In its December 2020 open meeting, FERC surprised regular observers by voting to reject a proposed order that would have instituted a Section 206 investigation of the August 14-15, 2020 California heat wave and ensuing rolling blackouts.
FERC’s rejection of a complaint was highly unusual because FERC open meetings are rigorously scripted affairs where all of the decisions are made before-hand. Federal law prohibits a quorum of FERC Commissioners from meeting to discuss business outside of an open meeting, so the FERC Commissioners instead have their assistants meet with Staff well before the open meeting and decide which orders will go forward. When an order does not have the votes to go forward, that becomes apparent in those discussions. Because all business is pre-agreed in this fashion, draft orders are rarely if ever presented for a vote, unless and until it is already understood that the vote will be successful.
This time was different, reflecting the deep political ramifications of the issues presented by the investigation. FERC Staff presented a brief report on the blackouts. Afterward, a draft order was proposed for vote, which would have instituted a Section 206 investigation. Chairman Danly spoke and explained his view that getting to the bottom of the California rolling blackouts in August 2020, and ensuring that similar blackouts would not happen again, was central to FERC’s purpose. The other Commissioners spoke, indicating that they did not disagree with that sentiment, but said that they would prefer to let entities in California address this issue first. Those opposing an investigation also argued that they would prefer that the Commission could informally participate in discussions about solutions – an option that would be foreclosed if a formal investigation imposed ex parte restrictions on Commission discussions with interested parties. The Commissioners voted to reject the proposed order.
Most of what is publicly known about California’s rolling blackouts can be found in two reports: the Preliminary Root Cause Analysis, issued by the California ISO, California Public Utilities Commission, and California Energy Commission in October 2020; and the November 2020 Report on System and Market Conditions, prepared by the ISO Department of Market Monitoring (“DMM”). From these reports, we learn that the ISO and its market monitors have identified three main issues:
1. The State of California has established capacity obligations known as Resource Adequacy (“RA”) to ensure system adequacy. Those obligations are generally tied to the ability of resources to meet load at time of system peak. But in California, because of its deep reliance on renewable resources, it has been understood that the most difficult period may be the time of “net peak,” defined as the time when resources net of solar and wind are most needed. In general, this comes in the early evening, when temperatures may remain high, when demands may remain high even if not quite at peak levels, but when solar resources have dropped off because the sun is no longer shining, and when wind resources that often peak in night-time hours may not have picked up. The evening is doubly problematic, because demand response resources (which typically means an industrial facility capable of reducing its demand) may not be online or at full capacity in the evening and thus do not have demand at that hour to back down. Both reports suggest that RA requirements should be updated to ensure that resources are available that can meet extreme net peaks – factoring in which resources can actually be available in that situation. While the issue is simple to state, it does not lend itself to easy solutions to the extent it requires devaluing solar, wind, and demand response, which are the primary categories of capacity that had issues at time of net peak.
2. A second issue that came up in both reports is a mistake in the ISO market software. In the ISO market, there is a day-ahead market called the Integrated Forward Market (“IFM”). However, it is part of the basic ISO market design that the IFM results might not obtain adequate supply to meet expected demand for the following day. To address this potential shortfall in the IFM, there is a subsequent day-ahead market called the “Residual Unit Commitment” (“RUC”) market. As spelled out in the ISO Tariff at Section 31.5, the express purpose of the RUC is to address those circumstances where “the IFM did not commit sufficient resources to meet the CAISO Forecast of CAISO Demand” and to “commit additional resources and identify additional RUC Capacity to ensure sufficient on-line resources to meet Demand for each hour of the next Trading Day.” On the days in question, there were indeed IFM shortfalls – the IFM market did not ensure sufficient supply to meet predicted demand. It turns out that a “market enhancement” that had been programmed into the ISO software erroneously ensured that the ISO software did not do what the tariff spelled out and that the RUC essentially ignored the known capacity shortfall in the IFM market. The ISO appears to have addressed the problem going forward, with a revision to its software implemented September 5, 2020. But it remains to be seen whether more issues will arise from the error.
3. As explained in the DMM Report, the ISO software error led to needed California capacity being sold in the RUC and exported out of the ISO, exacerbating the blackouts. Even with the September 5, 2020 software fix, the DMM explains that questions remain about the relative priorities of ISO exports and native load. The DMM suggests that those priorities should be considered and potentially revised. A host of market and operational issues turn on such priorities. And, of course, those outside of the ISO area might well argue that reforms that benefit ISO native loads will hurt native loads outside the ISO. To the extent that the priorities are a zero sum game, solutions may be difficult.
Because no formal proceeding was established, it is difficult to say what the next steps at FERC will be.
Stan Berman
Eric Todderud