5. Discussion

We commend the Standardization Team for conducting a comprehensive analysis of measure cost-effectiveness and NGAT procedures-efforts that represents the first of their kind for the LIEE programs in California. In addition, we commend the Team for its responsiveness to public workshop and written input, which undoubtedly contributed to the fact that the final reports were uncontested. As described in Section 2.1 and in the workshop report itself, the public had several opportunities to address the Team's approach to analysis and recommendations, and the Team was responsive to those comments.

5.1. LIEE Measure Cost-Effectiveness

We have carefully reviewed the Team's recommendations regarding which LIEE measures to continue to offer, and which to drop for PY2004 based on cost-effectiveness considerations. We find that the Team has applied the cost-effectiveness guidelines adopted by the Commission in D.02-08-034 consistently, and reasonably. In those instances where the Team recommends retaining measures that do not meet the specific cost-effectiveness thresholds established by those guidelines, the Team has offered persuasive reasons for doing so based on policy or program considerations. In particular, we share the Team's concern, also voiced by interested parties during workshops, that a strict adherence to cost-effectiveness results would eliminate several infiltration-reducing measures without sufficient consideration of the potential understatement of benefits and overstatement of costs associated with the analysis. We will adopt the Team's recommendations for PY2004, and until further order by the Commission.

At the same time, we believe that further work is needed to improve our capability to assess the benefits and costs of specific program measures and the LIEE program overall. As we discussed in D.02-08-034, program measurement issues should be revisited periodically:

"To this end, we will initiate an examination of savings measurement issues for the LIEE program sometime during 2003, as time and resources permit. We will examine the utilities' current methods for estimating energy efficiency program and measure savings, as well as the types and frequencies of the utility's measurement studies. Our consideration of these issues will need to be coordinated with the AEAP.... We delegate to the Assigned Commissioner the task of developing the scope and schedule for this review as a separate phase of this rulemaking, in conjunction with the PY2004 LIEE program planning process or by other means (i.e., a new proceeding), as appropriate."³⁷

We will be embarking on the PY2005 LIEE program planning cycle in the coming months. We encourage the Assigned Commissioner, in consultation with Energy Division and the assigned administrative law judge (ALJ), to initiate this examination at the earliest opportunity. As directed in D.02-08-034, the examination of methods for estimating energy efficiency program and measure savings should be coordinated with the AEAP.

In addition, we note that the Team's study (Appendix B) indicates sizable disparities across utilities in the cost for certain measures installed in the same climate zone. Further examination of the utilities' estimating methods should seek to explain these disparities more fully. To this end, the utilities should submit a per measure cost break-down of materials, labor, administrative and travel (or "windshield time") costs as part of their PY2005 LIEE program plan applications. The submittal should include a joint explanation of the reasons for these disparities, with joint recommendations for further evaluation work, and adjustments to the manner in which administrative costs and travel are allocated across measures, as appropriate. As a model for this explanation, the utilities should refer to the Bill Savings Standardization Report, which analyzes in some detail the causes for the variability in overall program cost-effectiveness across utilities.³⁸

Despite the fact that further examination of LIEE measure savings (and costs) is warranted, we are satisfied that the cost-effectiveness results presented in the Team's report are reasonable for the purpose of modifying current LIEE program design at this time. The thoroughness of the report itself and responsiveness to public input, as discussed above, persuades us that the Team prepared its recommendations using reasonable assumptions and in full recognition of the limits inherent in any attempt to quantify the cost-effectiveness of specific measures installed within a comprehensive LIEE program. Moreover, we note that the Team used the results of a load impact study that also underwent considerable public scrutiny, as described in Attachment 5. In sum, we find the Team's recommendations regarding which measures to include in PY2004 LIEE program plans reasonable, and will adopt them until further Commission order.

5.2. Natural Gas Appliance Testing

As the Team acknowledged in the NGAT report, the issues related to CO testing are complicated and the empirical evidence is not always conclusive. In particular, the empirical evidence does not clearly indicate that installing infiltration-reduction measures will increase CO levels in the home. Since this is the major hypothesis underlying the policy for testing CO levels in LIEE-eligible homes, the results of the NGAT study call into question whether such testing is necessary at all.

Nonetheless, we agree with the Team that some type of CO testing is warranted for the LIEE program for the following reasons:

_ Some homes covered by the NGAT survey were measured to have ambient CO levels above at least some of the current standards and thresholds.

_ The LIEE Program is unique in that it entails the provision of a comprehensive set of energy-efficiency measures spanning both building envelope infiltration reduction and appliance repair and replacement. Given this comprehensive treatment, the potential for adversely affecting CO levels is greater than in other programs.

_ Low-income households are generally less financially capable of maintaining their natural gas appliances, and may be less likely to know that gas appliance testing and services are available at no cost from their local gas utility.

In terms of timing, we concur with the Team's recommendation that CO testing be conducted only after weatherization. Conducting CO tests both pre- and post-weatherization would, in our view, be excessive in light of the study findings. By conducting these tests after weatherization, we can at least guard against the possibility that infiltration-reduction measures do have some influence on drafting or longer-term concentrations of CO in the home. We also agree with the Team's recommendation that preliminary combustion air ventilation evaluations be conducted as part of the initial home assessment. This would identify those instances, which the Team acknowledges are rare but potentially very costly, where major work to add combustion air venting is required to pass inspection.³⁹ In those instances, infiltration-reduction measures would not be installed under the program.

With respect to further standardizing the CO testing procedures, we find that the Team generally proposes a very sensible approach. In those instances where the NGAT survey data points to areas of improvement in the Minimum Standard, the Team has recommended modifications or additions to that Standard, across all utilities. For example, the new procedures expand room ambient CO testing to take into account not only space heating appliances, but water heating and kitchen appliances as well. (See Attachment 4.) Where the NGAT survey has indicated that certain assessments are superfluous (e.g., ambient or flue tests for dryers), the Team has eliminated such procedures. We also concur with the Team's assessment that CO alarms not be used in combustion appliance testing at this time.

However, we are not persuaded that the decision to conduct flue tests or room ambient tests for space and water heating should be left to the utility's discretion, as the Team recommends.⁴⁰ This approach was proposed because the individual utilities could not agree on the relative importance of flue tests. For the utility that conducts flue tests as the standard practice, this means that many customers' appliances would be "red-tagged" (disconnected) pending repair or replacement, even when CO levels in the room are well within safety thresholds and all appliances pass the draft tests. This is because there are far more test "fails" at the flue level, than in the room ambient air. Under the Team's recommendation, flue testing for CO levels could become standard procedure at the sole discretion of the utility--even though the NGAT study does not yield clear evidence that weatherization even increases CO levels in the home.

Proponents of flue testing of CO levels argue that appliances that fail this test may present a problem at a later date if venting problems occur. However, the Team presents no empirical evidence or research to indicate whether, or to what degree, CO levels in the flue that exceed certain thresholds may present health problems in the future. Implicit in this argument is the presumption that CO testing procedures under the LIEE program should guard against any future potential safety problem associated with combustion appliances in the home. As the Team acknowledges in its report, NGAT procedures cannot make such guarantees. NGAT procedures are designed to evaluate CO levels and test for adequate drafting at a particular point in time. These conditions may change over time due to a variety of unpredictable factors.⁴¹

For the reasons discussed above, we believe that the results of the NGAT study support some continued testing of CO levels in the home after weatherization. However, we find nothing in those results to warrant the use of flue tests as the standard procedure for assessing space and water heating appliances, as allowed for under the Team's proposed testing protocols. Instead, we will adopt the room ambient CO tests and the other standard testing procedures (olfactory and visual tests, smoke draft tests, combustion air evaluation) proposed by the Team. If CO levels in a particular room (or rooms) are at or above the 10 ppm action level, then the utility may (but is not required to) conduct flue tests as one diagnostic tool to identify the source(s) of the problem, as appropriate. This is the current manner in which SDG&E and SoCalGas utilize flue tests in combination with ambient tests in their NGAT procedures, and we adopt this approach for all utilities. For the purpose of conducting this secondary flue CO testing, the utilities will need to utilize standardized protocols and threshold levels. As indicated in Table 3-2 of the NGAT final report, there is a wide disparity between the threshold that PG&E utilizes to test flue CO levels (i.e., a consistent 100 ppm for each appliance) and the threshold utilized by SDG&E and SoCalGas (i.e., one that vary between 25 and 275 ppm depending upon the appliance). We direct the Standardization Team to report back to the Commission within 90 days with recommended flue CO threshold levels that are consistent across the utilities. For now, the utilities may continue to use their current flue CO testing thresholds when they conduct flue tests as a diagnostic tool under today's adopted NGAT procedures. ⁴²

We now turn to the Team's recommendation regarding the treatment of homes that contain a non-IOU fueled combustion appliance. Under this recommendation, the utility would no longer provide infiltration-reduction measures to any eligible low-income home (one that utilizes a IOU-fuel for space heating) if it also uses a non-IOU combustion fuel for one or more other end use. For example, a home with electric heat and electric cooking appliances would not receive infiltration-reduction measures (e.g., weatherstripping and caulking) under the LIEE program if it also uses a propane water heater.

Currently, PG&E performs CO tests on propane and other non-natural gas combustion appliances in these customer homes that could qualify for infiltration-reduction measures (i.e., that utilize electric heat). If a CO-problem associated with non-natural gas appliances is identified, the customer is referred to the supplier and receives only non-infiltration reduction LIEE Program measures. Otherwise, PG&E will provide infiltration-reduction measures to these homes. SoCal and SDG&E currently test ambient CO levels after weatherization with only the natural gas furnace/space heater operating, consistent with the current Minimum Standard. Under this testing approach, infiltration-reduction measures are installed irrespective of the fuels used by non-space heating appliances. Attachment 6 summarizes how the Team's recommendation would modify current practices with respect to the provision of infiltration-reduction measures, depending on appliance fuel-use.⁴³

Clearly, some number of customers in the utilities' service areas that are currently eligible for infiltration-reduction measures under the LIEE program would no longer be eligible for these measures under the proposed policy. However, the report does not provide sufficient information to allow us to assess the relative impact of the Team's proposal on program services, or to fully consider policy options.

At the request of the assigned Administrative Law Judge, the utilities provided supplemental information on this issue. The utilities provided estimates of the percentage of all residential homes in their service territories that use an IOU space heating fuel but use a non-IOU fuel for another end-use (e.g., cooking, water heat, pool heater). These estimates are presented in Attachment 6.

PG&E estimates that 5-7% of all its customers fall into this category, but states that the percentage in rural ZIP codes is probably higher. Based on LIEE program data for PY2002, PG&E estimates that 16.3% of LIEE participating customers in rural ZIP codes without PG&E gas accounts had propane appliances. (See Attachment 6.) In addition, PG&E estimates that over 20,000 customers would fall into this category in just the eleven rural counties of Alpine, Amador, Calaveras, El Dorado, Lake, Lassen, Mariposa, Plumas, Sierra, Trinity and Toulumne.⁴⁴

SoCalGas, SCE and SDG&E only provide estimates based on the entire population, stating that information on the low-income population in their service territories was not available. They estimate that approximately 1% or less of their entire population use an IOU fuel for space heating and a non-IOU combustion fuel for one or more other end-uses. (See Attachment 6.)

At least for PG&E, the data available to date indicates that the impact of this new policy would be significant over time in rural areas, where many electric customers do not have natural gas available for space heating, and therefore use electric heating and non natural gas combustion fuels (e.g., propane) for water heating and other end uses. Given this impact, we believe that it is prudent to obtain additional information and explore alternative options, with input from the Low Income Oversight Board, community-based organizations and other interested parties, before finalizing our policy on this issue.

First, we need better information on the relative impact of the Team's proposed policy regarding non-IOU fuels on the LIEE program in the Southern California service areas, and a presentation of data on a more consistent basis across utilities. In addition to the percentage of eligible LIEE homes that fall under this category, we need to see what percentage of homes actually treated over the last few years would have been affected by this policy, i.e., use IOU-fuel for space heating and non-IOU combustion fuels for other end uses. This will give us an idea of the potential impact in terms of annual program numbers, by service territory. The information to date indicates that the vast majority of the impact would be in rural counties, but we need additional data to support this assumption.

Second, we need to explore alternative policies in greater depth than presented in the Team report. In doing so, we need to carefully take into account both the study results with regard to the health effects of infiltration-reduction measures as well as the impacts of each alternative on meeting the needs of low-income customers. For example, the Team dismisses one option that the Southern California IOUs have implemented for many years and that is also consistent with today's determination to only conduct CO testing post-weatherization. This is the option of restricting NGAT assessments to IOU natural gas appliances, but installing infiltration reduction measures in homes with other combustion appliances.⁴⁵ We believe that this option warrants further consideration in light of the NGAT study results.

We also want to explore alternatives that would tailor the Team's proposed policy to a more specific set of circumstances, depending upon the type and/or location of the non-IOU fueled appliance. For example, infiltration-reduction measures might still be offered to homes with electric space heating and kitchen appliances, as long as the appliance that uses non-IOU combustion fuel (e.g., gas water heater) is located in a mudroom, garage, or other area that is unlikely to affect room ambient CO levels. In considering ways to fine-tune this policy, we will need information on the location of such appliances in the home affected by the Team's proposal, based on survey data and/or experience in the field.

Finally, we want to explore a leveraging option in conjunction with the Team's recommendation for homes that take service from the IOU for space heating, but use non-IOU combustion fuels for other end-uses. In its report, the Team recommends that these homes be referred to LIHEAP or non-IOU gas utilities for combustion appliance testing and the installation of infiltration-reduction measures (if they pass those tests). As we understand the Team's proposal, LIEE funding would no longer be used to provide those homes with infiltration-reduction measures.

This represents a significant decrease in LIEE-funded services to homes that are currently receiving such services from the program and contribute to the costs of LIEE through the public goods charge on their electric bills.⁴⁶ According to PG&E, homes that pass the CO tests receive approximately eight measures on average through the LIEE program. Under the Team's proposal, these same homes would only receive (on average) approximately three non-infiltration-reduction measures.⁴⁷ In effect, the Team's proposal would shift the costs of those additional (on average) five measures per home from LIEE to the LIHEAP program or non-IOU gas utility.

While we concur with the Team that IOU ratepayers should not be responsible for conducting safety tests on non-IOU fueled appliances, or bear the costs of replacing or repairing those appliances if they do not pass the tests, we are not persuaded that the costs of installing infiltration-reduction measures in those homes should no longer be funded out of the LIEE program. As noted above, these low-income households use IOU-fuels for space heating (generally electric heat), and often also for air conditioning, and contribute to the cost of the LIEE program through the public goods charge. Therefore, we believe it is appropriate to explore an approach that would use LIEE funding to "leverage" the provision of infiltration-reduction measures to these homes by LIHEAP or non-IOU gas utilities. Under this leveraging approach, LIEE funds would be used to pay for the LIEE infiltration-reduction measures that are feasible in these homes, provided that they pass either the IOU, LIHEAP or non-IOU gas utilities' safety testing procedures for combustion appliances. The infiltration-reduction measures would be installed by the non-IOU gas utility or under LIHEAP, as the Team proposes, but in addition the IOU would reimburse the non-IOU gas utility or LIHEAP on a per measure basis from LIEE program funds. The reimbursements would be contingent upon the receipt of appropriate documentation that the home passed the LIHEAP or non-IOU gas utility's combustion appliance tests, and could be subject to the same pre- and post- inspection procedures the IOUs currently implement to monitor their LIEE contractors.

Coupled with this form of leveraging, the Team's proposal would continue allocate LIEE funding for infiltration-reduction measures to low-income homes based on their space heating fuel, as is currently done, rather than based on the fuels used in the home for other end uses. At the same time, this approach provides for combustion appliance safety testing by entities other then the IOUs for homes that have appliances that use propane, kerosene or wood (or non-IOU natural gas), as proposed by the Team. We think that this leveraging approach has appeal, and should be explored as a potential modification to the Team's proposal.

These and other alternatives to the Team's proposal regarding the treatment of homes that contain a non-IOU fueled combustion appliance should be explored as a further task for Phase 4 of the Standardization Project. Consistent with prior Commission direction regarding the Standardization Project, the Assigned Commissioner shall direct this task with respect to the scope of work, budget and schedule.⁴⁸

In order to obtain broad input from the low-income community on this issue, Energy Division should schedule and notice a presentation by Team consultants to the Low Income Oversight Board (LIOB), with notice to all parties in this proceeding and to all community-based organizations (CBOs) that currently participate in the LIEE programs. The utilities should provide the additional data discussed above for inclusion in the presentation, along with any additional information that Energy Division deems appropriate. The presentation may occur at a regularly scheduled LIOB meeting, a workshop, or other public forum. It may involve a single meeting, or multiple meetings. The Team consultants should include a summary of the comments and recommendations of the LIOB, participating CBOs and other interested parties in their supplemental Phase 4 report, along with Team recommendations. The report should include a discussion of the pros and cons of each policy alternative, along with the estimated impact on services to low-income customers relative to current policy.

We emphasize that the public discussion and the supplemental report should address the treatment of homes that use non-IOU combustion fuels in the context of the NGAT policies and procedures that we adopt today, e.g., post-weatherization testing only, room ambient CO testing as the standard procedure, rather than flue testing, etc. We will not consider further argument or proposed modifications on these issues. In order to provide sufficient time for this effort, the final report on this issue shall be due 120 days from the effective date of this decision. Comments on the report will be due 30 days after the report is issued, and reply comments are due 15 days thereafter.

In the meantime, the utilities should implement the NGAT assessment procedures adopted in this decision for all homes that use IOU fuels for space heating and other end uses. These procedures are summarized in Attachment 7. For homes that utilize an IOU-fuel for space heating, but non-IOU fuels for other end uses, we authorize the utilities to continue their current CO testing procedures to qualify these homes for infiltration-reduction measures until we finalize our policy in the coming months. For SoCalGas and SDG&E, this means that they would continue to conduct ambient CO tests for these homes based on the current Minimum Standard, but in all other respects incorporate the new NGAT procedures we adopt today. (See Attachment 7.) PG&E would continue to conduct both pre- and post-weatherization CO testing of these homes using its current testing procedures, until further notice.

When appliances are found to have problems, the utility should take the actions recommended by the Team, as described in Section 4.8.2. We agree with the Team's assessment that natural gas water heaters failing one or more of the tests should be repaired or replaced, as is the current practice for natural gas space heaters. We recognize that this will increase program expenditures on repairs/replacements, relative to the status quo (by an estimated $1.9 million in 2003 for the three natural gas utilities). However, now that we are requiring more extensive tests (e.g., ambient CO testing) for water heaters using natural gas, it makes no sense to treat these appliances differently from space heaters in terms of authorized replacements. The continued operation of both types of natural gas appliances (water and space heating) are important to the well-being of low-income residents and should be replaced, rather than shut off, by the natural gas utility if they are tested and found not to be functioning safely, and cannot be repaired. We note that this change will not increase LIEE program costs overall for 2004, in light of our decision to eliminate certain measures from the program.

Today's decision describes the general policies and procedures we adopt for NGAT. As discussed above, we have adopted final policies and procedures for homes that utilize IOU-fuels, and have adopted interim NGAT procedures for homes that utilize non-IOU fuels until we finalize our policy with respect to offering infiltration-reduction measures to those homes in the coming months. Translating these policies and procedures into specific instructions will require some further work by the Team. For example, the addition of ambient CO testing for cook tops will require a write-up of specific protocols or instructions on how far the instrumentation should be placed from the cook top, the number of minutes the cook top should operate before measurements are taken, etc.

We direct the Standardization Team to complete the documentation of CO testing protocols and revisions to the Policy and Procedures Manual to reflect today's decision, and to submit this material as a compliance filing within 20 days. The filing should include protocols for the preliminary combustion air ventilation evaluations discussed above. Protests and comments will be due 10 days thereafter, and the Team should respond to protests or comments within 5 working days. If there are no protests to the Team's compliance filing, then the Assigned Commissioner or Administrative Law Judge, in consultation with Energy Division, may issue a ruling approving the testing protocols and revisions to the manual without further action by the Commission. Changes to NGAT testing procedures in the field will need to await approval of the protocols and manual revisions by ruling or Commission decision.

As the Team notes, one of the objectives of the NGAT study was to support the standardization of pre-approvals of the installation of LIEE Program measures. The Team's recommendation on CO testing essentially decouples measure pre-approvals from CO testing, and this is a major step in the direction of standardization. However, the Team did not make recommendations with respect to any other changes to the utilities' pre-approval procedures. Instead, the Team presents the utilities' rationales for differing pre-approval procedures in the final report, and recommends that they be permitted to retain their current option of conducting appliance assessments in-house or contracting with third parties to provide these services.⁴⁹

We are not prepared to conclude our standardization efforts without further consideration of whether pre-approval practices or other LIEE policies or procedures should be made more consistent across utilities. Within 90 days from the effective date of this decision, the utilities and interested parties should comment on what additional areas of standardization in LIEE procedures or policies should be addressed under the Standardization Project in the future, if any. The comments should discuss the reasons for (or against) further standardization. Reply comments are due 30 days thereafter. We will establish the scope of further standardization efforts, as appropriate, by subsequent Commission decision.

Finally, with respect to PY2004 program funding levels, we note that the utilities are currently authorized to recover in rates the following amounts for LIEE activities, per D.02-12-019: ⁵⁰

PG&E: $56,530,000

SCE: $15,893,500

SoCal: $34,521,502

SDG&E: $13,368,093

We make no changes to these authorizations for PY2004. Overall, the cost-effectiveness policies and NGAT procedures adopted today do not suggest that significant changes to current LIEE funding levels are warranted. Consistent with our direction in D.02-12-019, any unexpended LIEE PY2004 program funds shall be carried over and made available for funding LIEE program activities in PY2005 or subsequent years as an augmentation to the amounts authorized above.⁵¹ In their PY2005 program applications, due on July 1, 2004, the utilities should present estimates of PY2005 program costs based on the measures authorized today, along with year-to-date LIEE expenditures and estimates of any carryover funds, by budget category.