Back in the 1990s, when I was new to the industry, I was working for a competitive generator. I recall one morning, I was in the office early; the sun was bright that day and I had my lights off in my office. The CEO Joe Kearney, a gregarious but thoughtful guy, came by, turned my lights on and said, “We sell electricity”.
Before I get nostalgic for pre-pandemic days working in an office with a congenial group of colleagues, recent events suggest it is time to revive the old line that the “cleanest megawatt is the one that is not consumed”. This is an important point to make on the heels of the blackouts recently endured by Californians and the high prices pervasive across the West as the aggressive policies to incorporate cleaner resources march ahead. These recent Stage 2 Emergence rolling blackouts have correctly focused renewed attention on Resource Adequacy (RA) policies in the region. But the near-term need for resources that are “dispatchable” has – curiously – not had the effect of spurring much renewed interest in demand response.
The dog that doesn’t bark
In a famous Sherlock Holmes story (“The Adventures of Silver Blaze”), a key part of the solution to the mystery revolved around a “dog that did not bark” when the theft and murder occurred. However, as it pertains to efficient markets, we have always known that the “dog that doesn’t bark” is the megawatt that is never generated. In many contexts, we have focused too much on mitigating perceived market power rather than providing market mechanisms for demand to choose not to consume – and either get paid or save money.
While it is too early to say conclusively what caused the blackouts in California – conflicting authorities seems to have a role – the shortage of resources that can be “dispatched” at any time as the state is shifting toward renewable resources is surely part of the story. The term “dispatchable” here is meant to describe a resource that can, within a reasonable time, produce power or reduce demand at any time the need arises. There is no doubt that we need resources to produce power – some mix of thermal resources and variable resources like solar, geothermal, and wind, and yes, storage. But, the megawatt that “doesn’t bark” can be deployed as fast or faster in the market than any other resource, and can usually be developed more quickly than generating resources. To extend the metaphor, since it is arguably “cleaner” than even batteries, these are megawatts that don’t “bite”.
Remain calm. All is well…
There is a report on the recent blackouts authored by the California Energy Commission (CEC), the California Public Utilities Commission (CPUC) and the California ISO (CAISO) that has landed on the Governor’s desk as I write. We can assume it will be chock full of facts, data, etc. on load, temperatures, load curves and other representations of the August 14-15 events. I hope that it will demonstrate that there has been a significant reduction in dispatchable resources in the last few years. It should note the increase in renewable resources which provide energy but are not able to supply power in the net-peak hours of 4:00 pm to 8:00 pm.
I don’t expect the report to demonstrate that there were few outages of the thermal fleet on those days – why give anyone praise? It likely will not say anything about the saving nature of imports – California can’t seem dependent. It would be good to discuss the relationship between the CPUC and CAISO, in particular their accounting methodologies and respective authorities when it comes to ensuring sufficient capacity to meet load. That might not occur, however, because it is a thorny topic and it might take away from the likely emerging narrative the authors of the report and the audience in Sacramento want: California needs more storage.
In the transition to the renewable future, a sober assessment would be that the grid will need to rely on some portion of the thermal fleet for some time. Will this be discussed? I have my doubts. The need for resources that can be dispatched may not come up at all except in the form of an articulated need for storage. While storage is needed, the current state of the art in duration over time is not able to meet the needs of the system in the immediate years when once-through-cooling (OTC) units are retired and Diablo Canyon is decommissioned. Al Gore might refer to this as an “inconvenient truth”.
So, what is a regulator to do? This is where we can remind ourselves of the megawatts that don’t bark… I mean, aren’t consumed. Demand response currently exists in involuntary form when power is cut off to customers by utilities at the direction of the grid operator. A more voluntary form can be the calls to conserve by the Governor, the CAISO, or other officials such as Flex Alerts. Most customers will heed these calls – for a time – but then the effectiveness of such voluntary calls will wane. But if the customer negotiates a deal with its provider to get paid for occasional reductions in power consumption, the demand response becomes a resource that can be used just like a generator. It is “dispatchable” in a way that renewable resources currently are not. Further, these could be aggregated far faster than other resources. It is merely a contract negotiation and a sale of a product.
Come Watson! The CPUC is about to lose an opportunity!
While customers have been migrating in California for a few years now into Community Choice Aggregators (CCA), the state has been virtually closed to customers having “direct access” to competitive retail providers. An unfortunate reaction to the Power Crisis of 2000-01 was the shutting down of most direct access (DA). However, the legislature asked the CPUC in 2018 in SB 237 to conduct a study regarding opening DA and to report the findings by June 2019.
Finally, late last month, the CPUC delivered a report which suggested a robust testing protocol for load serving entities to follow to demonstrate the ability to procure RA, meet local needs and to meet Commission directives on procurement of renewable resources. It then suggested that if the LSEs met these requirements, an additional 10% of California customers could be eligible for DA each year after 2021. Currently about 5% of California customers are eligible for DA. Clearly, the CPUC was focusing on meeting RA in a very traditional fashion.
There was no indication of an analysis of DA providing demand response which would serve the dual purpose of meeting RA needs but doing so in a way that would not emit in any way. The conservative thought process apparent in the CPUC report D.19-05-043 gives no indication that anything other than meeting the CPUC’s past directives were considered. No doubt, this is because the report was in process long before the blackouts. It also suggests that the CPUC staff never appreciated the opportunity of DA, demand response, and the relationship of these tools to meeting reliability in a carbon free manner.
A happy ending is possible
It is understood that California intends to continue its leadership in decarbonization. It is also understood that to do so will require a reliable grid comprised of renewable generation, storage, some level of thermal generation to balance the grid in times of stress, and demand response.
But what must Scotland Yard, I mean, the CPUC do?
- Revise its ill-considered rule concerning imports of RA that will make designation of resources outside of California for import less attractive.
- Consider developing RA rules in collaboration with the CAISO that will make it easier for trading RA between California and its neighbors thus making more resources available over a broader area; and
- Revisit DA eligibility considering the rapid contribution that demand response can provide both for reliability and for achieving environmental goals.
Actions like these will contribute to meeting the state’s goals in the long run and might even help avert these disasters in the short run.
And now, I take up my violin and pipe. Watson, see if Mrs. Hudson has the London evening papers… (music fades out).