Most utility bankruptcies are caused by bad financial decisions. But Pacific Gas & Electric’s bankruptcy last week was driven by its massive potential liabilities for causing some of California’s most deadly and destructive wildfires of the past two years.
And while PG&E’s bankruptcy reorganization will take years to play out, it faces a major short-term challenge: doing its best to ensure that its power lines and equipment don’t start any more wildfires.
But how can PG&E best accomplish this goal? What standard should PG&E, and other California utilities, be held to when it comes to their responsibility for extreme weather, drought and other climate change-driven fire risks? How can state regulators ensure that it’s doing enough to prevent grid conditions that could cause fires? Should PG&E start de-energizing large parts of the grid during times of high fire risk, despite the disruption and danger such “public safety power shutoffs" can cause? And how can it afford to spend the money for all these fire prevention efforts, while it’s simultaneously going through bankruptcy?
These were some of the key questions PG&E attorneys faced during a heated hearing last week before U.S. District Court Judge William Alsup, who oversees PG&E’s probation for criminal convictions related to its role in the 2010 San Bruno pipeline explosion. Alsup said he is also prepared to use his authority as PG&E’s probation overseer to order the utility to undergo a massive new grid-inspection, tree-clearing and de-energization regime by the start of the 2019 fire season in July. His goal is for PG&E to start “zero fires” this year.
Alsup’s plan, which could be taken up in the coming months, has drawn a good deal of support from fire victims and long-time critics of PG&E’s safety record, who have been calling for a state takeover of the utility. But according to PG&E, while Alsup’s goals are commendable, his plan is both impossible to carry out and potentially harmful, since it could throw a wrench into its ongoing and expanding fire prevention efforts. The California Public Utilities Commission (CPUC) agreed, telling the judge in a filing this week that his proposed order “will in all likelihood conflict with and frustrate the extensive federal statutory and state Constitutional and statutory regulatory scheme.”
These conflicting views of PG&E’s proper role and responsibility in preventing wildfires have become an important subtext to the utility’s broader bankruptcy process. If PG&E can’t meet these aggressive — some would say impossible — goals set by the judge overseeing its criminal probation, what is it doing to prevent its equipment from causing more fires? And beyond the target of “zero fires” for 2019, what are the best ways to measure its progress toward its goals?
The challenge of clearing trees to a "zero fires" standard
Let’s start with the first half of Alsup’s proposal. His plan calls for PG&E to physically re-inspect the entirety of its power grid, and identify and fix “any other condition anywhere in its grid similar to any condition that contributed to any previous wildfires.”
In Wednesday’s hearing, Alsup cited last year’s report from the California Department of Forestry and Fire Protection (Cal Fire), which found that PG&E equipment started 16 wildfires in 2017, 12 of them caused by tree limbs contacting overhead wires, to justify his proposal. “You can see to your ordinary U.S. district judge, and I think to the ordinary mortal, there’s quite an obvious pattern. High winds, trees blown onto the lines,” he said. “Twenty-two people killed, one month, October.”
But according to PG&E, Alsup’s proposed remedy to these tragedies is simply not possible. The first barrier is the sheer amount of work required to inspect and clear PG&E’s network of nearly 100,000 miles of power lines in such a short time. PG&E has estimated it could cost between $75 billion and $150 billion, and require up to 600,000 qualified tree-clearing workers.
“Even if the state of California could find $100 billion to fund this work, there aren’t enough qualified personnel in the country” to do the job in time, PG&E attorney Kevin Orsini told Alsup in Wednesday’s hearing.
To back up this assertion, Orsini cited PG&E’s experience with its Enhanced Vegetation Management programs, which are focused on clearing the highest fire risks from its overhead power line network. These include overhangs, or tree limbs and vegetation that’s directly above power lines — a safety risk the utility can clear without getting permission from property owners, since it’s in their right of way. Even so, PG&E expects it will take eight years to complete this work across its highest fire risk areas, largely because “we haven’t been able to find the qualified tree personnel” to get it done faster, he said.
Alsup challenged Orsini’s claim that there weren’t enough qualified workers, suggesting that the utility could ask Cal Fire to recruit firefighters willing to work during the winter and spring off-season. “I don’t buy that there’s not enough people.” he said.
But PG&E’s assertion was backed up by a statement from Tom Dalzell, business manager of International Brotherhood of Electrical Workers Local 1245, who called Alsup’s plan “impractical to the point of impossible and costly to the point of impossible.” To illustrate the point, he noted that it cost PG&E about $9 million a week to hire about 600 line workers from across the country to inspect and repair about 50,000 transmission structures in high fire danger areas — a cost that would equate to nearly $1 billion per week, if PG&E were to hire the 600,000 workers it says it would need to do the work Alsup has proposed.
Orsini also pointed out that Alsup’s proposal for PG&E to remove any tree in “any condition that contributed to any previous wildfire” was far too broad a standard to apply to mass vegetation management. He cited several Cal Fire reports on past fires caused by PG&E equipment, including some that started during windstorms. Those fires caused limbs on trees more than 20 feet away to snap and blow onto power lines, and destroyed many “very robust, healthy trees that otherwise would not have been something that, before October 2017, any utility or vegetation program would have taken out.”
If PG&E is forced by court order to undertake Alsup’s plan, it could be forced to divert workers and funding away from its ongoing efforts, Orsini said. Specifically, since the Camp Fire in November, PG&E has launched a Wildfire Safety Inspection Program that targets the roughly 30,000 miles of its transmission and distribution system within so-called Tier 2 and Tier 3 high fire risk areas, as designated by the CPUC. Using both on-the-ground workers and drones, PG&E plans to have its transmission towers inspected by March, and its distribution poles inspected by May — an important step, given that evidence indicates that an equipment failure on a PG&E transmission tower may have caused the Camp Fire.
More utility fire prevention plans are set to be unveiled next week, as PG&E and the state’s other investor-owned utilities submit fire mitigation plans as part of the CPUC’s overhaul of fire safety regulations last year. The CPUC’s plan relies heavily on prioritizing hazards based on its expanded map of high fire risk areas of the state, which covers more than 70,000 square miles, or more than two-fifths of the state.
De-energization: Balancing the risks and rewards of blackouts to prevent fires
Alsup’s proposed plan accounts for the possibility that PG&E may not be able to inspect and repair the scope of potential fire risks it encompasses, and it offers a solution — turn off the power. Specifically, Alsup’s order would force PG&E to newly inspect its entire grid between now and July, and “rate each segment’s safety under various wind conditions (from little wind to high wind).”
Then, once the 2019 fire season begins, “PG&E may supply electricity only through those parts of its electrical grid it has determined to be safe under the wind conditions then prevailing,” and “must de-energize any part of its grid not yet rated as safe by PG&E for the wind conditions then prevailing until those conditions have subsided.”
De-energizing power lines as a fire prevention measure has been part of the California utility playbook since San Diego Gas & Electric got permission from the CPUC to deploy it in 2012. PG&E and Southern California Edison have had the authority to institute so-called public safety power shutoffs since last year.
PG&E first used its authority in October, shutting off power for about 60,000 customers over two days of high winds. The move yielded a large number of customer complaints over spoiled food, damaged equipment and other effects of the blackouts, as well as some significant safety problems, such as loss of cellphone service as de-energized towers ran out of backup generator fuel. After the shutoff, however, PG&E crews found more than 20 problems, such as tree limbs fallen across uninsulated wires, that could have been safety hazards had the lines not been de-energized.
But in the days and hours leading up to the Camp Fire, PG&E considered, but decided against, de-energizing the power lines in the region of the fire. This decision may not have changed the course of events, given that the scope of PG&E’s de-energization plan did not extend to the higher-voltage transmission line that has been linked to the Camp Fire’s start. Still, Alsup noted in Wednesday’s hearing that the decision could be considered, “in retrospect, a tragic error.”
Orsini replied that PG&E had followed the CPUC-established procedure for assessing the fire risk, which yielded “the conclusion that the risk of turning off the power was higher than the risk of a fire occurring.” But he also noted that the methods of determining this risk are in the midst of being refined, through a newly launched CPUC proceeding to examine the role of de-energization in light of the past two years’ deadly fires.
For example, since the Camp Fire’s destruction of the town of Paradise, Calif., which had only four roads that were blocked by fleeing residents during the fire, PG&E has begun taking into account a region’s potential for evacuation bottlenecks and traffic jams as part of its evaluation of fire risk, he said.
“There has to be an analysis of factors on the ground — and there are a lot of factors involved — to determine [if it] is fundamentally safe to leave the system on,” he said.
But Alsup’s plan doesn’t account for those factors, according to PG&E and the CPUC. In fact, his proposal specifically states that “PG&E may not take into account the need for reliability of service, the inconvenience to customers resulting from interruption in service, or its impact upon PG&E’s revenues and profits,” when making the de-energizing decision. Only the current wind speed and the safety rating the plan envisions creating for every circuit on the utility’s system can be considered.
Tool of last resort
PG&E noted that de-energization has long been considered a tool of last resort, since it can cause as many dangers as it helps avoid. Not all first responders, critical medical facilities and essential municipal services like water, street lights and communications systems have backup power systems, and those that do may not be equipped to run for days at a time.
That’s why the CPUC requires utilities to do significant public outreach and coordination days before doing a safety power shutoff. But under Alsup’s plan, PG&E might be required to power down large parts of its grid at a moment’s notice, potentially destabilizing large parts of its distribution system, or even the larger Western U.S. transmission grid.
There are also significant challenges to applying de-energization in a more targeted fashion, PG&E’s Orsini noted. For example, Alsup suggested in Wednesday’s hearing that PG&E use 20 mile-per-hour winds as a “red flag” for shutting down the power grid — a figure he came up with by studying the Cal Fire data on the 17 fires PG&E equipment caused in 2017, and noticing that 14 of them showed winds over that speed.
But Orsini responded with several objections to using that kind of rule-of-thumb standard for making de-energization decision. First, “if we de-energize every time wind speeds reach 20 miles per hour, we’re going to be turning off the lights a lot,” he said. Second, the wind speed data attached to the 2017 fires was not necessarily accurate, given that the data is collected from just a handful of sensor locations, he said.
Third, wind speed is just one of many factors that go into determining the risk involved in allowing power lines to stay energized. “It may be that if the wind is over 40 miles per hour tomorrow, because of the vegetation state and ground fuel state in a particular area, the wildfire risk is quite low,” he said. Conversely, even a slight wind could fan a fire in hot, dry conditions, he said. The challenge PG&E faces is, “what is the best, most risk-informed way to make that decision?”
PG&E has been making significant investments into its de-energization program. On the data side of things, it installed 200 weather stations last year, and plans 200 more this year, to collect more accurate and granular data on such factors as local wind speed and humidity, and is installing 60 high-definition cameras this year, and nearly 600 by 2022, to give it visual coverage over its most high-risk grid assets. In terms of managing its grid with these new inputs, PG&E is working on better weather forecasting and fire ignition and spread modeling, and adding new sectionalizing devices to more specifically target certain parts of the grid for powering down.
Finally, PG&E noted that following Alsup’s de-energization directive would put it in the position of having to frequently power down the grid in ways that violate both state and federal regulations. That’s more of a problem when it comes to the transmission network, which is regulated by the Federal Energy Regulatory Commission, and where de-energizing could require shutting off power to tens or hundreds of thousands of people far away from any fire risk.
But it’s also a problem at the state level, where the CPUC has currently limited the use of de-energization to deal with “an imminent and significant risk that strong winds will topple its power lines onto tinder dry vegetation or will cause major vegetation-related impacts on its facilities during periods of extreme fire hazard,” and requires utilities to consider “multiple factors including fuel moisture, existing fires and firefighting capabilities, meteorological conditions and humidity in addition to wind when making a de-energization decision.”
Of course, at the distribution grid level, it’s possible to talk about distributed energy resources taking up some of the slack for communities facing fire prevention-based blackouts. Michael Wara, the head of Stanford University’s Climate and Energy Policy Program who was recently named as one of five members of Gov. Gavin Newsom’s Wildfires Blue Ribbon Commission, has been a loud public advocate for far more aggressive de-energization policies, and has suggested that the state could incentivize about $30 billion backup power systems for 1 million of the state’s most at-risk residents, if it used mechanisms like folding the payments into monthly utility bills.
“Spending billions on alternatives to utility-delivered power might have seemed like an unaffordable luxury once, but now it’s clear what doing nothing will mean: more lives, livelihoods and communities lost in what the governor calls the ‘new abnormal,’” Wara wrote in a December op-ed piece.