This article is the second section of my critique of the December 2013 review of the Metrolinx Big Move Plan written by Michael Schabas for the Neptis Foundation. It should be read in conjunction with Part I and following sections.
3. GO Transit
Schabas observes that although The Big Move cited an improved GO Transit network as a first priority, that priority has not been embraced.
Our analysis confirms the original conclusions of The Big Move: there are massive benefits to be gained from upgrading the GO Rail system into a two-way system offering fast and frequent all-day services. Indeed, the financial benefits of reduced operating costs and higher fare revenues could offset most or all of the costs. Compared with the subway extension and LRT schemes, the capital costs are actually fairly modest. Taking account of the time savings and other benefits to transit riders and road users indicates very high benefit:cost ratios. Unfortunately, GO does not seem to have recognized this opportunity. [Page 25]
As I mentioned in the first article, at the time The Big Move was published, GO was still a separate organization from Metrolinx and the two would not be consolidated until a year later. Even then, GO tended to pursue its own agenda partly thanks to funding constraints, and partly because some of the Metrolinx schemes were not practical, especially in the short term.
As Schabas notes, the 2009 Metrolinx Benefits Case Analysis (BCA) and the 2010 GO Transit Electrification Study came to different conclusions about the viability of electrification on the Lake Shore corridor. In this context, it is important to remember that even a project with a favourable BCA will not necessarily be built depending on competing demands for capital and political priorities.
Schabas then makes a curious comment:
GO management and provincial decision makers hesitated to take on a massive investment that the study seemed to show to be barely worthwhile in economic terms and which would, apparently, need to be funded mostly by public investment. [Page 26]
There is a basic reason why electrification would be a public rather than a private investment. For portions of the route that GO Transit does not own, it would be difficult if not impossible to create a private ownership (except by CNR) structure to offload the capital costs and debt from provincial books. Moreover, if the benefits were only marginal and included imputed values of time saved, not just real dollars and profits, these could not be used to attract private capital.
This brings us to a fundamental problem with transit financing as it was seen by Queen’s Park in the wake of the 2008 fiscal crisis. At all costs, there would be no new net public debt. New facilities would either be built and operated on a turnkey basis through public-private partnerships, or assets built with public money would stay in provincial hands where they would appear on the books as a counterbalance to any new debt. This has produced no end of problems with The Big Move including the need to resolve ownership of LRT corridors built in the middle of public roads.
No doubt one reason some at Metrolinx would prefer totally grade separated lines is that they could be more easily hived off to a private owner.
With the recent publication of the Transit Panel’s report, Queen’s Park appears to have a new view of debt: borrowing is fine provided that there is a dedicated revenue stream to pay it down in a reasonable time, 30 years tops. This change in outlook would allow projects to be undertaken with large capital investments up front so that the benefits can be obtained as soon as possible.
What remains is the problem that soft benefits such as travel time saving cannot be monetized (unless fares are increased, a counterproductive measure). We can’t pay off a faster trip unless the “extra” time generates real, not imputed, value and that value is subject to tax.
Any discussion of GO priorities must be seen with these two factors in mind: GO prefers incremental expansion with whatever money is available, not “big bang” construction projects; and the spending rate for capital projects was constrained by a “pay as you go” outlook that avoided debt.
As for a GO master plan, sadly that depends a great deal on what the Minister and Premier want to announce from time to time. Kitchener-Waterloo has service not because it was in the plan (indeed the scope of The Big Move was cut back to exclude KW in the final version), but because Dalton McGuinty announced it. GO scrambled to “make it so” on a very short deadline. With the recent announcement of massive upgrades to transform GO into an all-day 15 minute service throughout its territory (again a political event, not a proposals that had gone through the mill of public review at Metrolinx), we are at the other end of the scale. The “master plan” is to build everywhere with little concern for the specifics of each route.
Schabas then claims:
The initial enthusiasm for upgrading the GO Rail system, which was “Priority Action #1” in the Big Move, dissipated quickly. Money that could be spent to turn the GO Rail system into a true regional metro, with frequent all-day services, has been diverted to LRT projects, mostly in the City of Toronto, at least for the “First Wave” of projects. [Page 27]
This misrepresents both the content and the structure of The Big Move. “Action #1” does not imply “do this first”, and the network to which it refers is the entire Big Move, not just GO Transit. Although this would include all-day GO service, it would also include the rest of The Big Move including the LRT projects. The nine “Big Moves” are:
- Build the regional rapid transit network.
- Higher order transit connectivity to Pearson Airport from all directions.
- An expanded Union Station.
- Complete walking and cycling networks.
- Transportation demand management.
- An information system for travellers.
- A region-wide integrated fare system.
- Build a system of connected “Mobility Hubs”
- A comprehensive strategy for goods movement.
If one wishes to review Metrolinx’ achievements, one would be pressed to show major advances on some of these points while others have received substantial funding and show good progress. They obviously have different costs and time frames, but these items were treated as equals in The Big Move.
To say that money “has been diverted to LRT projects” is false: GO had not even begun planning on many of the proposed expansion schemes when Transit City was announced, and a major one (the Georgetown corridor) will be finished long before any Toronto LRT routes open. Moreover, it is an example of the denigration of LRT sprinkled through the Neptis paper with its implication that much more extensive GO expansion would have happened if only we didn’t waste our money on those LRT lines.
By Schabas’ own account, GO has never embraced electrification, and even upgrades to the diesel network have come slowly. It’s a real stretch to suggest that some pro-LRT plot is responsible for the slow pace of change at GO.
It is no secret that, regardless of technology, the Eglinton-Crosstown line would take a huge bite out of funding resources for a tunnel through its central section. Politically, this line was vital to show “progress” on the centrepiece of new rapid transit within Toronto. Other LRT lines were pushed back in the plans, and there is good reason to doubt they will ever be built thanks to delays at Queen’s Park and the LRT-vs-subway politics of Toronto Council.
… until the GO Rail system is upgraded, there is little chance of delivering the Metrolinx objectives. The regional rail system is the “backbone” of the GTHA public transit system. LRT and BRT services have a local and feeder role, but are too slow to compete with the road system for longer-distance trips within Toronto and throughout the region. [Page 26]
As I have said many times, it is not the purpose of the LRT services to complete for longer-distance trips. If we build rapid transit only to serve regional trips, we will ignore the substantial demand for local transit. On this Schabas and a lot of early planning at Metrolinx part company with me.
With this sideswipe at LRT and BRT out of the way, Schabas turns to the electrification study. He covers much of the same ground as objections raised during the study’s workshops by various advocates: GO took a very conservative approach that rested too much on what diesel power could do without fully exploring the benefits of electrification.
The study did acknowledge some benefit from the (slightly) faster acceleration possible with electric locomotives, and noted that this feature would attract more passengers and allow fewer trains to carry the same peak capacity. But they made no other changes to take advantage of the opportunities electric traction opens up. EMUs were rejected before the final screening of options, because GO found that they cost about 40% more than locomotives combined with un-powered bi-level cars and could not see any offsetting benefits. [Page 28]
Among the shortcomings of the study, Schabas notes:
- The failure to consider the benefits of frequent service with shorter trains during the off-peak (such as would operate to the airport) on the Lake Shore corridor.
- The assumption of an instantaneous changeover of fleets from diesel to electric rather than a staged implementation.
- The primary comparison was between diesel and electric locomotives hauling existing bi-level coaches in full-length trains.
- More frequent and faster service would attract more riders whose fares would partly offset higher operating costs.
The electrification study purported to start with a huge range of alternatives (basically every conceivable permutation of every option) , and quickly whittled this down to a much smaller set for further study. This process was a sham intended to to give the impression of an exhaustive search.
Statements made at public meetings established that some members of the study team were out of touch with the range and longevity of electric operations (I could not help remembering some of the half-baked comments in the GO-Urban days), notably with concerns about whether electric trains would work in snow. This is precisely the type of remark guaranteed to undermine the credibility of a process however well-intentioned it might have been.
It is ironic and disappointing to read:
After reviewing an earlier draft of this report, Metrolinx staff told us that they now recognize that different rolling stock strategies might generate better business cases, but at first chose to test only a standard operating plan in the electrification study. They are now studying further strategies, and are looking specifically at electrifying the Georgetown corridor and possible use of EMUs for the Union Pearson Express service. [Page 35]
This is precisely what was suggested at the study workshops, but rejected as it did not fit into a preconception that only locomotive-hauled trains would be suitable.
On ridership, Schabas observes:
Carrying passengers is, of course, GO’s primary purpose and increasing ridership should be the main objective of any investment scheme. For some reason, the GO Electrification Study Summary Report does not give an explicit figure for the additional ridership that might be attracted by electrification, with faster and more frequent trains. This would seem to be the most important criterion, alongside cost, in evaluating any scheme. [Page 32]
Despite this, he works backward from other published figures to determine that the added ridership would be only 4% (or a mere 2.5% relative to the 2031 base case). That’s a rather sad increase considering that GO Transit has already seen a 30% rise in off-peak riding on Lake Shore with a move to half-hourly service. To be fair, however, the off-peak demand was very low, and 30% is not a huge achievement on that base. It is much smaller relative to all-day demand including the peak.
Even at a 30 minute frequency, riders have less concern about “just missing the train”, and this would be further reduced at 15 minutes. The real benefit is that GO trains could attract better shoulder-peak riding from those whose schedules are unpredictable or just don’t fit into the conventional commuter pattern. On routes with no off-peak service at all, the change would be breathtaking.
A problem here is to disentangle the benefits of:
- adding off-peak service where it does not now exist,
- operating more frequent off-peak service, something that can be achieved without electrification,
- operationally linking routes that now terminate at Union as through services on a common headway (e.g. Richmond Hill and Barrie), and
- the benefits of shorter and faster trains.
The recent provincial announcement bundles a great deal of this together. The goal is laudable – better service for all – but we need a rollout plan that will bring benefits as soon as possible to the most riders. What is particularly missing is any update on capacity studies for Union Station in light of the effect of frequent service on various GO routes. The 15 minute off-peak headways won’t strain the station, but much more frequent peak service will. Schabas has not allowed for this in his cost estimates nor recognized the knock-on effects if the passengers will not all fit into Union.
This is not to say that better service is a bad idea, but rather that we must go into such a scheme recognizing where there are system limitations. It would be odd to spend billions on electrification and yet be unable to handle the passenger loads at Union this implies.
Schabas cites GO’s calculation of travel elasticity in which the time saved on a faster electric train accounts for only a small part of a door-to-door trip. This explains GO’s conclusion that the overall saving is slight and therefore the effect on ridership would be small. What is missing, of course, is the benefit of the shorter wait time although this should not all be claimed as a benefit of electrification. Coupled with a claimed elasticity of -1 (every 1% decline in off-peak travel time will produce a 1% increase in ridership), Schabas notes:
According to the Passenger Demand Forecasting Handbook, going from a 60-minute frequency to a 30-minute frequency is equivalent to about a 10-minute reduction in journey time, on a typical 60-minute journey. On the GO Lakeshore route, this is about the same as a 16% reduction to the average journey. So introduction of half hourly all-day services should bring a 16% increase in non-peak riders. [Page 33]
(A point that is lost in the shuffle here is that station access by auto during the off-peak is very dependent on parking availability which tends to be low, on having someone drive you to the station, or on local bus service that is as frequent as the trains.)
Schabas reports that the initial effect of the 30-minute off-peak headways was a 15% bump in ridership (in line with the claim above), and recent comments by Transportation Minister Glen Murray give a 30% figure. The problem with any ridership projection is that the effects of change can be non-linear if a new service has a better perceived value than its numeric magnitude.
Continued growth is not surprising and it would accelerate with a move to 15-minute service to the point where, just like the Toronto subway, GO carried more passengers during the off-peak than during the peak periods.
Outside downtown Toronto, most GO stations are surrounded by large parking lots, often in industrial areas with few jobs or other destinations within walking distance. Many stations do not even offer frequent regular bus connections to local destinations. We see this as a chicken-and-egg problem. At the current level of service, this situation is likely to persist. But with improved service and more people using it, change begins to make economic sense. Station car parks can be redeveloped at higher densities. Feeder bus services can be improved. Already, GO has attracted impressive all-day traffic onto the Lakeshore line. With coordinated planning policies, and a proactive operating strategy, GO can do the same on the rest of the network. [Pages 34-35]
This certainly appears to be the attitude underlying Queen’s Park’s recent dedication to frequent service on all routes within a decade.
Schabas claims that the cost of electrification on the Lake Shore could be entirely paid for from additional fare revenue and cost savings of operating shorter off-peak trains. While this is an attractive argument, it is a dangerous one because it could commit GO revenue streams to debt service using revenue and savings it does not actually have. You cannot spend imputed savings (the difference between the projected cost of EMU operation and what otherwise would have been spent on diesel-hauled trains) because you never actually “spend” at the higher rate in the first place. As for the revenue, that depends on passengers actually showing up to ride, something that requires feeder services that have not been costed into the equation.
Unlike the Queen’s Park announcement, Schabas argues for electrification where the level of service warrants it, but that overall service improvements should proceed. I agree. The worst thing we could see is a “promise” high-jacked by a lengthy debate about “who goes first”, the operational effects on each line and the financing of an overall electrification project. This is not a pre-requisite to all-day service on 30-minute headways anywhere.
4. The Union-Pearson Express
Any discussion of the UPX must take account of the difficult history of this line rather than simply treating it as a recent, well thought-out scheme.
The original idea dates back to the days of a Liberal government in Ottawa when David Collonette was the Minister of Transport. His dream was that an airport link stopping only at Union and Pearson would be built and operated by the private sector, one of those magical arrangements in which a public benefit appears without any apparent cost. SNC-Lavalin took on the project, but they would not stay the course.
In time, the Feds washed their hands of the project, and ever-eager Ontario, under then-Premier McGuinty, took up the challenge. The line became a Pan Am Games project – how, after all, could Toronto welcome the world without a direct rail link to downtown from the airport? (Never mind that many of the games venues were nowhere near downtown or even in Toronto, or that the athletes and officialdom would travel on private buses or limos to their accommodation.) Don’t be the last city in the world without an air-rail link!
As projected costs rose and the economic viability of a premium-fare point-to-point service fell, SNC-Lavalin sought financial assistance from Queen’s Park for this no longer quite so profitable undertaking. McGuinty had the good sense to show them the door – what part of private enterprise and risk didn’t they understand, after all – but then made a crucial error of assuming the project, as is, for the provincial government.
I won’t rehash the fights over just what the UPX should be, and whether it could better have been designed as a limited stop service at a much lower fare premium than the originally mooted $22 or so. In the end, the line has grown at least two more stops – one at Bloor to connect with the BD subway, and one at Eglinton to connect with the Crosstown LRT. The fate of a station a few kilometres up the track in Weston is uncertain. (It may exist temporarily before the Eglinton Crosstown line at Mt. Dennis becomes a much more important station location in 2020.)
Schabas embraces the UPX because it has a positive “Net Present Value”, that is to say that it will make money even allowing for the capital investment. That’s a bit of a stretch on a few counts.
- We already know that the Provincial Auditor’s review found the whole scheme wanting and questioned the viability of the plan (just as the private sector “partner” had done before).
- At a one-way fare over $20 from Union, there is no hope that this route would perform any local benefit, nor would it attract commuters to the airport district. Schabas suggests that this could be overcome with reduced fares for regular users, but by doing so he undermines his own cost analysis.
- Buried in Appendix A3, we find that the presumed average fare is $15. This barely covers a reasonable allowance for lower fares at stations closer to the airport, let alone a substantial discount for airport workers (to $4.60 per trip for 40 trips/month). You can’t have it both ways: either the riders are cash cows, or we are giving away the service to make it more politically attractive.
- The cost (or foregone revenue) of other incentives to UPX use are not included in the calculation.
- Also in Appendix A3, we find that the “value” of a diverted road trip is set at $10. In other words, not only would a UPX rider contribute $15, on average, in fare revenue, the benefit calculation would include a further $10 because of the presumed reduction of auto traffic to the airport. Unfortunately, as I have said before, we cannot claim profits or pay off debt with the imputed value of a diverted trip. This is not real, bankable money.
- Benefits to riders of the UPX who formerly took the downtown express bus are calculated relative to the cost of that service. Again this is a “saving” that does not accrue to the UPX itself and cannot be counted as part of the “profit” of the service. Moreover, riders diverted from the bus contribute little to relief of road traffic.
- The 192 Airport Express bus, a regular fare service from Kipling Station, is not mentioned at all although it is very convenient for riders who live along the BD subway. Not all airport users want to go to Union Station, or pay dearly for the privilege of transferring off of the subway at Dundas West to take a more expensive ride (with a non-trivial walking transfer) to the airport.
- There is no mention of the possible role of the Eglinton or Finch LRT routes (or whatever technology one might prefer) as alternative ways for travellers to reach the airport. After all, if a transfer connection to the UPX will be provided at Mt. Dennis LRT station, why wouldn’t a rider simply stay on the LRT line all the way to Pearson?
These and other shortcomings of Schabas’ analysis suggest that his goal is to present this project in the best possible light, and this undermines the credibility of much other work in the paper.
5. Using the GO System to Relieve the Subway
Schabas and I agree on the basic premise that the GO system is underutilized as a means of relieving demand on the subway, but we differ quite substantially in the approach we would take to achieving this goal.
The basic problem is described in this paragraph:
There seems a very good case for using the GO system to relieve congestion on the subway into downtown Toronto from Scarborough, as several transit commentators have already pointed out. Currently there are 41,000 transit trips in the a.m. peak, by subway, from east and northeast Toronto into the downtown core. Mostly these passengers use the Danforth subway and change onto the Yonge line at Bloor-Yonge station.
The Bloor-Yonge station does not have the capacity to handle the passengers comfortably, and adding new platforms would cost hundreds of millions of dollars and cause massive disruption. In any case, the Yonge subway is approaching capacity. Although capacity can be increased with new signalling and longer trains, lengthening the line, including a planned extension to Richmond Hill, will also add substantially to demand. [Page 38]
It is amusing that in making this observation, Schabas does not see the irony that GO improvements could undermine the need for a Scarborough subway, but we will come to that can of worms later.
Schabas prefers to solve the subway problem by offloading passengers to GO at strategic points, notably Main Station, but also possibly Dundas West, Kipling and Kennedy. By contrast, my aim would be to ensure that, to the greatest degree possible, riders who might otherwise clog the subway never get on it in the first place.
There are, broadly speaking, three types of subway rider (current or potential):
- The local “in town” rider who probably lives close enough to a station to walk to it, or has a short feeder bus trip on a frequent route.
- The suburban rider who faces a considerable bus journey just to arrive at a subway station.
- The ex-urban rider who could use GO if service were available, but who instead travels on the TTC because it is (a) cheaper and (b) the only available service.
The first group is not a market for GO because their trips are short, they would probably have to travel out of their way to use GO, and the extra fare could not be justified given the relative inconvenience.
The second group depends on the subway, but faces increasing competition from the third group as we can see daily at locations such as Finch and Sheppard stations where the “local” subway is filled with “regional” commuters.
The third group should, to the degree possible, never get on the subway in the first place, at least not until they have used GO to reach downtown. GO service patterns and fares work against this.
Schabas proposes a shuttle service from Danforth Station to Union that could offload passengers from the Danforth subway at Main Station. The real question here is where did these riders come from in the first place? If from northern Scarborough, why are they not on a service in the Stouffville corridor. If from eastern Scarborough, why not on the Lake Shore? On the Yonge and Spadina subways, why not on the Richmond Hill and Barrie GO lines?
The answer is simple: most of those lines don’t have much service on them, and GO fares are disproportionately high for shorter trips. Moreover, within Toronto, the bus network is not organized to feed GO Transit even presuming a joint TTC/GO trip could be made at a reasonable cost.
Schabas proposes an underground connection between Main and Danforth Stations where trains coming from a western service (the Georgetown line) would terminate instead of at Union. This proposal makes several assumptions:
- Space is available for a platform and turnaround track.
- Trains can be turned quickly enough to achieve a headway of 5-10 minutes.
- A reliable supply of trains from western branches to feed an eastern shuttle is available.
- Track time is available on top of the through services to Lake Shore East and Stouffville.
There is no description of how the PM peak service would work, but presumably it would be the reverse of the AM pattern.
I will not get into the problems of turning trains around as these have been discussed at length here in other threads, but there is a basic issue with current operating regulations (federal) about the process of reversing trains.
Schabas presumes that this service would divert about 10k riders, or 200 passengers off of each of the roughly 50 trains over the two peak hours if the rate were sustained over that period. That’s a stretch as any transit planner knows there is a peak-within-peak and the maximum value is rarely held for the full interval. We would likely see 5k in the peak hour, and 7.5k over the peak two hours.
If the Danforth Shuttle operated every 10 minutes, one GO train would collect riders from about four subway trains, or 800/train at peak. More GO service might attract more riders, but there is a limit to how much service can be operated and how many would make the trek from one line to the other.
Although there are 41k riders going from the northeast to downtown (Schabas’ quoted value), only those whose destination is close to Union will be attracted by this scheme.
There is also the small matter of comparative journey times. The transfer from subway to GO will take about 10 minutes (2 from the train to the mezzanine, 3 to walk 250m to Danforth station, and an average 5 minute wait for a 10-minute GO service). Add in a 10 minute ride to Union and access time from GO to the destination in the core, and a trip time of 25-30 minutes would be common. Unless Bloor-Yonge Station is so crowded as to be impassible, this is comparable to the time one would consume merely by staying on the subway. The question obviously would be whether the extra fare and the extra walks just to use GO would be “worth it” for riders to avoid Bloor-Yonge.
Schabas goes to some lengths to justify his proposal when he would do better to emphasize the benefit of much more frequent GO service in diverting riders who originate in the outer parts of Toronto and in the 905 (a topic he covered in an earlier chapter).
Other questionable parts of his analysis include:
- A presumption that the large mezzanine at Main Station anticipated a future GO connection. This is completely false. Originally, this station straddled the fare zone boundary with suburban “zone 2” buses sharing the surface platform with “zone 1” bus and streetcar routes. Fare control was at the mezzanine level, and its size reflects the simple fact that this is a large box station and the vertical accesses are at its outer edges.
- Metrolinx is already working on passenger diversions at other subway stations. Well, no. Dundas West is not planned as a direct subway-to-GO connection, even though plans for this have existed for decades. The “link” at Dundas West will be a covered walkway ending on the east side of Dundas opposite the subway entrance. This will require the addition of a traffic signal at Edna and Dundas simply to get pedestrians to the subway entrance without J-walking. Connections at Kipling and Kennedy would not offload the subway unless we can convince those who now take bus feeder routes into them to use GO as an alternative access to downtown. Again, there are issues with transfer times and service frequencies on GO offsetting much of the saving in travel time one might have.
- Schabas completely misses the “Big U” concept of improved service on the Stouffville and one of the western corridors as a way to get longer-haul riders into the core without loading them onto the subway.
- There is no comment on the status of unused lines that are in The Big Move, but languish in the 25-year plan, notably the CPR corridor through northeastern Scarborough to North Pickering.
Again, we are faced with an analysis that is full of holes, and this affects the credibility of other parts of the report.
The Danforth Station proposal is a “solution” looking for a problem, and a much better alternative is suggested elsewhere in the report – run better GO services to handle long-haul trips rather than dumping them on the subway. Using GO for short-haul shuttles close to downtown wastes track time in the congested Union Station corridor and violates the basic premise that GO should be a regional carrier.
[This article continues in Part III with discussion of subway and LRT schemes.]
[This article was edited on April 25, 2014 to correct typos, make some minor stylistic changes, and to clarify the status of a Weston UPX station.]