LRT project delivers expensive lessons

We love building things like light-rail lines with publicprivate partnerships but have some amazingly screwed-up rules for them, says a report on the lessons the city learned from planning the first phase of rail construction in Ottawa.

Cities have to beg for provincial and federal money before they have a clear idea what their plans will cost, and sometimes those estimates are too low when they get locked in. The system pushes for cheap over good, and also gives bidders an incentive to squeeze as many costs as they can out of construction budgets and into decades-long maintenance agreements, where they’ll get much less attention.

The report is a co-production of consulting firms Deloitte and Boxfish Group, part of a thick bundle of paper laying out the city’s plans for the next, $3.6-billion stage of rail and related projects. Boxfish’s proprietor, you’ll recall, is Brian Guest, a former aide to Bob Chiarelli and Paul Martin who’s made good business out of consulting for the city, including on both phases of rail. He’s the brother of Robyn Guest, a senior city official who is married to Chris Swail, who’s the head of the rail planning office.

So keep in mind that this assessment is in part an inside job. The biggest foul-up it identifies on the city’s part is an error in handling the price of electricity: forgetting that big power consumers, such as a 12-km electric rail line, are charged based on the maximum electricity they demand at once, not just the total amount they use in a day. Won’t do that again.

Nevertheless, there’s a lot of insight into the huge deal the city made with Rideau Transit Group to build and maintain our rail system.

The work makes clear what a mess it is to scrape together funding for such a big project, largely because federal and provincial governments have way more money, but they won’t pay for any expenses the city incurs before they’ve formally agreed to kick in. The city has to get its request in as early as possible.

“On the Confederation Line project, the initial estimate of $1.8 billion was developed and submitted without allowance for inflation, and appropriate amount of contingency ... or other financing costs typically associated with (public-private partnerships). This estimate was then used as the basis for the federal and provincial funding announcements, which capped their contributions at one-third of these estimated costs,” the report says.

Finer engineering work took the estimated price up to $2.1 billion even before the provincial funding announcement came through, and then even more work led the city to conclude that it couldn’t build the rail line for even that much.

It took a major rethink, including rerouting the line under Queen Street, to pull it off.

This rigidity was arguably a consequence of Jim Watson’s hoping to kneecap a rail plan associated with Larry O’Brien, whom he was about to challenge for the top job at city hall after quitting provincial politics. But $600 million is still what the city had asked for, and that’s what the city got.

The same system puts a lot of pressure on the upfront construction cost but much less on long-term expenses. In the case of Ottawa’s rail project, that’s a 30-year contract to maintain the train system after it opens in 2018, which was worth $2 billion when the city and Rideau Transit Group signed their agreement.

The consultants think Ottawa got a good deal, all things considered. “However, it is possible for a bid that is far more expensive on the capital cost to have the lowest cost overall,” the report says, and the financing system wouldn’t handle such a case well.

One big thing that worked: telling bidders they’d be dumped if they didn’t agree to take on certain elements of construction risk, like unexpected ground conditions that led to cave-ins in the downtown tunnel, but other bidders did.

If you’ll take on all the risk, period, that’s perfect, the city told the bidders. If you’ll take on the risk but only subject to what’s in a geotechnical report we’ve had done, that’s second best. And so on, down several more steps. If another team agrees to take on more risk than you do, bidder, you’re probably out.

At first, all the bidders said they couldn’t get private financing — which the province demands, to get private-sector discipline injected into the project — if they took on all the risk for the tunnelling because no lender would be crazy enough to invest in a project with such a condition. But somehow all the bidders managed to talk their lenders into it.

“(E) ven though all the bid teams did not want to assume the full risk, they could not convince themselves that their competitors would not find a way to accept the risk,” the report says.

Ontario is deeply in love with public-private partnerships like this one but it’s still learning how to do them well. These are multibillion-dollar lessons, so the sooner we grasp them, the better.