In the first part of this series, I discussed The New Geography of Office Location, 2011, and then in the second part, its successor A Region in Transition, 2013. Now, I will turn to The Business Case for the Regional Relief Line, October 2013. All three papers were produced by SRRA (Strategic Regional Research Associates).

Only a 17 page summary version of the Relief Line report is available online, compared to the full versions of the first two. Considering the clear influence this series of reports has had on transit policy and the recent election campaign, the idea that

Detailed research is available to Investment Partners of SRRA [Page 1]

leaves a big hole in the range for public comment and review. I hope that Metrolinx will rectify this situation as part of whatever studies might take place.

My thanks to those of you who slogged through the first two articles.

Section 1. Background

The report begins with a clear statement of its origins:

The “Regional Relief Line” (RRL) was first developed in 2009 by Real Estate Search Corp and the CUI resulting from the research conducted in the study known as the “New Geography of Office Location”. This research showed that the location of employment had shifted from the centre of the City to several suburban nodes and employees in these nodes had no alternative to the automobile. [Page 1]

[For clarity, “Real Estate Search Corp” and SRRA have overlapping membership and should not be considered as separate entities in this context.]

From the first two reports, we know that the problem of how to serve suburban office development was a great concern, but no specific solution was put forward even though, according to this text, one was already being developed.

In a truly breathtaking claim, we read:

The new approach to solving congestion had to include reaching beyond the 416 and relieving congestion on the highways in the 905 as well as relieving transit congestion downtown. By identifying the route which forms the RRL the authors believed that the majority of the suburban auto congestion and the downtown transit congestion could be addressed in one high value project. [Page 1]

This is one of the more creative pieces of writing I have seen from a planning consultant during my long years watching transit issues. The idea that any one line can solve highway congestion in the 905 and transit congestion downtown presumes that it also addresses many choke points on the regional transportation network. Quite bluntly, that is a physical impossibility, and it should not require many paragraphs of detailed explanation as to why. In brief there are large areas in the regional network that are nowhere near the catchment area of this “relief line” and they cannot be affected by its implementation.

In another interesting claim, the report states:

The findings of the research as of October 2013 were provided to the Transit Investment Strategy Advisory Panel (TISAP) headed by Anne Golden co-chaired by Paul Bedford and included SRRA co-founder Iain Dobson. Much of the core evidence and conclusions from this study helped form the many recommendations of the panel which address the issue of creating high value, high capacity, and high ridership transit. [Page 1]

Well, no, actually, it did not. From my own conversations with Paul Bedford, I know that there was some discussion of “The Big U” which was the name used to describe the proposed Relief Line (to distinguish it from “The Little U” which we all know better as the “Downtown Relief Line”). However, the Advisory Panel’s mandate did not include making alterations to the Metrolinx Big Move, only to discuss how whatever transit projects might be built would be funded. Iain Dobson is now, but was not then, a member of the Metrolinx Board, and was an unofficial advisor to former Minister of Transportation Glen Murray.

Section 2. Introduction

We begin with another sweeping claim:

The “Regional Relief Line” (RRL) connects the majority of the Region’s office employment clusters, offering employers and employees an unprecedented degree of mobility choice, considerable reduction in commute time, significant congestion relief in the 905 and the downtown core. It further permits growth of employment in established nodes creating the necessary conditions for the Region to remain competitive.

The RRL delivers extraordinary value for public investment by creating solutions to congestion in by transit terms and has the potential to be implemented quickly without disrupting other transportation options. It will generate over 500,000 new riders within 5 years of commencing operations. It generates this high value by transforming a series of individual rail-based corridors into a cohesive, high speed, high frequency transit system. The RRL will connect to and incorporate the following existing projects or projects under construction to create the most important interconnected network possible for relieving congestion. [Page 2]

As we know from the earlier reports, the focus was on a number of nodes notably the 404/407 office cluster, the Airport Corporate Centre, the core and its satellite “brick and beam” districts. We also know that other potential nodes were systematically ignored on the basis that they were stagnant, even though they might already have substantial transit demand. The overwhelming consideration was to provide more transit to make office parks more viable as sites of greater development.

The claim of half a million new riders is quite a stretch, but I will leave that until we come to more details later in the report. To put this in context, daily TTC ridership is about 1.8 million distinct rides, and this probably translates to something like 750-900k individual riders. Getting 500k new transit riders (never mind those who may already be on transit) would be quite a feat.

Important words here are “high speed, high frequency transit system”. However, what is proposed is not a system, but one line, and both the speeds and frequency proposed (a) bear no relationship to what is practical or reasonable, not to mention (b) being significantly more aggressive as a transit operation that what was proposed in SmartTrack. That’s an important distinction because some of the claims for the RRL found their way, unchanged, into claims for SmartTrack, notably ridership estimates.

The line is purported to give many new options for travel on a variety of routes including some local streetcar lines. In fact, the RRL only crosses these lines, and not necessarily at a station. The degree to which it would provide an alternate path for the local type of travel taken on these routes is minimal.

The unique opportunity to serve employment destinations at either end of the line will result in highly productive passenger loads on trains in BOTH directions. This has only been successfully achieved in a few transit projects in the history of transit. [Page 2]

Without question, the holy grail of any transit planner is a route with bi-directional demand where what might otherwise be a low volume counterpeak move actually carries passengers. We can see this any day on the King car that serves a variety of neighbourhoods and does not exist simply to bring people downtown.

However, it is one thing to sit at the bottom of a funnel in the core of Toronto at Union Station collecting riders from several widely-space origins, and quite another to connect outward to a suburban node. For starters, many (most?) of the people working at those nodes will originate either beyond the end of the RRL, or they will come from directions it does not serve (e.g. east and west of a node in Markham). They will certainly not go out of their way to take a route that does not serve their trip unless it offers a massive advantage.

RRL will provide high frequency, high capacity and high speed transit service to car-dependent office employment clusters in Mississauga and Markham. These two clusters already have enough ridership potential (150,000 jobs in each) to make the RRL an instant ridership success and provide immediate congestion relief in the two busiest suburban areas of the Region. [Page 2]

However, the 300,000 jobs (and note that these are potential jobs, not existing ones) will not all be commuters from south of Markham and from east of Mississauga.

As for the core:

In Toronto’s Financial Core – already well serviced by transit – the line will offer significant relief to overcrowding on the Yonge line. Transit commuters will have the choice to go directly downtown from the ends of the Danforth and the Bloor subways in less than half the time or transfer from the busy Queen and King Street car or use the new crosstown LRT at Eglinton to access the core from the east or west – all for the same fare. The line will also allow employment growth to occur at “shoulder” nodes such as Liberty Village and the Lever Bros site, permitting continued growth of the core’s employment districts.

For starters, the reference to the “ends” of the Danforth and Bloor subways implies that the route will take the CPR/GO Milton corridor passing by Kipling Station. This is quite different from the Weston/Eglinton route proposed by SmartTrack and shown in the RRL map later in the report. Next, the viability of a transfer to RRL at Kennedy or Kipling presumes that (a) the service is frequent enough that a less direct transfer connection is tolerable and (b) that riders are destined for somewhere served by the RRL. We all know of the anguish caused by the SRT to subway transfer at Kennedy, and moreover, once the Scarborough subway opens, riders would need a strong incentive to get off trains westbound at Kennedy to transfer to the RRL. Kennedy will no longer be the end of the line.

(By the way, a Kipling connection aligns better with the “Big U” discussed by Paul Bedford that would have operated via the GO Milton corridor in the west end. Possibly someone had second thoughts about this idea, and switched to the Eglinton alignment, problems and all, closer to the publication date.)

Next, we see that the Lever site finally makes an appearance. Between it and Liberty Village, we now have potential relief for office space growth in the core. This was an explicit concern of the earlier reports that treated the core as if it were almost “full”.

Transferring from the Queen or King lines to the RRL imposes a time penalty both on the rider for transferring, and on the RRL itself for providing more local stops. Moreover, the riders have to be bound directly for Union Station because otherwise they would be better off just staying on their streetcars.

The RRL enables the network to add over 250,000 commuters during both the morning and evening rush hours, … [Page 3]

These are added riders, but we are also diverting traffic off of the existing network, and so the total ridership must be even higher. That 250k showed up in the SmartTrack campaign material, but for a service with comparatively short trains running every 15 minutes. Some basic math is in order.

• If we assume a four-car train with a design capacity of about 650 (that’s roughly equivalent to a train on the Sheppard subway), we are looking at 2,600 passengers per hour at the peak point for the proposed SmartTrack headway.
• Assuming that there is both a west side and an east side “peak”, and that we are able to attract an equal demand outbound, we might get up to around 10k/hour if the trains are full on both legs in both directions.
• Generously, there are six peak hours per day, and so that gets us to 60k riders. If the off peak period carries (over more hours) an equal demand (the TTC actually carries about 60% of its riders off peak), we are now at 120k daily riders. This is far short of 250k, let alone the combined total of the new and existing demands that appear to be presumed.

Then we come to the question of travel times:

Most significant will be the time saved by commuters coming to the core from places like Rexdale and Etobicoke in the northwest and Scarborough in the northeast; all these commuters will enjoy a 50% reduction in travel time and convenience. [Page 3]

The exact calculation of this saving does not appear in the report, but it is likely that it applies only to the leg of the journey actually taken on the RRL, not to the full trip that will almost certainly include a bus feeder. Also, as we will see later, the claimed speed for RRL greatly exceeds what is practical given the station spacing and track geometry of the lines it would use.

In this Nirvana, the RRL will have, among other benefits:

Created two-way peak traffic where fare revenue will exceed operating costs allowing for the elimination of operational subsidies. This will afford transit planners greater financial ability to continue to expand the system instead of subsidizing transit solutions with sub optimal ridership.

Reduced commute times by at least 15 minutes and in some cases over 45 minutes for over 250,000 commuters improving the quality of life for employees.

Given high speed access (maximum commute of 45 minutes) to employment for over 750,000 people, a benefit to employers making expansion more competitive in the Region. [Page 3]

Anyone who tells me they are going to make money on transit is immediately suspect, and the challenge here is that this is a claim that can blind gullible readers (for which read politicians) who leap at the thought of a project they can announce, but won’t have to pay for.

Partial capital funding will come from beneficiaries of its services. [Page 3]

This is an echo of the financing scheme for CrossRail, and it makes big assumptions that there will be buy-in from affected property owners.

And by the bottom of page 3, we have:

The RRL will have unlocked the capacity of businesses to grow, people to get to work with many choices, and increase the capacity of the entire Region Transport system by over 20%. Estimated new ridership is 500,000 passengers per day.

Sounds marvellous (but is that 500k passengers, or 500k riders — they are not the same thing), but it all depends on running a service that can actually attract and carry them.

In previous articles, I have not reproduced the illustrations because they were too numerous, but it’s worth putting the RRL map here for reference. [Source Page 4, click for a larger version.]

Here we clearly see that the RRL does not go near to Kipling Station, and using it would require transferring closer to the core at Dundas West where the time saving would be rather less than a transfer at Kipling. The RRL also makes that turn west onto Eglinton in Mt. Dennis which has been the subject of some discussion in the SmartTrack context.

Section 2.1 About the Regional Relief Line

This type of transit solution is called a “surface subway” because it moves quickly between stops unimpeded by other transit or pedestrians, and carries as many as 70,000 people per hour. [Page 4]

70k/hour you say! Amazing! And with what technology? Service frequency?

The RRL will operate through a 62.8 km of rail on the surface. Two additional parts of the system can be added at a later date to further improve the project and they will be underground. Those options are underground subway components (10.47 km) under central Toronto and the airport. These two additional high value parts of the project can be phased in as required. [Page 4]

During the election campaign, there was some debate about whether “SmartTrack” included any tunnels, but this question focussed on the difficulty of the Eglinton West leg. Now we can see that RRL has tunnels, but not on Eglinton West. They are an airport link north from the Airport Corporate Centre, and a future tunnel through downtown as a bypass for congestion at Union Station. The link from Weston’s rail corridor to the airport is quite definitely on the surface on the RRL map.

We also learn that the Meadowvale to ACC leg of the RRL would unlock the growth potential in the Airport Corporate Centre. This presumes that the majority of the net addition of commuters could reach the ACC via the RRL. More generally, we learn that the RRL will open up many areas to redevelopment along its route. [Pages 4-5] That is possible only for land adjacent to stations, and more stations means slower trains, a tradeoff the report does not explore.

Where Do Those Capacity Numbers Come From?

I have to jump out of sequence here because so many of the report’s conclusions depend on the speed of the line, the frequency of the service and its total capacity. For this we must jump ahead to section 2.5 New Trains and Stations starting on Page 11.

The rolling stock, a new high speed electric train which looks and feels very much like the new subway cars on the TTC, will be assembled in Ontario, built tested and delivered in a similar manner as the rolling stock for Crossrail in the UK which allows for the cost effective delivery of the rolling stock over time and the provision of new capacity as the system grows. Benefits of the new Rolling Stock are speed, comfort, flexibility and safety. They have the capacity to be lengthened as need be. Initially they will run on 5 minute intervals at peak hours but increase frequency easily and cost effectively as demand grows.

The new trains are being designed to meet the needs of passengers Greater Toronto Area. They will travel at speeds of up to 160kmh thereby delivering fast cross town service. Critical to increased speed is the spacing between stations, which is why stations are on average more than 2.5 kms apart. They will be electric and organized by the world`s most advanced signalling system allowing the frequency of service up to every 90 seconds if demand requires. Initially frequent (less than 4 minutes between trains) service can be delivered at peak demand times, while less frequent as demand reduces.

With Crossrail rolling stock as the model, we launch into an operating model of 5 minute headways initially (12 trains/hour, or 3 times the capacity proposed for SmartTrack) or is it a less than 4 minute headway as claimed in the next paragraph?

As for 90 second headways, that will not happen unless these trains have completely exclusive right of way for reasons that have been discussed at length here before. Conventional rail traffic cannot co-exist with a rapid transit operation, especially an automated operation needed for very frequent service.

The experience will be similar to the new TTC subway cars, only they will carry slightly more passengers with more seating and travel at twice the speed. [Page 11]

As for speed, 160kmh is faster than GO trains except, possibly, on a few stretches with widely spaced stations and a high track speed. At the RRL station spacing, especially east of Union, it would be impossible to run this fast because trains would be stopping before they reached top speed. Also, operation at this speed on the shared GO network would have implications at stations RRL did not serve, and would place demands on track geometry that may not be physically possible to build.

Now let’s go back to where we left off:

Section 2.2 Benefits of Phased Delivery

The report proposes that RRL be implemented in stages:

1. Markham to Union

The report claims a travel time of 21 minutes from what is now Ellesmere Station to Union via RRL. The overall travel time is about 40 minutes from Markham to Union (implied by the text) as compared with 47 minutes for GO trains today. The authors compare this to a small saving available to users of the Scarborough Subway (amusingly setting up a competition between the two routes which, in today’s political climate, will both be built). Of course, the last time I looked, Ellesmere Station is not at Scarborough Town Centre, and one must add the access time for the roughly 2km jaunt eastward.

A major issue for the Markham end of the line is that the rail corridor is somewhat distant from the actual job centre. This link would depend on the York VIVA BRT network, a rather odd way to deal with inbound traffic to a major employment node, and a chunk of the commuting time about which the report is silent.

2. ACC to Union (2019)

This is the most problematic for RRL and SmartTrack because the authors don’t know what they are talking about here.

This phase will take a little longer to complete than the Markham phase because there are no tracks on the section from Mt Denis to the Airport Corporate Centre. This right of way will take longer to design and obtain approval in an EA process. It is owned by the Province and was intended to be an expressway.

There is ample room for the dedicated right of way. [Page 6]

The right of way is not owned by the province, but by the City of Toronto which has been selling it off for new housing developments. There is not ample room for the dedicated right-of-way.

Note that this claim implies that the line would not be underground because otherwise it would not require a right-of-way. This is the same location where surface LRT operation was met with great hostility, in part because of bad design for LRT/traffic intersections. They should have lots of fun with a high speed RRL train that will need a well-protected corridor.

At no point is it explained how the line will get from the Weston rail corridor to the former Richview Expressway lands on Eglinton that begin west of the Humber River. I discussed this in some detail in my review of John Tory’s SmartTrack line.

3. Meadowvale to Airport Corporate Centre (2021)

This corridor from ACC to Meadowvale runs parallel to the 401 highway and connects the Hurontario LRT and Go Lines to employment areas currently containing over 100,000 commuters. The 401 highway is the most congested roadway in the Region, with all commuters completely auto-dependent. This is also a heavily travelled trucking corridor which needs to be freed up for the movement of goods and services. [Page 6]

If capacity is needed for trucking, this must occur on more than a comparatively short stretch of the 401. Only commuting traffic that is actually diverted onto RRL counts against such a benefit, and one must ask whether the highways would simply backfill with latent demand.

4. Wellington Street Union Relief Subway (2022)

This phase would shift the RRL into a dedicated subway from west of Bathurst to Broadview. For the benefit of readers, this is similar to the alignment I have often advocated for the DRL subway.

This tunnel, whatever runs in it, will be one of the most complex parts of any new cross-downtown line. It is intriguing that it magically can be in service in less than a decade under this proposal when a similar structure for the DRL almost certainly would have a much longer claimed delivery date.

5. Subway Connection to Pearson (2024)

This phase would see an underground branch off of the RRL to directly serve Pearson Airport. Of course the same service, as a surface operation, was proposed as part of the Eglinton LRT.

And so, yes, gentle reader, we could have all this in a mere 10 years if only we started today.

Section 2.3 Future Demand

The forecast growth in the GTA is quite substantial with 3 million more people living here by 2036. This represents a huge growth in travel both for work and for other purposes. Transit demand is expected to rise by 45%. This sounds impressive, but not quite so when put beside a projected 50% increase in regional population.

Even with RRL and its claims, we would not be keeping up. Moreover, many parts of the GTA would not be served by RRL. It is designed primarily around a handful of development nodes, and if it serves anything else, that’s o-by-the-way.

If we are really talking about that scale of population growth, and we intend to increase transit’s share of the market, we will need much more than one new RRL route. Moreover, it can only make a dent in the demand provided it operates as a rapid transit line, not as a now-and-then commuter rail system. There lie the two challenges: changing the operating rules for the rail corridors in general, and recognizing that large-scale improvements to transit will be needed throughout the region.

Unfortunately, thanks to the absence of details, the claims of large scale reductions in present congestion are unsupported by information such as demand models, and there is no discussion of the degree to which vacated capacity will be backfilled rather than bringing an actual reduction in congestion.

Preliminary calculations put the number of cars which will be initially taken off the road on a daily basis at 70,000. This is a number which will have an exponential impact on mobility of the highways of the Region and will reduce congestion significantly. [Page 11]

We are going to carry 250k riders in each of the peak periods, but we are only going to reduce car traffic by 70k? The key word is “initially” and it gives some sense of the early days riding and benefits that are anticipated rather than the much grander numbers for a full buildout.

Section 3. The Economic Case for Regional Relief Line

This section depends on some of the questionable techniques we have seen in Metrolinx “benefits/business case analyses”. The first and most obvious is that the construction cost and effort is counted as a benefit. That is a misleading premise because it completely avoids the question of whether spending $X billion is the best use of the money, and not just within the limited scope of transit alternatives. I can create economic stimulus simply by handing out $20 bills, a tactic already employed by one recently retired Toronto Councillor.

Spending extra billions because someone has their heart set on a more expensive option is not necessarily the best investment of public funds.

The report claims that there is

a cost benefit ratio of 2.5 to 1 making it on a par with the most valuable transit projects in the world [Page 12]

At this point we are at pure fantasy because the assumptions behind this claim are not published. With the grandiose claims made for the success of this project, it is no wonder it has a high ratio. Is this genuine, or are we talking Florida swamp land? This claim really needs detailed review, and the numbers should be in the public domain. There should be no “commercial confidentiality” about estimating the cost and benefits of a public investment on this scale.

There is no question that improved service levels and quality on the GO network will have a major effect on its attractiveness and on travel patterns within the GTA. However, there is no one magic solution, and claiming that the RRL or whatever we call it can be the one project that makes all others unnecessary is a simplistic attempt to highjack the discussion for one favoured project and for the land developments it will support.

Metrolinx is supposed to be planning regionally for all of the region, not just for those corners where a rail corridor conveniently strings together development sites of interest.

After the detailed planning background in the Geography and Transition reports, the so-called Business Case for the Regional Relief line is almost a parody, the kind of sloppy, half-baked paper one expects from amateurs, not from professionals. Metrolinx and Mayor Tory should be wary, and should review the claims made for RRL/SmartTrack with care.