Skip to main content
analysis
Open this photo in gallery:

Electric company workers fix broken power lines and poles after Category 1 Hurricane Larry hit St. John's on Sept. 11, 2021.GREG LOCKE/Reuters

When your power goes out, there could be any number of culprits. High winds may have brought trees and branches down over power lines. Equipment can fail due to corrosion or water infiltration. Perhaps, somewhere deep in the bowels of the nation’s electrical system, a squirrel has been fried to a crisp. Even weather in outer space – geomagnetic disturbances arising from the sun – have been known to knock out electricity transmission networks.

The darkest possibility is that your local utility skimped on maintaining its equipment. Pacific Gas and Electric Co., California’s largest utility, has become a poster child for this. Many of the state’s biggest fires in recent years have been linked to mishaps on its network: wires downed by failing hardware, aging towers collapsing and weakened trees blown onto power lines.

PG&E has been accused of pursuing profit at the expense of proper maintenance. In 2019, The Wall Street Journal said the utility didn’t know about the condition of much of its equipment, or even the age of many of its towers – things a well-managed utility ought to know.

Electricity is a critical service for most households, and with rising adoption of electric heating and vehicles, its importance is growing. So it would be helpful to identify Canadian utilities pursuing a “run-to-failure” philosophy, along with those that are investing in maintaining their networks or even improving reliability.

To that end, The Globe and Mail turned to arcane utility reliability benchmarking data. This data is available from utilities themselves, and regulators in some provinces also publish them. Of the utilities we studied, Saskatoon Light and Power stood alone in refusing to provide it. To our knowledge, this is the first time the data has been published nationally.

One of the most common benchmarks is known as the System Average Interruption Duration Index (SAIDI). It measures the total duration of outages experienced by the average customer during the year. There’s also the System Average Interruption Frequency Index (SAIFI), which represents the average number of outages customers experience each year. In both cases, lower numbers are desirable.



The Globe compiled SAIDI and SAIFI data for 15 major Canadian utilities. These statistics reveal that overall, Canadians enjoy reasonably reliable electricity service – and the situation hasn’t changed much for a decade.

Yet the performance of Canadian utilities is far from uniform. Customers of Newfoundland and Labrador Hydro suffer the most unreliable electricity service by a wide margin. They can only dream of the level of service enjoyed by citizens of major cities such as Calgary, Toronto, Ottawa and Edmonton.

The SAIDI and SAIFI figures reported for Newfoundland and Labrador Hydro reflect unusually high unreliability only for its 38,000 direct customers, who are concentrated in remote areas. Most Newfoundlanders (roughly 274,000) are customers of Newfoundland Power, the island’s primary distributor, which purchases nearly all of its power from Newfoundland and Labrador Hydro. Newfoundland Power’s SAIDI in 2022, including the impact of loss of supply from Newfoundland and Labrador Hydro but excluding major events, was 3.58; its SAIFI was 2.82.

The two metrics are influenced by many factors – some of which lie outside utilities’ control. For instance, they can’t deflect the tracks of hurricanes, nor govern how often lightning, ice storms or wildfire strike. For this reason, SAIDI and SAIFI figures reported by utilities often exclude “major events.”

There are also attacks from foreign actors. A report released in late June by North American Electricity Reliability Corp., a non-profit based in Atlanta, warned that energy transmission networks face “growing attack surfaces” and an “ever-changing threat landscape.”

“Hostile nation-states persist in targeting North American critical infrastructure, constantly evolving their methods to compromise the grid’s reliability, resilience and security,” it noted.

Geography may help explain Newfoundland and Labrador Hydro’s poor showing. Francis Bradley, president and chief executive officer of Electricity Canada, which represents Canada’s utilities, said that province’s electricity systems are exposed to extreme conditions commonly found in Canada’s Far North.

“They probably get more ice loading on their systems than anywhere else in the country,” he noted.

Reliability stats from city utilities like Toronto Hydro and Calgary’s Enmax Corp. are invariably superior to utilities that serve entire provinces, such as Manitoba Hydro or New Brunswick Power. Studies have found a direct relationship between reliability and circuit length: unsurprisingly, longer circuits suffer more interruptions. Utilities that have large amounts of infrastructure in remote or mountainous terrain may face greater maintenance challenges, including restoring power after extreme events.

Providers with expansive networks have “a lot more exposure to potential tree contacts, or defective equipment, because they need a lot more equipment to serve these customers,” said Harneet Panesar, chief operating officer of the Ontario Energy Board.

A city utility, on the other hand, typically has a shorter drive to fix damaged infrastructure. And a greater proportion of electrical systems lie underground, shielding them from many hazards.

“Comparing reliability statistics amongst different companies is fraught, because there are no two companies that are the same,” said Mr. Bradley. “Simply because the geographic differences between companies are so enormous, to be able to find true peers is a challenge.”

Even so, in their public reporting, some companies acknowledge the obvious: If they fail to maintain their networks, equipment performance will suffer, which can compromise reliability and lead to service interruptions. They can regularly service and upgrade substations. They can trim trees along rights-of-way. They can replace rotten poles. They can install new technologies that enable rapid detection of outages and pinpoint their location, which can speed up service restoration.

Some utilities may be inclined to direct money to other priorities, such as lavish executive compensation and dividends to shareholders. A utility that takes a “run to failure” approach with equipment may face declining reliability in the years ahead.

“We know that reliability and cost are the two main drivers for customer satisfaction,” Mr. Panesar said. “There’s the old saying that what gets what gets measured gets managed. We expect utilities to manage their reliability.”

Mr. Panesar, who supervises the OEB’s compliance and enforcement team, said worsening SAIDI and SAIFI can be early indicators of underinvestment in reliability. There’s no simple threshold that signals trouble, but declining performance could spark conversations about distribution system planning and the condition of assets, and trigger demands for maintenance records.

In 2021, the OEB inspected a licensed electricity distributor, E.L.K. Energy Inc., that provides power to the towns of Essex, Lakeshore and Kingsville in Ontario. The utility couldn’t provide records demonstrating that it had conducted regular inspections of its assets – because it didn’t have any. E.L.K. blamed this on its former management, vowed to improve and agreed to pay a $5,000 administrative penalty.

Though used internationally, reliability statistics are a Canadian strength. Roy Billinton, a professor who joined the University of Saskatchewan in the 1960s, specialized in power systems reliability and wrote volumes of influential books and papers on the subject.

“He really has been the fellow who laid all the academic foundation of how everybody around the world now measures and categorizes their reliability statistics,” Mr. Bradley said. “To people who work within the reliability statistics community, this guy’s a god.”

Publicly reported reliability stats typically are highly aggregated. There are exceptions: The OEB recently published a dashboard online that citizens can use to get more detailed data. For instance, users can see how much of a utility’s outages are caused by lightning, tree contacts, defective equipment and other causes. Mr. Panesar said SAIDI and SAIFI were intended for use within the industry, but there’s a need for more consumer-focused reporting as electricity reliability grows in importance.

Mr. Bradley said no Canadian utilities are likely to emulate PG&E’s behavior.

“I don’t see us going down that route,” he said. “If they did attempt to neglect their maintenance, pretty quickly the regulatory authorities in this country would be on top of it.”

Editor’s note: This article has been updated to explain why Newfoundland and Labrador Hydro appears to have such unusually high unreliability as reported in the charts above. The SAIDI and SAIFI figures reflect the service for its 38,000 direct customers only.

Follow related authors and topics

Authors and topics you follow will be added to your personal news feed in Following.

Interact with The Globe