4. History of SRAC Energy Pricing
4.1. Background of the Formula
The Commission has set SRAC energy prices using a variation of the following formula for 25 years:
SRAC Energy Price = Fuel Price x IER Heat Rate + O&M Adder
Each element of the formula has a lengthy history of CPUC proceedings and decisions. The formula reflects the fact that a fossil fuel - oil or natural gas - has always been the predominant marginal resource for producing electricity in California. The components of the SRAC formula reflect costs averaged over periods from one month, at a minimum, to as long as several years. Thus, SRAC prices will likely not equal IOU avoided costs on a day-to-day basis.
Since the outset of the QF Program, SRAC energy prices have always been set on a prospective basis. With respect to retroactive adjustments of these prices, the Commission has generally declined to make retroactive downward adjustments47 and we decline to do so here. Refinements to the SRAC methodology do not, in and of themselves, indicate that prior iterations of the SRAC calculations were wrong. The SRAC methodology provides an estimate of avoided cost and although we believe each refinement may increase the accuracy of the estimate, invariably whatever number is produced by the SRAC methodology will be off the mark by some, unknown amount. Constant ex ante adjustments to past payments, without any demonstration that such adjustments were necessary to comply with PURPA, create uncertainty and adds a great deal of complexity to an already complicated process.
Until the mid-1980s, fuel oil was the predominant marginal fuel. Avoided fuel costs were revised quarterly, based on the IOUs' actual costs. When natural gas largely displaced fuel oil in the mid-1980s, the avoided fuel cost was based on the fully bundled tariffed rate that the electric IOUs paid to the gas utilities for natural gas supplied for electric generation.
With restructuring of the natural gas industry in the late 1980s, the electric IOUs began to buy their own gas supplies, with the gas IOUs providing only transportation and storage services. Unbundled gas commodity markets opened first in the producing basins and later at natural hubs along the major interstate pipelines, such as Topock, Arizona and Malin, Oregon. The natural gas trade press began to report price indices for these markets.
In 1991, the Commission approved an "index methodology" to determine the avoided fuel cost, using published producing basin indices to track the electric IOUs' actual natural gas costs on a timely basis. SRAC postings changed from quarterly to monthly, to coincide with the reporting of monthly "bid week" gas prices.
From 1991 - 1996, the Commission adjudicated numerous issues concerning the index method, as gas markets continued to develop and the electric IOUs' gas purchases became more diversified and complex. The electric IOUs began to buy significant volumes in the border markets to take advantage of low border prices that resulted from the then-present excess pipeline capacity to California.
In 1995 and early 1996, it became clear that the California electric industry would be restructured. In an effort to simplify the transition to a restructured market in which electric market prices would set SRAC, and to reduce the contentiousness of the index method, the IOUs and QF parties agreed in early 1996 to move to simplified SRAC "transition formulas" to set SRAC prices until the PX market was functioning properly.
The Commission-adopted SRAC "Transition Formula" for each utility, pursuant to Pub. Util. Code § 390(b), prescribes the basic elements for determining energy prices to be paid to QFs. D.96-12-028 adopted specific formula values for each of the IOUs. Each IOU's Transition Formula includes a starting energy price, a starting gas price, a utility-specific gas factor (or factor), California border gas price, and intrastate gas transportation costs to approximate a burnertip gas price.48 The Transition Formula provides for the starting energy price to be adjusted monthly to reflect changes in assumed fuel costs, as reflected in percentage changes to certain border gas price indices. The specific 'factor' for each utility was "necessary to yield a fair representation of the historical values required by AB 1890." (D.96-12-028, mimeo., p. 14.)
The original transition formula values adopted in D.96-12-028 were based on regressions of 1994 - 1995 SRAC prices versus border gas prices, and were driven entirely by changes in border gas prices. The SCE and SDG&E formulas used 100% Topock border prices; the PG&E formula reflected a 50/50 mix of Malin and Topock border prices.
The Transition Formula was expected to be used for a relatively short "transition period" until energy payments could be based on California PX prices. (See Pub. Util. Code, § 390(c).) The PX ceased market operations at the end of January 2001, so the Transition Formula remains in use. At the time of the PX demise, the Transition Formula for each utility had remained unchanged for four years.
In the wake of the 2000-2001 energy crisis, and in response to numerous SCE requests, the Commission modified the Transition Formula for SCE in D.01-03-067, although PG&E and SDG&E remained on the original Transition Formula approved in D.96-12-028. D.01-03-067 also replaced the Topock gas price index in the SRAC energy formula for each utility with a Malin index plus an off-system transportation rate.49 The SRAC energy Transition Formula adopted in D.96-12-028 is shown here:
Original SRAC Transition Formula
Pn = [ Pb + Pb x [(GPn-GPb)/ GPb] x (utility factor) ] x TOU
Pn = calculated SRAC energy price, cents/kWh
Pb = starting energy price (as required by Section 390), cents/kWh
GPn = current gas price, $/MMBtu
GPb = starting gas border price (as required by Section 390), $/MMBtu
Utility Factor for SCE = .7067 (unitless -- all units cancel out)
TOU = time of use multiplier (no units)
In D.01-03-067, the Commission modified SCE's Transition Formula by replacing SCE's fixed factor of 0.7067 with a `floating' factor that changes in value from month to month. The `floating' factor is actually a formula unto itself, employing an updated burnertip gas price, an IER, and an O&M adder. Shown below, first, is the `floating' factor adopted in D.01-03-067 at page 6 (with the omitted division line now included). Note that all the units cancel out rendering the factor unitless:
SCE Factor = [IER x (GPn + GTn)/10,000] + O&M - Pb
Pb x (GPn - GPb)/GPb
Sample Factor Calculation for November 2001 for SCE
0.4932 = [9140 (3.3439 + 0.2777)/10,000] + 0.2 - 2.0808
2.0808 (3.3439- 1.3975)/1.3975
GTn = intrastate transportation costs, $/MMBtu
IER = Incremental Energy Rate (utility heat rate) Btu/kWh
O&M = operations and maintenance costs, Cents/kWh
10,000 = [$1/100 Cents] x [1,000,000 Btu/MMBtu]
SCE's Modified Formula
===============utility factor============
Pn = Pb + [Pb x (GPn-GPb)/GPb] x [IER x (GPn + GTn)/10,000 ] + O&M - Pb
[Pb x (GPn - GPb)/GPb]
When the floating factor is inserted into the Transition Formula, a number of the components algebraically cancel out, resulting in the following:
Pn = (IER x (GPn + GTn)/10,000 ) + O&M
Sample Calculation for April 2006 for SCE
PApril-2006 = 6 .4597 cents/kWh = (9140 (6.3205 + 0.5282)/10,000) + 0.2
The IER, a heat rate in British thermal unit (Btu) per kWh, is intended to reflect the efficiency with which the IOUs could obtain the energy that they would have to produce (or purchase) "but for" QF production. IERs reflect the fact that fossil generation is not always on the margin. IERs increase as demand increases, as less efficient plants are needed to supply the marginal kWhs.
Traditionally, IERs have been calculated through complex production cost computer modeling of the IOU systems both with and without QFs, and have generated issues that have been difficult, at best, for the Commission to adjudicate.
The general formula for the IER has been:
IER = [ (QFOUT Costs - QFIN Costs) / QF Energy ] / Avoided Fuel Cost
The IER is expressed in units of Btu per kWh, as follows:
IER = [ (Costs in $) / (QF Energy in kWh) ] / Fuel Costs in $ per Btu
= [( $ / kWh ) / ( $ / Btu) ] = Btu / kWh
IERs were originally determined in general rate cases. In the late 1980s, the Commission moved IER issues to annual Energy Cost Adjustment Clause (ECAC) cases. Due to the complexity of IER issues, the IOUs, DRA, and QF parties tended to settle IER issues outside of the hearing room, with the Commission reviewing and approving those agreements.
Commission-adopted IERs have been in the range of 9,000 to 10,000 Btu per kWh over the two decades of the California QF Program. The SRAC transition formula factors approved in D.96-12-028 are based on regressions of 1994 - 1995 SRAC prices, and thus reflect 1994 - 1995 IERs.
The Operation and Maintenance (O&M) component of the Transition Formula is designed to capture the IOUs' generating costs (except for fuel and capital costs) that vary with the amount of power purchased from QFs. Historically, these costs have been limited to consumables such as chemicals and lubricants and to O&M costs that vary with the amount of power produced in IOU-owned gas-fired power plants (such as the costs of certain maintenance activities that are scheduled based on plants' production or operating hours, as well as the O&M costs avoided if QF power allows an IOU to place older units on standby). Variable generating costs today also include air emission credit costs and periodic costs to replace expensive catalysts in air emission control equipment.
Commission-adopted O&M adders have ranged from $1 to $3 per megawatt hour (MWh). D.01-03-067 adopted an O&M adder of $2 per MWh for SCE.
4.2. Proposals for SRAC Energy Pricing
Five parties (PG&E, SCE, SDG&E, TURN, and CCC) have proposed SRAC energy pricing methodologies that utilize implied market heat rate (IMHR) figures derived from Day-Ahead power price indices at NP15/SP15 and spot bid week natural gas indices at border trading points or at the burner-tip. For example, the IMHR for $56.00/MWh power at NP15 or SP15, and $7.00/MMBtu gas at the border is $56.00/MWh ÷ $7.00/MMBtu = 8,000 Btu/kWh. While the respective PG&E, SCE, SDG&E and TURN proposals differ in overall mechanics, they all use unadjusted IMHRs. The CCC's proposal derives IMHRs in a manner similar to SDG&E and SCE, except that CCC uses forward prices as opposed to historical prices. CCC then grosses up the result with a proposed adjustment factor to reflect an estimated aggregate value of QF generation.
In contrast, two parties (CAC/EPUC and IEP) recommend keeping PG&E's existing Transition Formula. IEP also recommends keeping SCE's existing Modified Formula. However, CAC/EPUC recommends moving SCE from the Modified Formula adopted in D.01-03-067, back to the original Transition Formula approved in D.96-12-028. The QF parties generally argue that there are many problems with the existing Day-Ahead market that prevent Day-Ahead prices from accurately reflecting the utility avoided cost. In particular, the QF parties explain that the Day-Ahead market is very small and the utilities' transactions in the market represent only about 4% of their load. The QF Parties also complain that the Day-Ahead market price doesn't reflect the cost of higher priced units that are dispatched through reliability-must-run (RMR) contracts or CAISO must-offer waiver denial (MOWD) provisions. The QF parties are also concerned that, since the utilities are the dominant participant in the markets, they have the ability to artificially depress market prices.
It should be noted that most parties recommend the use of burner-tip gas prices in their proposed SRAC energy equations, while PG&E recommends the use of a border gas price, and TURN recommends the use of the PG&E City Gate trading point price. An illustration of these gas price differences appears in Table 2, Party Positions on SRAC Energy Pricing.50 Although these prices, and their relative differences, will fluctuate over time, it is imperative to clearly identify the proposed price inputs for comparison purposes.
SCE proposes that "the Commission abandon the [Transition Formula] methodology adopted in D.96-12-028 in favor of an approach that compares monthly electricity prices in the wholesale electricity markets to natural gas prices to compute an implied market heat rate..." (Exhibit 1, p. 61.) It also recommends that we adopt a heat rate pricing methodology that compares SP15 Day-Ahead prices to natural gas prices to compute an implied market heat rate and multiplies that IMHR by a monthly bid week natural gas price. SCE's SRAC energy pricing proposal functions essentially the same as the Modified Formula. SCE proposes that the Commission calculate SRAC energy each month using the following formula:
SCE's Proposed SRAC Energy Formula
Where:
A = Monthly average of daily Day-Ahead SP15 prices (DJ / ICE / MWD), where DJ = Dow Jones, ICE = Intercontinental Exchange, and MWD = Megawatt Daily.
B = Variable O&M ($2.00/MWh)
C = Topock bid week gas price average (NGI, NGW, Btu Daily Gas Wire)
D = So Cal Gas Intrastate Transportation51
E = Burnertip Gas Price (C + D) in $/MMBtu
HRm = Monthly Heat Rate [ ( A - B ) / E ] * 1,000 Btu/kWh
HRCap = 9,864 Btu/kWh
HRFloor = 5,864 Btu/kWh
HRc = Collared Monthly Heat Rate ( HRFloor <= HRm <= HRCap )
HR12mthMAvg = 12 month Moving Average of Capped Monthly Heat Rates
= [ Σ (HRc1 ... HRc12) ] / 12 months]
IER = HR12mthMAvg
According to SCE, "using this approach, the IER for December 2004 would have been 7,837 Btu/kWh.52 Over the three-year period from August 2002 through July 2005, the average implied market heat rate was 7,864 Btu/kWh. (Id.)
Given the fact that SCE's Modified Formula will yield the same SRAC energy result as SCE's Proposed SRAC Energy Formula (when the same inputs are used), the actual difference between the two is in the development of the IER heat rate. The IER in SCE's Modified Formula is a heat rate that is tied to the 1994-1995 time period and was adopted in D.96-12-028, whereas the heat rate in SCE's Proposed SRAC energy formula is derived from a twelve-month rolling average of historical Day-Ahead market price data with a "collar" around the possible IER values to provide a cap and a floor for possible IER values. SCE states that its proposed SRAC energy formula is designed "to reflect wholesale market conditions..., includes a `trigger' that provides for expedited review of the methodology in the event of persistent and significant changes in the SP15 market relative to gas prices, and ... in order to mute volatility and to account for seasonality, SCE's proposal employs rolling averages of market data and collars on permissible monthly data points." (Exhibit 1, pp. 60-61.)
SCE states that it developed the "collar" for the implied market heat rate by reviewing the monthly implied market heat rates in SP15 from August 2002 through July 2005. According to SCE, during this period, 100% of the implied market heat rates in SP15 fell within the range of 5,864 to 9,864 Btu/kWh, therefore, SCE recommends that "collars" of these numbers be adopted and if the implied market heat rate hits or exceeds the collars in four successive months, any stakeholder may seek modification of the SRAC formula.
With regard to the gas component in the SRAC formula, SCE "proposes that the Commission adopt a Topock burnertip price for natural gas in lieu of the Malin burnertip price currently used in the transition formula." (Id., p. 64.) According to SCE, adopting a Topock burnertip price would result in SRAC energy prices that are "approximately 17% lower than the price produced using SCE's current SRAC transition formula:
Using the December 2004 IER of 7,837 Btu/kWh shown in Figure 10 and replacing Malin with Topock yields an illustrative SRAC price in December 2004 of 5.5640 cents/kWh as compared to SCE's posted SRAC of 6.6827 cents/kWh. In this example, SCE's formula results in a price approximately 17 percent lower than the price produced using SCE's current SRAC transition formula. (Id., p. 66.)
PG&E proposes to update its original SRAC Transition Formula to account for current market conditions. More specifically, PG&E proposes to "update the `factor' in the SRAC energy formula so that SRAC energy prices for existing QFs approximate NP15 day-ahead prices" (Exhibit 28, p. ES-2).
PG&E proposes to revise the factors such that when the current natural gas border index price is put into the Transition Formula, the resulting SRAC energy price will reflect the monthly NP15 Day-Ahead price.
To derive the revised factors, PG&E performed a regression analysis using bid-week border gas index prices53 and monthly NP15 Day-Ahead prices.
PG&E provides this overview and other observations:
...the transition formula includes gas "factors" that reflects the relationship between the historical border gas and SRAC prices. The Commission initially derived the factors from a regression analysis.54 The Commission has previously confirmed that it has authority to modify a utility's transition formula factor to arrive at a price that better reflects a utility's avoided cost and complies with PURPA. Four years after originally adopting a factor in Southern California Edison's (SCE) transition formula, the Commission modified the factor, at SCE's request, to lower SCE's SRAC prices.55 QF groups petitioned for review of Decision 01-03-067, claiming that revising SCE's factor violated Section 390(b). The California Court of Appeal affirmed the Commission's decision to adjust the SCE's transition formula factor to comply with PURPA's avoided cost cap.56 The Court of Appeal expressly rejected the QFs' contentions that the Commission lacked authority to revise the factor to adjust to changes in the market. (Exhibit 28, p. 3-2.)
PG&E further notes that "under PG&E's proposal, the starting energy and border gas prices used in the formula remain unchanged.57 The transition formula factors would be modified, however, to yield energy prices that reflect PG&E's avoided costs." (Exhibit 28, p. 3-5.) Thus, PG&E's formula would continue to be the original Transition Formula:
Pn = Pb + Pb x [(GPn-GPb)/ GPb] x (utility factor) x TOU, as already described in detail above.
As stated, it is PG&E's goal to calibrate its SRAC Transition Formula using revised utility factors (one for summer and one for winter) so that "SRAC energy prices for existing QFs approximate NP15 day-ahead prices." PG&E derived its proposed factors through regression analysis, the same method used to compute the original Transition Formula factors in D.96-12-028, however, PG&E's proposal would base its new factor on the correlation between NP15 Day-Ahead prices and border gas prices instead of the original correlation between pre-1996 SRAC energy prices and border gas prices. PG&E compared factors from several different time periods and compared the revenues earned under the SRAC Transition Formula with each set of revised factors with the revenues that would have been earned using monthly NP15 Day-Ahead prices. PG&E then selected the factors which most closely matched the total revenues that would have been earned by a QF with a price based on monthly NP15 Day-Ahead prices. To ensure that the revised factors continue to yield SRAC energy prices that closely track NP15 Day-Ahead prices, PG&E notes that the Commission must establish a process to periodically compare SRAC energy prices with corresponding NP15 Day-Ahead prices and provide for further updates of the revised factors as needed, either monthly, yearly, or seasonally.
PG&E states that the NP15 Day-Ahead price is very transparent, because there are at least three different providers of NP15 indices approved by FERC. PG&E also notes that the Day-Ahead price is used as the benchmark price for settling financial and physical contracts for trading hubs across the United States, including NP15 energy. As an example, PG&E states that the New York Mercantile Exchange (NYMEX) Dow Jones NP15 Electricity Price Index Swap Contract (on-peak) is settled by cash payment based on the contract price and the so-called "Floating Price" which is the arithmetic average of the Dow Jones NP15 Day-Ahead on-peak indices for the contract month.58
In this rulemaking, SDG&E has requested that the Commission approve "the same formulation of the variable factor as the Commission adopted for SCE in D.01-03-067." More precisely, SDG&E has requested to be put on the Modified Formula:
"Since the Commission stated in D.01-03-067 that all elements of the transition formula should be updated, SDG&E is proposing to use the same formulation of the variable factor as the Commission adopted for SCE in order to update the IER, intrastate gas transportation rate, and the variable O&M in this proceeding." (Exhibit 85, pp. 6-7.)
SDG&E's proposal for SRAC pricing is based on the Transition Formula, as required by § 390(b), but converts the fixed factor to a formula, consistent with Commission precedent and policy. SDG&E recommends determining an O&M adder and then deriving an IER from two years of historical Day-Ahead market prices using the O&M adder and historical natural gas prices. SDG&E proposes to use daily market prices at SP15 less O&M divided by burnertip priced gas for the day. The daily values would be averaged over two years to create a forecast market IER for the year. SDG&E suggests updating the various components of the SRAC formula for 2006 along with an automatic recalculation process for subsequent years. SDG&E's proposed 2006 IER would be 7,782 with a $2.60 /MWh variable O&M adder. The variable O&M adder would be updated annually for inflation while the IER would be updated based on the most recent two-year average of historical information on gas prices and SP15 prices. SDG&E recommends an as-available capacity price of $68.93 in 2006, to be adjusted in subsequent years, depending on resource adequacy and already acquired reserves.
TURN recommends basing SRAC payments on actual electricity market prices, using publicly available on-peak and off-peak pricing date from the Intercontinental Exchange (ICE) or Dow Jones until the CAISO's MRTU project becomes operational, at which point hourly prices from the CAISO's day-ahead electricity market should be used. TURN believes that this approach would ensure that SRAC payments would be equal to the price of energy in the market, ratepayers would not be subject to systematic overpayments, and the Commission would be relieved of the responsibility to adjudicate an administratively determined cost.
In its reply brief, TURN provides us with another option in which we would retain, as a temporary measure, the SRAC Transition Formula with a new heat rate developed from real electricity market price data, until CAISO's MRTU reforms are implemented. This approach relies on forward market prices for electricity and natural gas, as suggested by CCC witness Beach, but would look no more than one year forward. According to TURN, on a yearly basis, the utilities would compile publicly reported forward market prices for electricity and natural gas for the upcoming year. The average forward price of electricity would be divided by the average forward price of gas to derive the incremental heat rate, which would be fixed for the following year (with appropriate peak/off-peak and seasonal differentiation.) The SRAC would be set prior to the beginning of each month, using the month-ahead price of gas. No O&M or other adders should be used, because the forward market price already reflects all the underlying components of the price of electricity. TURN recommends this alternative approach as a temporary measure until MRTU is implemented and robust locational day-ahead market prices are available from the CAISO. The use of forward market prices would eliminate the concerns that QFs raise regarding the use of day-ahead market prices to determine SRAC.
For short-term purchases, DRA recommends a one-year contract similar to an SO1 contract, but with updated terms and conditions. DRA recommends basing SRAC on market prices and supports PG&E's proposal to replace its Transition Formula by revising the fixed factor. DRA notes, however, that setting SRAC prices to market prices would expose ratepayers to price fluctuations of the market, should the market fail to function correctly, but suggests that the Commission could mitigate this risk exposure by placing a cap on SRAC prices as recommended by TURN.
Alone among the QF parties, CCC recommends that the Commission revise SRAC energy prices for all three utilities using the Modified Transition Formula adopted in D.01-03-067, updated to reflect current market conditions. CCC states that updating the Modified Transition Formula is reasonable because it complies with § 390(b), it can be updated periodically as necessary to keep pace with changing market conditions, and it is flexible enough to accommodate all of the SRAC energy pricing proposals.
CCC recommends updating the IER as well as gas prices and the variable O&M adder. CCC states that IERs for 2006 -2010 can be estimated using the market heat rates, or spark spreads, reflected in forward market prices for natural gas and electricity in the California market, dividing the forward electric prices (in $ per MWh ) by the forward gas price (in $ per MMBtu) to yield a heat rate (in Btu/KWh). However, CCC states that because forward heat rates do not include the least efficient generators, they do not reflect the utilities' full avoided cost and must be adjusted to meet PURPA requirements. CCC states that market heat rates must be adjusted to reflect the absence of many of the least efficient generators as well as the price elasticity benefits of QFs in lowering market-clearing prices.
In support of its position, the CCC compared SRAC energy prices to CAISO Competitive Market Clearing Price (CMCP) values for the years 2002 through 2004. CCC states that the CAISO's CMCP represents "an estimate of the market-clearing price in a perfectly competitive market for California's energy."59 CCC explains that the CAISO uses all in-state thermal generation priced at each unit's full lead heat rate times the daily burnertip gas prices, plus a variable O&M adder. The CCC reports that the CMCP values were very close to SRAC values in 2002 through 2004. The CCC also compared the heat rates implicit in the CMCP data, using daily burnertip gas prices and an assumed variable O&M adder of $2.50 per MWh to the heat rates implicit in a weighted average of posted SRAC energy prices, using bid-week delivered natural gas price indices and an O&M adder of $3.00 per MWh.
CCC's proposal derives an updated implied market heat rate using forward day-ahead electricity market prices, divided by forward gas prices. The forward electricity prices are developed using publicly-available NYMEX data for monthly on-and off-peak NP15 and SP15 electric forward prices for 2006-2007 and annual broker quotes for 2008-2011. The forward gas prices would use NYMEX gas futures prices, NYMEX Clearport basis differentials for the PG&E city-gate and the southern California border, plus intrastate transportation costs on the PG&E and Southern California Gas Company (SoCalGas) systems. CCC then applies an elasticity factor to the forward market heat rates to develop IERs that reflect the aggregate value of QF generation. The elasticity factor used by CCC is similar to the elasticity factor developed and used by Energy and Environmental Economics, Inc. (E3)60 for use in avoided cost in Phase 1 of R.04-04-025. The CCC's formula for the IER is as follows: IER = FMHR × (1 + Σ ×RNS%) where FMHR is the Full Market Heat Rate, Σ represents price elasticity, and RNS is the utility's residual net short.
CCC believes that using forward market prices is superior to using historical prices because forward prices reflect actual transactions and anticipated market conditions during the time the QFs will actually deliver energy to the utilities. According to CCC, the use of forward market prices is especially important when market prices are trending upwards and historical markets have been affected by the market flaws and gaming behavior that the QF parties argue has existed and currently exists.
CAC/EPUC and IEP are opposed to pricing SRAC energy at market levels and support a continued reliance on a largely administratively determined formula that requires periodic adjustment via protracted litigation. They argue that the Commission should reject the utilities' SRAC energy pricing proposals and continue to set monthly SRAC energy prices using the Section 390(b) formula. They advocate changes to the capacity payments, as well as a change to SCE's factor, but no change to the SRAC energy pricing formula for SDG&E and PG&E. Their primary objections are summarized briefly below.
First, CAC/EPUC and IEP argue that the current SRAC energy price formula fairly reflects the short-run avoided costs of the utilities and should be retained. In their opinion, not only is there no basis in the record on which to find that the current SRAC formula should be replaced, but they contend that it is impossible to verify the proposed market price proxies without actual utility data, and conclude that the current SRAC energy price formula should not be changed.
Second, CAC/EPUC and IEP argue that the utility proposals are unlawful, both because they do not accurately reflect the price "at the time of delivery" and because they do not represent the market clearing price that would result in the absence of QF generation. Each of the QF parties argues that a fundamental flaw in the assumption that market prices reflect full avoided costs for utilities is the assumption that the market price remains unchanged with or without the QF capacity. The CCC agrees with the other QF parties in this respect, and includes an "elasticity adder" in its long-run proposal to reflect the aggregate impacts of QF generation. CAC/EPUC testified that prices in a well-functioning market can reveal the value of the marginal unit of electricity, but note that to the extent energy payments to QFs depend on the cost of energy avoided as a result of the aggregate value of energy provided by multiple QFs, the market price in even a well-functioning market will not reflect this higher cost necessary to replace QF-provided energy in the face of an upward-sloping energy supply curve. (Exhibit 42, p. 7, fn. 8.)
Third, CAC/EPUC/IEP argue that the use of a day-ahead market price cannot represent the costs a utility would incur "but for" QF purchases because it does not include utility costs incurred outside that market. They argue that the market price benchmarks proposed by the utilities are not liquid, do not reflect all sources available in the market, are artificially depressed, and are extremely subject to market manipulation due to the monopsony power of the utilities.
In particular, CAC/EPUC and IEP argue that the day-ahead market price is "artificially depressed" because it does not account for the impact of the CAISO dispatches of energy from RMR contracts and FERC MOWD units (also known as FERC MOO). These units trade "out-of-market," meaning the cost of power from those units is not reflected in the price of energy that actually trades in the NP15 and SP15 day-ahead markets. IEP testifies that "...as a result of these out-of market dispatch actions, the CAISO adds a significant supply to the market place that is generally not eligible to set market clearing prices. This results in observed prices that do not accurately reflect the actual generation supply resources that are dispatched to meet demand. The exclusion of the resources from the price-setting process significantly lowers the market-clearing prices." (Exhibit 42, pp. 17-18.)
They point out that underscheduling and infeasible scheduling can result in significant volumes of out-of-market energy to replace or make up for energy not purchased and scheduled by the scheduling coordinators. IEP suggests that the utilities would have a huge incentive to manipulate the market if QF energy payments were tied to day-ahead prices. "Suppression of the day-ahead price of only $1 per MWh... would result in $48.5 million dollars of savings to the utilities if SRAC payments to QFs were based on this price." (Exhibit 42, p. 34.) In addition IEP notes that "strategic generation or dispatch would entail the production of energy at times either to replace energy that would be purchased in the short-run energy market, or to add supply to short-run market to suppress prices. Strategic behavior could take the form of substituting higher cost retained generation or purchased energy for energy that would otherwise be purchased in the market."
CAC/EPUC agrees, and claims that the utilities chronically underschedule in the NP15 and SP15 day-ahead markets. They note that SCE witness Silsbee acknowledges that "Some parties may have been underscheduling in order to take advantage of the lower prices in the real time market ... [and] amendment 72 was designed to prevent that kind of gaming opportunity by requiring accurate scheduling in the day-ahead market." (RT 19, p. 2,698.) According to IEP, SCE's purchases from the SP15 Day-Ahead market are never more than 3% of its total supplies, while SCE's QF purchases are typically 28% to 35% of SCE's total supplies.
The QF parties state that the current design of the California electric market provides the opportunity and economic incentive for utilities to submit day-ahead schedules with inadequate quantities of energy and infeasible energy, resulting in load that cannot be effectively served by day-ahead scheduled energy even if it were balanced when scheduled. The CAISO has been forced to procure contracts for thousands of megawatts of higher-cost generation capacity, out of market on a day-ahead, hour-ahead , and real-time basis to make up for shortfalls of usable scheduled energy. The cost of these out-of-market energy purchases is socialized and allocated in various ways by the CAISO among multiple parties. The QF parties note that anticipated changes by the CAISO to move to a system of locational prices with balanced and feasible day-ahead commitments may correct some of these problems and provide usable signals of marginal energy costs. However, at this point, they maintain, the day-ahead market prices of SP15 and NP15 will not reflect marginal costs.
According to CAC/EPUC, to the extent high-cost energy (through the RMR or MOWD) is effectively prepaid through a long-term contract or converted into a non-energy charge through the socialization of cost through a central purchasing entity, the true marginal price of energy may never actually be observed. Under this theory, all existing utility contracts and agreements must be participating in the market in order to determine the true marginal price of energy. Indeed, IEP argues that unless and until the CAISO and other California electricity markets can and do meet all demand and supply with energy that is in the market, California energy markets cannot be used to establish prices reflective of utilities' marginal costs of generation, or SRAC.
CAC/EPUC explains that they have analyzed SRAC energy prices using four approaches:
(a) QF-in/QF-out computer simulation similar to those performed in past ECAC proceedings, but without some confidential data provided by the utilities on loads and resources,
(b) an analysis of forward market energy prices (without some confidential utility data),
(c) an analysis of CAISO FERC Form 714 data, and
(d) an examination of CAISO Market Surveillance data.
CAC/EPUC report that prospective IERs resulting from the four analyses range from 9,067 Btu/kWh to 10,689 for SCE and 9,177 Btu/kWh to 10,730 Btu/kWh for PG&E. CAC/EPUC maintain that the results of these analyses demonstrate that a continuation of the current § 390(b) Transition Formula is reasonable.
The Renewables Coalition recommends that the Commission adopt an as-available contract option for QFs. According to the Renewables Coalition, the contract term should be for up to 15 years and should be terminable by the QF upon 30 days prior notice by the QF. The Commission should update the utilities' as-available capacity prices to equal the full cost of a combustion turbine, or $110 per kW-year61 and escalate this price annually to reflect changes in the Consumer Price Index.
The Renewables Coalition maintains that capacity is very tight on SCE's system and the Energy Reliability Index (ERI) should be higher than 10%, at which it is currently set. Furthermore, the Renewables Coalition argues that updating as-available capacity prices is necessary to ensure that QFs with expired or expiring long-term contracts stay online.
4.3. Should the SRAC Energy Formula
be Updated?
As a threshold issue in this proceeding, we must first determine whether it is necessary to update or revise SRAC pricing to ensure that it continues to represent the utilities' short-run avoided cost. The current SRAC formulae are dated, which is not inherently problematic; nonetheless, certain components of the formulae contain hard-wired values that are based on pre-electric restructuring markets and utility portfolios. For example, PG&E and SDG&E have been on the SRAC energy Transition Formula since it was originally established in 1996 per D.96-12-028, and include unchanged IERs and utility factors. The latter utility factors are a result of "regression [analysis] describing the historical relationship between changes in border gas costs and ... [an IOU's] calculated avoided cost" (D.01-03-067, p. 5). The regressions were based on 1994-1995 data. With regard to SCE, the utility was on the Transition Formula until 2001 when it was effectively replaced by the Modified Transition Formula per D.01-03-067. Although SCE's fixed factor was replaced by a dynamic factor that changes monthly, the SCE SRAC formula still contains an original 1996 IER.
In D.04-07-037, we clarified our observations and intent on the issue of SRAC in Ordering Paragraph 1:
1. The discussion under heading (2) "Revision of SRAC Prices" on pages 56 to 58 of D.03-12-062 is deleted and replaced with the following discussion:
All three utilities contend that revision of the current SRAC methodologies for determining QF energy and capacity payments is needed. For many years now, SRAC has been approximated through time-differentiated energy prices (set once a month) and time differentiated capacity prices (set annually). There is evidence on the record in this proceeding for some time periods the current SRAC energy pricing methodology has yielded prices in excess of spot market prices.
Although, the evidence presented here raises questions and supports the need to revisit SRAC pricing system, the utilities' have not demonstrated the SRAC formula is inadequate or that it exceeds avoided costs in violation of PURPA. Moreover, this procurement proceeding is not the appropriate forum to review the SRAC pricing formula. The current SRAC formula was considered and adopted in D.01-03-067 and D.02-02-028, and this formula was upheld on appeal. (Southern California Edison Co. v. Public Utilities Comm. (2002) 101 Cal. App. 4th 982.)
The concern exists, however, that the SRAC pricing formula may need to be revised in light of the current energy market. Therefore, the Commission should carefully consider how to modify the SRAC methodology and whether to seek legislative changes to Pub. Util. Code § 390. Because it is important that current methodologies to establish SRAC be critically evaluated and modified where necessary, we are directing Commission staff to immediately begin work on a draft OIR that will examine and propose appropriate modifications to the SRAC methodology." (D.04-07-037, mimeo., Ordering Paragraph 1.)
Energy pricing, under both existing long-term QF standard offer contracts and the revised Standard Offer 1 (RSO1) five-year contract extensions mandated by the Commission, is based on the current SRAC Transition Formula (unless a different pricing term is provided in the contract between the utility and the QF, such as the five-year fixed energy price amendments entered into by many QFs during 2001). Currently, as noted above, the SRAC Transition Formula is based in large part on the current cost of gas times an assumed heat rate or IER. Under the formula, if the assumed heat rate (existing IER) is greater than the utility's incremental heat rate, then the SRAC formula results in a price that exceeds avoided cost (all other factors being equal).
PG&E, SCE, SDG&E, DRA, and TURN argue that the IERs adopted in D.96-12-028, for PG&E and SDG&E, and D.01-03-067 for SCE, currently exceed actual market heat rates, resulting in SRAC energy payments that exceed the utilities' short-run avoided cost of energy. PG&E, SCE, SDG&E, DRA, and TURN assert that day-ahead market prices more accurately reflect the utilities' avoided cost and should be used to determine SRAC energy payments.
All parties seem to agree that generally, in a well-functioning market, the price of energy established in the market is equal to the marginal cost of the incremental unit of energy where the quantity of energy supplied and the quantity of energy demanded at that price are equal. The market-clearing price for energy is then determined by the bid for and marginal cost of the last unit of energy in the market. They disagree, however, on whether the market clearing price accurately reflects the utilities' avoided cost and whether the market-clearing prices that are available are the result of a well-functioning market.
As argued by the QF parties, SRAC energy prices should exceed current wholesale market prices. Specifically, they maintain that market prices must be adjusted to reflect the estimated increase in the market clearing price that would result from removal of a block of QF generation and that the market price indices recommended by the utilities are not sufficiently liquid or competitive enough to represent the utilities' avoided cost. In addition, the QF parties claim that the utilities have not provided sufficient data to assess whether SRAC prices exceed actual avoided costs.
In support of their argument, the QF parties state that they analyzed SRAC energy prices using several approaches and that each demonstrated that the SRAC Transition Formula results in SRAC energy payments that are in line with, or lower than, current avoided costs. CCC, CAC/EPUC, and IEP each present comparisons of SRAC energy prices to the CMCP. First, CCC, CAC/EPUC, and IEP maintain that the heat rates implicit in the CAISO CMCP values demonstrate that the SRAC Transition Formula continues to reflect avoided cost "when viewed from the perspective of a broad, competitive market that includes the thermal generation that is operated outside of today's limited wholesale market." (Exhibit 102, p. 31.) The CCC estimates a 2002 through 2004 average CMCP implied heat rate of 9,449 Btu/kWh and compares it to a statewide average SRAC heat rate of 9,776 Btu/kWh. CAC/EPUC performed a similar calculation, then extrapolated the CAISO 2003 and 2004 CMCPs to calculate QF-out heat rates using the price differential between QF-in and QF-out electricity prices using the AURORA production cost model. CAC/EPUC calculates that PG&E's current SRAC energy price understates avoided costs by approximately 1.3% in 2003 and by 17% in 2004. (Exhibit 134, pp. 71-74.) IEP also compares the CMCP with SRAC energy prices from 2002 and 2003. IEP contends that SRAC energy prices in these years were about 6.8 % higher than the CMCP. (Exhibit 95, pp. 58-62.)
Thus, all three QF parties maintain that the current SRAC energy prices accurately reflect avoided costs for this period. However, there are several problems with the various QF analyses. For example, according to the CAISO Department of Market Analysis, the resources used to establish the real-time CMCP:
[c]ompares real-time market prices to estimates of system marginal costs. The analysis only includes resources that were actually dispatched for real time energy by the CAISO, therefore it excludes resources or certain portions of resources that were unable to respond to dispatch instructions for reasons such as physical operating constraints.62
Moreover, the CCC and IEP did not state that the CAISO, in its 2004 report to FERC, itself noted that the CMCP or "system lambda" data reported to FERC was not actually system lambda data:
The CAISO operates its control area through forward energy scheduled and operation of an imbalance energy market, plus reserve/ancillary service markets (to cover generation and transmission contingencies). Suppliers provide the CAISO real-time energy bids that are used by the CAISO to match supply and demand every 5 minutes in a least cost manner. Because energy bids do not necessarily reflect system marginal costs, the CAISO does not have true system lambda information. Therefore, the CAISO will not be submitting system lambda data as part of this FERC 714 filing. (In previous years, the CAISO had provided a formulated estimation of system lambda data; upon closer review of Form 714 instructions, this was not appropriate.) Though not a true system lambda, historical real-time energy price information is available on the CAISO's OASIS website at http://oasis.caiso.com, under Real Time Information.63
In addition, as explained by SCE witness Silsbee, the competitive market clearing price data utilized by QF parties is only for purchases by the CAISO (the incremental, or "INC" market) and does not include sales by the CAISO (the decremental, or "DEC" market). The incremental CMCP represents the cost of increasing generation from thermal units designated by the CAISO. The incremental CMCP does not include units that were asked to reduce generation by the CAISO in the "DEC" market. PG&E reports that the incremental market is roughly 2.5 times that of the decremental CMCP.64 As an example, SCE calculates that market performance based on a weighted average of the two markets resulted in an average IER of less than 6,000, well below the IER of 9,140 currently embedded in the SRAC Transition Formula.65
According to CAC/EPUC, the only way to accurately estimate utility avoided cost is to perform a QF-in/QF-out production cost simulation to calculate the cost that would have been incurred by the utility in lieu of QF generation. (Exhibit 134, p. 48.) This complex computer simulation would calculate the system costs based on the economic dispatch of the available generating resources. Two production cost simulations (or runs) would be performed: (1) with the QFs included in the utility's portfolio (QF-in) and (2) without the QFs (QF-out). The difference in cost between these two production costs simulations provides an estimate of the utilities production costs avoided by QFs' provisions of short run energy. The difference is then divided by the incremental fuel cost resulting in an IER.66 In other words, in the QF-out run, the utility would have to either generate more power or buy more power from the market. The cost of this incremental amount of power and the amount of this incremental power can be expressed in $/kWh. Dividing this $/kWh by the gas price in $/MMBtu leaves a heat rate figure in Btu/kWh. Under the QF-in/QF-out approach, the IER is a measure of the thermal efficiency for the entire system that would have been required to serve the load absent the block of QF resources. The O&M cost is then independently determined and added to the IER for an SRAC energy payment.
CAC/EPUC claim that a market price does not reflect a utility avoided cost of energy unless "the QF capacity normally supplied to the utility is assumed to be unavailable." (Exhibit 134, p. 41.) The QF parties rely on FERC's regulations referring to the "aggregate value of energy and capacity from qualifying facilities on the electric utility's system" to support their argument that QFs should be treated as a block in determining a utility's avoided cost. (See Exhibit 102, p. 4.) The language cited by CCC and others appears in a subsection of the regulations entitled "Factors affecting rates for purchases" (18 CFR 304(e)). This subsection lists a number of factors that should be taken into consideration "to the extent practicable." One such factor is "The individual and aggregate value of energy and capacity from qualifying facilities on the electric utility's system. (18 CFR 304(e)(2) (vi).)
Given that the majority of the utilities' resource procurement efforts involve competitive solicitations, we agree with the utilities and TURN that it is neither reasonable nor practical to base avoided costs solely on a "QF-out," or "aggregate value" methodology. The "aggregate value" is only one of several factors that FERC suggests should be considered and is not determinative. One of the other factors to be considered is the "individual" value of QFs. Furthermore, as PG&E points out, to read this section as supporting the treatment of QFs in the aggregate is inconsistent with many of the other factors listed in the section, which refer to the characteristics and capabilities of an individual QF. (PG&E Opening Brief, p. 20.)67
The utilities point out that in comparing the QF-in or marginal cost pricing approach, and the QF-in/QF-out or incremental approach, the Commission found that while "the QF-in/QF-out, method meets the PURPA requirements for QF pricing, to the extent that changes in ECAC and possibly other developments create a more competitive environment and move this industry closer to a true spot market, it is appropriate and consistent with PURPA to reconsider marginal cost energy cost pricing for short-run QFs." (27 CPUC 2d p. 576, D.88-03-079.)
As we stated in D.92-01-018, using QF-in/QF-out methodology involves hundreds of modeling assumptions and forecasts.68 We concluded at that time that the IER is not just a "somewhat" artificial concept, but a "totally" artificial concept.69 This conclusion has become even more-true as QF generation has become a larger percentage of the utilities' resource portfolios. The continuing long-term obligations to thousands of megawatts of QF power mean that QFs as a block will never be "out."
Furthermore, we find no right through any contract term or fair market expectation that the Commission must adopt the QF-in/QF-out approach. As TURN points out, "[N]o supplier anywhere can expect to capture the higher price that would have prevailed had that supplier not offered its product to the market." (TURN Opening Brief, p. 2.) Even CAC/EPUC admit that, in light of electric restructuring, the utilities and QF parties developed a simplified SRAC energy pricing approach that was ultimately adopted by the Commission in D.96-12-028. Although this simplified method initially utilized IERs that were based on 1994-1995 data developed using a QF-in/QF-out method, a revision to this method that does not rely on the pre-1996 IERs was adopted for SCE by this Commission in D.01-03-067 and approved by the Court. As CAC/EPUC correctly note, changing from a fixed factor to a dynamic factor through the use of an algebraic expression in D.01-03-067 results in a formula without a "starting energy price" or "starting gas index price" and therefore does not utilize pre-1996 results of a production cost simulation. Although CAC/EPUC assert that the formula is inconsistent with § 390(b), the Court upheld our revision in Edison II.
The QF parties' primary objection to revising SRAC energy prices is that in the current market, the Commission cannot find that the SRAC energy prices exceed the utilities' avoided cost and as a result, cannot make a finding supporting modifying SRAC. They claim that since the IERs embedded in the current SRAC energy formula have been shown to be lower than the IERs calculated using certain market data on occasion, the SRAC formula should not be revised. We disagree.
The utilities contend that the since their dispatch decisions are based on the prices in the day-ahead markets, these day-ahead markets represent a reasonable proxy for SRAC. Standard of Conduct (SOC) 4 was adopted in D.02-10-062 and modified in D.02-12-069, D.02-12-074, D.03-06-076 and D.05-01-054. SOC 4 requires an IOU to dispatch its portfolio of existing resources, allocated California Department of Water Resources (CDWR) contracts, and new purchases to meet its electric load obligations in a least-cost manner. D.04-07-028 requires system reliability and deliverability of power to be included as part of least-cost dispatch. For example, PG&E states that "the wholesale power market, and in particular, the NP-15 Day-Ahead market, is PG&E's short-run avoided cost and guides PG&E's dispatch decisions." (Exhibit 28, p. 1-3.)
Existing resources in PG&E's portfolio (i.e., utility retained generation, CDWR, and those contractual obligations which allow economic dispatch) are regularly compared to the market price, with power being either bought or sold at that price. Regardless of the resource stack, the utility's avoided cost for a given hour becomes the market price. The market price that PG&E uses to determine what resources are dispatched in northern California is the NP15 price. If the dispatch decision is made day-ahead, then the price is the day-ahead NP15 price. If the dispatch decision is made hour-ahead, then the price is the hour-ahead NP15 price. PG&E's traders are active in the market and are keenly aware of current prices at which sellers are offering, buyers are bidding and the price at which the most recent transaction was executed. Price discovery is available through voice brokers, electronic trading platforms, such as the ICE, and direct contact with trading counterparties. (Id., p. 3-10.)
According to PG&E, the Day-Ahead spot market is both an obvious and conservative (i.e., erring on the side of overpayment) measure of PG&E's true short-run avoided costs. PURPA's definition of avoided cost clearly and correctly envisions that utilities may satisfy their short-run incremental energy needs either through spot purchases or by increasing generation under their control. Since the divestiture of most of its fossil plants around 1998, PG&E has been a net buyer of power, meaning that the main sources of additional power in both the long and short run have been purchases, not self-owned generation. For short-run spot market purchases in NP15, there are three common product types: bilateral Day-Ahead, bilateral Hour-Ahead (HA) and the real time imbalance energy from the market run by the CAISO. While the markets for these products are linked to a substantial degree by the arbitrage activities of participants, the attributes of these products do differ and market prices do differ from day to day. (Id., p. 3-14.)
PG&E further contends that the NP15 Day-Ahead price is very transparent, based on the fact that there are at least three different providers of an NP15 Day-Ahead index approved by FERC, including the ICE and Dow Jones indexes that PG&E is using. (Id., p. 3-16.)
The utilities further argue that the day-ahead markets are "workably competitive," claiming that they meet the FERC liquidity criteria set forth in FERC's November 19, 2004 Order Regarding Future Monitoring of Voluntary Price Formation. PG&E analyzed price levels in the NP15 Day-Ahead power market, as reported by ICE from 2002 to 2005 and compared them to other Day-Ahead power prices delivered in related markets or trading hubs: South-of-Path 15 (SP-15), the California-Oregon Border (COB), and Palo Verde (PV).
My analysis, presented in more detail in Appendix A, concludes that the NP-15 DA hub is within a larger market that is workably competitive. I find that DA prices are nearly identical across the CAISO control area and also close in the other two nearby trading hubs during the vast majority of all hours. Thus, NP-15 is almost always part of a larger market, either SP-15, COB or PV, depending upon season. Historical prices of these hubs during the 2002 to 2005 period are at levels that show that the market is sufficiently robust and well-functioning. There have been few "price separations" within the CAISO control area and also few price spikes across the Western U.S. The CAISO's automatic mitigation program, designed to capture and mitigate excessively high sales bids in the RT market, has not been triggered once in the last three and a half years. (Id., p. 3-17.)
The QF parties disagree, stating that day-ahead market prices cannot serve as a proxy for avoided costs because they are thinly traded, and are used infrequently by the utilities. IEP reports that SCE purchases from the SP15 market are never more than 3-4% of their total supplies, while SCE's QF purchases are typically 28% to 35% of their total supplies. IEP also notes that the current trading volumes of 33,500 MWh per day in the NP15 market and 44,600 MWh per day in the SP15 market are less than total QF deliveries.
TURN compared the northern California NP15 daily electricity prices and PG&E natural gas prices from the ICE to calculate market heat rates for the summer and winter on-peak and off-peak periods during the one-year period from August 1, 2004, through July 31, 2005. TURN also reports that the actual market heat rate averaged approximately 8,300 Btu/kWh during that period (TURN notes that the results using burnertip gas prices would be somewhat lower). TURN compared those prices with PG&E's current SRAC formula, which yielded an implicit heat rate of approximately 10, 840 Btu per kWh averaged across the same period (9,360 in summer and 12,324 in winter), resulting in SRAC payments that were approximately 30% greater than market prices. TURN notes that since many QFs are continuing to operate under the fixed price amendments, SRAC payments did not exceed market prices to the same degree, but with those fixed price amendments due to expire in 2006 and 2007, there will be a substantial increase in cost if the SRAC formula is not revised to reflect actual market prices. (TURN Opening Brief, p. 6.)
SCE provided a similar example, comparing posted SRAC energy prices to monthly average prices reported by Dow Jones, Megawatt Daily (MWD), and ICE for day-ahead electricity in SP15 from August 2002 through July, 2005. During this period, SCE states that the monthly average day-ahead price in SP15 was $45.47/MWh, while the average posted SRAC energy price was $55.76/MWh, 23% higher than SP15 prices.70
SCE also compared the embedded IER in the Modified Formula with an implied market heat rate calculated by taking a monthly average of SP15 day-ahead electricity prices expressed in $/MWh, subtracted $2/MWh for variable O&M, and divided this result by a Malin-based burnertip price for natural gas for the same period.71 SCE states that implied market heat rates in SP15 were consistently below the 9,140 Btu/kWh heat rate in the Transition Formula.
PG&E states that not only do QFs receive SRAC energy prices that are approximately 30% above prices in the NP15 Day-Ahead wholesale power market, but many QFs also receive capacity payments pursuant to the standard offer contracts, resulting in all-in SRAC payments that are well above the utilities' actual avoided cost.
All parties acknowledge that, from its inception in D.96-12-028, the Transition Formula was intended as a temporary measure, to be used to calculate utility avoided costs until energy payments could be based on California PX prices pursuant to § 390(c). D.96-12-028 adopted factors, consistent with § 390(b) designed to "yield a fair representation of the historical values required by AB 1890." (D.96-12-028 [69 CPUC 2d 546, 553].) Those factors were derived from a regression analysis and were based on pre-1996 data, a time when the utilities owned their own generation resources or purchased from QFs, and there was no market mechanism available for use as an avoided cost benchmark. Since then, electric restructuring, the energy crisis, and the resulting shift in the utilities' procurement practices have made the determination of avoided costs much more dependent on market activity.
As we are all aware, the PX will never be fully operational because it is defunct, yet we are mandated to calculate avoided costs pursuant to § 390(b). Although the PX ceased market operations at the end of January 2001, Day-Ahead markets for electricity continue to exist.
The evidence suggests that the existing Transition Formula by itself may no longer serve as the most reasonable proxy for determining avoided costs. Therefore, we find that it is time to update the SRAC methodology to ensure that it continues to reflect utility avoided costs. Moreover, we find that the variable factor formulation of the Transition Formula and updates to the formula are legal and permitted by § 390(b). This belief was upheld in Edison II, which affirmed the Commission's finding, stating, "to the extent that CCC is arguing that the Commission is forever wedded to the pre-1996 figures and cannot take current prices into account, CCC is in error."72 Even CCC does not dispute that SCE's proposal can be implemented through the MIF consistent with § 390(b). (CCC Opening Brief, p. 9.)
In upholding our discretion to modify the factors, as needed, to reflect changing conditions in the market, the Court, in Edison II, stated that the Commission not only has the power to alter the factors, but has the duty to do so in appropriate circumstances, finding:
The Legislature did not prescribe a specific formula. Rather, it prescribed a general formula to be transitional until such time as the PX was up and running properly...[I]t is now becoming obvious that the PX will never properly function. Thus it was up to the Commission to arrive at a formula that met the requirements of section 390 and also complied with PURPA.73
Although some QF parties may view certain proposed SRAC revisions as too extreme, our goal is to price QF energy at avoided cost, not based on QF economics. The primary difference between the Transition Formulas adopted in D.96-12-028 and D.01-03-067 and the formulas proposed in this proceeding is the IER. The IER used in the existing formula has remained unchanged for almost ten years, and is based on data that is 11-12 years old. In D.01-03-067, we found that an update of both the IER and the O&M adders were necessary, but would require additional information. A proceeding to update these factors was held, but a decision was never issued as the testimony in that proceeding quickly became outdated as a result of the ongoing energy crisis and its aftermath.
The evidence in this case demonstrates that the Commission should adjust the factors in the Transition Formula such that the SRAC energy prices resulting from the formula continue to appropriately reflect the utilities' short-run avoided cost.
4.4. The Market Index Formula
All of the proposals to update SRAC energy prices recognize that current market prices, whether historical or forecast, should be taken into account when setting avoided cost. As discussed above, PG&E, SCE, SDG&E, TURN, and CCC have each proposed SRAC energy pricing methodologies that utilize IMHR figures derived from day-ahead power price indices at NP15/SP15 and spot bid-week natural gas indices at border trading points or at the burner-tip.
The NP 15/SP 15 power prices are currently the only "market based" pricing available. In our effort to transition to pure market based pricing for QFs we agree that SRAC energy prices should incorporate power prices as reported at the NP15 trading point for PG&E, and the SP15 trading point for SCE and SDG&E. These prices reflect an element of the cost that would otherwise be incurred by the utilities in the short run to replace QF power.
However, we do not believe it would be appropriate to base SRAC pricing solely on the NP 15/SP 15 markets. The fact that these are the only market based prices available does not mean that they are the right prices. It is significant that FERC has declined to make a finding that QFs have nondiscriminatory access to competitive wholesale markets in California. EPUC/CAC and CCC have shown these markets represent less than 5% of the total power purchases by the utilities, may be subject to manipulation, and reflect only lower cost products.74 Further, reliance on a single trading point to derive an overall market clearing price as a proxy for the utilities' marginal costs ignores the existence of out of market purchases where higher prices may prevail. In particular, the CAISO relies on reliability-must-run contracts and must offer obligations, to address local market power and reliability concerns. 75 To the extent that some portion of market demand is satisfied by these out-of-market purchases, the remaining demand will intersect the supply curve at a lower point, yielding a lower market clearing price than what would result were all demand met through market purchases.76
In a well-functioning market, the market clearing price reflects the cost of the marginal resource. Currently that market does not exist in California. While NP15/SP15 prices may provide a reasonable starting point for developing SRAC prices, we are persuaded by the evidence offered by CCC and CAC/EPUC that these prices would likely understate utility avoided costs. With regard to QFs, the NP 15/SP15 markets' failure to reflect these out-of market transactions is particularly troublesome given the role that QFs play in reducing local market power.77 For example, many cogenerating facilities may be located within transmission constrained load pockets and may reduce the need for RMR or other high priced and less efficient resources. In testimony CCC notes that cogeneration resources are a "vital component of the in-basin resources dedicated to serving Edison's loads". Furthermore, in determining the need for RMR facilities, the CAISO compares anticipated load with the availability of generating resources to serve that load, and the availability of transmission capacity. In conducting this assessment, the CAISO explicitly assumes that all QF facilities are operating.78 As such, QF facilities logically reduce the need for RMR facilities by reducing the amount of energy that, absent these facilities, would need to be delivered to a given load pocket. Because the relatively high energy prices paid and associated heat rates under RMR contracts are not reflected in NP15/SP15 prices, these prices cannot reflect the value QFs provide in terms of avoiding the need to enter into these types of contracts.
In addition to the general failure of NP15/SP15 prices to adequately reflect out of market purchases, we are also reluctant to wholly embrace a proxy price based on a market over which the utilities themselves can potentially exert significant influence through their purchasing decisions and role as Scheduling Coordinators. As observed by the QF parties, if the price the IOUs pay for QF power is based on NP15/SP15 prices, they may have an incentive to engage in strategic behavior that could yield a lower price in the NP15/SP15 market. QF parties argue that such market gaming could take several forms including deliberate underscheduling ,the submission of infeasible schedules, in which the scheduled energy is undeliverable due to transmission constraints, as well as strategic generation and dispatch. In the case of underscheduling and submission of infeasible schedules, the CAISO would enter into additional out-of-market contracts to ensure reliability. As we have learned during the 2000 Energy Crisis, the potential ability to manipulate market prices is harmful to ratepayers and the overall energy market. In this instance, we are concerned that if the IOUs were to exert market power, this will put downward pressure on the observed market clearing price to the disadvantage of QF generators and despite the fact that the resources ultimately dispatched may in fact be higher cost, not lower cost resources. 79
Based on the record, we believe that using NP15/SP15 prices alone would likely result in SRAC prices that understate utility avoided costs, as they do not include the full range of generation resources in the electricity industry today and do not include out of market transactions. At the same time, we recognize that continuing to use the administratively set heat rates may result in SRAC prices that exceed utility avoided costs. Despite the shortcomings of both proposed methodologies, we recognize they each provide certain benefits. The market-based approach, while understating utility avoided cost, reflects a portion of the current energy market, while the administratively-based approach, which could potentially exceed utility avoided cost, would include consideration of higher cost out of market and RMR requirements.
Consequently, we believe that adoption of an interim hybrid approach would result in SRAC prices that more closely reflect utility avoided costs. The hybrid approach we adopt here involves calculating an average heat rate that, in effect, combines a market derived value with the administratively determined approach adopted in prior Commission decisions. We believe this approach is consistent with information presented in the record, as well as with the general evolution of QF pricing in the state, in which market-based factors will play an increasing role. This approach also fulfills our obligation to develop a formula that meets the requirements of section 390(b) and federal law.80
Each of the parties offered a different proposal for deriving a market based IER. PG&E's proposal preserves the D.96-12-028 Transition Formula, simply updating the pre-1996 factor in the Transition Formula. Specifically, PG&E's proposal would link the SRAC energy prices to the Day-Ahead trading points. However, if the market conditions underlying the data used in the regression analysis differ from current market conditions, the resulting SRAC price may not accurately reflect a utility's avoided cost. Moreover, PG&E's proposal would require formal Commission update immediately and on an ongoing basis. PG&E agrees that its factors require revision before they can be used and would continue to require continuous updates. (RT, p. 3,567.)
SDG&E's proposal uses a two-year average of daily Day-Ahead market prices at SP15 for SDG&E for the past two years, less the proposed O&M divided by the burnertip price of gas. This average would be updated automatically on an annual basis. SCE's proposal is also based on historical market price, but proposes to use a twelve-month rolling average. The CCC expresses concern that SCE's backward-looking proposal tends to result in IERs that are lower than appropriate when costs are rising.
CCC's proposal uses two years of forward market prices, along with an "elasticity adder" to adjust the forward prices to reflect the price increase if the "aggregate" amount of QF energy production on the utility's system is withheld. While we believe use of forward market prices is appropriate, we do not agree with CCC's proposed use of the E3 methodology. The elasticity adder proposed by CCC was originally developed by E3 in their avoided cost report to the Commission81 to account for changes in avoided generation costs as a result of a change in demand (i.e., energy efficiency). CCC assumes a high net short position of 7% to 13% that persists and increases through 2011, whereas E3 assumes a much smaller net short position of 5% that declines to zero by 2008. In short, CCC did not apply the E3 elasticity methodology. Instead, CCC applied a significant variant of the E3 methodology which produced a significantly different outcome. We are unpersuaded that the price effects associated with a decrease in demand would be the same as the impacts associated with an increase in the supply of electricity, as proposed by CCC.82
Parties have also expressed reservations on the use of forward market prices. For example, SDG&E argues that forward prices should not extend beyond 15 - 18 months due to its belief that data beyond that time would not reflect a sufficiently liquid and robust market. (SDG&E Opening Brief, p. 34.) While we recognize these concerns, we believe that SRAC prices based on historical market prices would not best reflect utility avoided cost. Further, SRAC prices are not expected to track utility avoided costs on a real time or day-to-day basis. During particular time periods SRAC prices may be higher or lower than actual, real-time avoided costs, but, as the Commission has recognized, such differences balance out over time. (D.04-07-037.) Consequently, we believe that use of forward market prices would result in SRAC prices that will more reasonably reflect avoided costs.
Table 3 illustrates a sample derivation of the market heat rate using a 12-month rolling average of forward SP15 prices. This is based on SCE's proposed methodology in Exhibit 1, but deducts the variable O&M from prices as proposed by SDG&E. We note that by using a 12-month rolling average of forward prices, there is little, if any, difference between a collared and an uncollared heat rate. Thus, SCE's rationale for utilizing a collar around the IER does not appear to be present, as a rolling average of forward prices serves to mitigate excessive price volatility.
We find that it is to the benefit of all interested parties to adopt a solution that relies to a greater degree on market-derived prices, namely, the Market Index Formula (MIF) and at the same time corrects for the failure of the existing markets to reflect the full cost of the total generation mix available in California. The MIF is based on the Modified Formula adopted in D.01-03-067. This formula complies with § 390(b). The IER or heat rate in the MIF shall be calculated by taking an average between an NP15/SP15 - derived value as generally proposed by SCE, and the existing administratively determined heat rates pursuant to prior Commission decisions. For PG&E and SDG&E, these are the heat rates adopted in D.96-12-028 corresponding to the values of 9,794 Btu/kWh and 9,603 Btu/kWh, respectively. For SCE, we adopt the CCC proposal of a heat rate of 9,705. This value represents the average administrative heat in effect for SCE under the Transition Formula adopted in D.96-12-028 and modified in D.01-03-067. In calculating the market heat rate using NP15/SP15 indices, rather than using historical prices, we will use a 12-month rolling average of the weighted average price of the forward market prices for NP15 (for PG&E) or SP15 (for SCE and SDG&E). We agree with the comments of TURN, CCC, IEP and SDG&E that we should not rely on a 24 month forward price as the prices may not be reliable in the second year. Additionally, variable O&M shall be deducted from the market prices used to calculate the market based heat rate. The MIF is shown below and in Table 4.
Market Index Formula (MIF)
Pn = [IER x (GPn + GTn)/10,000] + O&M
IER = (.5 x MHR + .5 x AHR)
Pn = calculated SRAC energy price, cents/kWh
IER = Incremental Energy Rate
GPn = gas price, $/MMBtu
GTn = intrastate transportation costs, $/MMBtu
MHR = Market Heat Rate Btu/kWh
AHR = Adminstrative Heat Rate (PG&E = 9,794 Btu/kWh, SCE = 9,705 Btu/kWh, SDG&E = 9,603 Btu/kWh
O&M = operations and maintenance costs, Cents/kWh
10,000 = [$1/100 Cents] x [1,000,000 Btu / MMBtu]
We direct Energy Division to host a workshop on the technical issues related to calculating the market heat rate above, and subsequently we direct PG&E, SCE and SDG&E to file a joint advice letter specifying the exact data sets used to calculate the market heat rate component of the IER, as described above. This advice letter should also include a description for how the IER will be calculated once MRTU is operational and the administrative heat rate component of the calculation is eliminated, as described below.
Finally, while we find that the MIF, as defined above is the best, currently available estimate of the utilities' avoided cost, we decide today that this formula will change when the CAISO's MRTU is operational. We provide a six month transition after MRTU is operational before the MIF will change. The CAISO's day-ahead market should be sufficiently robust and all-encompassing to reflect the full range of generation resources used to meet the state's energy needs. The CAISO's MRTU market power mitigation features allows resources to bid into the market and obviates the need for RMR and FERC MOO. This should allow for full transition to market based pricing, as the shortcomings of existing market-based proxies for utility avoided cost will be largely eliminated. When both the CAISO's day-ahead market is fully functioning for purposes of deriving SRAC prices we will adjust the MIF accordingly. These changes will apply on a going forward basis to the prices paid under both existing contracts as well as new QF contracts.
Six months after the implementation of the CAISO's day-ahead market the MIF shall be revised to remove the administrative heat rate component and base the IER exclusively on MRTU market prices. By this time we anticipate the existence of the CAISO markets will make the forward markets sufficiently robust to eliminate the need for an administrative component.
We direct the Energy Division to monitor the operation of the CAISO markets, in close consultation with the CAISO's market monitoring group. If the Assigned Commissioner in consultation with the Energy Division and based on the CAISO's market monitoring reports, determines that the market price does not fully reflect utility avoided cost, then the Assigned Commissioner shall delay the methodology change from the initial MIF (which includes the Administrative Heat Rate in calculating the IER) to the revised MIF (which eliminates the Administrative Heat Rate part of the IER calculation) for up to six additional months. If applicable, the Energy Division will notify the service list of any delay and will continue to monitor the CAISO's market.
The MIF has a Variable O&M component. The O&M adder accounts for the variable O&M expenses incurred by the utility to produce energy and is a relatively small component of costs in the SRAC formula. SCE has proposed $2.00/MWh (see Figure 1 above), and IEP concurs. SDG&E's proposed "$2.50/MWh in 2004 dollars was adopted in D.05-04-024 and implemented for Energy Efficiency by SDG&E Advice Letter 1687-E. Escalated to 2006 at 2% per year, this value would be $2.60 in 2006."83 (Exhibit 85, p. 7.) For purposes of establishing SRAC energy prices, TURN does not recommend the use of a Variable O&M adder.84 Likewise, PG&E does not propose a variable O&M adder value because the Transition Formula does not contain that component.85 CCC recommends a Variable O&M adder of $3.00/MWh, and also recommends an automatic adjustment in future years.
Given the uncertainty in formulating such estimates, all three utilities will now be on the MIF as described herein. With regard to our consistency goal in this avoided cost rulemaking, there is no compelling reason to not adopt the same variable O&M adder for all three utilities. As SDG&E notes in its direct testimony, the Commission has adopted variable O&M figures for other purposes:
SDG&E proposes the variable O&M component be based on the variable O&M of a Combined Cycle Gas Turbine (CCGT). This level of variable O&M is consistent with the type of power that would replace QF power, baseloaded power supplies as provided by a CCGT. In the decision in phase 1 of this proceeding, D.05-04-024, the Commission recommended using the data developed in R.04-04-026 for the costs of operating a CCGT. For consistency, SDG&E proposes to use the 2004 value for the variable cost of a CCGT adopted in Phase 1. (Exhibit 85.)
We concur with this approach and adopt it for use in the SRAC energy formulae for the three utilities. However, the O&M shall be escalated by 2% per year, consistent with Advice Letter 1687-E.
Overall, as is shown in Table 2, eight parties in this rulemaking are recommending the use of three different gas prices: border, burner-tip, and the trading point at PG&E City Gate. For this illustration, the respective prices in Table 1 are $6.33, $6.53, and $7.00/MMBtu.
Border prices are recommended by PG&E and CAC/EPUC, while burner-tip gas prices are recommended by SCE, SDG&E, and CCC. IEP supports the status quo, which for PG&E is border, and for SCE and SDG&E is burner-tip. As noted, TURN recommends the use of the PG&E City Gate trading point. All parties advocate the use of the Topock border point in lieu of the Malin border point adopted in D.01-03-067.
For PG&E in its May 2006 SRAC posting, the utility takes (1) the average of three Malin bidweek gas indices as reported in Gas Daily, Natural Gas Intelligence, and Natural Gas Weekly which is $6.1167 per MMBtu, (2) then PG&E adds $0.377 per MMBtu for intra-state transportation and $0.0551 for shrinkage to the Malin average to get $6.5488/MMBtu to approximate the formerly unrobust Topock border point per D.01-03-067, and (3) then PG&E averages the $6.1167 and $6.5488 to get $6.3328 per MMBtu. For SCE in its May 2006 SRAC posting, the utility takes (1) the average of three Malin bidweek gas indices as reported in Gas Daily, Natural Gas Intelligence, and Natural Gas Weekly which is $6.1167 per MMBtu, and (2) then SCE adds $0.377 per MMBtu for intra-state transportation and $0.0555 for shrinkage86 to the Malin average to get $6.5492 to approximate the formerly unrobust Topock border point per D.01-03-067. SDG&E makes the same calculation as SCE.87
SDG&E proposes to update its intrastate gas transportation rate based on the current Schedule EG tariffs for electric generators using more than 3 million therms. According to SDG&E, this rate is the intrastate transportation rate for most electric generators in SDG&E's service area; the value is presently 36.98 cents per decatherm.
CCC summarizes the burner-tip gas price in the proposed MIF as the sum of: (i) the bidweek Topock gas prices as published in the three publications currently being used in SCE's postings, (ii) the tariffed SoCalGas Schedule GT-F5 "Sempra-wide transportation rate for large electricity generators, including Interstate Transition Cost Surcharge (ITCS), and (iii) SoCalGas' tariffed schedule G-MSUR, the transported gas municipal surcharge." CCC explains that this is essentially the same approach adopted in D.01-03-067, with the exception of the use of Topock border gas prices instead of Malin gas prices.
Because burner-tip gas prices include intra-state transportation costs, on top of border gas prices, burner-tip gas prices are necessarily higher. With regard to avoided cost, whether the utility bought the gas to run its own plant or whether the utility bought the power from a merchant plant fueled by natural gas, burner-tip gas would be required. Therefore, we adopt a burnertip gas price for use in calculating SRAC. We will allow SDG&E and the other utilities to annually update the intrastate transportation rate to the most recent value in their gas tariffs, as necessary. For example, if border gas at Malin is $6.00/MMBtu and intra-state transportation is $0.50/MMBtu, the burner-tip gas price is $6.50/MMBtu which, in this example, is 8% higher than border gas.
We also agree with the parties that the Topock border point is now sufficiently robust and should once again be utilized calculating SRAC. SCE provides a succinct description of the changes that have occurred with respect to the Topock border point since D.01-03-067 was issued. (See Exhibit 1, pp. 64-65.) Therefore, for SCE and SDG&E, SRAC shall be based on the Topock border price, while SRAC for PG&E shall be based on a 50/50 weighting of published border prices at Malin and Topock.
In accordance with D.96-12-028, SRAC energy prices are time differentiated to reflect the different value of power on the utilities' systems throughout a given day. Time-of-Use (TOU) or Time of Delivery (TOD) factors convert annual or seasonal prices into intra-day, time-period specific prices.
SDG&E proposes to change both the TOD factors and the TOD periods. SDG&E proposes to use the current TOD hourly time periods going forward but to change the current May through September summer period to a summer season of June through October. TURN recommends changing the summer period for capacity to exclude May and October.
Since existing QF contracts have incentives tied to performance during different TOD periods, keeping the hourly time period definitions roughly the same will reduce problems related to changing the terms of existing contracts. However, SDG&E proposes that the energy price be the same for the on-peak and the semi peak periods going forward. Similarly, SDG&E proposes that the prices for off-peak and super off-peak be the same. Under SDG&E's proposal, the four TOD period definitions would remain the same (with the exception of the summer period change described above), but there would be only two prices.
SDG&E would update the existing summer and winter price differentials and on-peak and off-peak price differentials using the same two years of recent historical data used to forecast the IER. However, SDG&E notes that if the Commission decides that the capacity payment derived from the transition formula should constitute the entire payment to a QF, no added adjustments to the TOD factors are necessary. If the Commission continues to provide a separate capacity payment, there would be potential double-counting since the market price includes some contribution to fixed costs. SDG&E therefore proposes two sets of TOD factors, for use with and without a separate capacity payment.
CCC believes that the PG&E and SDG&E factors are "quite `flat' across TOU periods, and thus do not value on-peak generation substantially more than off-peak power." (Exhibit 102, p. 54.) CCC notes that both PG&E and SDG&E use significantly "peakier" TOU factors in their RPS solicitations. CCC recommends that the Commission update PG&E's and SDG&E's TOU factors to reflect either the allocations in their recent RPS solicitations or those contained in PX price data from the 1998-2000 period when the PX market functioned well. The CCC notes that this PX data was used by PG&E witness Strauss and by E3 in the development of avoided costs for energy efficiency programs adopted in D.05-04-024. CCC remains silent regarding SCE's TOU factors.
The CCC proposes that the Commission adopt updated TOU factors based on the E3 TOU price profile utilized and adopted in D.05-04-024 for the development of avoided costs for energy efficiency programs.
DRA also recommends an update of the utilities' TOU factors given the length of time since the factors have been examined, but does not provide a specific proposal and instead suggests that the Commission convene workshops to update the IOUs TOU/TOD factors and periods using more recent load profile data.
PG&E argues that it is constrained by Section 390(b) and cannot change TOU factors.
As noted above, the Legislature did not adopt a specific formula, nor did it adopt specific TOUs factors. Therefore, it is appropriate to update the TOU or TOD factors periodically. The evidence in this proceeding clearly demonstrates that the TOU/TOD data is outdated. Unfortunately, the parties recommending specific changes to the TOU/TOD factors and periods did not provide a sufficient showing to support their recommendations. Nevertheless, we believe that updating the IOUs' TOU/TOD factors and periods to be consistent with the TOU factors adopted in other procurement proceedings is reasonable and as pointed out by CCC, the TOD factors are too flat to adequately reflect the differential in prices in peak and off-peak periods. In light of this, we believe it is appropriate to adopt TOU factors that are consistent with the adopted TOU factors for the Market Price Referent (MPR). The MPR is a benchmark price for a new combined cycle combustion turbine and is used to evaluate whether or not a given renewable project, submitted in response to a Renewables Portfolio Standard solicitation, is priced above market. TOU factors are used in the RPS to ensure that the time differentiated value of energy is appropriately taken into account when comparing projects against the MPR. TOU factors used for purposes of this proceeding fulfill fundamentally the same role. In light of these parallels, it is reasonable to adopt here, as an interim approach, the TOD factors used in calculating the MPR, until we consider updates to the TOU/TOD factors and periods in a subsequent proceeding.
FERC's regulations require the Commission to take line losses into account in determining avoided cost.88 Line loss adjustments to QF prices are currently determined in accordance with the methodology adopted in D.01-01-007, which is based on the CAISO generator meter multipliers (GMMs). PG&E recommends that the Commission modify the GMM that is used to estimate line losses associated with QF power and replace the current formula of (GMMqf-GMMsys) with GMMqf.
Since the MIF we adopt today is based on the Transition Formula, we decline to modify the GMM calculation at this time.
47 See, e.g., Biennial Resource Update Plan [D.96-07-026] (1996) 66 CPUC2d 780; Order Instituting Ratemaking No. 2 [D.82-12-120] (1982) 10 CPUC2d 553.
48 The Transition Formula does not contain a variable O&M adder, but SCE's Modified Formula does contain an O&M adder.
49 The change from Topock to Malin was made due to concerns that the Topock gas prices were being manipulated and were no longer robust for purposes of pricing SRAC energy.
50 Table 2 is a modified version of a table that appears in Exhibit 104. "Table ES-1 summarizes the principal SRAC recommendations of the parties to this case, and expresses those recommendations as a "spark spread" between natural gas and SRAC prices" (Exhibit 103, p. ii). However, the actual table was not included but was submitted with the errata in Exhibit 104.
51 This rate is calculated in the same manner as in SCE's Short Run Avoided Cost Energy Price Update for Qualifying Facilities (SRAC posting). In SCE's SRAC posting, the Intrastate Transportation is referred to as GTn and is currently derived from applicable So Cal Gas rates from tariffs GT-F5, ITCS, G-MSUR and G-CPA.
52 Exhibit 1, Figure 10.
53 PG&E used a 50/50 mix of Malin and Topock border prices.
54 D.96-12-028, mimeo., p. 14. For PG&E, the CPUC adopted two factors, one for summer, one for winter.
55 D.01-03-067, mimeo., p. 11.
56 Southern California Edison v. Pub. Util. Comm'n, 101 Cal. App. 4th 982, 992-93 (2002).
57 Section 390(b) mandates the use of the starting energy and border gas prices. These starting values were derived using a 24-month average of pre-January 1, 1996 values as originally adopted in D.96-12-028.
58 Exhibit 28, at p. 3-17, citing the "New York Mercantile Exchange Inc. Online Rulebook," Chapter 644, NYMEX Dow Jones NP15 Electricity Price Index Swap Contract.
59 Exhibit 102, p. 29.
60 "Methodology and Forecast of Long Term Avoided Costs for the Evaluation of California Energy Efficiency Programs," prepared for the California Public Utilities Commission's Energy Division, dated October 25, 2004.
61 Exhibit 90, p. 5.
62 CAISO Department of Market Analysis, 2004 Annual Report on Market Issues and Performances, at p. 2-17 and p. 2-18.
63 CAISO 2004 FERC Form 714 Filing, Part IV, Notes to Page No. 43.
64 Exhibit 29, p. 3-18.
65 Exhibit 2, p. 41.
66 The QF in/QF out IER calculation can be illustrated as follows:
IER = { [(QF-out Market Costs) - (QF-in Market Costs)] ÷ Gas Market Price } ÷ QF Volumes
Alternatively, the equation can be stated in terms of net market costs as follows:
IER = { [ Net Market Costs to the utility when QFs are out ] ÷ Gas Market Price } ÷ QF Volumes
67 18 CFR § 292.304(e) lists the following factors, among others:
(1) The utility's system cost data;
(2) The availability of capacity or energy from a QF during the system daily and seasonal peak periods, including: (i) the ability of the utility to dispatch the qualifying facility; (ii) the expected or demonstrated reliability of the qualifying facility; (iii) the terms of any contract or other legally enforceable obligation, including the duration of the obligation, termination notice requirement and sanctions for noncompliance; (iv) the extent to which scheduled outages of the qualifying facility can be usefully coordinated with scheduled outages of the utility's facilities; (v) the usefulness of energy and capacity supplied from a qualifying facility during system emergencies; including its ability to separate its load from its generation; (vi) the individual and aggregate value of energy and capacity from qualifying facilities on the electric utility's system; and (vii) the smaller capacity increments and the shorter lead times available with additions of capacity from qualifying facilities.
68 D.92-01-018, mimeo. at pp. 8-9.
69 Id. at pp. 11-12.
70 Exhibit 1, p. 57.
71 Id., p. 58.
72 Southern California Edison v. Pub. Util. Comm'n, 101 Cal. App. 4th 982, 993 (2002).
73 Southern California Edison v. Pub. Util. Comm'n, 101 Cal. App. 4th 982, 991-992 (2002).
74 See, e.g., CCC/Beach Ex. 103, at 19-20, 24-26, Table 4.
75 While the CAISO has released some RMR resources, CAC/EPUC points out that these resources continue to operate under out-of-market resource adequacy contracts. (CAC/EPUC Reply Comments, p. 4).
76 IEP/CCAC/EAP/CCC Ex. 42 pp. 11-12.
77 CCC/Beach Ex. 102 p. 16.
78 See, for example, "Local Capacity Technical Analysis - Overview of Study Report and Final Results" p. 11; submitted as Attachment 1 to the "Proposal of the California Independent System Operator Corporation Regarding Local Resource Adequacy Requirements," R.05-12-013, January 1, 2006.
79 IEP/CCAC/EAP/CCC Ex. 42 pp. 33-38.
80 Southern California Edison v. Pub. Util. Comm'n, 101 Cal.App. 4th 982, 991-992 (2002).
81 Methodology and Forecast of Long-Term Avoided Cost(s) for the Evaluation of California Energy Efficiency Programs, E3 Research Report submitted to the CPUC Energy Division, October 25, 2004. ( http://www.ethree.com)
82 Exhibit 102, pp. 41-42.
83 The 2% escalation was also adopted in Advice Letter 1687-E.
84 With regard to variable O&M, TURN does present recommendations on variable O&M, not for the purpose of calculating SRAC energy but, instead, for the purpose of "capping market energy prices at the costs of generating energy from such a new CT." (Exhibit 149, p. 1.)
85 CCC (in Exhibit 104) impute a variable O&M adder value for PG&E based on its proposed factors and is useful for illustration, but it is not a value recommended by either PG&E or CCC.
86 It is not clear why the shrinkage rates are reported differently by PG&E and SCE.
87 SDG&E reports the same shrinkage rate as PG&E which results in a slightly lower gas price than SCE.
88 18 CFR § 292.304(e)(4).
