12. Eligible Population for LIEE
12.1. Introduction
The Plan and D.07-12-051 set forth the following programmatic initiative:
By 2020, 100 percent of eligible and willing customers will have received all cost effective Low income Energy Efficiency measures.
In order to obtain this vision, the IOUs must reach approximately 25% of all eligible and willing households in 2009-11. The assigned Commissioner issued a guidance document for the IOU budget applications on April 1, 2008, asking the IOUs to calculate the number of eligible and willing LIEE customers based on utility's proposed "standard means of deriving the number of LIEE customers on which to base 1/4 of the Commission's programmatic initiative."83
In their budget applications, the IOUs proposed a joint methodology designed to obtain a base point of eligible and willing customers for each utility's service area. This base point represents the population to be reached in order to meet the 2020 goal, 25% of which should be served in the next budget cycle. The proposed methodologies for calculating the base point were basically the same for each utility, with one specific difference. The methodology used to calculate the base point for the programmatic initiative can be simplified as follows:
A: Total number of households estimated eligible for LIEE Program in 2008
B: Total number of households treated by LIEE from 2002 - 2008 (actual plus estimate for 2008)
C: Total number of households treated by LIHEAP from 2002 - 2008 (actual plus estimate for 2008)
D: 10% of A who are estimated as unwilling to participate
A - B - C - D = Base point for calculating 25% of the programmatic initiative, as proposed by SDG&E and SoCalGas
A - D - B - C = Base point for calculating 25% of the programmatic initiative, as proposed by PG&E and SCE84
Step A of the IOU proposed methodology includes obtaining an estimate of the overall eligible population. The IOUs estimated the number of CARE eligible customers, using a methodology developed in D.01-03-028 and designed to be used as an interim methodology until 2000 Census data became available. Using the 1990 Census as a foundation for household size and income estimates, the methodology included data adjusted for growth in households and income using other sources. Based on the methodology outlined in D.01-03-028, the IOUs presented an overall population estimate for 2009 of 5.404 million customers.
Step B of the proposed methodology involves subtracting the number of households that have been served by the LIEE program between 2002 and 2008. The IOUs selected this time frame based on a reexamination of the "10 Year Go Back" rule, in which the utilities do not serve households that have been treated within the last decade. However, given the Commission's directive to reconsider this rule, the utilities concluded that households served after 2001 should not be included in the base point estimate. The year 2002 marked an increase in measures included as part of the LIEE program, as authorized in D.01-05-033. The utilities state that households treated from 2002-08 received similar measures to households to be treated in the upcoming budget cycle and should not be included as part of the base point population estimate.
Step C includes estimating the number of households treated by the LIHEAP from 2002 through 2008. The joint utilities claim that customers served by this federal program should not be included as part of the households to be treated in order to reach the 2020 goal. LIHEAP offers a majority of the measures available through LIEE, if not more measures. For this reason, homes treated by LIHEAP are determined to be ineligible for service under LIEE and should not be included in the base point calculation, the utilities assert.
In Step D, the utilities propose subtracting the number of customers who are likely to be unwilling to participate in LIEE. The utilities estimate this number to be 10% of the eligible population, based on the KEMA Report, which states, "Using the information collected through the onsite survey, we estimated that 10 percent of all low income households would be unwilling or unlikely to participate in CARE."
Based on this methodology, the final IOU population estimates are as follows:
The IOUs recognize that the estimate needs to be better developed by tracking the number of homes unwilling to participate. In doing so, they plan to track this information based on two sub-components. The first sub-component includes customers who affirmatively reject LIEE program participation. The second sub-component includes customers who have been contacted on multiple occasions but do not indicate whether or not they would like to participate in the LIEE program. The utilities question at what point to consider such a customer as "unwilling."
12.2. Parties' Positions
DRA makes several claims that the utilities are underestimating the willing and eligible population. Based on various revisions to the proposed methodology for calculating the programmatic initiative, DRA recommends that the utilities treat 1,135,189 households over the next budget cycle, 2009-11. In contrast, the utilities plan to treat 928,729 households in the same period.
According to Greenlining, the proposed methodology contains "fuzzy math" and could potentially underestimate the base point. Greenlining recommends that the adopted methodology be as simple as possible.
Beginning with the eligible population estimates, represented as A, DRA points out that the utilities' estimate of 5.404 million for 2009 is lower than other estimates. Specifically, DRA cites the estimated population of 5.633 million households for 2006 from the KEMA Report. DRA does not believe that the eligible low income population will "shrink" over time, as the utilities seem to suggest, given the current economic downturn and population growth likely to have occurred from 2006-09. DRA states that a record number of California customers are in arrears and that various state agencies are escalating services for the low income community to confront the expected increase in this population.
DRA also questions the constancy of the utilities' population estimates from 2009-11. As part of the eligible population estimate, DRA believes that the utilities should include a projection of population growth from year to year, which would serve as a better estimate of the eligible population and allow for the inclusion of low income households in the program's target market. Specifically, DRA recommends an annual growth rate of 1% to take population growth into consideration. Accordingly, this conservative figure is based on the average change in population claimed by the IOUs between 2005 and 2008.
To further justify these modifications, DRA submitted an email sent by John Peterson, a consultant who provides the IOUs with annual estimates of the CARE and LIEE eligible population. Peterson sent an email on June 22 to various utility representatives, stating "I wish that we had used 5.63 million and a growth rate, but I understand the reasons for the approach taken."
PG&E stated that it does not know the reasons for the decrease in the eligible population but offers the following:
The disjunction between a given year's CPUC guideline setting based on previous years' national CPI estimates and the current year's empirical midyear CPI from the California Department of Finance; the methods and sources used by the vendor supplying current year small area (block group) marginal distributions on household income, household size, and householder age; and that 2007 was the first year in which Census PUMS data was supplemented with accumulated 2005-2006 American Community Survey estimates, a method that should generate more robust partners.
SoCalGas and SDG&E claim they did not use the population estimate of 5.633 million from the KEMA Report because the study contained inaccuracies and has not been authorized for use by the Commission. In contrast, they claim, the Commission endorsed the methodology presented in D.01-03-028. In addition, the use of the methodology from D.01-03-028 allowed for the inclusion of more recent and relevant data than that which was used in the KEMA Low Income Needs Assessment, according to the utilities.
PG&E does not support the growth rate suggested by DRA, asserting that the eligibility estimate should serve to pinpoint a moment in time as opposed to forecast future population estimates. If the Commission decides to forecast population growth on a yearly basis, PG&E requests that a commensurate level of funding be provided.
SoCalGas and SDG&E propose that the IOUs annually update their population growth rates based on actual annual meter growth. According to them, this more accurate indicator of overall customer growth is easily accessible.
DRA contends that households treated by LIHEAP in future years should not be subtracted from the population, as the Commission does not guarantee LIHEAP service. According to DRA, subtracting homes that have not yet been treated is premature. Instead, the Commission should allow utilities to retroactively subtract homes treated by LIHEAP from the eligible population.
LIF asserts that the households treated by LIHEAP should not be excluded from the estimate of the eligible and willing LIEE population. LIF believes that the Commission cannot be certain that weatherization and/or energy efficiency services were provided to these homes. LIF also believes that customers treated by LIEE and LIHEAP more recently than 2002 should be included in the base point calculation. LIF asserts that low income residences tend to wear out at a faster rate than other homes because they are located in dilapidated environments, house multiple families in single units, and experience significant turnover in occupancy given the transient nature of the low income population.
Greenlining also expresses its concern with the 10% unwillingness factor. They believe that the figure is unsubstantiated. Greenlining argues that currently, a rising number of customers are defaulting on their energy bills, making very few likely to be "unwilling" to participate. It believes that the lack of participation may be due to the confusing and difficult participation requirements, as opposed to outright unwillingness.
The utilities assert that the LIHEAP program provides participants with all feasible measures, with the majority of the measures available under both programs. For this reason, the utilities believe customers treated by both LIEE and LIHEAP since 2002 should be excluded from the base point. According to SoCalGas and SDG&E, the DCSD provided the IOUs with the actual number of households treated by LIHEAP from 2002-07 and the estimated number of households to be served by LIHEAP in 2008. The utilities argue these numbers are valid and should be subtracted from the eligible and willing population, argue the utilities.
DRA does not agree with the subtraction of 10% of eligible customers who are deemed as unwilling to participate. DRA claims that the utilities are inappropriately using information concerning the CARE optimal penetration rate to calculate the unwillingness factor for LIEE. According to the KEMA Report, only 5% of customers are unwilling to participate in LIEE (p. 5-69). This dissimilar estimate should not be used, especially in a time of rising energy costs and greater importance placed on emissions reductions.
DRA claims that the IOUs have committed to tracking customers who are unwilling to participate in LIEE. As part of this process, DRA believes the IOUs should retroactively subtract any unwilling customers tracked through their efforts from the total population estimate, calculated as part of the base point. The tracking mechanism should distinguish between those already treated and other utility customers. According to them, this will encourage accurate tracking on the part of the utilities as well as identify a more specific and verifiable understanding of the willing LIEE population.
LIF believes that the inclusion of a 10% unwillingness factor would diminish the goals of the LIEE program. LIF believes the program should account for all eligible households and that the IOUs can retroactively subtract households who are verifiably unwilling to participate. LIF notes that the KEMA Report estimated that only 2% of those surveyed were deemed as "not at all willing to participate in CARE."85
The utilities state, in rebuttal, that additional onsite surveys incorporated into the CARE unwillingness factor led to their IOU estimate of 10% as "unwilling and unlikely to participate in CARE." They state that the KEMA Report did not include a similar estimate for customers both unwilling and unlikely to participate in LIEE. The utilities believe the LIEE 5% unwillingness factor is deficient and should be higher, especially given that LIEE presents greater participation barriers than CARE. The utilities believe that a 10% unwillingness factor serves as a reasonable, and perhaps conservative, estimate of customers who are unwilling or unlikely to participate in LIEE.
The utilities state that they aim to treat all eligible and willing customers by 2020. They believe the 10% unwillingness factor enables them to project realistic three-year budgets that are not too burdensome for all ratepayers. Additionally, they claim that the treatment of additional customers would substantially impact ratepayers and require additional budgets to do so.
12.3. Discussion
The Commission has carefully considered the information presented by the parties regarding this issue. We recognize that the actual population of willing and eligible customers is difficult to definitively ascertain. We use the information presented by the IOUs and the parties as a starting point for devising a CPUC methodology for determining the eligible and willing LIEE population. Taking each position and estimate into consideration, we have devised a standard methodology and subsequent estimates of the number of households to be treated over the next budget cycle. We direct the utilities to plan to treat the number of households we have calculated for each respective utility.
We agree with DRA that it is cheaper now to treat the homes than it will be three years from now, given expected inflation and the urgency of reducing energy costs in the present. However, we do not want to place an additional strain upon the ratepayers in the process of serving more homes. Therefore, our methodology strikes a balance between the utilities' proposed methodology and the recommendations of DRA. Moreover, we take Greenlining's suggestion into consideration, attempting to make the methodology for calculating the programmatic initiative as simple and clear as possible.
The following graph illustrates the proposed estimates of households remaining to be treated by LIEE for 2009, demonstrating the middle-of-the-road approach we adopt today.
Under our adopted methodology, the number of households eligible for LIEE is first estimated. The Commission agrees with DRA that the utilities' overall population estimate of 5.404 million customers for 2009 should not be used, given that this number is significantly lower than previous years' estimates. Though we are not in a position to verify DRA's claims that the utilities are underestimating the eligible population, it is unlikely that the low income population diminished by 4% from 2006-09. However, we also recognize that data collection can lag and processes for collecting such information may vary, which could explain the decrease in the eligible population estimates.
To resolve this issue, we turn to the IOUs annual reports, in which the utilities present estimates of the number of eligible households for the current year. On May 1, 2008, the utilities submitted their 2007 annual reports to the Commission, including eligible household estimates of 1,868,598 for PG&E; 1,368,584 for SCE; 354,489 for SDG&E; and 2,046,086 for SoCalGas. In total, approximately 5,637,757 households were eligible for the LIEE program in 2007. We use this number as the starting point for the estimate of eligible LIEE customers.
While the utilities presented a static estimate of the eligible population, DRA's estimate is factored up annually by 1%, thereby accounting for population growth and economic conditions. We agree with DRA that population growth should be taken into consideration. Therefore, the estimate of 5.637 million households is factored up by 1% annually, resulting in an estimated population of 5.751 million households for 2009.
In the next step of the methodology, we consider the number of households unwilling to participate in LIEE. Various parties expressed their concerns with the subtraction of 10% of the customers based on a deemed unwillingness. The Commission agrees that the IOUs have inappropriately used this estimate from the KEMA Report.86 The assessment clearly states that 5% of customers are unwilling to participate in LIEE. As this latter number is a more valid estimate for our purposes, we take 5% of the given population from each annual eligible population estimate, resulting in 287,554 customers in 2009 estimated as unwilling.
In doing so, we agree with the PG&E/SCE proposal to use the total population as the basis for this estimate, given that any estimate of unwillingness takes the entire population into account. Future estimates of willingness may be more precise and may be considered for the 2012-14 budget cycle.
After estimating the number of unwilling households, the Commission methodology calculates the number of households treated by LIEE from 2002-08. In their quarterly and annual reports, the utilities present estimates of the number of households treated from 2002-2007, amounting to a total of approximately 961,176 households. The number of homes treated in 2008 should also be included. Approximately 191,208 households are projected to be treated in 2008, according to the IOUs' budget applications. In sum, the IOUs treated a total of 1,138,349 homes from 2002-2008.
We obtain this information from the projections of homes to be treated in 2008, obtained from the budget applications.87
Households treated under the LIHEAP program should also be counted as treated, given that LIHEAP offers most of the same measures offered by LIEE, as well as additional measures not offered by LIEE. Additionally, under LIHEAP, more money is spent to treat a home than under LIEE. With data obtained directly from the Department of Community Services and Development (DCSD), approximately 224,387 homes were treated by LIHEAP from 2002-07. Also administered by DCSD, the Weatherization Assistance Program funded through the Department of Energy treated 183,171 homes from 2002-2007.88 We do not include the estimated number of households treated in 2008, given that this data is not final. The number of households treated in 2008 can be subtracted from the estimate of households to be treated in the 2012-14 budget cycle.
Under our adopted methodology, the estimates of unwilling households, households treated by LIEE from 2002-08, and households treated by LIHEAP from 2002-07 must be subtracted from the estimate of the eligible LIEE population. After conducting this calculation, we must estimate the number of households to be treated in the next budget cycle. To do so, we divide each annual estimate by 25%, leaving the utilities to determine how many households are treated annually in order to reach the estimate for 2011. Overall, the utilities must treat a total of 1,052,651 households over the next budget cycle to meet 25% of the programmatic initiative. Specifically, utilities must treat the following number of households over the next budget cycle - PG&E at 340,885 customers, SCE at 250,336 customers, SoCalGas at 400,279 customers and SDG&E at 61,152 customers.
Improved data will likely be available in the next budget cycle, via the 2010 census and information from utilities' proposals to improve the tracking of customers. Such information will improve the validity of the population estimates conducted in the 2012-14 budget cycle.
If the population is lower, the higher estimates of eligible and willing customers in this budget cycle will make the utilities' 2020 goal easier to reach. However, we conclude the methodology described above is the best way to ensure that we target the correct goal.
83 Assigned Commissioner's Ruling Providing Guidance for Low Income Energy Efficiency 2009-11 Budget Applications, filed in R.07-01-042 on April 1, 2008.
84 The calculation we perform below demonstrates the A-B-C-D methodology, rather than PG&E's and SCE's proposal. Obviously, the point at which one subtracts "unwilling" customers has a significant effect on the bottom line. We prescribe a consistent approach across all IOUs, as we discuss below.
85 LIF cites the KEMA Report, p. 5-68.
86 The 10% figure refers to the number of customers unwilling to participate in CARE.
87 When calculating the number of homes treated by LIEE from 2002 - 2008, we originally referred to the information submitted in the IOUs' 2007 4th Quarter Report, submitted on February 1, 2008. In the Final Decision, we used the more up-to-date information submitted in the 2007 LIEE Annual Reports, submitted May 1, 2008. This resulted in the modification of the number of households treated by LIEE in 2007. According to the 2007 Annual Report, the IOUs treated 9,782 more households in 2007 than was originally reported in the 2007 4th Quarter Report.
88 We have received various figures from DCSD, as DRA points out in its comments on the proposed decision, but figures shown here are the ones we received from DCSD and the IOUs.
