- High-speed rail construction — Canada's progress against international high-speed-rail benchmarks.
- Pipeline construction — how Canada builds major energy linear infrastructure.
- Electricity generation capacity — whether the grid is being built fast enough for demand.
Every figure on this page is capital construction cost. Long-term operations, maintenance, and lifecycle payments under DBFM and BTO concessions are excluded where they can be separated from the construction figure. The Eglinton number is Crosslinx's latest construction-cost disclosure (CAD 12.58 B), not the all-in DBFM concession value (CAD 13.08 B including 30 years of maintenance).
Eleven recent OECD/G20 light rail and light metro projects, plotted three ways: capital cost per kilometre, calendar years per kilometre, and capital cost against share of line tunnelled. Toggle between the three views below.
Capital cost only — long-term O&M under DBFM/BTO concessions excluded. Eleven recent OECD/G20 projects, sorted ascending. Eglinton highlighted in red.
Eglinton lands at $662 M/km in capital cost — roughly 4× Ottawa's Confederation Line and almost 9× Seoul's Sillim Line.
All eleven projects sorted with Eglinton first, then by ascending capital cost per kilometre within each tier.
| Project | Country | Tier | Length (km) | Tunnel % | Build period | Total cost | Capital/km (CAD) | Yrs/km |
|---|
The Eglinton figure is Crosslinx's latest construction-cost disclosure (CAD 12.58 B), not the all-in DBFM concession value (CAD 13.08 B incl. 30-yr maintenance). Costs are converted to CAD at period-average exchange rates and rounded to two significant figures.
Honolulu's fully elevated metro (CAD $444 M/km), Sydney's surface light rail ($238 M/km), and LA's partly-tunnelled K Line ($203 M/km) all came in below Eglinton on capital cost, while Seoul's fully-tunnelled Sillim Line came in at $76 M/km. The factor driving Eglinton's per-km cost above its peer set is procurement, scope evolution, schedule extension, and labour rather than engineering geology. The NYU Marron Institute's Transit Costs Project, which controls for tunnel share across hundreds of projects worldwide, reaches the same conclusion: alignment mix explains only part of cost variance, and the North American premium survives the control.
Eglinton's 0.76 years per kilometre sits in the bottom quartile of the set. Only Sillim, which built fully underground, was slower, at 0.83. Surface-only projects delivered at 0.25–0.45 yrs/km; Korean fully-tunnelled lines averaged ~0.5 yrs/km. Eglinton needed roughly double the years per km of comparable OECD light-rail builds.
Sillim, Gimpo Goldline, and Montréal's REM all delivered fully grade-separated automated light metro at CAD 76–124 million per kilometre in capital cost within the last seven years. That range, not Eglinton's $662 M/km, is the defensible benchmark for new Canadian projects (Calgary Green Line, Hurontario LRT, Surrey-Langley SkyTrain).
Of the seven projects that built at 0.5 yrs/km or faster, all but Sydney ($238 M/km) stayed under $200 M/km. Eglinton is alone in the upper-right of the cost-time plane, with high capital per km and a slow build. The two axes track each other, suggesting a shared underlying cause (procurement structure, contract management, scope discipline) rather than two independent problems.
Eglinton cost CAD 662 million per kilometre, above every project in the eleven-project set and almost nine times Seoul's Sillim Line, and took 14.5 years to build.
- Compared with
- Eleven recent OECD and G20 light rail and light metro projects, tiered by grade separation, with Toronto's Eglinton Crosstown as the Canadian case.
- Where Canada sits
- At $662 M/km in capital cost, Eglinton sits above its whole peer set, roughly 4× Ottawa's Confederation Line. Its 0.76 years per kilometre is in the bottom quartile of the set.
- Which way it is moving
- Flat. The page covers one completed project, so there is no trend line. Montréal's REM is cited within the CAD 76–124 million per kilometre range for automated light metro.
- What the verdict follows
- Position in the peer set. Eglinton is alone in the high-cost, slow-build corner, and the page attributes the premium to procurement, scope, schedule, and labour rather than geology.
- Data basis
- Office of the Auditor General of Ontario (2018 Annual Report, §3.07), Crosslinx project filings, and comparator project sources. Costs are nominal, converted to CAD at period-average rates, and rounded to two significant figures. Page last reviewed July 2026 · Data current to February 2026, the latest construction cost Crosslinx has disclosed.
Methodology notes
Cost basis — capital construction only. All figures on this page are capital construction cost; long-term operations, maintenance, and lifecycle payments under DBFM/BTO concessions are excluded where they can be separated. Eglinton uses Crosslinx's latest disclosed construction cost of CAD 12.58 B rather than the CAD 13.08 B all-in DBFM concession value that includes 30 years of maintenance. Ottawa Confederation Line uses the ~CAD 2.1 B capital cost reported by OC Transpo; the 30-year maintenance is paid separately at ~CAD 5 M per month. Sillim's KRW 560 B and Gimpo's KRW 1.65 T are the construction costs under each Korean BTO concession; operating revenue is collected separately by the concessionaire. Canberra's AUD 675 M is the final construction cost; the 20-year O&M is contracted separately. LA K Line, Honolulu Skyline, REM, and Utsunomiya are reported as capital construction cost in their source filings.
Currency and vintage. Costs are nominal in local currency converted to CAD at period-average exchange rates and rounded to two significant figures; no adjustment to constant 2026 CAD has been applied. Honolulu Skyline is in phased opening with completion projected to 2031; the full FFGA project cost over physical length is used, which is conservative against the completed comparators. Sydney's cost is the NSW Auditor-General's revised total project cost (Aug 2020) of AUD 3.1 B, which includes early enabling works, business assistance, and financing costs attributable to project delays; the originally tendered construction budget was AUD 1.6 B. Edinburgh's reported cost is GBP 776 M for the truncated 14 km Stage 1; with related business-disruption costs the Lord Hardie Inquiry recorded approximately GBP 835 M. Construction-start dates use the start of major civil works.
Alignment mix (surface / elevated / tunnel). Cost per km over a whole line depends partly on how much of it is tunnelled or elevated, which is why the comparator set is tiered by grade separation and a dedicated cost-vs-%-tunnelled view is provided. Element-level costs (a project's tunnelled kilometres priced separately from its surface kilometres) are not published by any project in the set, so the comparison controls for alignment at the project level rather than decomposing it. Independent corroboration: the NYU Marron Institute Transit Costs Project, whose database spans hundreds of projects and explicitly models tunnel share, finds alignment mix explains only part of cross-project cost variance and that the North American cost premium persists after controlling for it.
Companion workbook. The editable series and underlying calculations live in Building_Light-Rail-Transit_Data.xlsx. Only series the workbook actually contains are described here.
Primary sources by project
