Outlook for Utility-Scale Renewables in California
Near-term surpluses of renewable energy, a sharper focus on costs and heightened concerns over environmental impacts are the new realities in the California market for utility-scale renewable power.
California’s largest investor-owned utilities are expected to slow the rate at which they procure renewable energy in the near term as they meet or draw near to meeting their regulatory mandates under the state’s renewable portfolio standard.
As the cost of renewable resources has fallen, regulators and utilities have both sharpened their pencils when it comes to new projects, and only projects that are competitive with the new market realities are winning in utility solicitations.
Finally, regulators are more rigorously evaluating the environmental impacts of large-scale solar and wind projects, and projects with significant environmental impacts face an uphill battle to win regulatory approval.
California is not turning away from renewable energy, but developers are likely to find a more competitive marketplace in the near term. Projects that can offer a cost or technology advantage will fare better in this tight market. Looking farther out, demand could rebound once regulators and legislators define the post-2020 renewable portfolio standards.
Meeting the 33% RPS Mandate
California’s renewable portfolio standard of 33% renewable power by 2020 has led to a decade-long boom in renewable energy project development. However, the state’s largest utilities — Pacific Gas and Electric, Southern California Edison and San Diego Gas and Electric — have over-procured renewable power for the near term and claim to have enough projects under contract to meet most or all of the 2020 RPS mandate.
Presently, the utilities are exceeding annual renewable energy targets and are banking or selling off their surpluses to draw upon in later years. PG&E anticipates that it will not need to draw on its banked RPS credits until 2019 and will have enough banked credits to meet a 33% RPS requirement through late 2023. To continue meeting a 33% annual RPS target beyond 2023, PG&E forecasts that it will need an average of 9,500 additional GWh per year from 2024 through 2030 (see Figure 1 on page 4).
SDG&E similarly anticipates that it will continue to contribute surplus RPS credits to the bank through 2019 and that, between ongoing contracts and banked RPS credits, it has enough contracted resources to meet a 33% RPS requirement through 2025. SDG&E estimates that it will need an average of 2,000 GWh of incremental renewable power each year from 2026 through 2030 (see Figure 2 on page 4).
Southern California Edison’s surplus is not as large as PG&E’s or SDG&E’s. SCE anticipates tapping into its banked reserves by 2017, exhausting its balance in 2020, and needing an additional 7,300 GWh of renewable energy in 2020 to meet the 33% RPS requirement for that year. SCE forecasts an increasing procurement deficit post-2020, with a need, on average, for an additional 13,000 GWh of renewable procurement per year to meet a 33% RPS requirement from 2021 through 2030 (see Figure 3 on page 4).
The utilities’ assessments suggest limited contracting opportunities for renewable projects coming on line before 2020. However, these numbers do not tell the whole story because uncertainty associated with the utilities’ forecasts may increase or decrease the forecasted need for additional renewable procurement. These uncertainties affect both the demand and supply sides of the equation.
On the demand side, the primary uncertainty is the level of future electricity sales. If sales (i.e., consumption) are higher than anticipated in the RPS assessments, then RPS requirements will be correspondingly higher and the utilities will draw down banked credits more quickly. The need for new procurement would occur earlier than currently anticipated. This is a symmetrical risk, as lower electricity sales would reduce the RPS requirement and delay the need for new procurement.
In addition to the utilities’ preferred RPS procurement forecasts presented above, the utilities also developed alternate forecasts that use sales assumptions from the California Public Utilities Commission. Under the alternate sales forecasts, PG&E would have less need for incremental renewable procurement than in its preferred forecast (with PG&E’s RPS procurement deficit delayed from late 2023 to 2025), and SCE would have greater need for incremental renewable procurement than in its preferred forecast (with SCE’s RPS procurement deficit starting in 2019 instead of 2020).
On the supply side, there is the risk that some contracted projects will fail to achieve commercial operation or will be delayed. Projects under development face any number of hurdles in financing, permitting, interconnection and completion of construction. Delays and cancellations are not uncommon. Historical project failure rates have been as high as 30% to 40%. While failure rates appear to have fallen significantly in recent years, project delays and failures remain a concern.
Many of the projects included as existing contracts in the utilities’ procurement plans remain under development. For example, as of December 2013, only about half of the 74 renewable energy projects included in SDG&E’s plan to meet its 2020 RPS were operational, with nine projects under construction and 27 projects in the pre-construction phase. SDG&E has acknowledged that some of these projects are experiencing project development-related issues that may affect their ability to meet commercial operation deadlines or even to come on line.
Development risk is accounted for in the utilities’ procurement plans to varying extents. SDG&E assigns a probability of success to each individual project, with an average success rate of 75% for approved projects that have not yet begun delivering energy. SCE uses project-specific, risk-adjusted success rates for large, near-term projects that are not yet on line and a success rate of 50% for projects with commercial operation dates more than three years out. PG&E assigns a success rate of 0% to high-risk projects and assigns a success rate of 100% to all other projects. PG&E defines high-risk projects as those that have failed to meet contractual deadlines or have certain known issues that place them at risk for doing so, as well as projects that were operating in the past but have ceased operation. Accordingly, it appears that PG&E would assign a success rate of 100% to a newly-contracted project that had not yet received CPUC approval as long as that project had no known financing, permitting or interconnection issues. To the extent that this assessment or the other utilities’ risk assessments underestimate project failures and delays, there may be a need for additional renewable procurement to replace contracts that do not deliver as planned.
There is a possibility, as well, that the CPUC will modify the risk assessment approach that is used in the calculation of need for new renewable procurement. The CPUC is concerned that the utilities’ assumptions of project risk are insufficient. The utilities’ confidential assessments have not been benchmarked against actual project success, and the utilities have been unwilling to provide data publicly that would allow such benchmarking.
In February 2014, the CPUC staff proposed formal benchmarking of utility risk assessments through an independent analysis of projects under development using a public methodology that assesses a project’s risk based on the following weighted project viability categories: project technology (10%), the developer’s experience (15%), site control status (25%), permitting status (25%) and interconnection progress (25%).
Under the proposal, the CPUC staff would assign each project a viability score based on a standard rubric that assesses each of these elements using pre-determined metrics. (This rubric would be a simplified version of the existing “project viability calculator.”) For example, the score for developer’s experience would be assessed as follows: 50 points for no demonstrated experience developing renewable energy projects, 75 points for any demonstrated experience developing renewable energy projects, 90 points for demonstrated experience developing renewable energy projects of similar size and technology, and 100 points for demonstrated experience developing renewable energy projects of similar size and technology in the utility’s service territory. The CPUC staff would use the project viability score to adjust a utility’s entire portfolio of RPS projects under development for risk. Staff would then benchmark the staff’s risk adjustment scores against each utility’s own risk adjustment to determine if there are any outliers that the utility would be required to justify in its annual RPS plan.
The CPUC is expected to issue a decision on this matter in the second quarter of 2014. It is too early to predict whether the decision will increase contracting opportunities.
Additional contracting opportunities could also emerge if the utilities sell some of their surplus renewable power to third parties with near-term need for renewable energy credits. For example, if an entity with the need for RECs in 2015 purchases some of PG&E’s banked RECs, PG&E’s need for new power contracts could advance by several months or more in the early 2020s when it currently anticipates relying on banked credits to meet its RPS requirements. This situation would open up new opportunities for competitively-priced renewable energy projects that are not already operational (i.e., projects that could not meet the near-term REC need directly but could meet the replacement power need in the early 2020s). The utilities have said that they will sell banked credits only if the sales price is higher than the replacement power cost. This is possible given the steep decline in renewable prices in recent years; however, opportunities are likely to be limited.
Focus on Price
The cost of the renewable energy contracts that make up the current RPS portfolio has prompted both concern and optimism in California.
The concern is that expensive renewable energy will lead to higher retail electricity rates for consumers at the same time that other factors are already driving up power costs. For example, Energy and Environmental Economics, Inc. forecasted in 2012 that rates in 2020 will be 8% higher than they would be under an all-gas scenario due to the 33% RPS, while prices will be more than another 11% above 2011 rates in real terms for non-RPS reasons such as the need to replace aging transmission and distribution infrastructure, pay for Smart Meter projects, and repower or replace generators to comply with once-through cooling requirements.
On the other hand, there is room for optimism due to the decline in renewable energy prices over the last few years. While the weighted average price of bundled renewable contracts approved from 2003 through 2011 was 12.2¢ per KWh for PG&E, 10.1¢ per KWh for SCE and 11.6¢ per KWh for SDG&E (in nominal dollars), bundled renewable contracts approved in 2013 had declined on average to 6.7¢ per KWh for PG&E, 8.9¢ per KWh for SCE and 7.5¢ per KWh for SDG&E. This decline reflects lower bid prices in the 2010 to 2012 RPS solicitations, consistent with industry-wide cost reductions.
Given these cost reductions, regulators are now able to exercise some cost discipline and greater selectivity in approving modifications to existing contracts, knowing that modifications to contracts from past solicitations that are denied are likely to be replaced by lower-cost contracts in future solicitations. So far, however, the CPUC has been very selective in exercising this option, with the rejection in October 2012 of three of BrightSource’s proposed solar thermal projects being the notable exception.
Despite the downward trend in prices, legislators have expressed concerns with the upward pressure on retail electricity rates resulting from RPS procurement. As part of the 2011 legislation that increased the RPS from 20% to 33%, the CPUC is required to implement a “procurement expenditure limitation” in order to impose some cost discipline on the RPS procurement process. The CPUC is currently considering methods for establishing such a limitation. The CPUC staff has proposed a method that would establish a ratio of RPS procurement expenditures to a utility’s total revenue requirement over a 10-year period. The ratio would provide a benchmark to indicate whether the forecasted RPS procurement is likely to put upward pressure on retail electricity rates. Other parties have proposed alternative methods.
According to an illustrative example provided by CPUC staff, SCE’s annual “procurement expenditure limitation” ratios under the staff’s proposed methodology would range from 14.6% to 21.2% over the 10-year period from 2014 through 2023. The ratio would essentially set an overall budget for SCE of $26.9 billion to spend on procuring RPS-eligible energy in that time frame. In 2013, SCE spent $1.4 billion, or 11.9% of total revenue requirement, to achieve an RPS level of 23.2%. Adjusted to account for the higher 2014 to 2023 RPS requirement, this level of expenditure — $17.4 billion over the 10-year period — would still remain well within this illustrative budget. While these results are merely illustrative since a final methodology has not yet been adopted, given this outcome, it remains to be seen whether the procurement expenditure limitation methodology will impose real price discipline or will serve only as a high ceiling price.
Regardless, price discipline will continue through competition among renewable energy developers. Market competition and reduced project costs have driven down the cost of newly-approved renewable contracts by more than 25% since 2011 and are likely to continue putting downward pressure on prices, particularly if new contracting opportunities remain limited in the near term.
Minimizing Environmental Impacts
Environmental concerns over the impacts of large-scale renewable energy projects are moving to the foreground as well. This reflects to some degree the knowledge and experience gained as the initial wave of renewable projects complete construction and begin operations.
In December 2013, a California Energy Commission siting committee released a proposed decision recommending that the CEC deny BrightSource Energy’s application to convert the proposed 500-MW Palen project from a solar thermal parabolic trough project to a project that uses BrightSource’s solar thermal power tower technology, in large part due to concerns over avian mortality.
The Palen project’s power tower system would create steam by using a field of 85,000 elevated mirrors known as heliostats to focus the sun’s rays onto a steam generator that sits atop a 750-foot tower near the center of the heliostat field. As proposed, Palen would consist of two adjacent 250-MW fields.
The CEC previously approved a different BrightSource power tower project, the 377-MW Ivanpah project, which consists of three 459-foot power towers and 173,500 heliostats. The CEC approved the Ivanpah project in September 2010 and concluded that the clean energy benefits of the project outweighed its significant impacts on cultural, visual and environmental resources, and that no feasible site or generation technology alternatives to the project existed that would reduce or eliminate the project’s significant environmental impacts.
Concerns about the impact of power tower technology on avian mortality began to surface during construction of Ivanpah, when BrightSource’s monthly compliance reports filed with the CEC listing avian deaths indicated possible increased mortality, particularly during the migratory months. BrightSource reported 23 avian deaths at Ivanpah in January 2014, up from the 13 deaths recorded in December 2013 and 11 reported in November 2013, but still less than the 52 reported in October 2013.
In the proposed decision denying Palen, the CEC siting committee, consisting of Commissioners Douglas and Hochschild, concluded that, as proposed, Palen would result in significant, unmitigable impacts on local environmental, visual and cultural resources, and that the solar flux generated from the project’s solar towers would probably harm birds. The committee said the original solar trough project or a conversion to photovoltaic technology would be the preferred options for the project site. In an effort to avoid a CEC decision denying the project, BrightSource requested that the commission postpone voting on the proposed decision until at least the spring of 2014, to allow the company more time to present additional data on avian mortality being gathered at Ivanpah and from other projects employing alternative solar technologies.
The difficulties faced by BrightSource are, to a certain extent, technology specific and are not indicative of a wholesale change in sentiment against large-scale solar. At a January 2014 conference on the proposed decision regarding Palen, the CEC noted that BrightSource still has the option to build Palen as the solar thermal parabolic trough project that has already been approved or to propose a different project on the site. Commissioner Hochschild specifically asked concerned parties not to read the proposed decision as a strike against solar thermal and emphasized the benefits of the technology, stating that he believed it has a role to play as California expands its clean energy portfolio.
That same month, the CEC also demonstrated that significant environmental impacts will not necessarily undermine a renewable project, as it unanimously approved an amendment to modify the proposed Blythe project from a 600-MW solar parabolic trough project to a 485-MW solar PV plant, even though significant environmental impacts were identified. The CEC concluded that the project would result in benefits that outweighed these impacts and that there were no feasible alternatives to the project that would reduce or eliminate any of the impacts.
Environmental impacts are also a concern with wind projects, and avian mortality issues in particular have come to the fore in this context as well. The US Department of the Interior recently began granting wind developers eagle “take” permits lasting up to 30 years that, under the Bald and Gold Eagle Protection Act, shield projects from liability for unavoidable bird deaths at wind plants. (In the past, the Interior Department only issued take permits that lasted for up to five years.) To be eligible for these extended permits that will be subject to review every five years, wind plant operators must agree to regular monitoring and adaptive conservation measures. This approach provides greater certainty for renewable energy developers while offering some measure of protection to threatened species.
These decisions, both at the CEC and the Department of the Interior, show how government agencies are trying to find a balance between renewable energy development and environmental protection. The agencies are still trying to find the right balance, and this creates risk for developers. While most projects that are thoughtfully sited are not likely to be rejected on environmental grounds, as BrightSource found, the risk of rejection is all too real, particularly for less-tested technologies.
Potential New Opportunities
Despite the slowing growth in demand for renewable energy projects, downward price trends and more stringent reviews of environmental impacts, opportunities for new utility-scale renewable projects still exist in California.
The utilities’ assessments of when they will need to ramp up procurement of renewables and how much to procure are based on a 33% RPS mandate. The likelihood is quite high that there will be a need for a greater level of renewable resources after 2020 as California continues to pursue its goal of reducing greenhouse gas emissions to 80% below 1990 levels by 2050. As part of that effort, the California Air Resources Board has recommended that the RPS target for 2030, expected to be above 33%, be set in 2016 to allow enough time for contracting and development.
In addition, the state legislature recently granted the CPUC the authority to require utilities to buy more renewable energy than required under the RPS requirement. While the CPUC has not indicated its intention to do so on a universal basis, this could open up opportunities in specific circumstances. For example, a March 2014 decision that directs Southern California Edison and SDG&E to procure 40% (600 megawatts) of the power needed to replace the closed San Onofre nuclear power plant from preferred resources may lead in the near term to opportunities for new renewable power development above the RPS-driven requirements.
A similar opportunity would likely emerge in the event that the PG&E Diablo Canyon nuclear power plant licenses are not extended beyond their current expirations in 2024 and 2025. The CPUC has already put PG&E on notice that the utility will need to justify the economic costs and benefits of the large nuclear plant before the CPUC authorizes any ratepayer funding for a federal relicensing application. Should the plant not be relicensed, the carbon-free power that Diablo Canyon currently generates is likely to be replaced to a large degree from renewable resources and other preferred resources.
Additional opportunities could also open up in the near term due to utility load growth (which triggers the need for additional RPS procurement since the RPS target is a percentage of load), unanticipated contract failures, a change in the methodology for predicting contract failure rates, or utility sales of banked RPS credits to third parties. These opportunities are likely to be limited.
Renewable projects that incorporate energy storage technologies may have an advantage in upcoming solicitations. California faces a significant challenge in balancing the increasing share of variable energy resources on the grid, and the CEC and the CPUC have both made it clear that they are looking to storage as part of the solution. For example, CEC Commissioner Douglas indicated that the addition of thermal storage capability would greatly strengthen BrightSource’s application for the Palen project. Similarly, when the CPUC rejected three BrightSource solar thermal contracts in part on economic grounds, the CPUC at the same time accepted an uneconomic BrightSource contract for a solar thermal project with accompanying storage and even accepted a second uneconomic BrightSource contract for a solar thermal project on the grounds that the project was needed to lay the groundwork for a more advanced project with storage to be financed and built.
Renewable projects with storage may be eligible to bid in the solicitations that the utilities are preparing to issue by December 2014 to procure additional storage capacity toward meeting a CPUC-mandated target of 1,325 MW of storage by the end of 2024.
Thoughtful project siting will also remain key. The US Department of Energy and the US Bureau of Land Management jointly established the solar energy zones program in 2012, which identified 17 solar energy zones in the western US, defined as areas with few impediments to utility-scale production of solar energy where BLM would prioritize solar energy and associated transmission infrastructure development. In addition to the 285,000 acres in the 17 solar zones, BLM identified roughly 20 million acres outside of the zones that are available for right-of-way or lease applications if developers apply for a “variance.” Projects in these zones will have permitting advantages over projects located outside of these preferred areas.