OTTAWA — The winning bid for Via Rail’s high-speed project was so low that government evaluators checked repeatedly whether it might be impossible, according to documents obtained by The Logic.

In one of his last acts as prime minister, Justin Trudeau announced in February that the consortium called Cadence would take on Canada’s biggest infrastructure project ever. Cadence is to work with Via to plan, build and operate a new fast passenger rail line between Toronto and Quebec City. It’s to replace Via’s poky service on tracks mostly leased from CN, from which passenger trains are often shunted in favour of freight.

The group includes the Caisse de dépôt et placement du Québec, AtkinsRéalis (formerly SNC-Lavalin) and Air Canada, along with several companies affiliated with France’s national railway, SNCF. It was up against bid groups featuring Germany’s national rail company Deutsche Bahn and Spain’s Renfe.

First conceived as “high-frequency rail,” or “HFR,” the initial idea was that reliable, frequent trains on dedicated tracks would be enough to revive passenger rail in the densely populated corridor between Ontario’s and Quebec’s provincial capitals. The government has since upgraded the plan to a European-style high-speed rail, promising trains that reach 300 km/h.

Alto, the Via subsidiary championing the project, has estimated the cost to be between $60 billion and $90 billion. The planning phase alone is to take six years and cost more than $4 billion.

The documents, released in response to an access-to-information request, are materials prepared last fall, summarizing the bid evaluations for the deputy minister at Transport Canada and the top procurement official at Public Services and Procurement Canada (PSPC), the department running the competition.

It gave codenames to the three bid groups—the winner was labelled “Sleeper Car,” and it beat out “Locomotive” and “Diner Car.”

“All three proposals were high quality and reflected extensive consideration of market, technical, operational and customer experience factors in developing a realistic, defensible and actionable Train Service Proposition,” the briefing materials said. “All bidders demonstrated extensive real-world experience and knowledge they could apply in developing diverse, customer-oriented strategies to provide a strong customer experience.”

The winning bidder has “design experience with very relevant project experience, understanding of interdependencies between key functions from alignment identification and associated works, methodologies and understanding of the risks, accessibility, governance, and journey times,” the document went on.

All the scores except the best bid’s total were redacted from the documents the government released. It scored 107 out of a possible 120 points. But it stood out in one particular way.

“The preferred bidder’s total bid score is well above the other teams,” the document said. “This is due in large part to their very competitive commercial package.” Any details of that package are redacted.

PSPC’s spokesperson Nicole Allen confirmed to The Logic in an email what the documents don’t say directly: the “commercial package” covers “multiple financial aspects of the submission”—in other words, how much Cadence would charge.

The commercial proposal was worth up to 30 of the 120 points, and Cadence’s was so good it made the team at PSPC nervous, the briefing note said: “PSPC asked [subject-matter experts] and evaluators to confirm that the low commercial bid would not put the delivery of HFR at risk, and we were told it is not a concern.”

A few pages later, the document said the experts went over the numbers more than once to make extra sure: “In addition to the evaluators reviewing the bids individually, then in consensus, there were two series of reviews by the SME that confirmed the very competitive response did not put the project delivery at risk.”

Through spokesperson Martin Croteau, Cadence declined to say how it could so dramatically outbid the other groups, citing restrictions in its agreement with the government.

Matti Siemiatycki, director of the University of Toronto’s Infrastructure Institute, said in an interview that bid outcomes like this aren’t uncommon in major project procurements: accomplished teams of engineers and project managers typically come up with pretty similar technical proposals, but they win or lose on the dollars.

“You bid for the lowest technical score that will still keep you in the game, and then you aggressively drive down your financial bid,” he said. “The variation in the technical scores tends to not be all that wide, but the variation in the financial scores is really wide.”

In the best case, that just means the buyer gets a good deal.

“But what also can happen is that’s sort of an initial financial offering, and then the prices start going up, either before a contract is signed or as the project is going along,” Siemiatycki said. “Over time, it starts to look much more like what some of the higher bid prices were.”

The contents of the bids have not been released, and PSPC’s Allen said they won’t be: “To protect confidentiality, we cannot provide financial details of the bids,” she wrote. 

Without knowing those, Siemiatycki said, it’ll be impossible to track whether costs are creeping up.

Alternatively, a cheaper bid can be reflected in workmanship. Siemiatycki pointed to problems with a new Toronto courthouse built through a public-private partnership, and a lawsuit over allegedly shoddy floors in a marquee hospital project.

“In practice, there’s all sorts of ways that risk, even though it’s transferred on paper, is not necessarily transferred in practice, and it can start to boomerang back to government in all sorts of ways,” he said.