In the spring, a series of price spikes forced several Retail Electric Providers (REPs) out of business, and led briefly to dramatically higher rates for some customers.
In September, Hurricane Ike—the third costliest hurricane ever to hit the United States— slammed into the Texas coast, wreaking havoc on utility infrastructure. CenterPoint Energy alone announced repair costs totaling $750 million.
Then there’s the global economic crisis. The utility sector, as the third-largest borrower after the financial sector and the federal government, already faces massive exposure to the credit crunch. But that exposure is amplified for Texas REPs, which require large amounts of capital to cover their positions in the Electric Reliability Council of Texas’s (ERCOT) various markets (see sidebar, “Texas Credit Quality”).
Such developments set the stage for a November 18 pre-session hearing of the Texas Senate Business and Commerce Committee, which oversees regulated industries. Committee Chairman Sen. Troy Fraser billed the hearing as an annual forum in which to ask the central question, “Is our deregulated market working?” The first person to testify responded by invoking Dickens.
“It was the best of times, it was the worst of times,” stated Barry Smitherman, chairman of the Public Utilities Commission of Texas (PUC). “Things are positive here in our state. We continue to see development in generation and transmission. Nevertheless, there are storm clouds on the horizon, coming in from outside Texas, that are beginning to have an effect on the Texas market.”
The crises that buffeted Texas in 2008 tested the state’s deregulation model, now in its seventh year, on several fronts. And though the challenges largely were met, companies and consumers paid a heavy cost. Experts hope the lessons learned will strengthen the market and provide a better framework going forward.
“From my perspective, the punch line of all this is that every time we do something we learn that restructuring in this industry is harder than we thought it would be,” says Lynne Kiesling, an economics professor at Northwestern University who has written extensively on the Texas market.
Kiesling says it’s important to understand the ways the Texas experience differs from deregulation efforts elsewhere, differences that extend well beyond the simple fact that the state is an energy island—ERCOT being the only North American grid contained within the borders of a single state.
First consider the culture.
“There is a deep cultural element of independence, meritocracy and self sufficiency that goes back to the Republic of Texas,” she says. Then there’s political will. The deregulation movement in Texas was accelerated by the state’s governor in the 1990s—George W. Bush—a self-described “decider,” who was keenly focused on developing a retail power market in the state.
“There wasn’t a specific, clearly defined notion of what that meant,” Kiesling explains. “But that’s a feature, not a bug. California was very defined, but they fell on their face.”
Finally, there was infrastructure and experience. The industrial community in Texas, particularly the oil and gas sector along the Gulf Coast, had been using distributed generation, cogeneration and waste-heat recovery technology for decades. Put it all together and it created an ideal market for deregulation.
In 1999, then-Gov. Bush signed Senate Bill 7, Texas’s deregulation statute, and it took effect in January 2002. The wholesale market was created first, followed by retail competition. Industrial customers had lobbied hard for retail choice, and their robust participation was anticipated. Smaller customers initially were slow to switch providers, but by early 2008, 67 percent of residential customers had either switched to a different REP or to a new rate package with their existing provider. Deregulation was looking like a success.
However, underlying flaws in the relationship between wholesale and retail markets were about to emerge.
“I’ve been ranting for more than a decade about dynamic pricing and demand response, and how if you reduce barriers in retail markets to entry you’re going to get this array of products and services,” Kiesling says. “I’m not happy about it, but I’m willing to admit that even I oversimplified things. It’s not that easy.”
“Wind is an uncontrollable resource,” says Dan Jones, the independent market monitor for ERCOT. “It’s a piece of the supply mix unlike anything we’ve traditionally had.”
In other words, it blows when it wants to. A strong wind can push thousands of megawatts onto the grid, and then drop off to almost nothing. Although modern forecasting systems can model wind generation patterns with a great degree of accuracy, ERCOT’s rudimentary forecasting methods didn’t prepare the market for ebbs and surges in wind power coming from the flatlands—culminating in a dangerous frequency sag when wind generation didn’t deliver as expected on Feb. 26, 2008.
Moreover, ERCOT’s zonal model didn’t correlate to the actual location of generation and congestion. The result was several weeks of tremendous volatility and price spikes in the spot market. At times, electricity prices in southern Texas almost doubled those in northern Texas.
“We got a good dose of what the limitations on the system are,” Jones says. “Even though we had the same model for the first seven years of the system, we obviously knew problems needed to be solved. It was just a matter of time before something like that occurred. We put in some changes, but the ultimate solution is to get the market redesign in place.”
In 2006, Texas passed California to become the national leader in wind power generation. The Texas deregulation scheme provides significant incentives for renewable power development, and the state’s wind resource is sufficient for it to become a competitive energy source. But the development has happened so fast that wind generation already has exceeded transmission capacity (see Figure 1, “Winds of Change”). Because most wind turbines in the state are constructed in the vast, sparsely populated regions of West Texas and the panhandle, there simply aren’t enough wires to move the power to market. The result is congestion in regions with comparatively light infrastructure.
The ERCOT wholesale markets rely on what’s called zonal pricing to accommodate congestion cost—prices reflect the average congestion in each of more than 30 zones. But zonal congestion pricing largely has become discredited, according to the latest thinking about wholesale electricity markets. Even CAISO, the California Independent System Operator—the most well-known prior adherent to that idea—decided to scrap zonal pricing in favor of nodal pricing in its MRTU initiative (Market Redesign and Technology Update).
“To the PUC’s credit, they saw this coming and were already working really hard to develop nodal pricing market platform for ERCOT,” Kiesling explains. “It’s just a massive endeavor.”
In fact, ERCOT and the PUC long have contemplated a move away from a zonal congestion pricing model, with the PUC launching an investigation to study ERCOT’s wholesale market design as far back as April 2002. That move came on the heels of a sharp spike in natural gas prices in the fall of 2000, which in turn produced an explosion of wind farms in West Texas in 2001. But with the ERCOT market lacking the nodal-locational price signals then common to such wholesale markets as PJM and New York, Texas saw developers locate 1,000 MW of new wind capacity in West Texas on the wrong side of a 500-MW transmission constraint in the Pecos River area. That left the new wind turbines partially stranded, lacking grid access to load centers farther east.
Moreover, Texas officials began voicing concerns about phantom congestion and gaming of decremental bids—the very same market distortions seen earlier in California, at the peak of the Golden State’s notorious power crisis, due in no small part to CAISO’s reliance on imprecise zonal approximations for congestion pricing.
Beginning in 2002, proposals started to emerge to move the ERCOT market toward a more granular design, with unit-specific bidding in place of portfolio bidding, and locational marginal pricing (LMP) with prices specific to individual buses (termed “nodes”) on the grid system. The PUC’s market oversight division suggested a hybrid, middle way, based upon flow gates and physical grid rights. But even then, industry leaders warned that such changes would come only at very high cost—and with customer benefits uncertain at best (see “Winds of Change in Texas,” Fortnightly, April 1, 2003).
Now, some six years later—on November 26, 2008 to be exact—ERCOT submitted a revised timeline to the PUC, reporting that it will implement nodal pricing in December 2010, at a projected project cost of $660 million.
Other changes are coming in Texas as well—some involving barriers to entry in the retail market.
“The market spikes caused five small REPs to go out of business,” says John Fainter, president of Association of Electric Companies of Texas. “It affected 40,000 customers—less than one-tenth of 1 percent of the market. I don’t want to minimize anything, because I think it’s going to take time to work through, but I think to a large extent it’s stabilized.”
Stabilized, yes. Resolved, no.
“The commission, as a result of what happened earlier this year, is undergoing a rulemaking modification to strengthen the financial assurance requirements of retail providers,” says Terry Hadley, a spokesman for the PUC.
Those changes include beefing up financial certification standards for REPs, improving disclosure practices so customers are better informed if their provider goes under, and revising the so-called provider-of-last-resort process.
The early days of deregulation were all about drawing new players to the table. Market barriers, traditionally almost insurmountable in the electric business, were intentionally left low. The strategy succeeded in giving customers choice—dozens of REPs now operate in Texas—but not all the new players could afford the game.
When the wholesale market went haywire in May, five of the smaller REPs couldn’t meet their credit requirements. In that situation, under SB7, the customer account passes to a provider of last resort. Electric service is not interrupted, but rates almost always go up, sometimes significantly. Hadley says the crisis demonstrated a need to improve the system.
“Most troubling was the effect on customers who essentially did the right thing,” he says. “They shopped for a provider, looking for value. However when the provider left the market suddenly, the provider-of-last-resort rule as it currently exists dramatically increased their electricity rate until they were able to switch.
“They can switch at any time, but the dilemma is that these smaller providers that go out of business often do not provide necessary notification, so there’s a delay in the customers finding out. The process worked in terms of keeping the juice on, but the notification aspect did not.”
Neither did the low table stakes.
“Right now the lowest barrier to be a retail provider is $100,000 cash,” Hadley says. “That will be significantly strengthened.” The PUC is considering new rules that would require prospective REPs to have an investment grade rating or tangible net worth of at least $100 million, with liquid capital of at least $1 million.
“Early on the intent was to encourage as much market entry as possible,” Hadley says. “Now that the market has matured, the current commission has realized it’s time to strengthen the minimum financial requirements for providers.”
The challenges of 2008 revealed a lot about Texas’s deregulation scheme—both what works, and what doesn’t.
“I think it showed the evolution of [deregulation],” Hadley says. “We’re now in our seventh year. At first the key issue was getting customers to realize that they had to shop for electricity if they wanted the best price. Now we’ve evolved to the point where a significant number of customers have to deal with the fallout when smaller providers can’t meet financial requirements.”
The ongoing maturation process will reshape the PUC’s approach to reducing transaction costs and increasing consumer protection. Plus, wind generation presents growing challenges for load forecasting and market management.
“We have to get smart about predicting wind generation,” Hadley says. “If you think you’re going to have 5,000 MW of wind blowing tomorrow and it doesn’t show up, that’s a big number. And vice versa, if you decide not to count on any of it, then you might bring on a lot of units that you don’t need, which is costly. It’s a real balancing act. We need to get a lot better about forecasting the wind output.”
Toward that end, ERCOT has been testing a state-of-the-art wind forecasting system from AWS Truewind, and expects to implement such a system when it converts to a nodal market. Further, the Texas PUC has come to be seen as the best-practices leader on the question of assuring enough electric transmission capacity to bring renewable energy from its often-remote production areas to the big population centers where it’s needed.
Under the Texas solution, known as the CREZ model (Competitive Renewable Energy Zones) and implemented by a PUC order in October 2007, the state follows a top-down process that begins by mapping out geographic areas most suitable for wind energy development, and then conducts region-wide planning to get new grid lines in place—ideally before the wind developers come forward with new projects and find themselves waiting in a lengthy interconnection queue for permission to hook up to the grid.
The Texas CREZ model has won praise as a logical way to break the chicken-and-egg impasse that can force delays in wind project development for lack of transmission capacity. California, the nation’s acknowledged leader in renewable energy development, has sought to emulate the Texas CREZ model under its Renewable Energy Transmission Initiative (see www.energy.ca.gov/reti).
Nevertheless, as always, the devil is in the details, and the details almost always have to be worked out in practice. But despite the difficulties, the solutions that emerge in Texas might show more about what’s right with ERCOT than what’s wrong with it.
“Deregulation in Texas hasn’t been a slam dunk,” Kiesling says. “There’s been price volatility. Every time you turn over a stone you find a new set of complex issues. But communicating price signals from the consumer all the way back to the generator part is partly how you get more efficient resource use, lower environmental impact, and reductions in cost.
“That’s the integration of retail and wholesale markets,” she says, “an important idea that’s reflected in the Texas market design.”