7. If Ratepayer Funding of Pre-Deployment Activities is Allowed, What is the Proper Scope of Pre-Deployment Activities for Funding?

How "pre-deployment" is and should be defined is an important consideration in this case. Because PG&E has already performed its system requirements analysis, request for proposals, evaluation of bidders, and conducted its due diligence, PG&E defines pre-deployment differently than San Diego Gas & Electric Company and Southern California Edison Company who also filed applications for AMI "pre-deployment" funds in March 2005. Because PG&E is farther along in its internal decision making process than the other utilities and believes its proposed investment to be essentially cost-effective on the basis of operational benefits, the activities it has defined as pre-deployment consist of preparing its existing legacy systems to accept data from its proposed AMI system, establishing and testing processes for meter and communication system installation and billing. PG&E witness Corey describes the funding that it requests as "specifically to test the incremental benefits of the technology, and to test the end-to-end data integrity, and to test the installation processes and the other business processes associated with deployment [of AMI]." (RT 75:22-25.) Likewise, witness Vahlstrom describes the activities to be performed during PG&E's defined pre-deployment period as follows:

"PG&E has long said that the intention of this test is not strictly to test the viability of the technology, but the processes of installing it and that all the hookups can be made properly through all of its systems, and that it will bill properly." (RT 54:18-22.)

PG&E's pre-deployment activities do not consist of a significant amount of testing that the AMI system selected physically works or meets the minimum functionality criteria. Under cross examination by ORA, PG&E witness Vahlstrom stated: "I would say the costs represented by meter deployment [testing] is a small percentage of the [$] 49 million [requested]." (RT 59:12-13.) Meter testing and functionality testing is an activity that both ORA and TURN argue is more appropriate for funding prior to a Commission determination that pursuing AMI is cost-effective for ratepayers. In essence, ORA and TURN argue that pre-deployment costs should, at most, be limited to costs associated with physical testing and minimum functionality analysis.

PG&E argues that the scope of pre-deployment activities it has proposed is appropriate and critical to allow PG&E to be in the position of having "a viable demand response tool in the summer of 2007," which it argues is required by the Commission's demand response goals. (PG&E Brief, p. 1.) PG&E states that the Commission has a clearly articulated goal of implementing the capability for customers to respond to price signals by 2007, citing the vision statement described in D.03-06-032, and that to meet that goal, it must complete its systems integration and testing work that it describes as its pre-deployment activities as promptly as possible. "PG&E believes there is a strong likelihood that the Commission will permit PG&E to move forward with its overall AMI Project. For these reason [sic] and others, there is virtually no risk associated with authorizing PG&E's pre-deployment funding request." (PG&E Brief, p. 5.)

TURN, on the other hand, argues that PG&E's justification regarding meeting the 2007 demand response goals "is unsupportable as a rationale for the level of spending sought for pre-deployment costs." TURN argues that in other proceedings (specifically A.05-06-006, PG&E's 2006-2008 demand response program application) that PG&E is seeking modification of the demand response goals the Commission has established in order to account for the lack of residential participation in demand response programs. TURN also points out that PG&E has adequate resources to meet its 2007 demand, it justifies its AMI deployment plans primarily on the basis of operational savings, not demand response savings, and modifying the timeline of deployment would not impact the program costs. (TURN Brief, p. 9.) Finally, TURN notes that even if PG&E's aggressive timeline were pursued, very little meter deployment would actually occur by 2007, limiting the demand response that could occur by 2007.

ORA also takes issue with PG&E's position that the Commission has indicated a "mandate" for the utilities to deploy AMI by 2007. ORA notes that the Vision Statement language relied on by PG&E explicitly does not prejudge the cost-effectiveness of specific proposals and states it should be viewed as a starting point. (ORA Brief, p. 5.)

7.1 Discussion

Based on the record developed, we agree with TURN and ORA that the investments that PG&E has identified as pre-deployment activities are activities that, for the most part, provide limited value to ratepayers if the Commission ultimately decides AMI should not be pursued. These are activities that would be necessitated only if PG&E deploys a new metering and communication infrastructure. As described in its Reply Brief PG&E's "pre-deployment is not simply analysis or research and development. PG&E's pre-deployment activities are an integral part of its actual AMI Project." (PG&E Reply Brief, p. 3.) A small portion ($3 million) of the identified funding is to install and test 5,000 meters, which will include testing that enhancements of the installed meters meet PG&E's requirements. (RT 80:2-9.)

When asked by the ALJ whether PG&E expects "any system benefits associated with the pre-deployment expenditures, independent of installation of AMI", PG&E witness Corey responded:

"a certain component of the pre-deployment costs are going to the [interval] billing capability, which is the deployment of the 1.5 [CorDaptix] system upgrade for our billing system.

And even though we might not pursue that at this time [absent the AMI project], we would likely...pursue that at some later time...outside the AMI project." (RT 121:4-14.)

Thus, PG&E identifies only one component of its proposed pre-deployment activities that it would pursue, independent of the decision to deploy AMI. The remainder of the proposed pre-deployment activities relate to developing the system for processing meter data into a format to render bills, preparing and installing the hardware and software to send and receive data from the AMI system, developing interfaces between PG&E legacy systems and the AMI system, system design work to protect customers service and billing accuracy while fulfilling the benefits of the system, and project management, reengineering of internal processes and training. These are clearly activities that must be undertaken to implement AMI; however, the question is whether it is reasonable for the Commission to authorize ratepayer funding for these activities prior to a finding that ratepayers should fund PG&E's AMI Project.

7.2 Conclusion

We conclude that it is only reasonable at this time to authorize ratepayer funding for a narrow set of activities: 1) those that provide value to ratepayers regardless of whether the Commission ultimately decides that ratepayers should fund the AMI project, a decision that will be rendered in A.05-06-028, and 2) activities designed to test minimum meter functionality, prior to committing to full scale deployment are reasonable for ratepayers to fund, given the past direction to explore the cost-effectiveness of installing AMI.

This is not to say that the costs or activities PG&E has proposed within its definition of pre-deployment are unreasonable per se; in fact, PG&E has presented a compelling case that these activities are necessary to any AMI project's success. Rather, that it would be unreasonable for the Commission to commit ratepayer funding to these activities at this time without having a record on the full project. Despite PG&E's statements that it has filed its evolving business case several times in R.02-06-001, there has been no record developed yet through the evidentiary process to test, rebut, or verify PG&E's assumptions or results. Because the AMI Project cost-effectiveness is highly dependent on the assumptions about costs, benefits, energy and capacity prices, and demand response, an evidentiary record is needed in order to ensure that the Commission has a sufficient basis to establish cost-effectiveness. While we do not doubt that PG&E believes it has presented an accurate picture of the costs and benefits for its ratepayers, the Commission has not yet made a finding that it is cost-effective for ratepayers to fund. We note that in general, cost recovery for capital additions occurs through general rate cases, as confirmed by PG&E witness Bottorff, so the fact that we do not pre-approve ratepayer cost recovery at this time does not foreclose PG&E from pursuing these activities. In fact, PG&E already has authorization to record costs associated with all of these activities in a memorandum account, which further preserves its ability to seek future cost recovery.