As California's population looks heavenward waiting for winter rains to arrive, the state is grappling with what now might be a new normal of regular major wildfires in the summer and fall—and the ongoing preventive power shutdowns instituted to avert them. Residents, utilities, and other businesses are adapting, as are the independent power suppliers in California. While S&P Global Ratings acknowledges that wildfires and preventive shutdowns are a new risk for power suppliers, and that the agreements they have to sell power to investor-owned utilities could leave them vulnerable, we believe that at least in the near term, the financial and credit impact on these power suppliers will generally be modest.

While 2019's wildfires to date have not been as devastating as those in 2017 and 2018, the Oct. 23 Kincade Fire nevertheless destroyed about 175 homes and burned nearly 80,000 acres. And in a nightmarish déjà vu, the fire may have been caused by Pacific Gas & Electric Co.'s (PG&E) equipment. If so, the latest fire complicates PG&E's emergence from Chapter 11 by next June. Moreover, days later a series of fires ignited in Southern California. One of these, the Getty Fire, started when winds blew a eucalyptus tree branch from outside of a brush clearance zone into a Los Angeles Department of Water and Power (LADWAP) power line. The city's mayor called the incident "an act of God". On Nov. 13 we placed LADWAP's long term 'AA'-rated power system revenue bonds on CreditWatch with negative implications due to the public utility's wildfire risk, which is higher than its peers. Significant portions of the service area have an elevated fire risk and have seen recurrent wildfires, while we think that LADWAP is apparently unwilling to initiate preventative shutoffs.

The fires occurred despite preventative power outages the investor-owned utilities invoked in October and November, which left around three million Californians without power for days on end, mostly in PG&E's territory. Beyond utility risks and customer ire, the economic costs to California loom large. According to estimates released in mid-November by the chief economist at Bank of the West, this year's fires and the forced blackouts will cost $11.5 billion. Considering the state is the fifth biggest economy in the world, the impact, according to Marketplace, is expected to slow the state's growth rate to below that of the U.S. economy for the first time since 2010.

And with no sustained rain in the forecast (rain starting Nov. 18 in Southern California is not expected to end the fire season), the 2019 fire season could stretch into December, raising the possibility that dangerous Diablo (in Northern California) and Santa Ana (in Southern California) winds are not over. Indeed, after cooler but still dry weather, on Nov. 17 PG&E warned that due to forecasts for unusually warm weather and gusty offshore winds, there is an 'elevated' risk of a power shut off this week—the first of three levels, with the last, a 'warning' indicating a shut off is imminent. On Nov. 18, the watch moved to a 'watch' risk, and on Nov. 20 to a 'warning', with power shutoffs beginning this morning.

A Decade Of Public Safety Power Shutoffs?

Amidst these events, the state is trying to adjust to what this means for the future of power service in California. No one knows with any level of precision how often or how long Californians can expect to see preventative black outs (knows as public safety power shut offs, or PSPS). But fears are widespread that extended summer and fall outages could become normal for the next ten years. PG&E CEO William Johnson told regulators at the California Public Utilities Commission (CPUC) on Oct. 18 that customers could expect ten more years of intentional power outages.

California Governor Gavin Newsom and state legislators expressed outrage at the widespread nature of the outages and the possibility of years of interrupted service. But few expect any large and systemic improvements in the grid anytime soon, especially with PG&E trying to emerge from bankruptcy. Much of PG&E's service territory is deemed high risk for wildfires, with the head of PG&E's utility subsidiary, Andrew Vesey, likening its operating environment to a "tinderbox". Warmer summer temperatures and the wind itself dry out vegetation, creating abundant fuel for the fire season. And if 2018 and 2019 are any guide, the fire season itself could extend into early winter. However, Governor Newsom recently stated that this is "not a climate change story as much as a story about greed and mismanagement over the course of decades". At the same time, state politicians and regulators are calling for investigations into the shutoffs and possible reforms to the shutoff rules after considerable customer complaints. In short, it is difficult to know what is in store for power customers, especially those in PG&E's territory, where the ability to target small sections of the grid to minimize shutoff disruption is some ways away.

PSPS: A New Curtailment Risk For Power Suppliers

As many sectors of the economy contemplate what the wildfires and public safety power shutoffs mean, we take a look at our rated power projects that sell power long-term to investor-owned utilities under power purchase agreements (PPAs). Many of these projects have already seen their credit profiles decline due to relying wholly on PG&E for revenues. All of the projects we rate in which PG&E is the sole revenue counterparty are now rated 'CCC+'. Our rating on Topaz Solar Farms LLC, in fact, actually fell seven notches after PG&E announced it was exploring a restructuring a couple months after the November 2018 Camp Fire.

Typically our rating on a project is no higher than that on the revenue counterparty given the above market rates in the PPA. (Project ratings wholly relying on PG&E as their offtaker are currently rated higher than the issuer credit rating on PG&E because the project default risk depends on whether PG&E, in bankruptcy, will try to reject any PPAs. To date, it has pledged not to.) But assuming PG&E eventually emerges, and the recently passed wildfire legislation AB 1054 provides credit protection to the three major California investor-owned utilities, intermittent power shut offs during the fire season could alter power supplier curtailment risk. For example, the Kincade fire, which started near the Geysers Power Co. LLC geothermal power plant in Geyserville, Calif., shut down that plant for several days. And Crockett Cogeneration had a scare in October when the Sky Fire near the town of Crockett forced the evacuation of most of the town's residents. By coincidence, as this was occurring, Crockett Cogeneration was already offline as part of PG&E's public safety power shutoff.

Given the overall small number of power projects we rate in California (see table below), and either their location away from the highest fire risk or revenue structure, we view the immediate risk of curtailment due to wildfire prevention power shuts offs on our projects to be minimal. Crockett was the only project we rate that was shutoff during a PSPS.

But there is an interesting question as to whether utilities who buy power will argue the PSPS programs fall under force majeure and as a result, not liable to compensate a power supplier curtailed during a preventative outage. There are arguments for and against this view. Ultimately, we think that the issue will be determined through litigation.

If power suppliers are not protected and these outages continue or increase, we could see some erosion in their debt service coverage levels. PSPS programs are a new and potentially important emerging risk for California power suppliers more generally because most PPAs were not structured with an assumption that power suppliers could be significantly curtailed by force majeure, and it is unclear who bears these costs. Preventative power shutdowns now raise important questions for any entity pursuing a California power project—especially if it aspires to be investment grade. Unless this risk is specifically mitigated in the PPA, we would expect any investment-grade project to be able to withstand a shutoff with minimal impact to debt service coverage ratios. What is difficult to know now is how large of a curtailment we should factor into any future ratings, and this will depend on a variety of factors.

Table 1 provides details for our rated California projects. In reviewing the PPAs for each project, it's clear that curtailment risk is not the same for each asset, and depends on its revenue source, technology, location, and the power project's contractual provisions.

Table 1

Curtailment Explained

Curtailment is the reduction of output from a power supplier. It is generally involuntary in the sense that the power supplier is willing and able to produce, but is constrained from doing so by another party or event. In take-or-pay PPAs, in which power suppliers are supposed to be paid for whatever they produce, curtailment arrangements in a PPA are a big deal. Power suppliers typically negotiate the terms and conditions of their compensation for common instances of curtailment. Given that preventative blackouts are not clearly a force majeure event in PPAs, there is likely to be litigation and the need for legal opinions to determine whether the utility or the power supplier bears the cost of the foregone revenues when a power supplier is curtailed.

Curtailment is typically due to oversupply and transmission constraints. Power suppliers are curtailed when power output exceeds demand (and therefore supply needs to be shed to maintain system balance) or congestion occurs on transmission lines. Power plants with PPAs will typically be compensated for this type of lost output. This is often called "economic curtailment" because oversupply pushes prices to uneconomic levels. In some cases, buyers are protected up to a certain point (via a cap) if curtailment is caused by system maintenance or a technical issue with the grid. However, buyers don't compensate the power supplier if curtailment is caused by an emergency or "force majeure."

Curtailment has already been an important emerging issue in California over the last several years with the rapid increase of renewable energy capacity. The issue is demonstrated well with the so called "Duck Curve" in which abundant solar capacity (roughly between noon and 3 pm) combined with power plants unable to ramp down exceed demand, requiring that solar energy be curtailed. Perhaps because of this, there has been a trend toward PPAs now excusing a limited amount of economic curtailment. But while economic curtailment has mostly been quantified and thus can be sized in take-or-pay PPAs, preventative wildfire curtailments have not been specifically addressed, creating a new form of curtailment risk.

Wildfire Related Curtailments In PPAs

We reviewed several California utility company PPAs dating as far back as the 1980s. Preventative curtailments to limit wildfires were not considered or specifically defined when these contracts were written. It therefore requires interpretation to assess which part of the PPA best covers a PSPS when one leads to a supplier being curtailed. Because this is subject to interpretation, we expect both the power suppliers and utilities will be able to find language supporting their claim for and against compensation.

We think that the most contentious issue is whether a large scale preventative power shut off would be deemed a force majeure event, in which case the utilities are generally not liable. That would leave the power supplier to absorb the lost revenues if they are otherwise able to generate. There are two parts of this issue: 1. Is a PSPS a force majeure-like event? and 2. Do the utility's actions (or lack thereof) disqualify its ability to claim force majeure?

Establishing Force Majeure

Many of the PPAs have a general definition of force majeure. The oldest PPA we reviewed has the most opened-ended definition ("unforeseeable causes") and the least examples, but including "acts of God" and "sudden actions of the elements". Others we reviewed define it as an event or circumstance that impedes the performance of a material obligation and is out of the control of the party claiming force majeure. Some typical and relevant examples are "unusual" fires or droughts, or "other natural catastrophes". Some also contain "unusual or extreme adverse weather-related events". We believe a good case could be made for claiming PSPS is a force majeure event. And, equally, that it is not. General language like "extreme adverse weather" is likely the most supportive language allowing a PSPS to be a force majeure event.

The fires of 2017 and 2018 especially demonstrate the destructive potential of dry and windy conditions in Northern California, and therefore lend credence to the argument that the prevention of a devastating fire is equivalent to a real fire. On the other hand, dry and windy conditions are an annual occurrence, and a precautionary action could be meaningfully different from an actual destructive event outside of the utility's control. Ultimately, we think this would be contested in court, and that the verdict would likely set a precedent that could be applied to all PPAs, regardless of whether all the language was exactly the same.

But even if the circumstances of a PSPS met the definition of a force majeure event, most PPAs state the precondition that the event was not directly or indirectly the result of the party seeking to have their obligations excused. Furthermore, they state the party must have taken all reasonable precautions, measures, and due diligence to prevent or avoid the event, and that the event is not a direct or indirect result of negligence, which is already being questioned by regulators, politicians, and customers in regards to PG&E. Thus, the question is whether the utility's lack of investment in hardening the system or its inability to achieve its own vegetation management plans disqualifies PSPS from utility claims of force majeure. Although this too will be up to the courts, in the case of PG&E, that utility's troubled safety record and questions about whether it has sufficiently hardened its grid or managed surrounding vegetation may not help its case to stop paying power suppliers due to a force majeure event.

If PSPS is not an event of force majeure, some of the PPAs include language that could make it an uncompensated curtailment. For example, in some the utility is not obligated to pay for interruption of deliveries due to the utility's "system emergency". Others have the concept of "dispatch down periods", which can vary from one PPA to another. Clearly those with language stating the utility's decision to curtail due to a system emergency or to a "warning of an imminent condition or situation, which jeopardizes Buyer's [the utility] electric system integrity…as determined by Buyer in Buyer's sole discretion", supports uncompensated curtailment. But we note the spirit of dispatch down is more related to transmission issues as opposed to a downed power line sparking a fire. Other definitions of dispatch down exclude this language altogether and only associate it with actions from the California Independent System Operator (CAISO), thereby supporting the power supplier. In addition, even if a PSPS dispatch down period is declared, it is usually capped, typically to a couple of days of outages per year. Anything longer would be compensated.

Wildfire Curtailment Risks In Project Financings

Some of the project-financed power projects we rate do have some risk of curtailment due to PSPS over at least the next decade. The level of risk, however, is specific to each project, so we do not apply a one-size fits all assumption. Furthermore, we view this type of curtailment as event risk, and unlikely to reoccur at the same projects—although there are certain plants, such as the Geysers geothermal plant (which we do not rate), that have much higher frequency risk or probability of fire given their high-risk locations. On the other hand, some projects, such as CSolar IV South LLC, which is in a desert, are not directly impacted by fires. But when curtailments are based on transmission line shutoffs affecting population centers, it is still a guess which generation resources the utility will choose to shut off. Most importantly, we view the risk to be minimal to modest for our currently rated power projects because of a low probability of being shut off and because of the likely limited duration of shutoffs, even if we assume there would be no compensation from the utilities. (Of course, that could change if we see more sustained interruptions.)

Curtailment risk is most evident in projects whose margins are based on selling electricity volumes rather than those that receive capacity payments. Projects like Panoche Energy Center LLC and the Redwood Portfolio within WG Partners Acquisition, LLC make almost all of their margins on capacity payments, which in turn are based on the power plant achieving availability thresholds. A PSPS would not affect availability. In contrast, projects that rely almost exclusively on selling power would be most at risk if no compensation were provided because a reduction in output leads to lower margins. Most of these types of projects are solar and wind farms. Even so, only one of our rated projects were affected by the power shutoffs in October and November. For now, we are viewing these events to be infrequent, akin to event risk.

A Modest Financial Impact

Considering the length of interruptions we see, the actual financial impact may be relatively modest, especially at those PPAs with higher prices, signed before renewable energy costs were at today's lows. For example, a 300 MW solar farm with about 30 hours of lost output at $100/MWh could stand to lose $900,000, or about 1%-2% of annual expected revenues if the PSPS is deemed a force majeure. While the PPAs are unclear about how to treat a PSPS in terms of compensation, the amounts are minimal relative to the project's lifetime earnings, and are often capped. A PSPS curtailment is thus less likely to impact lenders than equity, and it is more likely that project sponsors rather than lenders will raise the first challenges.

There are also cases when PSPS can coincide with negative spark spreads, and thus the project would not have wanted to dispatch anyways. Crockett Cogeneration was forced offline for about 59 hours during the second PSPS in October. Although it makes most of its revenues from capacity payments, it still earns meaningful gross margins from selling energy at short-run avoided cost (SRAC) prices. It does not expect to receive compensation from the outage, but spark spreads were already negative.

Thus, because we view such events as unlikely to repeatedly occur at any given project and to be limited in duration, we do not intend to modify our long-term forecasts to account for higher outages, even if they were all assumed to be uncompensated. However, if we view one of our projects to be vulnerable to shutoffs, we would incorporate the curtailment risk into our forecast given it could have a rating impact. In particular, we would expect any investment grade projects, and especially those rated 'BBB-', to be able to withstand additional curtailment related to a PSPS without any impact to credit quality.