4. Implementation Issues
In the workshop and workshop report comments, a number of parties discussed questions related to the implementation of a GHG cap on IOU procurement. Those implementation issues include the following: (1) GHG emissions baselines; (2) adjustments to GHG emission reduction requirements (and associated caps) over time, relative to those baselines; (3) allocation of emissions allowances; (4) flexible compliance mechanisms; (5) potential penalties; (6) requirements for registration with the CCAR; (7) continuation of the GHG or carbon adder adopted in D.04-12-048; and (8) treatment of GHG emissions from the provision of natural gas for purposes other than electricity generation.
Below we discuss the key implementation issues. We also make a number of preliminary determinations to guide our next steps in implementing a load-based cap.
4.1. GHG Emissions Baselines and Adjustments to Reduction Requirements Over Time
Significant issues surround the question of how to establish GHG emissions baselines against which to set a GHG cap and make future downward adjustments to that cap. Baseline options include multi-year averaging of historical GHG emissions or selection of one single historical baseline year. Another option suggested in the workshop was to develop the emissions cap based on the emissions profile of the IOUs' procurement plans going forward. The method selected has the potential to reward or penalize entities for their prior performance. In addition, significant technical issues exist, such as how to account for weather variability among potential baseline years.
UCS suggests that the selection of a GHG emissions baseline be guided by both principle and practicality. In principle, UCS believes that the chosen baseline should represent each utility's predominant existing pattern with respect to GHG emissions. UCS submits that using an average of historical years is the best method to achieve this goal. Solargenix agrees.
In the alternative, UCS proposes that the Commission could adjust a single year's data to reflect average-hydro-year conditions. In terms of practicality, UCS recommends using a historical period for which the most comprehensive and accurate data can be obtained. In UCS's view, the use of historical year or years avoids much of the potential for gaming that is inherent in using a prospective year. Whatever baseline is chosen, UCS encourages the Commission to work with the CCAR and the CEC to establish methods of assessing out-of-state emissions.
Once the baseline is established, UCS recommends that the trajectory of emission reductions be based on reasonable assumptions about the technical potential of innovations in the GHG emissions reduction area. UCS believes that existing energy efficiency and renewable commitments should not be assumed to exhaust the potential for these resources. UCS suggests creating a supply curve of such resource options, in order to better evaluate the future potential for GHG emissions.
GPI recommends establishing 1990 as the baseline year along with a 7% emissions reduction requirement by 2010, in order to be consistent with the Kyoto Protocol. In GPI's view, this approach has the advantage of harmonizing California's baseline with international efforts. GPI would only enforce this requirement after the emissions profiles of the utilities have been established for the period since 1990. GPI also recommends that targets be reduced over time by some reasonably achievable margin. Further, GPI recommends that hydroelectric variability be addressed in the evaluation of IOU performance.
NRDC recommends a series of joint workshops with the CEC Climate Change Advisory Committee to develop an appropriate baseline and requests that early actors not be penalized in whatever baseline is adopted. PG&E agrees that early action should be recognized in the baseline. NRDC recommends that emission reduction requirements be established over a long period of time in order to send a clear market signal, and that flexible compliance mechanisms should be allowed.
SDG&E recommends that the Commission limit its consideration of these issues to a pilot program, in order to gain experience with GHG cap issues over time. SDG&E also recommends that any GHG cap adjustment take into account factors that are outside of the utilities' control, such as population, economic activity, and pre-existing contracts. SDG&E also suggests building in off-ramps in case costs escalate. SDG&E opposes any approach that would calculate a baseline and associated emissions cap from the emissions profile of an adopted procurement plan. In SDG&E's view, this would not truly be a baseline because it would be calculated using various assumptions and emission factors and could not be relied upon to gauge true changes from year to year by comparing certified emissions.
We agree with UCS and others that a historical reference point, rather than a prospective one based on procurement plans, should be used to establish a GHG emissions cap for LSEs. As UCS points out, the use of a prospective year has the potential for creating a perverse incentive for LSEs not only not to take immediate measures to start reducing GHG emissions, but to take measures that would actually increase their GHG emissions. The use of a historical baseline avoids this perverse incentive as well as the reliability issues identified by SDG&E. Moreover, using a historical baseline is consistent and compatible with efforts underway on the state and international level to address climate change.
As discussed in Section 2, subsequent to the workshop and the filing of post-workshop comments in this proceeding, Governor Schwarzenegger announced statewide GHG emission targets that establish 1990 as the historical baseline year against which emission reductions for 2020 and beyond will be gauged. As GPI points out in its comments, the Kyoto Protocol also uses 1990 as the emissions reduction baseline. The selection of 1990 as the reference year for a load-based GHG emissions cap clearly allows the greatest harmonization with the Governor's Executive Order and with existing international efforts to address climate change.
Therefore, it is our preference that 1990 be used as the baseline for developing a load-based GHG emissions cap in this proceeding. Our final determination on this matter will await further discussion of implementation issues associated with using this particular year as the reference, including the availability of adequate historical emissions data for the LSEs.31
We also leave to that discussion the consideration of the appropriate level of emissions reductions (and associated cap) over time, relative to the baseline year. For example, we could cap the emissions of each LSE at 1990 levels by 2020 and at 80% below 1990 levels by 2050 to be fully consistent with the statewide GHG reduction targets-or adopt an alternative trajectory of emissions reductions to serve as the load-based cap for LSEs. We will also need to adopt emissions reduction requirements (and associated caps) for the years between now and 2020.
We believe there is considerable merit to UCS's recommendation that this process be informed by an assessment of achievable potential in GHG reductions over the reduction period. During the implementation phase, we will explore UCS's suggestion that a "supply curve" of GHG reduction measures associated with each utility's resource portfolio be developed for this purpose.32
We also agree with a number of parties that we must account for the variability of hydroelectric resources in any given year. We leave to the implementation phase of this effort the determination of the best manner to account for hydro variability.
In addition, we recognize that the CCAR is essential to this effort. We note that CCAR participated in the workshops in this proceeding by describing the emissions data collection efforts already completed and those underway. CCAR has also offered to work closely with the LSEs on the further development of emissions data and with this Commission in exploring the implementation options associated with a load-based cap.33 We appreciate CCAR's constructive participation in this proceeding. We will work closely with them, as well as the Governor's Climate Action Team, in our efforts to establish baselines and associated GHG emissions caps.
Finally, in order to facilitate further rigorous assessment of current performance in the establishment of GHG caps, we will require that all LSEs subject to the Commission's 2006 procurement process file information about their GHG emissions performance in their procurement plans.
As suggested by NRDC in its comments, 2006 procurement plans should include an integrated strategy for reducing GHG emissions over the timeframe addressed in the long-term plans. The plans should also include detailed information about the resource types (including different fuel types) planned for and the emissions characteristics of the preferred resource plans, as well as the various other resource scenarios.
In addition, the 2006 procurement plans should also include detailed information about the existing GHG emissions characteristics of the utilities' portfolios without the new resource additions proposed in the procurement plans. These will offer a starting point for further consideration of how to establish GHG reduction requirements that will most effectively reduce the absolute level of GHG emissions over time.
4.2. Allocation of GHG Allowances
An "allowance" refers to a permit provided to the LSE within the scope of the GHG emissions cap to emit one unit of emissions (e.g., ton of CO2 equivalent). There are basically two options for distributing GHG emissions allowances to LSEs. The first option is to have an administrative allocation. The second is to have an auction where LSEs with obligations bid for the GHG emissions allowances. Opinions vary on the appropriate manner in which to allocate allowances, particularly for the first time. The initial staff Sky Trust proposal advocated an auction structure in order to provide additional revenues to fund energy efficiency and potentially other EAP preferred resources. The modified staff proposal developed during workshops stepped away from recommending an auction.
UCS advocates administrative allocation of the GHG emissions allowances in order to avoid potential problems with handling large revenue streams that would result from an auction. However, UCS believes that a limited auction could provide flexibility under the cap. In addition, UCS recommends further analysis of utility GHG emission profiles before answering this question.
NRDC recommends that an administrative allocation approach be used that ensures (1) no large windfalls, (2) no penalties for early action, (3) LSEs are motivated to make investment decisions that will reduce emissions, (4) administrative burdens are minimized, and (5) updating mechanisms do not penalize action. NRDC also advocates that allowances be allocated to LSEs on behalf of their ratepayers and not their shareholders. NRDC suggests that a limited auction could be useful to raise additional funds as contemplated under the staff Sky Trust proposal, with the qualification that proceeds from such an auction not be used to replace dedicated funds for existing programs.
In addition, NRDC lists three key indicators that should be considered in determining the LSE-specific allocation, namely, number of customers, percent of statewide retail sales and historical emissions. NRDC advocates an initial allocation based on number of customers, in order to encourage energy efficiency by customers. NRDC also recommends further analysis of the option to weight allocations by customer class.
As discussed above, SDG&E prefers that a GHG emissions cap not be adopted at all. However, if one is established, SDG&E recommends that allowances be allocated administratively. SDG&E requests that such allocation ensure inter-utility equity and account for the variability in GHG emissions outside of the control of the LSE. PG&E also prefers that allowances be allocated administratively if the Commission decides to move in this direction.
Based on the record in this proceeding, our preference is to allocate allowances administratively, based on some combination of the factors listed by NRDC: number of customers, percentage of statewide retail sales, and historical emissions. As discussed during workshops, an auction with so few buyers (as would be the case with a load-based cap for LSEs under CPUC jurisdiction) would be economically inefficient and prone to market power abuses. Allocation, rather than auction, also avoids the need for the Commission to undertake the set-up of an auction structure and rules. In addition, an administrative allocation of allowances is more conducive to the existing regulatory process we have been using to address procurement-related issues.
We are, however, certain that the manner in which we allocate GHG emission allowances will require a great deal more thought and analysis by the Commission and the parties. Therefore, we intend to have further discussion, perhaps in workshops, on this issue in the next phase of our investigation into implementation of the GHG emissions cap.
4.3. Flexible Compliance
In workshops and comments, parties discussed a number of issues related to flexible compliance with a GHG emissions load-based cap. These issues included: the use of offsets, trading of GHG emissions allowances, and banking and/or borrowing of allowances. The modified staff proposal presented at workshops included a proposal to allow limited offsets associated with utility-related activities within California, at least initially.
"Offsets" refer to a reduction in one unit of emissions outside the scope of the cap, which in turn allows an increase in emissions within the scope of the cap. In other words, offsets would allow an LSE to exceed its allocated GHG emissions allowances under the load-based cap, provided that it reduced a comparable level of emissions elsewhere. For example, if an LSE made approved investments in activities outside of the scope of the GHG emissions cap, such as investing in forestation projects, it would receive an offset credit.
GPI and UCS agree with the staff proposal that, at least initially, offsets should be allowed in a limited manner, and only for activities directly resulting from utility activities (for example, diesel pump electrification). They believe that it is important that IOUs focus immediately on reducing GHG emissions from their own energy portfolios, to ensure that the majority of emissions reductions originate from that portfolio and operational changes by the IOUs.
GPI would allow offsets that provide real net reductions in GHG emissions. UCS would allow offsets only if they are of high quality, independently verified, and with Commission oversight. In addition, they recommend limitations both quantitatively and geographically.
NRDC would prefer no offsets at all, with the possible exception of in-state programs directly connected to IOU operations. NRDC is also concerned that developing offset rules now would substantially delay work on other aspects of the GHG emissions program.
Solargenix is similarly concerned that counting offsets will create emission savings that will be difficult to track and maintain.
CAC/EPUC offer their cogeneration sources as potential in-state offsets for utilities.
PG&E, SCE, and SDG&E are unanimously in favor of allowing offsets, especially for all actions directly associated with IOU procurement (PG&E). In their view, offsets can encourage least-cost attainment of emission reduction goals.
Trading in the context of flexible compliance generally refers to the trading of allowances among entities subject to the GHG emissions cap, although offset credits can also be sold (traded) on the market. UCS would limit trading, at least initially, to inter-utility transactions. NRDC would do the same, but allow limited sales outside of California in order to provide the allowance incentives proposed under the modified staff proposal.
CAC/EPUC and Sempra recommend full market-based trading of emission allowances and offsets, working up to a regional or national trading system.
SCE and PG&E believe that unlimited flexible trading should be allowed. SDG&E, however, only advocates trading if purchasing is allowed beyond California borders. Otherwise, SDG&E contends that the market power of the two larger IOUs will place SDG&E at a consistent disadvantage. SDG&E also believes that allowing selling outside of California will likely be of little benefit, since California's GHG reductions are likely to be achieved at a relatively high cost. In SDG&E's view, banking is a better approach than allowing sales outside of California of emissions allowances.
All parties commenting on this issue indicate support, to one degree or another, for banking of emissions allowances (also referred to as emission "credits"). These parties include GPI, UCS, NRDC, Solargenix, SDG&E, and PG&E. According to SDG&E, this type of compliance flexibility is prudent because of the long-range nature of the GHG emissions problem. UCS would limit the amount of banking to encourage continued action. Both UCS and NRDC suggest options for consideration to discourage overbanking, such as expiration dates for allowances, an "accelerator" that would steepen the rate of decline in the cap if a threshold condition is met, or other discounting methods that would decrease the value of banked emissions credits over time.
GPI and SDG&E also advocate for borrowing of emissions credits from future years to be allowed.
All parties recognize that flexible compliance options are an integral component of a GHG emissions cap, but differ with regard to the scope and type of options that should be available, at least initially, as the Commission implements a GHG emissions reduction program. We are persuaded by the workshop comments and those submitted on the draft decision that there are significant drawbacks to adopting specific limitations to compliance options at this time, even with the intention of broadening those options over time.
Instead of dictating the type and location of offsets or the scope of trading at this juncture, we will focus our efforts during the implementation phase on ensuring that the compliance options that we do permit are credible, verifiable and administratively feasible, and that we further explore the pros and cons of alternate proposals for offsets, trading, banking and borrowing and other compliance options before making our final determinations. We will also evaluate the costs and benefits of the framework that emerges from the implementation phase. Throughout this process, we will continue to coordinate our efforts with the Governor's Climate Action Team as well as other state, regional or federal agencies that are exploring design options for cap-and-trade programs.
4.4. Penalties
Comments about the enforceability of the GHG emissions cap centered around the potential imposition of penalties for noncompliance.
UCS was the main party arguing that penalties are essential for program success. UCS recommends structuring penalties as alternative compliance payments (ACPs), so that the funds may be used for future GHG reduction efforts. UCS advocates collecting ACPs from shareholders, after flexible compliance mechanisms have been allowed to be fully utilized by LSEs.
NRDC recommends that CCAR protocols be used to track and report emissions, as well as monitor compliance. In NRDC's opinion, further analysis is needed in order to determine whether penalties are necessary.
SCE and SDG&E do not believe that penalties should be assessed for LSE non-compliance with Commission goals.
We agree with UCS that some form of penalty structure is necessary or else the program will only be a voluntary one. At this juncture, based on the discussion of this issue in the workshop report and in UCS's comments, we prefer structuring penalties as ACPs. We do not have enough information, however, to determine the level or exact nature of an appropriate penalty mechanism at this time. This subject will require further work during the implementation phase in conjunction with our examination of compliance options, as discussed above.
4.5. Emissions Registration
The question here is whether resource suppliers to LSEs (as well as LSEs themselves) should be required to register their emissions with the CCAR. This is an important step in quantifying existing and future emissions. We note that SCE, PG&E, and SDG&E are already voluntary members of CCAR, and register their emissions using CCAR's reporting protocols.
Of parties commenting on this issue during the workshop process, only Sempra is opposed to the idea of requiring suppliers to register with the CCAR. Sempra contends that such a requirement could limit the pool of potential suppliers to LSEs and raise prices for consumers.
UCS and NRDC favor requiring supplier registration with the CCAR through provisions in contracts with LSEs. They echo the suggestion made at workshops that IOUs be required to make registration a condition for granting an IOU power purchase agreement.
SDG&E recommends supplier registration with the CCAR, and suggests that if suppliers do not register voluntarily, their unspecified sources of power should be assigned the emissions value of coal.
Our preference would be to require the immediate registration of emissions by all generation resources serving California load with the CCAR. We agree with UCS and NRDC that this should be a required element of a power purchase agreement with IOUs in California. However, we understand that CCAR's registration requirements are currently "entity based," that is, CCAR requires certification of both direct and indirect GHG emissions for the entire entity in California (or in the United States). Therefore, during the implementation phase, we will explore with CCAR ways in which their protocols can be modified to include generation/facility specific data to fit within a load-based cap, and establish a date by which all power purchase agreements that PG&E, SDG&E, and SCE sign for power should include a provision requiring supplier registration with the CCAR. We may extend this requirement to the smaller electric IOUs under our jurisdiction after further consideration of this issue in a proceeding to which these companies are also respondents.
That leaves a significant portion of the existing supply market, as well as the ESP suppliers, left to voluntary registration with the CCAR. In order to address this larger portion of the market, we are inclined to adopt the suggestion offered by SDG&E. For any non-renewable supplies of electricity with fossil fuel emissions that are unregistered with the CCAR, SDG&E suggests that we require that those supplies automatically be assigned the emissions value of coal. In this way, LSEs purchasing those supplies will be encouraged to negotiate with suppliers for CCAR registration without the need for an explicit requirement by this Commission. However, we are persuaded by the comments on the draft decision that this approach and alternatives to it should be further explored during the implementation phase for our consideration.
Finally, we will take CCAR up on its offer to work with LSEs on developing appropriate proxy emissions factors for more accurate reporting of emissions.
4.6. Continuation of GHG/Carbon Adder
The issue here is related to the GHG/carbon adder adopted in D.04-12-048. Parties commented on whether the continued use of the adder would be necessary after a GHG emissions cap was in place.
In its comments, SDG&E characterizes a GHG procurement cap as a quantity limitation on the production of GHG emissions that produces a price to achieve that reduction, and a GHG adder as a price that leads to long-term resource choices that produce a certain reduction of GHG. If there were a GHG procurement cap with no price cap, SDG&E argues that the GHG adder is irrelevant. If there is a price cap, SDG&E suggests that the GHG adder be set at the price cap.
UCS believes that unless the GHG emissions allowance price is explicit, an adder will still be needed to shift procurement away from GHG-intensive resources. NRDC also argues that a forecast of allowance costs will still be necessary to allow for the possibility of future regulations that are stricter than the Commission's GHG cap.
PG&E and Sempra argue that the adder would be redundant with a GHG emissions cap.
Based on the comments in this proceeding, we are inclined to eliminate a GHG or carbon adder once a working GHG emissions cap is in place. However, the time to discontinue the use of the adder is only after we have successfully implemented a GHG emissions cap, and have considered the value of continuing with a carbon adder in that context. Until further notice from this Commission, the GHG/carbon adder should still be used in procurement resource evaluation.
4.7. Treatment of GHG Emissions From Natural Gas for Purposes Other Than Electricity Generation
Utility customers' direct use of natural gas is substantial compared with the amount of natural gas used to produce electricity for utility customers.34 NRDC estimates that approximately 15% of carbon dioxide emissions are associated with end-use consumption of natural gas, making it a substantial contributor to the State's GHG emissions.35 Both the Sky Trust and the modified staff proposals suggest that the Commission consider the issue of setting limits on GHG emissions associated with natural gas distribution for purposes other than electricity generation. The workshop report and subsequent ALJ ruling posed the following questions for comment:
(1) Can or should GHG emissions from non-electric generation usages of natural gas be addressed in a GHG cap and overall procurement incentive framework for the IOUs? If so, how?
(2) Are there important differences between a load-based and a generation-based approach in this regard?
(3) If the Commission focuses initially on GHG emissions associated with the production of electricity, what steps should it take to ensure that GHG emissions associated with customer use of natural gas can be incorporated in the future?
As a general principle, we believe that any long-term effort to limit carbon emissions should address natural gas use both for electricity production and directly by customers. To this end, the logical corollary to a load-based cap on GHG emissions for electricity procurement would be a GHG limitation program that includes emissions from the natural gas sector. However, we are persuaded by the comments and workshop discussion that we should refrain from limiting emissions from non-electric generation usages of natural gas until the requisite emission reporting and certification protocols become available.
In the meantime, however, we will continue to establish aggressive goals for energy efficiency in both the natural gas and electric sectors. We will also continue to support the CEC in its efforts to improve building and appliance efficiencies through codes and standards, and take other steps to reduce end-use consumption of GHG-emitting energy sources over time through our energy efficiency, demand response and renewable energy programs. During the implementation phase discussed below, we will further define the steps this Commission should take to ensure that GHG emissions associated with customer use of natural gas are incorporated into a procurement incentive framework in the future.
31 That discussion will also need to consider appropriate adjustments for energy service providers and community choice aggregators to take account of the fact that these entities did not exist as of 1990.
32 Post-Workshop Opening Comments of The Union of Concerned Scientists on Procurement Incentive Framework Workshop Report, May 2, 2005, pp. 8-9.
33 Letter dated May 23, 2005 to President Michael Peevey from CCAR.
34 The CEC reports that direct utility customer natural gas consumption is currently about 58% of total usage, with the remaining 42% used to power in-state electricity generation. This does not include gas usage associated with imported electricity. See http://www.energy.ca.gov/naturalgas/statistics/natural_gas_consumption_electricity.html.
35 Post Workshop Comments of the NRDC on a Procurement Incentive Framework, May 2, 2005, p. 10.
