Hydroelectric dams like Muskrat Falls are particularly vulnerable to cost and schedule overruns, an expert testified Monday as a public inquiry into the Labrador hydro megaproject got underway.
Oxford University economics professor Bent Flyvbjerg, who is described on his faculty page as “the most cited scholar in the world in megaproject planning and management,” took the stand in Happy Valley-Goose Bay, N.L., to outline common mistakes and causes of cost overruns of such projects.
He said hydroelectric projects are particularly risky, seeing average cost overruns of 96 per cent and schedule overruns of 32 per cent, with little change over time since the 1950s.
The report also noted that dams doubled their budgets on average.
“The context is that building a hydroelectric dam is very risky,” Flyvbjerg said.
“It seems to be a very persistent phenomenon that has been with us for decades.”
The massive Labrador dam and powerhouse harnessing the lower Churchill River near Happy Valley-Goose Bay still isn’t complete, but its price tag has climbed to $12.7 billion.
The 824 megawatt hydroelectric dam, being developed by the Crown-owned Nalcor Energy, will eventually send power to Newfoundland and later Nova Scotia through subsea cables.
The independent inquiry, led by provincial Supreme Court Justice Richard LeBlanc, is examining how the project was approved and executed, and why it was exempt from oversight from the Public Utilities Board.
Flyvbjerg, who has also consulted on megaprojects in several European and Asian countries, was commissioned by the inquiry to compile a report giving global context and best practices for success in megaprojects, but not specifically looking at the case of Muskrat Falls.
The report, co-authored by Alexander Budzier, surveyed 274 megaprojects around the world, including “several” in Canada. Flyvbjerg defines a megaproject as one that costs more than $1 billion.
Data from the report shows that hydroelectric dams are “riskier than all other projects, except nuclear.”
Flyvbjerg said many projects are hit by underestimated costs and schedule, overestimated benefits, and lack of independent oversight.
The report cited optimism bias and political bias as the root causes of high cost overruns.
With political bias, he said, project managers intentionally “overestimate benefits and underestimate cost and schedule” to ensure their projects get a green light.
Optimism bias involves the “systematic tendency for people to be overly optimistic about the outcome of planned actions, includes underestimating the likelihood of negative effect,” he said.
“The real cause of the cost overruns and schedule overruns is that planners were optimistic about all these things,” said Flyvbjerg.
The report recommends outside, independent oversight in the early stages to “de-bias” projects. Flyvbjerg said too many project managers take an “inside view” of projects.
Other recommendations include as much transparency of project development as possible, and assessing groups of similar projects to adjust estimates early on, as well as bringing in experienced advisers.
The summary of the report includes a note that “the maturity of leadership in capital investment projects is often perceived to be lacking.”
When questioned by co-counsel Barry Learmonth, Flyvbjerg said projects are generally more successful when stakeholders, like Indigenous groups and environmental groups, are engaged in the process early on.
“We’ve generally found that it does make sense to involve all stakeholders as early as possible,” said Flyvbjerg.
“It’s like with the biases, if you don’t do it, it’s going to come back to haunt you ... it’s much more expensive to take these things into account if you have to do it later on in the process.”
Flyvbjerg is the first of many witnesses to testify before the inquiry, the first phase of which is focused on the project’s sanctioning and the involvement of the Public Utilities Board. Phases two and three, scheduled for next year, will look at construction, oversight and policy moving forward.
The hearings will conclude in August 2019, with a final report expected by Dec. 31., just after the next provincial election.