Why it’s so hard to deliver major infrastructure projects on time and on budget

Brian Kelcey, vice-president of public affairs for the Toronto Region Board of Trade, talks to TVO.org about why large infrastructure projects hit delays and cost overruns — and what we can do to get them back on track
By Mary Baxter - Published on February 4, 2019
A new patient-care wing at Cambridge Memorial Hospital was supposed to open in 2016, but the project has been beset by delays. (cmh.org)



In November 2016, a new 240,000-square-foot patient-care wing was supposed to open at Cambridge Memorial Hospital.

But three years later, construction of the new wing — part of a larger $187 million project that includes renovations to the existing building — is only now wrapping up. (The larger project, which was scheduled for completion in 2019, has a revised end date of 2021.)

In November, Patrick Gaskin, president and CEO of the Cambridge Memorial Hospital, penned an apology to the community for the delay. “The challenge from our perspective is there has been the absence of the coordination of work and the sufficient resources on the site,” the hospital chief told the CBC in December. 

The expansion isn’t the only major infrastructure project in the province that’s gone off the rails. The Waterloo Region LRT, which was was supposed to be up and running by December 2018, now won’t be operational until spring 2019. And completion of the revitalization of Union Station in Toronto is expected to be this year — nearly four years after its original target and millions over budget.  

Brian Kelcey, the vice-president of public affairs for the Toronto Region Board of Trade — an organization that represents and advocates for the business community — knows a thing or two about managing large construction initiatives. He helped the mayor’s office in Winnipeg reform the city’s infrastructure procurement process in the mid-2000s, and he’s advocated for greater flexibility with construction and bid models (which could involve engaging more contractors to work on different project components at the same time) and for more “fast infrastructure” (for example, standardizing processes and incorporating pre-building modular components to make onsite assembly more efficient).

TVO.org spoke with Kelcey about why major infrastructure projects hit delays and how governments can get them done — on time and on budget. 

Why are major infrastructure projects vulnerable to cost overruns or delays?

It's the same cartoon relationship everybody jokes about — somebody who's doing home renovations on your kitchen comes in and says the estimate's going to be $20,000 to gut your kitchen. Then the contractor finds a surprise somewhere else and says now it's going to be, because of a pipe issue, $25,000. Dealing with that pipe issue creates a delay.

So the contractor goes off to another work site, and you're stuck asking yourself, “Do I give up midway through and a have to pay an ‘end of contract’ fee to this contractor for the work they've done? And then go through the struggle to go and find another contractor who's going to finish it the way I want? Or do I keep paying the money and hope the delay goes away?”

Now multiply that across a couple of electrical firms, an engineering firm, an architectural firm, as many as five or six different contractors and subcontractors. We’re spending, in some cases, millions or billions of dollars.

No matter how many different ways people come up with to structure deals to prevent these things, one in 10, one in 20 of these projects are just going to get much more complicated because something's going to go wrong. And complications cost millions when you're dealing on that scale.

So the problem isn’t necessarily the model used to manage or build the projects — it’s their complexity.

Yes. People often ask the question: Why does it take so much longer to build something now than it did 80 years ago? Part of it is how we build — work is pieced out to a lot more different firms. 

But another part of it, no question, is that we've got tougher safety rules. So the more variables there are, the more those variables can go wrong.

How do infrastructure buyers try to mitigate the risks?

Most of the energy in the last 20 years or so has been around finding different ways to use money and cash flow, and how you pay builders to manage those variables. That has delivered some success on a lot of projects.

Why don’t those strategies always work?

However much it doesn't look like it, I think Canadian governments and builders are both getting better at preventing problems. But high-profile problems are still going to pop up for the time being, though.

Payment-on-completion bid models can grind to a halt pretty quickly because delay on piece B in the process can then create delays on pieces C and D and E that cascade throughout the system.

If I'm a builder with a bunch of different subcontractors, I now have to bridge finance my construction of that particular project to meet those deadlines and pay the workers in order to earn the money that I'm going to get paid at the back end from building this. If the project reaches a point where the cost of carrying on and trying to meet those targets is so high for the builder, it's worth it just to slow down and be working on other things.

Often the choice that the infrastructure buyer is left with is, do you reward failure by throwing money at the problem? Or do you find some way to try and keep pushing the same contractors, who have already slowed down, to meet those targets?

It's a difficult choice. You don't want to pay more for the project that you thought you were going to get on time or on budget. But if somebody is willing to walk away on the other end, or if they're already working on other things, you can only do so much through the process to resolve so many different challenges.

What are some solutions?

In Asia and Europe, people are building subway lines, LRT lines, or bridges in a few years, at light speed. You're starting to get those stories in North America, as well. It's been because more thought and investment has been put into the planning and bidding stages to reduce the margins of error. They’re using more modular design and standardizing more steps to this process by getting more contractors engaged in one go, to try to build whatever it is more quickly, rather than piecing out the bid more slowly.

If we design these things to be built quickly, if we design the process to get to a decision quickly, if we make some decisions to shave off a few of the bells and whistles that we're sometimes prone to build, to make it easier to prefabricate components of these buildings or structures before they ever reach the construction site — the more of that that we can do, the more we winnow down that risk of elapsed time and cascading problems.

This is one in a series of stories about issues affecting southwestern Ontario. It's brought to you with the assistance of faculty and students from Western University’s Faculty of Information and Media Studies.

Ontario Hubs are made possible by the Barry and Laurie Green Family Charitable Trust & Goldie Feldman.

Related tags:
Thinking of your experience with tvo.org, how likely are you to recommend tvo.org to a friend or colleague?
Not at all Likely
Extremely Likely

Most recent in Southwestern