5. Issues Presented

This proceeding presents several key issues.

5.1 Will implementation of critical peak pricing tariffs
result in demand reduction for Summer 2005?

First and foremost, the intent in pursuing a new default tariff with critical peak prices and under the rapid schedule set forth in the December Ruling was to accomplish demand response for Summer 2005. Therefore, we must first assess what level of demand reduction might reasonably be expected before deciding whether to move forward. To do that, we assess the existing summer peak loads for customers over 200kW.

PG&E has bundled load (excluding non-firm load) of about 3,400 MW. (RT 166:14-17 and e-mail to ALJ Cooke.²⁹) SCE has coincident summer peak bundled load (excluding interruptible) of about 3,042 MW. (Exh. 12.) SDG&E has about 835 MW of peak load for customers over 300 kW. (RT 338:4-5.³⁰)

Next, we must determine the maximum projected demand reduction for these customers. PG&E utilized a set of assumptions to establish an upper bound of possible demand reduction. PG&E assumed 1,800 MW of the peak load would actively manage demand in response to prices and that entire 1,800 MW reduces load by 5% during the critical peak. The resulting upper bound of load reduction is 90 MW. (RT 166:23-167:12.) Using PG&E assumptions of approximately 50% of peak load actively managing demand in response to prices and load reductions of 5% during critical peak, SCE's expected upper bound of demand response is 76 MW. SDG&E assumes 2.4% to 3.4% load reduction from total peak load, resulting in potential reduction of 20-28 MW. (Exhibit 7, Chapter 1, Attachment A.) Using these upper bound figures, the total maximum load reduction we might expect from implementing a critical peak demand tariff for customers over 200 kW is 186-194 MW statewide, with 96-104 MW of that amount in Southern California.

If we were to narrow the applicability of the new default tariff to loads only between 200 kW and 500 kW, as suggested by many parties, the maximum, upper bound figure declines significantly. For example, PG&E bundled load between 200 and 500 kW is about 900 MW. (RT 166:17-19.) Using PG&E upper bound assumptions, we would estimate that 450 MW of load is actively managing demand in response to prices and that entire 450 MW reduces load by 5% during the critical peak with a resulting upper bound of demand reduction of customers between 200 and 500 kW of 22.5 MW. If we narrow the applicability for SCE and SDG&E to just 200 - 500 kW customers the summer peak load for these customers is 1,424 MW³¹ and 327 MW³² respectively, which results in a maximum projected load reduction of 35.6 MW and 8.2 MW respectively. Thus, using these upper bound figures, the total maximum load reduction we might expect from implementing a critical peak demand tariff for customers between 200 and 500 kW is 66.3 MW statewide, with 43.8 MW of that amount in Southern California.

Because of the structural framework for the proposed rates and the focus on energy rather than demand, customers who see a bill increase if they do not modify their on-peak usage have very little ability to mitigate that bill impact even with significant change in usage. For example, for SDG&E, only 9 of its 998 customers that are 300 kW or above would be expected to see their bills increase by more than 3% if they made no load reduction efforts. However, the customers whose bills do increase, would not see commensurate reduction in their bills, even if they made significant demand reductions. An example for SDG&E can be found in Exhibit 14 with Assembly Industry Sample 3. Using historical load data for this actual customer, SDG&E shows the customer would see an increase of 3.66% to its summer bill if SDG&E's proposed rates were implemented and it made no changes to its usage. If it reduced its usage on peak by 5%, its bill would still increase by 3.34% compared to current rates. For this customer, a 5% reduction in use results in a 0.32% reduction in its bill compared to taking no action, not a strong motivator to reduce peak demand on critical event days.

SCE's data shows similar results to SDG&E. For SCE, Exhibit 6, Appendix A shows that with the exception of agricultural, pumping and refrigeration loads, 94% of customers of all other building types would see bill increases of less than 6%, with most bill increases between 0-2%, when four critical peak events are called. However, those customers with significant bill impacts for SCE would generally only see their bill go down by less than 1% for each 5% reduction in peak load they make. (See generally Exhibit 12.) PG&E's analysis also shows that bill impacts for most customers under its proposal will be limited to +/-1%, with nearly all customer bill impacts limited to +/-2%. (See Exhibit 4, Attachment 4, pp. 5-7.) For PG&E, a 5% reduction in use generally results in less than a 0.50% reduction in its bill compared to taking no action. (See generally Exhibit 22.)

Given the bill impacts for customers whose bills increase if they do not modify their loads, we find that the upper bound assumptions that 50% of the applicable customer load will manage their demand in response to the rate, and that every customer that does actively manage their load will reduce their usage by 5%, to be quite aggressive assumptions. The timing concerns raised by parties, especially for customers in the 200 to 500 kW size that do not have account representatives, also factors into this finding. Because the bill impact from modifying usage is quite limited, the rate designs do not provide a strong motivation for customers to change their on peak usage, since they will still be worse off. More reasonable assumptions would result in lower expected demand reductions. If instead we assumed that 10% of the applicable customer load actively manages their load and achieves a 5% reduction, the expected load reduction for all firm bundled customers with load over 200 kW would be 17 MW for PG&E, is 15 MW for SCE, and 4 MW for SDG&E, or 36 MW statewide. If the customer applicability were limited to firm bundled customers between 200 and 500 kW, the expected load reduction would be 4.5 MW for PG&E, 7.1 MW for SCE, and 1.6 MW for SDG&E, or 13.2 MW statewide.

We emphasize that these findings relate specifically to implementing the rates proposed in these applications on the accelerated schedule that would be necessary to implement such rates by June 1, 2005. These findings have limited value for predicting longer term ability of customers to respond to rate changes or rate structures that would occur over time.

5.2 From a policy standpoint, should any customers
with loads of 200 kW and above be exempted from
critical peak pricing rates in Summer 2005?

Here, we consider whether any customers should be exempt from a default critical peak rate for Summer 2005. The December Ruling indicated an intent that all customers, including those on interruptible/non-firm rates would be subject to the new tariff for the summer. We continue to believe that all customers should receive the same price signals as similarly sized customers and that non-firm capacity is best compensated through a reservation payment rather than reduced rates, but we are convinced by the parties that to transfer existing non-firm/interruptible rate customers to the BIP reservation payment program now could compromise an important short-term reliability resource for Summer 2005.

BIP operates as a rider to the customer's otherwise applicable tariff. The BIP reservation payment ($7/kW/month) was adopted in D.01-04-006 and was designed to provide the same bill impact to customers as non-firm and interruptible rates do. Therefore, we do not understand how customers could be disadvantaged by switching to BIP from the non-firm or interruptible rates in the long run, but concede that even though the programs are nearly identical from an operational (trigger) standpoint, there could be some short-term confusion on the part of customers associated with the switch to BIP for Summer 2005. Therefore, we conclude it is prudent to exclude non-firm or interruptible load from the default CPP for Summer 2005. Because BIP is designed to serve the same purpose as the non-firm/interruptible rates, for Summer 2005 customers on BIP should also be excluded from participation on a default CCP rate. Thus, the load reduction figures provided by the utilities properly excluded interruptible load.

The DRP provides participants with both a capacity reservation payment and a payment for performance when called. Although the DRP is called the day ahead, it contains penalty provisions for non-performance, unlike other day-ahead demand response programs. Therefore, it operates more like BIP and should be treated the same way for exclusion purposes. Excluding DRP loads from default CPP would further reduce the amount of potential summer peak reductions³³ by 0.1 - 3.2 MW.³⁴

Many parties argue that customers over 500 kW should not be subject to a default critical peak pricing tariff because the load profiles of customers over 500 kW are basically flat; thus, although they do impose load at time of peak, their loads do not drive summer peaks like residential and commercial air conditioning loads. In the longer term, we believe that all customers should receive price signals, regardless of their load shape or size, that indicate when power is more expensive to procure. A properly designed CPP rate will most likely result in bill reductions for customers with stable load profiles that do not vary with temperature. Therefore, we would expect that any changes to default rates over time would apply to customers over 500 kW as well as those over 200 kW.³⁵ However, for Summer 2005, we agree that customers with flat load profiles are generally not well positioned to reduce load on-peak without significant impacts to their core business and therefore customers with 500 kW of load or greater should be excluded if the Commission implements new default rates for Summer 2005.

Customers served on agricultural schedules also seek exemptions from adoption of any default tariff for Summer 2005. As a preliminary matter, there is a significant difference between the types of customers served on PG&E and SCE agricultural tariffs. PG&E's agricultural tariffs include food processors but SCE's tariffs do not. The other types of customers served on agricultural tariffs are generally water pumpers and farming operations who use electricity to operate their irrigation systems. The most compelling arguments presented about these customers' inability to reduce load on-peak was that farmers rely on the State Water Project and the Central Valley Water Project for water for irrigation and that water is released to them on a schedule established by the Water Project control entities (Department of Water Resources and U.S. Bureau of Reclamation, respectively) that the customers cannot control.³⁶ Sometimes customers must use electricity during peak hours in order to perform their irrigation because that is when the water projects release water to them. The problem identified is a legitimate concern and we will alert the proper decisionmakers at the Department of Water Resources and U.S. Bureau of Reclamation about the impact water release decisions have on energy consumption and peak demand issues so that we can better coordinate response to expected high demand days throughout the state.

As we explained with respect to customers over 500 kW, we believe that all customers should receive price signals that indicate when power is more expensive to procure. Thus in the longer term, especially with coordination with the State Water Project and the Central Valley Water Project, we would expect that any changes to default rates would apply to agricultural customers over 200 kW.³⁷ In the short term, we agree that agricultural customers over 200 kW should be excluded from any revisions to the default tariff for Summer 2005.³⁸ This exclusion would further reduce the amount of potential summer peak load reduction by 0.6 - 3.4 MW.³⁹

Schools, hospitals, and oil pumping customers also seek exemptions as do customers who rely on the utilities for start-up power to restart their electric generators. Oil pumpers and generators both argue that their industries support California's energy infrastructure and reliability needs and should not be subject to revisions to the default tariff for Summer 2005. Schools and hospitals argue they provide vital services that cannot be curtailed. PG&E and SCE already did not include the rate schedules that serve most oil pumping and generation customers in their eligible customer groups. Any other pumping and generation customers that for some reason do not take service on traditional pumping or generation schedules are served under other relevant tariffs for that customer size.

Unlike customers over 500 kW and agricultural customers who are served on unique rate schedules, electric generators requiring start-up power, oil pumping customers, schools, and hospitals with loads between 200 and 500 kW cannot be easily excluded by simply excluding a tariff schedule from applicability. As we stated for agricultural customers and customers with load greater than 500 kW, in the long term all customers should receive price signals that send signals when power is more expensive to procure. Although we are sympathetic in the short-run to the economic problem that higher prices during a critical peak period might cause to a generator deciding whether to start-up, defining the start-up load to exempt presents practical problems for those served on tariff that were not already excluded. (RT 272.) Therefore, we do not establish an exemption for electric generators requiring start-up power, oil pumping customers, schools, and hospitals with loads between 200 and 500 kW for Summer 2005.

Given the exclusions we have identified for Summer 2005, using the aggressive assumptions of load reduction described in the prior section, the resulting upper bound load reduction potential would be 22.3 MW for PG&E, 32.3 MW for SCE, and 8.0 MW for SDG&E. Using our more conservative assumptions, it would be 4.5 MW for PG&E, 6.4 MW for SCE, and 1.6 MW for SDG&E.

5.3 Is there sufficient time for the targeted customers to be educated about the rates prior to implementation, to understand how to respond for this summer?

Customer representatives responded unanimously that if the Commission issues a decision in April 2005, there is insufficient time to make operational changes or capital investments to assist customers in responding, before the rates become operational in June 2005. Many large customers say that they make energy related capital investments as part of their ongoing business capital planning process and that the investment plans for Summer 2005 are already established. CRA lays it out well in their brief:

"There is simply not enough time available between now and the summer for program implementation, customer education, and any reasonable expectation that customers can make business adjustments that would be responsive to any CPP program. For those commercial operations that have not already taken all reasonable steps to improve operating efficiency and adjust operations in response to already high electricity rates, some level of investment in new equipment and energy management systems will most likely be required. If mandatory CPP is going to be effective in capturing demand response from buildings that have not yet implemented efficiency and load management practices, then CPP must induce new investment and acknowledge the lead-time required for such new investment. That simply cannot be expected to happen by this summer." (CRA Opening Brief, p. 6.)

In addition, both SDG&E and SCE, because of the structure of their proposed rates, require the customer to make a decision on opting out quickly, for SDG&E by May 15, and for SCE, by June 5. Although the utilities propose to provide educational materials to their customers, most customers between 200 and 500 kW do not have assigned account representatives who can provide individualized education or help them assess the bill impacts from the new rates. Given these facts, we conclude that most customers will not be well positioned to respond to new default rates this summer, even if the bill impacts justified their response.

5.4 Given the projected demand response for the targeted customers and issues surrounding short term customer education, should the Commission move forward with implementing a default critical peak pricing tariff for Summer 2005?

As is probably clear by now, there is a resounding groundswell of opposition to implementation of default critical peak pricing tariffs for Summer 2005 by customer representatives, even those who would see lower bills as a result of adoption of these rates. Customers whose bills would increase because of the design of the rates will have little ability to mitigate their increased bills even with significant reduction in their usage. By narrowing the applicability of the rates to non-agricultural customers between 200 and 500 kW in Summer 2005, we significantly reduce the potential demand reduction achievable, even assuming that customers are sufficiently educated to make demand reductions. For these reasons, it appears that implementation of the proposed rates would not accomplish sufficient demand reduction this summer to justify the expected implementation costs of $10.45 million or the disruption to customers. Therefore, we will not adopt a default critical peak pricing tariff for Summer 2005.

We also note that these proceedings provided interesting information regarding the contributions to system peak that indicate that the largest customers may not have the most discretionary load to remove from peak. For example, CLECA cited a California Energy Commission (CEC) report⁴⁰ that identifies that the summer peak is driven by air conditioning load (29% of summer peak load) with another 11% from commercial lighting. Industrial process load represented 5% of summer peak load. EPUC also presented historical load profiles of PG&E's residential, E-19, and E-20 customer classes (reproduced below) that make clear that the residential class places more load on peak than the largest customers. We provide these facts simply for information, without drawing any conclusions about the value that different customers place on their energy usage on peak, which will drive their elasticity of demand and therefore their response to critical peak prices. However, it does indicate that to achieve significant demand response during the critical peak, we will need to place special emphasis on reaching air conditioning load, which drives 29% of the peak load, whether through pricing or other types of programs.

(EPUC Opening Brief, p. 11.)