4. Economic Need

SDG&E initially identified need for the Miguel-Mission and Imperial Valley projects through a series of system impact studies performed by SDG&E in connection with generators' requests for interconnection.⁷ In this phase of the proceeding, SDG&E presents an economic analysis of the effect of these upgrades on energy costs to SDG&E ratepayers, as well as the effect on ratepayers throughout the CAISO control area. In the following sections, we briefly describe SDG&E's economic analysis and consider the results.

4.1 SDG&E's Economic Analysis

SDG&E hired Henwood Energy Services (Henwood) to analyze the effect of the transmission upgrades on annual energy costs to ratepayers. Using its Electric Market Simulation System (EMSS) and PROSYM production cost models, Henwood examined the operations of the power grid (generation dispatch and line loading) for the entire WSCC region, divided into 23 different market zones. The San Diego region was divided into two separate zones represented as San Diego North and San Diego South, with Miguel-Mission transmission as the path between. The modeling effort examined operations for all hours of a single year (2004), both with the transmission upgrade and without. In instances where the transmission between zones was not adequate to allow the most economic resources to meet load, the model redispatched to deliver more expensive energy via the unconstrained transmission paths that were available. The key assumptions used in Henwood's analysis can be summarized as follows:⁸

· The demand for electricity in WSCC is forecasted to grow at 2.1 percent per year over the next ten years.

· There is little retirement of existing generation between 2001 and 2004 other than that associated with repowering projects.

· Natural gas prices are assumed to revert from the higher price levels experienced in early 2001 to a lower long-term trend. After this reversion, the assumed real rate of growth in natural gas prices is 0.4 percent per year.

· Market clearing prices are equal to incremental production costs (fuel price multiplied by the incremental heat rate, plus variable operations and maintenance cost) plus a scarcity premium.

SDG&E provided Henwood with input data, which included load data and transmission path rating updates along with the following generation modeling scenarios:

· Case 1 (Base Case)--generation in the San Diego and North Baja areas equals the existing generation plus planned proposed summer peakers (343 MWs).

· Case 2--Base Case plus 510 MWs, representing the Otay Mesa Power Project.

· Case 3--Base Case plus 1360 MWs, representing the 510 MWs in Case 2 plus a 250 MW AEP Resources plant and 600 MW Sempra Energy Resources plant.

· Case 4--Base Case plus 2360 MWs, representing the 1360 MWs in Case 3 plus a second 250 MW AEP Resources plant and the 750 MW La Rosita Power Project (LRPP) planned by Intergen.

· Case 5--Base Case plus 3810 MWs, representing the 2360 MWs in Case 4 plus a 500 MW AES plant, 350 MW Intergen LRPP plant and 600 MW Sempra Energy Resources plant. The Path 45 rating is also increased from 800 to 1300 MWs.

SDG&E's planning studies indicate that the benefits from the Imperial Valley Project would not be seen without the Miguel-Mission Project. Therefore, the economic analysis assumes that the Miguel-Mission upgrade is already in place and uses the same cases to evaluate the Imperial Valley Project. According to SDG&E, the amount of generation in Case 3 would not require upgrades to the Imperial Valley transformer, but the amount in Cases 4 and 5 would require the upgrade (or congestion management).⁹ Therefore, the Imperial Valley Project was modeled in the Case 4 and 5 scenarios, and its costs were added to those cases in the calculation of net benefits.

Figure 1 presents the location of the proposed new plants that are referenced in the generation scenarios.¹⁰ Table 1 summarizes the results of SDG&E's economic analysis, as follows:

Table 1:

Henwood's model projects that the upgrades will produce positive net benefits in Cases 3, 4, and 5 by reducing congestion hours and improving the economics of the transmission in that area. For example, in Case 3, the additional transmission capacity provided by the Mission-Miguel Project is estimated to reduce the number of hours of congestion from 4,077 to 371 in SDG&E's service territory, thereby reducing total annual costs of energy from $843 million to $837 million, or by $6 million. For the entire CAISO control area (corresponding to SDG&E, PG&E, and SCE service territories), the Henwood model projects a reduction in annual energy costs of $13 million under Case 3. The projected benefits increase as additional generation comes on line in Cases 4 and 5, as indicated in Table 1.¹¹ The benefits presented in Table 1 above reflect Henwood's modeling efforts for a single year, 2004.

The projected annual cost in Table 1 represents the levelized annual revenue requirement associated with the Mission-Miguel Project in Case 3, and both the Mission-Miguel and Imperial Valley Projects in Cases 4 and 5. Total project costs were discounted to 2001 dollars (to be consistent with the benefit figures developed by Henwood) and multiplied by a 15% levelization factor. In presenting the net benefits (benefits minus annual costs), Table 1 allocates all of the annualized costs to SDG&E ratepayers.

As indicated in Table 1, SDG&E's economic analysis projects that SDG&E ratepayers would receive $3 million in net benefits from the Mission-Miguel upgrade in 2004, assuming that 1350 MWs of new generation is developed in the US-border region or Mexico for export to California. If 2360 MWs of new generation develops, these net benefits increase to $7.24 million, including the cost of the Imperial-Valley Project. They increase further to $26.24 million under a scenario with 3810 MWs in new generation.

The individual generators identified in the various generation scenarios are also projected to benefit from increased net operating revenues, as shown in Table 1. Net benefits to ratepayers within the CAISO control area, including those served by SDG&E, are projected to range from approximately $10 to $174 million in 2004, depending on the level of new generation in the California-Mexico border region and in Mexico that is sold into the California market.

ORA and Border Generation concur with the results of SDG&E's economic analysis and request a ruling from the Commission that, in view of the potential for net economic benefits associated with the Miguel-Mission and Imperial Valley Projects, these upgrades are in the public interest.¹²

4.2 Discussion

In examining the credibility of the results from SDG&E's analysis, we first consider whether the models and modeling approach used to simulate system operations are reasonable. We next consider alternative scenarios or input assumptions that might affect the magnitude of net benefits accruing to ratepayers, and whether the modeling results presented by SDG&E are robust with respect to those changes.

The models used by Henwood simulate hourly operation and dispatch of individual generation facilities and consider transmission line interconnections, ratings, losses and wheeling rates for the entire regional transmission system. While modeling complexity does not guarantee against algorithm errors, we nonetheless have increased confidence when the models used attempt to simulate the electric system as it actually functions.¹³ The record also documents Henwood's expertise in performing such analysis in the industry, as well as the wide use of its models to simulate electric markets by many other entities, including WSCC members.¹⁴ The modeling methods described in the testimony present a logical approach for evaluating annual energy costs with and without the upgrades. In sum, we have confidence that the modeling approach presented in this proceeding is a reasonable one.

The downside to using such complicated, resource-intensive models, is that it is difficult to quickly run alternate scenarios to test the robustness of the results. In its testimony, SDG&E presented the modeling results across all 5 cases for only a single year, 2004, the year in which the transmission upgrades are projected to come on line. However, without further analysis, it is impossible to determine whether or not the annual cost reductions projected for that single year are representative of future years. We must be confident that they are before we can conclude that SDG&E ratepayers realize net benefits from the investment in these transmission upgrades over time.

At the request of Energy Division, Henwood modeled Cases 3 and 5 for additional years, and presented the estimated total energy cost savings from 2004 to 2010. For Case 3, the estimated cost savings to SDG&E ratepayers is $50 million over the seven-year period, or an average of about $7 million per year. For Case 5, the estimated savings over that period to SDG&E ratepayers is $328 million, or an average of approximately $47 million per year. Witness Lauckhart testified that these savings were relatively constant on an annual basis over that timeframe, and that one could make a reasonable estimation that the benefits would continue over a longer study period.¹⁵ Based on this information, we conclude that the energy cost saving benefits projected for a single year are likely to continue at a relatively constant rate into the future.

We also examined the market pricing assumptions used in the modeling scenarios. One obvious area of inquiry is the "scarcity premium" used by Henwood. If employed indiscriminately, this premium could greatly overstate project benefits by imputed extremely high market prices in the modeling runs without the transmission upgrades. This, in turn, could overstate the economic need for the projects. However, based on the testimony of Witness Lauckhart, we are assured that this is not a problem. Because the Henwood study looked at normal weather conditions, there are very few hours in which the scarcity premium comes into play, i.e., in which loads are so high relative to resources that bidders bid a large amount towards their fixed costs and drive up market prices exponentially.¹⁶

We also carefully examined the plausibility of the amount of generation assumed to be coming on line in the modeling scenarios. The individual members of Border Generation are the sponsors of four of these generation projects in the Otay Mesa area (east and south of Miguel) and in Mexicali, in Baja California, Mexico. Once constructed, these four generation projects will represent more than 2000 MWs of generation that will move through the Miguel substation into the SDG&E load center and potentially beyond. All of the generation projects sponsored by members of Border Generation have commenced construction, and all are scheduled to commence operation by the third quarter of 2003. The project developers have made substantial financial and contractual commitments at this time. Some of the key record evidence is as follows:¹⁷

· Otay Mesa (510 MWs). Calpine's Project Development Manager, Mitchell D. Weinberg, submitted a declaration stating that Otay Mesa Generating Company, LLC (a subsidiary of Calpine) has, among other things, commenced construction and purchased its major equipment, including combustion turbines, a steam turbine, heat recovery steam generators, and an air-cooled condenser.

· Sempra Energy Resources Project (600 MWs). An affiliate of Sempra Energy Mexico, Termoelectrica De Mexicali ("TDM") is planning this 600 MW peaking capacity plant. According to Mr. Octovio Simoes, a Project Developer for Sempra Energy Resources, TDM commenced construction on September 1, 2001. TDM has also secured a 20-year firm gas transportation contract with Bajanorte Pipeline and awarded a $158.5 million contract for engineering and construction services.

· LRPP Projects (1060 MWs). Intergen and Coral are developing these two projects. Intergen's Manager, Business Development, Stephen A. Kaufman, filed a declaration stating that, for one of the projects, Intergen has: (1) spent in excess of $250 million, (2) entered into a 25-year power purchase agreement, (3) entered into fuel supply and natural gas transportation agreements, and (4) completed 14.3% of the project. For the other project, Intergen has (1) commenced construction, (2) spent approximately $65 million and (3) entered into fuel supply and natural gas transportation agreements.

Based on the evidence in this proceeding, we find it plausible that the new generation assumed in Cases 3 and 4 (1360 MWs and 2360 MWs, respectively) will develop in the California-Mexico border region or in Mexico for export to California. We concur with SDG&E's assessment that this range of new generation is more likely to develop than the higher level (3810 MWs) assumed under Case 5.¹⁸

However, the benefits associated with transporting this new generation from the California-Mexico border to SDG&E's load center will be offset to the extent that new generation develops within the transmission constrained area, i.e., in the San Diego North zone modeled in Henwood's analysis. SDG&E's economic analysis assumes that there will be no such generation development. At the request of Energy Division, SDG&E presented additional information on this issue in order to explore the impact of alternate assumptions on ratepayer benefits.

In Exhibit (Exh.) 105, SDG&E presents a status listing of projects currently in its interconnection queue for the San Diego North area. Although there are approximately 3000 MWs in the queue, most of them have not completed the very initial step in project development (a system impact study), none have signed interconnection studies or initiated with CEC any permitting activity.¹⁹ Assuming that one-fifth of this amount would actually be in operation by 2004, Henwood ran Case 3 from the years 2004 through 2010 adding a 600 MW plant in San Diego North. The results show that the total energy cost savings associated with the Mission-Miguel line decreases, but still produces net economic benefits over the seven-year period.²⁰ Given the uncertain status of project development in San Diego North and the relative robustness of SDG&E's scenario analysis if even 600 MWs of those projects come on line by 2004, it appears unlikely that the cost saving benefits associated with Miguel-Mission will be negated by increased development of generation within the constrained area. However, we note that those benefits will diminish to the extent that new generation comes on line in the constrained area.

This brings us to our final area of inquiry regarding the robustness of SDG&E's analysis: project costs. The record provides very little detail describing the basis for SDG&E's project cost estimates, particularly for the Miguel-Mission Project. The construction cost information provided under protective order provides some detail for the Imperial Valley project, in terms of assumed labor hours and material costs by task, but there are no cost details presented for Miguel-Mission. Project cost estimates on the record are characterized as "very conceptual," "preliminary" and "subject to any CPUC licensing requirements."²¹ For Miguel-Mission, SDG&E has done only feasibility engineering, not detailed engineering. SDG&E has not conducted any detailed environmental surveying or licensing activities, and states that the costs could significantly escalate if the proposed route is changed or the licensing is protracted. SDG&E's economic analysis does not include operation and maintenance costs for either projects.²²

Therefore, we are unable to assess the reasonableness or credibility of the project costs presented in SDG&E's economic analysis, based on the record in this proceeding. We can only find that the net benefits to ratepayers justify the construction of these projects provided that: (1) a threshold amount of new generation develops in the California-Mexico border area and in Mexico, and (2) project costs are contained at or below the preliminary estimates presented in this proceeding.

In sum, we conclude that construction of the Miguel-Mission and Imperial Valley Projects can be expected to reduce energy costs to SDG&E ratepayers by approximately $6 to $14 million per year, assuming that 1360 to 2360 MWs of new generation becomes operational in the California-Mexico border region and Mexico for export to California. CAISO ratepayers, as a whole (including SDG&E) can expect energy cost reductions in the range of $13 to $50 million per year. Assuming that project costs do not exceed the preliminary estimates presented in this proceeding, the net annual benefits to SDG&E ratepayers are projected at approximately $3 to $7 million per year, and to CAISO ratepayers at approximately $10 to $43 million per year. These net benefits will diminish to some degree if new generation is constructed in the San Diego North area.

Based on this level of economic benefits, we find that the Miguel-Mission and Imperial Valley Projects are needed and in the public interest. As discussed above, net benefits to ratepayers are contingent upon: (1) the development of at least 1360 MWs in new generation and (2) the reasonableness and accuracy of the preliminary cost estimates presented in this proceeding. Therefore, we will condition our finding of economic need in two ways. First, SDG&E is required to coordinate construction of the transmission upgrades with the construction of a threshold level of new generation in the border area. This will be accomplished through a set of milestone conditions discussed in Section 7 below. Second, we will cap project costs based on the estimates presented in this proceeding, subject to a reasonableness review. (See Section 5.) In this way, we can proceed with the licensing process with confidence that the projects are justified based on the economic analysis presented in this proceeding.

¹² Exh 108, p. 6. While concurring with the results of Henwood's modeling efforts, Border Generation contends that ratepayer benefits are actually greater than those identified in SDG&E's analysis because that analysis does not consider: (1) reductions in reliability-must-run (RMR) costs, (2) qualitative reliability benefits and (3) reductions in market power. However, Border Generation presents no evidence that the Miguel-Mission or Imperial Valley Projects actually produce such benefits. Nor does Border Generation attempt to quantify them. In fact, Border Generation's contention that SDG&E's RMR contractual requirement would change as a result of the new generation and transmission upgrades is not supported by the record. The ISO establishes RMR requirements based on unrelated criteria. (RT at 404-407.) Moreover, the future of RMR contracts beyond 2002 appears uncertain. (RT at 511-512; ORA Opening Brief, pp. 2-3.) We therefore have no basis in this record for considering RMR cost reductions, or the other benefits that Border Generation mentions, in determining whether or not the upgrades are in the public interest.