Tag Archives: stranded assets

The real lessons from California’s 2000-01 electricity crisis and what they mean for today’s markets

Leave a reply

The recent reliability crises for the electricity markets in California and Texas ask us to reconsider the supposed lessons from the most significant extended market crisis to date– the 2000-01 California electricity crisis. I wrote a paper two decades ago, The Perfect Mess, that described the circumstances leading up to the event. There have been two other common threads about supposed lessons, but I do not accept either as being true solutions and are instead really about risk sharing once this type of crisis ensues rather than being useful for preventing similar market misfunctions. Instead, the real lesson is that load serving entities (LSEs) must be able to sign long-term agreements that are unaffected and unfettered directly or indirectly by variations in daily and hourly markets so as to eliminate incentives to manipulate those markets.

The first and most popular explanation among many economists is that consumers did not see the swings in the wholesale generation prices in the California Power Exchange (PX) and California Independent System Operator (CAISO) markets. In this rationale, if consumers had seen the large increases in costs, as much as 10-fold over the pre-crisis average, they would have reduced their usage enough to limit the gains from manipulating prices. Consumers should have shouldered the risks in the markets in this view and their cumulative creditworthiness could have ridden out the extended event.

This view is not valid for several reasons. The first and most important is that the compensation to utilities for stranded assets investment was predicated on calculating the difference between a fixed retail rate and the utilities cost of service for transmission and distribution plus the wholesale cost of power in the PX and CAISO markets. Until May 2000, that difference was always positive and the utilities were well on the way to collecting their Competition Transition Charge (CTC) in full before the end of the transition period March 31, 2002. The deal was if the utilities were going to collect their stranded investments, then consumers rates would be protected for that period. The risk of stranded asset recovery was entirely the utilities’ and both the California Public Utilities Commission in its string of decisions and the State Legislature in Assembly Bill 1890 were very clear about this assignment.

The utilities had chosen to support this approach linking asset value to ongoing short term market valuation over an upfront separation payment proposed by Commissioner Jesse Knight. The upfront payment would have enabled linking power cost variations to retail rates at the outset, but the utilities would have to accept the risk of uncertain forecasts about true market values. Instead, the utilities wanted to transfer the valuation risk to ratepayers, and in return ratepayers capped their risk at the current retail rates as of 1996. Retail customers were to be protected from undue wholesale market risk and the utilities took on that responsibility. The utilities walked into this deal willingly and as fully informed as any party.

As the transition period progressed, the utilities transferred their collected CTC revenues to their respective holding companies to be disbursed to shareholders instead of prudently them as reserves until the end of the transition period. When the crisis erupted, the utilities quickly drained what cash they had left and had to go to the credit markets. In fact, if they had retained the CTC cash, they would not have had to go the credit markets until January 2001 based on the accounts that I was tracking at the time and PG&E would not have had a basis for declaring bankruptcy.

The CTC left the market wide open to manipulation and it is unlikely that any simple changes in the PX or CAISO markets could have prevented this. I conducted an analysis for the CPUC in May 2000 as part of its review of Pacific Gas & Electric’s proposed divestiture of its hydro system based on a method developed by Catherine Wolfram in 1997. The finding was that a firm owning as little as 1,500 MW (which included most merchant generators at the time) could profitably gain from price manipulation for at least 2,700 hours in a year. The only market-based solution was for LSEs including the utilities to sign longer-term power purchase agreements (PPAs) for a significant portion (but not 100%) of the generators’ portfolios. (Jim Sweeney briefly alludes to this solution before launching to his preferred linkage of retail rates and generation costs.)

Unfortunately, State Senator Steve Peace introduced a budget trailer bill in June 2000 (as Public Utilities Code Section 355.1, since repealed) that forced the utilities to sign PPAs only through the PX which the utilities viewed as too limited and no PPAs were consummated. The utilities remained fully exposed until the California Department of Water Resources took over procurement in January 2001.

The second problem was a combination of unavailable technology and billing systems. Customers did not yet have smart meters and paper bills could lag as much as two months after initial usage. There was no real way for customers to respond in near real time to high generation market prices (even assuming that they would have been paying attention to such an obscure market). And as we saw in the Texas during Storm Uri in 2021, the only available consumer response for too many was to freeze to death.

This proposed solution is really about shifting risk from utility shareholders to ratepayers, not a realistic market solution. But as discussed above, at the core of the restructuring deal was a sharing of risk between customers and shareholders–a deal that shareholders failed to keep when they transferred all of the cash out of their utility subsidiaries. If ratepayers are going to take on the entire risk (as keeps coming up) then either authorized return should be set at the corporate bond debt rate or the utilities should just be publicly owned.

The second explanation of why the market imploded was that the decentralization created a lack of coordination in providing enough resources. In this view, the CDWR rescue in 2001 righted the ship, but the exodus of the community choice aggregators (CCAs) again threatens system integrity again. The preferred solution for the CPUC is now to reconcentrate power procurement and management with the IOUs, thus killing the remnants of restructuring and markets.

The problem is that the current construct of the PCIA exit fee similarly leaves the market open to potential manipulation. And we’ve seen how virtually unfettered procurement between 2001 and the emergence of the CCAs resulted in substantial excess costs.

The real lessons from the California energy crisis are two fold:

  • Any stranded asset recovery must be done as a single or fixed payment based on the market value of the assets at the moment of market formation. Any other method leaves market participants open to price manipulation. This lesson should be applied in the case of the exit fees paid by CCAs and customers using distributed energy resources. It is the only way to fairly allocate risks between customers and shareholders.
  • LSEs must be able unencumbered in signing longer term PPAs, but they also should be limited ahead of time in the ability to recover stranded costs so that they have significant incentives to prudently procure resources. California’s utilities still lack this incentive.

The PCIA is heading California toward another energy crisis

1 Reply

The California ISO Department of Market Monitoring notes in its comments to the CPUC on proposals to address resource adequacy shortages during last August’s rolling blackouts that the number of fixed price contracts are decreasing. In DMM’s opinion, this leaves California’s market exposed to the potential for greater market manipulation. The diminishing tolling agreements and longer term contracts DMM observes is the result of the structure of the power cost indifference adjustment (PCIA) or “exit fee” for departed community choice aggregation (CCA) and direct access (DA) customers. The IOUs are left shedding contracts as their loads fall.

The PCIA is pegged to short run market prices (even more so with the true up feature added in 2019.) The PCIA mechanism works as a price hedge against the short term market values for assets for CCAs and suppresses the incentives for long-term contracts. This discourages CCAs from signing long-term agreements with renewables.

The PCIA acts as an almost perfect hedge on the retail price for departed load customers because an increase in the CAISO and capacity market prices lead to a commensurate decrease in the PCIA, so the overall retail rate remains the same regardless of where the market moves. The IOUs are all so long on their resources, that market price variation has a relatively small impact on their overall rates.

This situation is almost identical to the relationship of the competition transition charge (CTC) implemented during restructuring starting in 1998. Again, energy service providers (ESPs) have little incentive to hedge their portfolios because the CTC was tied directly to the CAISO/PX prices, so the CTC moved inversely with market prices. Only when the CAISO prices exceeded the average cost of the IOUs’ portfolios did the high prices become a problem for ESPs and their customers.

As in 1998, the solution is to have a fixed, upfront exit fee paid by departing customers that is not tied to variations in future market prices. (Commissioner Jesse Knight’s proposal along this line was rejected by the other commissioners.) By doing so, load serving entities (LSEs) will be left to hedging their own portfolios on their own basis. That will lead to LSEs signing more long term agreements of various kinds.

The alternative of forcing CCAs and ESP to sign fixed price contracts under the current PCIA structure forces them to bear the risk burden of both departed and bundled customers, and the IOUs are able to pass through the risks of their long term agreements through the PCIA.

California would be well service by the DMM to point out this inherent structural problem. We should learn from our previous errors.

Getting to the harder question about stranded assets

1 Reply

diablocanyon

John Farrell at the Institute for Self-Reliance argues that existing utility fossil-fuel plants should not be given “stranded assets” status. While his argument about “stranded assets” makes sense from a society wide economic sense, it doesn’t conform with the utility regulation world in which “stranded assets” is actually relevant. Having gone through California’s restructuring and competitive transition charge (CTC) (I’m the only person outside of the utilities to conduct a complete accounting of the CTC collection through 2001), it’s all about what’s on the utility’s books and what the regulatory commission agrees is an acceptable transfer of risk. And based on what happened with Diablo Canyon in 1996 (the correct treatment would have recognized that PG&E had collected its entire investment at the regulated rate of return by 1998—I did that calculation too), it’s not promising.

So I suggest focusing on the shareholder/ratepayer risk sharing arrangement and how that should be changed in the face of the oncoming utility 2.0 transformation as a more fruitful path. We have to change the underlying paradigm first. That there are benefits from retiring generation assets is not a hard argument to win—that was the case in 1996 in California. The harder one is that the past regulatory compact should be changed and that it won’t somehow impact future investment in the distribution utility.

Citigroup climate risk study part 2 – stranded assets

3 Replies

The CitiGPS study makes a unique contribution to the climate change risk literature: reducing GHG emissions will lead to stranded investment assets. These assets include both fossil fuel holdings and the equipment that uses those fuels. Protecting those investments is at the heart of much of the resistance to addressing climate change risk.  Removing political barriers is probably the single greatest difficultly in moving to implement policies to mitigate this risk; many policy proposals are at the ready so there’s no lack there. Given the apparent urgency of acting, perhaps it’s time to ask the question whether these asset owners should be compensated by those who will benefit directly, i.e., the rest of us? 

What’s behind the reluctance of political actors to propose this type of solution is the belief in the underlying premise of benefit-cost analysis. Economists have unfortunately perpetuated a misconception on the public that so long as total societal benefits exceed costs, a policy is justified even if those suffering those costs are not compensated for their losses. The basis of this is the Kaldor-Hicks efficiency criterion. In contrast, market transactions are presumed to only occur if both parties gain through Pareto efficiency--one party fully compensates the other one for the transaction. Public policy now casts aside this compensation requirement. Unfortunately this leads to significant redistribution impacts that are too often left unexamined. And of course, the losers resist these policies, with a ferocity that is accentuated by both loss aversion (where potential losses are felt more strongly than gains) and that these losses are usually concentrated among a smaller group of individuals than the spread of the benefits.

Too often public agencies are running over these interests to push for societal benefits without compensating the losers. A recent example that I was involved with was the adoption by the California Air Resources Board of the in-use off-road diesel engine regulations. CARB mandated the premature scrappage of construction equipment that had been purchased to comply with previous regulatory mandates from CARB and the US EPA. CARB claimed societal air quality benefits of $13 billion at the cost of $3 billion to the construction industry. Yet CARB never proposed to pay the owners of the equipment for their lost investments. GHG regulation is proceeding down the same path.

If the benefits truly justify adopting a policy, and GHG reductions certainly appear to meet that criterion, then society should be willing to compensate those who made investments under the previous policy environment that endorsed those investments. Certainly there’s questions about whether those investors truly had property rights in the resources they used, but that issue should be addressed directly, not as an implicit assumption that no such property rights ever existed. (This question about property rights has been raised in regulating California’s water use.) Too often policy proponents conflate a goal of an improved environment with goals to redistribute wealth. By jumping over the property rights question, wealth also can be redistributed implicitly. Societal equity issues are important, but they shouldn’t be achieved through backdoor measures that make all of us worse off. Requiring politicians and bureaucrats to consider the actual cost of their policy proposals will make us all better off, and maybe even remove obstacles to a better environment.