Journey times
Transit City, TTC, and Metrolinx documents make little mention of journey times or speeds.
Surface LRT can be ideal for relatively short trips: the Spadina line is only 3 km long; the St Clair line is 6 km long, but the typical trip is less than 3 km. Although the average speed is fairly slow, only 15 to 20 km/h, journeys still take only 10 or 15 minutes. These lines are close to downtown, and passengers can complete their trips on the subway in another 10 or 20 minutes.
Typical journeys on the Transit City LRT lines will be very different. With an average speed of about 25 km/h, a trip from the east end of the 13-km Sheppard Line to Don Mills station would take about 30 minutes. Total journey time from Malvern to downtown Toronto using the subway would be about 1 hour and 15 minutes, with two changes. This is not much faster than driving in rush-hour traffic and slower than the existing GO rail service, which takes 26 minutes from Agincourt Station to Union Station. The Transit City schemes will certainly improve local mobility, but will do little to further Metrolinx's regional transport objectives.
Where LRT runs on the surface, it will be a bit faster and more reliable than the current buses, because vehicles will not have to share a lane with cars and may have priority at intersections. Because LRT cars carry about three times as many passengers as a bus, and cost more to operate, passengers are also likely to face longer wait times, as TTC will likely operate fewer vehicles at longer intervals. Because LRT will run along main streets like Sheppard and Finch, many passengers in neighbourhoods like Malvern will need to change to a local bus or walk some distance to complete their trip home.
Along Eglinton between Laird and Keele, the LRT will run in a deep tunnel, and will be unaffected by traffic. However, it will still run at only 30 km/h on average. This is because TTC is building 13 stations, with an average spacing of about 700 metres. While older parts of the Toronto subway have stations spaced as little as 500 metres apart, normal practice is for metro stations to be at least 1 km apart even in city centres, and 2 km to 5 km apart in lower-density suburban areas. TTC's plans make sense only if Eglinton Avenue is substantially redeveloped, at much higher densities, a prospect that is at best, distant.
Ridership and Revenues
TTC has prepared forecasts of the ridership on each LRT, although it is not clear how these were developed.[1] TTC breaks the ridership down into "existing," on the current route, "diverted," currently using a different route, and "new" ridership to 2021. The forecasts do not, however, distinguish new ridership that would be generated by population growth and development, even if the LRT line is not built.
TTC's figures are usually for the full schemes (for example, for the 30-km Eglinton Crosstown line extended to Pearson Airport), and not necessarily for the somewhat shorter lines now envisaged. Generally, LRT lines serve local and feeder trips, so ridership is roughly proportional to route length. Some passengers may travel from Eglinton and Yonge to the airport, but with the slow speeds and long journey times, the service may not be attractive to many riders. Passengers making longer journeys might use GO Rail or GO Bus services, but more likely, they will drive.
Metrolinx provides some total ridership figures for the "Big 5" schemes in a series of project fact sheets, but has not specified the proportion expected to be new riders.[2] The Eglinton Crosstown BCA gives a figure of 105 million boardings in 2021,[3] which works out to 350,000 riders per day with an annualization factor of 300.[4] This is double TTC's forecast of 176,000 per day, and about three times as much as current bus ridership on the route. It is also two-thirds of the current traffic on the Bloor Danforth subway. We question the validity of this figure.
The Metrolinx BCA provides a figure of $5.6 million per year or $101 million NPV for incremental revenues on the Eglinton LRT. This figure implies new ridership of less than 10,000 per day. We do not understand how this figure can be consistent with Metrolinx's forecast of 105 million boardings on the line in the same year. Again, we find the figure questionable.
Figures in the Metrolinx BCA for the Scarborough LRT states are even less plausible: base-case ridership is forecast to be 112.7 million trips in 2031, or about 373,000 per day. This is almost 10 times current ridership, and three times the ridership implied elsewhere in the report. The figure is actually similar to ridership on the Danforth subway, and several times beyond the capacity that the BCA is assuming would be provided.
Bizarrely, the Metrolinx BCA then states that additional ridership with extension of the line to Malvern will be only 600,000 to 800,000 passengers per year, or about 2,000 per day. These cannot be simple typographic errors: the figures are repeated several places in tables and text. While most riders on the extended line would otherwise ride the bus a bit further before boarding the RT, we would be very surprised if extending the Scarborough line to Sheppard does not attract at least 15,000 new riders, beyond the underlying growth that is expected.
We use TTC's forecasts, adjusted pro-rata for route length, for the new LRT lines. To estimate the benefits, we need to know what share of projected ridership is new ridership, that is, riders who have switched from another mode of travel to transit specifically because of the scheme. Neither TTC nor Metrolinx have disclosed this figure. We think new ridership might realistically be about 25% of the projected total.
With complementary policies, the incremental ridership on the Transit City LRT routes could double to 2033. However, this will happen only with strong land-use planning policies. Higher-density development has been slow to materialize around the Spadina and Sheppard subways and the Scarborough RT, even though these are fully grade-separated lines with relatively fast average speeds. The Transit City schemes are relatively slow, local transit lines. Although they improve local accessibility, longer-distance trips are still overwhelmingly made by car.
Other passenger benefits
Surface LRT lines are typically 10% to 20% faster than surface buses in mixed traffic, so a typical 25-minute bus journey would become a 20-minute LRT trip. The 5-minute saving is worth about $1.12 to a typical passenger, at the rate of $13.52 per hour that Metrolinx has used in its calculations. The benefits may be reduced, however, if passengers using the LRT need to make an additional transfer to a local bus. We ignore this possibility, but in some circumstances, it will cancel out much or all of the benefit of LRT. We assume passenger benefits, not captured by fares, of $1.50 per existing rider for the Sheppard, Finch, and Eglinton lines, and $0.75 for new riders.
For the Scarborough line, Metrolinx estimates travel time benefits to transit users and motorists together worth about $1.3 billion NPV. While it does not split these, time savings to motorists must be very small, as with so few new riders, there cannot be many avoided car trips. According to the BCA, time savings are valued at $13 per hour. This implies time savings of about 14,000 hours per day, or about 60,000 daily riders each saving 10 minutes compared with riding the bus to the existing terminus at Scarborough Centre. We assume a benefit of $2.00 per trips for existing riders, and $1.00 per trip for new riders.
Metrolinx's BCA reveals that the Eglinton Crosstown scheme has a benefit:cost ratio of less than 0.4. In this case, Metrolinx has added in "reliability benefits," something that is not included in any of the other Benefits Case Analyses.
The value of the reliability benefit appears to be about $0.50 per passenger, which is credible, given the vulnerability of buses on Eglinton to traffic congestion. Metrolinx provides no empirical evidence to support this estimate, but we accept that existing transit riders will benefit and that this improvement may not be reflected fully in calculations of time savings.
Figure 15: This figure from the Eglinton Crosstown BCA includes "Reliability Benefits" in the Benefit:Cost ratio for the Eglinton Crosstown. Even so, benefits are substantially less than costs for Option 3, the scheme currently under construction. Benefits might exceed costs for Option 1, which would be entirely underground. The increased capital costs would be offset by greater benefits because there would be faster trips, with fewer stations between Laird Drive and Kennedy.[5]
However, Metrolinx has also forecast massive growth in transit ridership on the corridor, and seems to be applying the benefit to all riders, not just existing riders. This seems very optimistic. While existing riders will benefit from improved reliability at no cost, new riders consider reliability in their decision to switch to transit from automobiles.
Road traffic impacts
There will be some reduction in car traffic from new riders attracted onto the LRT lines. However, the LRTs do not seem likely to attract many new inter-regional trips, because of the slow journey times and relatively poor connections to commuter rail. Car trips diverted onto transit will mostly be fairly short, so this benefit might therefore be worth about $2.50 per trip, on average.
Because Toronto's streets are generally narrow, providing an exclusive median for LRT means that some sections of Eglinton, Sheppard, and Finch may be reduced from three lanes to two lanes. Even where it is possible to maintain the number of through traffic lanes, left-turn lanes will disappear except at intersections, and access to driveways and parking on either side may require a time-consuming U-turn, as it often does now on St. Clair West and Spadina Avenue. There will therefore be road user disbenefits. We estimate these at $1,000 per day for each kilometre of road affected. This rate assumes 10,000 motorists per day are each delayed (that is, they face longer trip times) by 27 seconds on average, because of the loss of a road lane.[6] Actual road user disbenefits will depend on details of road design that have not yet been disclosed.
The Metrolinx BCA for the Scarborough LRT scheme estimates that the scheme will take 60 million car-km off the roads. We find this puzzling. This implies that each of the 800,000 new riders would otherwise drive 75 km per trip, or 150 km per day. We don't think the extended Scarborough RT is likely to attract many new riders who would otherwise drive to Hamilton. In the same section of the BCA, Metrolinx estimates automobile operating cost savings to be worth $60m in 2031, which it says are calculated at $0.60 per km. Applying long division, this implies only 1 million car-km taken off the road in 2031. This is probably a gross underestimate, but is at least consistent with the estimate of 800,000 new riders per year.