VIII. Test Year Forecast Methodology
SoCalGas and SDG&E used the following method to estimate Test Year 2004 expenses: For both SoCalGas and SDG&E, a Base Year 2001 of recorded data was identified and then adjusted for known downward changes for one-time or non-recurring expenses; the residual Base Year was then escalated for inflation in 2002, 2003 and Test Year 2004. To this adjusted base, SoCalGas and SDG&E added any forecast for new activities begun in 2002, 2003 or forecast for 2004. This addition was in 2004 dollars. The escalated base year with the two adjustments for changes formed the basis for the test year estimate. SoCalGas and SDG&E then provided testimony to justify the level of activity and the purpose of the activity represented by the estimate. This is similar to a "budget-based" method that the Commission found to be reasonable if properly applied (see Southern California Edison, 64 CPUC 2d 241, 316 (D.96-01-001) or California Water Service Company D.03-09-021, mimeo., pp. 35 and 36.) As discussed in detail elsewhere, this methodology is a reasonable starting-point to meet the burden of proof for SoCalGas and SDG&E Test Year 2004 expense estimates. Where SoCalGas and SDG&E used a different method to forecast Test Year 2004 they provided a specific explanation and offered a justification for that method. The appropriateness of this method depends upon a thorough review of the supporting data.
ORA was less consistent in its methodology, and as discussed in detail where applicable, appeared to select a method based on outcome (i.e., a lower estimate) instead of looking for the most reasonable and consistent methodology to critique SoCalGas and SDG&E. In some instances, ORA used a three-year or seven-year trend of recorded costs without adequately distinguishing why a different method was superior or SoCalGas and SDG&E were using an inferior method. We find it inappropriate and unreasonable for ORA to pick and chose between forecast methodologies in order simply to recommend a lower estimate. Without shifting the ultimate burden of proof from SoCalGas and SDG&E, we note that ORA must meet its burden of producing evidence counter to that of applicants that is sufficient to support the adoption of the ORA position. ORA and all other intervenors are expected and encouraged to point out the defects they see in the applicants' methods and supporting data. Selection based on mere outcome is not supportable. The ratepayers' best interest is only served when the adopted test year estimate is adequate to allow the utility to provide the appropriate level of safe and reliable service.
TURN raised what it believed were serious concerns about the SoCalGas and SDG&E forecast methodology, including its development of Base Year 2001 costs. It cited the large number of adjustments and reallocations made by SoCalGas and SDG&E41 and pointed out that recorded 2002 data, available late in the proceeding, were different than the forecast 2002 expenses, one step towards estimating Test Year 2004. But we decline to constantly shift between the methods as filed and selective use of 2002 recorded data. We see this as a problem inherent in parties not having the opportunity to review an NOI, compounded by the overlap of other major rate cases. As we consider the test year estimates in this decision, we used the best information provided by the parties in the record and we add whatever safeguards we deem appropriate to the specific circumstances.
UCAN made similar objections to the Base Year 2001 and subsequent forecast method used by SDG&E; again, we will work with the data we have available. UCAN also argued that the rate application as filed is an inadequate base for rates to be in effect for five years.42
41 TURN opening litigation brief, pp. 9-13.
42 UCAN opening litigation brief, p. 42.
