful way of looking at network benefits, not simply those local to Sheppard, this is a ranking criterion that should have been broken out consistently through the analysis.

Again, it is impossible to dis-aggregate the contributions of the three components.  The scores for the options are 4.43 (LRT), 2.14 and 2.57.  This reflects the availability of funding for another project (Finch) if the LRT option is taken.


The LRT option is ready-to-go both from a funding and design viewpoint and therefore gets full marks for this component.  The subway and hybrid components score much lower because they do not have funding in place.  The scores are 5, 1.86 and 2.29 respectively.

Summary:  Funding & Economic Development

The scores as percentages of the possible maxima for the three options are 88% (LRT), 54% (subway) and 56% (hybrid).

Transit Service


This criterion measures the ability of each option to handle projected 2031 ridership.   It builds on and partly duplicates the measure of Cost Effectiveness above.  The LRT option rates highest because it provides sufficient capacity for the likely demand i n the corridor and frees up funds for a Finch LRT.  The scores are 4.57, 2.29 and 2.71 for the three options.

Network Connectivity

This score addresses the ability of each option to meet overall movements in the transit network for affected riders.  Figure 10 on Page 47 shows the destination of peak period flows out of the section of Scarborough north of the 401.  One third of the demand is local to the area, and a further quarter is oriented north to Markham or south across the 401.  About one tenth heads west beyond Victoria Park, and the remainder goes to midtown and downtown.

The LRT best supports this pattern because it supports the local travel and provides connectivity with other routes.

A subway east to STC connecting with the SRT would improve network links in western Scarborough, but would do little for Scarborough from Kennedy eastward.  The LRT option provides access to STC through a connection with the extended SRT, an extension that according to Metrolinx will not be built under the subway scenario.

The scores assigned to each option are 4.71, 3.14 and 3.29.  It is unclear why the hybrid option scores so low here unless there is an assumption about the presence or absence of an SRT extension north to Sheppard.

Level of Service

This component considers the overall travel times for riders under each option including stop access (out of vehicle) and on board times.  Part of the overall impression of alternates in travel includes the experience of getting to transit including the pedestrian environment, not a notable part of Sheppard as it stands today.  The LRT option brings access closer to more riders (because it goes all the way to Morningside and because the stops are spaced more closely than a subway would be), and the reconstruction of Sheppard will give an opportunity to improve the pedestrian realm.

Sadly, on a topic which gets much discussion, the raw data leading to these scores is not presented in the report.  The LRT option gets a score of 4.14, while the subway option gets 3.57 and the hybrid 3.29.

Transit Service Summary

The scores as percentages of the possible maxima for the three options are 89% (LRT), 60% (subway) and 62% (hybrid).

At this point, the LRT option has an overwhelming lead given the relatively low weighting (1.5X) assigned to the three remaining components.

Sustainability and Social Impact

Equity and Accessibility

This component measures various factors including

“social cohesion social cohesion and access to opportunity; transit safety and mobility; end user affordability (e.g. fares); equity in access to rapid transit across the City.”

These are difficult to measure and depend on assumptions about the network beyond Sheppard Avenue itself.  The issue of affordability is not germane to the technology choice unless Toronto eventually sees a change in fare structures making express routes like subways a premium fare service.

The panel ranks the LRT options higher than subway because they provide service to far more neighbourhoods (particularly to existing neighbourhoods).  Moreover, under the subway plan, the SRT would not be extended north of the 401 and, eventually, to Malvern.  An additional pro-LRT consideration is the potential improvement of safety for pedestrians through better streetscaping under the LRT plans.

Both of these considerations duplicate issues covered by earlier points and reinforce the relative rankings of the options.  The scores are 4.57 (LRT), and 3.14 for the other options.  It is unclear why the hybrid option ranks so low here.

Environmental Sustainability

This component considers

“Mixed-use, higher density, more walkable and bikeable neighbourhoods are an essential component in promoting healthier less auto-dependant lifestyles. Moving towards this type of urban form in Scarborough will be difficult under any option being considered.”

As with earlier components, the availability of funding for Finch under the LRT scheme expands the scope of environmental savings within the available funding.  The urban form of Sheppard East with an LRT to Morningside may evolve into a more walkable neighbourhood.  Both of these points have been made before.

Scores assigned for the options are 4.14, 3.57 and 3.43.

Community Impacts

This component looks at the effects of construction, but more importantly the community effects of intensification, housing affordability and the redesign of neighbourhoods (e.g. “placemaking” effects on businesses and community hubs).

The scores for this component are 3.86, 3.57 and 2.86.  It is unclear why the hybrid option ranks so much lower than either of the single-mode solutions.

Sustainability and Social Impact Summary

The scores as percentages of the possible maxima for the three options are 84% (LRT), 69% (subway) and 63% (hybrid).

Overall Summary

The weighted scores have a maximum possible value of 95 points, and they are factored up to obtain values out of 100.  The adjusted scores for each option are:

  • LRT 87.3
  • Subway 59.3
  • Hybrid 59.5

My feeling is that the ranking is fair, although there are cases where the hybrid version is oddly penalized with no obvious explanation.  Publication of the component scores might have clear up some of these issues.

Many components of the total score turn on related matters and there is a compounding effect including:

  • The LRT option serves more of Scarborough.  Any component that depends on maximising the number of people, particularly in existing neighbourhoods, with access to improve transit will benefit from the scope of the LRT.
  • The LRT option leaves funding available for Finch Avenue.  Any component that considers spending on a network basis, rather than a line basis, benefits from the larger reach of two LRT lines and their potential for rider, connectivity and environmental benefits.

If anything, I would expect the hybrid option to rank second, and the reason this did not happen is unclear.  If it is the LRT component that performs more strongly than the subway, then the LRT-based options should generally rank higher than a subway-based one.

In the next article, I will turn to the background papers for the SRT report.

9 thoughts on “The Sheppard LRT Report (Part II)

  1. The north-eastern extension of SRT was on the books for decades, well before the Sheppard LRT plan was conceived. So, they should have considered alternative options for the yard.

    Even if Conlins Yard is the only place to host vehicles for the extended SRT/SLRT, they could build non-revenue tracks on Sheppard between Progress and Conlins. I really can’t understand why they preclude the SLRT extension if Sheppard LRT is not built.

    Two other puzzling elements of the recent Sheppard panel report:

    1) On Page 12, they state that the subway extension from Don Mills to Vic Park (Hybrid option, “C”) would be 2.7 km long. But the distance between Don Mills and Vic Park is 2.0 km, even if one follows bends of Sheppard (on the straight line, it is 1.9 km between the intersections; I used Even if they included longer tail tracks at Vic Park, do they have to be 700 m long?

    Steve: I suspect that they are double-counting the underground LRT access to Vic Park Station. The section describes the LRT as being 10.3km of surface running, and this implies that the distance from the portal to the western end of the LRT starts at the 2.7km mark measured east from Don Mills. It could just be an error.

    2) In their ranking system, the Hybrid option “C” ranks below the LRT-only option “A” on the following criteria:

    Network Connectivity (LRT 4.71; Hybrid 3.29)
    Level of Service (LRT 4.14; Hybrid 3.29)
    Equity and Accessibility (LRT 4.57; Hybrid 3.14)

    How it is possible, is anyone’s guess. The Hybrid option would have exactly same list of stops and same network connections as the LRT, as follows from the comparison of Figure 1 with Figure 3.

    Steve: As I mentioned in my article, I believe that the hybrid has been unfairly marked down in some evaluations, and wish we could see the details underlying these scores. It would be rather embarrassing to have the subway option come dead last in the ranking, and yet this might have given comfort to those for whom the hybrid option was an acceptable compromise.


  2. I haven’t seen much discussion on the issue of the “unnecessary” transfer at Don Mills station in the panel’s report. Not that I’m saying that the Don Mills transfer is a more important issue than it really is, but skeptics would question the panel’s competency on how they can ignore such an obvious talking-point.

    Steve: It is an obvious talking point only to those who persist in counting it as a transfer, but ignore the need to get to the Sheppard subway, wherever it goes. Riders from northeastern Scarborough must still ride to either STC or to, say, Kennedy North Station. All that happens is that the transfer shifts further east.

    What’s a shame in the panel’s report is that they did address total trip time by various schemes, but included no information on a calculation of the times various subgroups of riders would have consumed to reach their destinations.


  3. As a former student of Professor Miller’s, and a “fiscal conservative” myself, I have every bit of faith in his providing a fair and thorough analysis. This is someone whose life work depends on maintaining respect amongst the experts in his field, and he is about as far from a political hack as you can get. It hurt me to see such a caring, intellectual guy get smeared by the politicians today, but I am pretty confident that, no matter how the vote goes, history will vindicate him.


  4. As I had meant to say earlier, it is also a shame that the panel didn’t use this report as an opportunity to explain why the transfer at Don Mills isn’t a bad thing. Regardless of whether the transfer can be eliminated or not (which you’ve explained earlier), still a big question on everyone’s mind.


  5. What seems to be missing from the report is any discussion of operational costs. People like to talk about building new stuff but when it’s all built and done with you still have to pay to run the system. Which option is the most cost-efficient operation-wise?

    Steve: Subways are much more expensive on the operating budget because there is so much more infrastructure. Only if you carry high, truly “subway” volumes, do you spread the costs over a enough passengers to get the per ride cost below LRT. Even then, about 25 years out, things start to break down, and there are large capital costs for rehabilitation. It gets even worse by 50 years. I agree that this issue needs to be addressed and killed off, but subway advocates are too busy telling us their chosen mode lasts for a century. Problems on both sides.


  6. Can someone explain to me how P3s are supposed to produce magical ‘free’ money or at least make projects cheaper for the taxpayer? I’m not a finance major so presumably must be missing something.

    Say a transit project costs $1 billion. The city could sell $1 billion in municipal bonds to get the money and then have to service those bonds going forward at some standard interest rate.

    In operation, the project could make an operating profit or require an operating subsidy. For simplicity, I’ll assume it generates enough revenues to cover its ongoing costs, leaving only the $1 billion + municipal interest rate on the city’s books.

    But if it was a P3 whereby half the capital costs were put up by the mighty private sector, we’d then have the city borrowing only $500 million at municipal interest rates. I’ll assume operating costs and revenues are the same as before.

    But why would the private sector put up that kind of money? By definition the private sector is there to make a profit. If they wanted a relatively low risk return on their $500 million, they could have just bought those municipal bonds and received the interest payments.

    Since construction of any large infrastructure project entails all kinds of risk, presumably they are expecting to receive a rate of return greater than that of the municipal bonds. So where is this extra money coming from?

    If the transit line isn’t generating any additional revenue, then wouldn’t that mean the city (taxpayer) is on the hook to provide them with their expected profit? Since that profit rate would be greater than the rate for municipal bonds, doesn’t that mean the city (taxpayer) is having to pay more for that private $500 million?

    Wouldn’t any other source of profit (‘air rights’, advertising revenue, etc) just be funds that end up going to the private sector instead of the public purse?

    Or is there some way that P3s are able to produce extra money out of thin air that results in us getting $1 billion in infrastructure without having to actually pay the costs of obtaining that $1 billion?

    Steve: Do not ask how the rabbit gets into the hat. One ridiculous situation with P3 proposals is that the Feds have a special fund for P3 projects. That’s right, they are so popular, so efficient, that if you go down that path, Ottawa will give you a subsidy you would otherwise not get. This neatly makes the P3 more financially attractive to, say, a city looking to minimize its costs. Perish the thought that the private sector would actually just do this sort of thing because they thought it was good business.


  7. Now that the decision has been made for LRT along Sheppard, I’m starting to wonder about the potential for LRT along Don Mills, at least down to Eglinton Ave. or Overlead Boulevard (south of Eglinton but not as far as the Don River).

    The TTC talked about a Don Mills LRT line but it had major weaknesses south of Eglinton and has now fallen by the wayside.

    Steve says a Downtown Relief Line should be built at least up to Eglinton, which is a natural interchange point. Metrolinx is talking about a DRL up to Eglinton as well. The Sheppard subway will be staying at Don Mills.

    At this point, what is the potential for a subway line along Don Mills extending past Eglinton and up to Sheppard*?

    *where it would either interline with Sheppard (we’d end up with a backwards “C” shaped Sheppard-Don Mills-Queen(ish) line from Sheppard & Yonge to Don Mills to River, to, presumably, the CNE) or just interchange (leaving us with the Sheppard line plus a new Don Mills line shaped like a letter “J”, from Don Mills & Sheppard down to River, then west to the CNE).

    Is there enough development potential (residential & commercial) to justify the cost of constructing a subway rather than an LRT along Don Mills between Eglinton & Sheppard, if there is a Downtown Relief Line to the south & Sheppard Subway to the north?

    Cheers, Moaz

    Steve: I suspect this might be possible, but not for quite a long time. I would prefer to see a proper study of what Don Mills might become rather than simply falling into the trap of playing “connect the dots” with subway lines. There are many other places we need subway and LRT service, and any considerations need to look at the whole list, not one superficially attractive option.


  8. I was disappointed at the dismissive attitude to the hybrid option – it seems like a sloppy insertion because it was the option Stintz proposed to Ford in 2011. I can’t see how diverting a bunch of buses away from the 404 overpass/exit (and so close to the 401) wouldn’t be helpful even if it was only 2-3 routes. You could then build a station where the LRT and subway terminated in a purpose designed two track per mode terminal rather than hacking the LRT into Don Mills.

    The fate which befell Ontario Northland today should be sobering to both the LRT advocates keen to spend “our” billions from Queens Park and embarrassing to the Fordites who claimed even more money would be forthcoming if we only waited.


  9. I would like to know your opinion on the feasibility and potential value of the following.

    An extension to the existing Sheppard east LRT that would see it turn south on Don Mills Rd, continue till Overlea Blvd and align with the Bayview Extension via Millwood Rd and Redway Rd. continuing south till it reaches an acceptable alignment downtown (perhaps Queen or Front St).

    I recognize that there are issues with this proposed alignment like the turn south on Don Mills from Sheppard, the multiple turns, and the grade elevation along Bayview. However that being said, could it be justifiable to build this type of one seat LRT ride from the north east corner of the city to the downtown, that could double as a modest downtown relief line (with a short turn at Eglinton Ave during rush hour).

    Steve: If one wants a line from the northeast corner of the city to downtown, there is a perfectly good rail line which does not, yet, have GO service on it. This would run from northeast Scarborough via the CPR to Leaside and then via the abandoned Don Sub (now owned by GO) to Union. As many others have pointed out here, we should not be building an LRT line for long-haul one seat rides.

    Yes, getting from Overlea and Millwood to Bayview via Redway is rather tricky, and it would make the link from Thorncliffe Park to downtown essentially an express operation without any connection to the Bloor-Danforth subway.

    This is not a practical scheme.

    A “downtown relief” line must, almost by definition, provide significant new capacity to offload or complement the existing subway system. This will likely occur on concert with better GO service to the outer parts of Toronto and the inner 905 so that riders from these areas do not look on the subway/LRT network as their only route to downtown.


Comments are closed.