Evaluating technology upgrades as a complement to traditional bill assistance programs

Evaluating technology upgrades as a complement to traditional bill assistance programs | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Evaluating technology upgrades as a complement to traditional bill assistance programs Sydney Forrester, Eric O’Shaughnessy, Galen Barbose This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8725239/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Increasing electricity prices have raised concerns about energy affordability. State programs to decrease energy burden (share of income spent on energy) could help mitigate these concerns. Many states offer some combination of bill assistance and rebates for installed measures such as weatherization and rooftop solar to income-qualifying households. Here, we analyze how these approaches may complement each other to improve energy affordability. Results illustrate tradeoffs between program costs and energy burden reduction as well as between a program’s upfront capital costs versus ongoing costs. Generally, the three strategies complement one another to improve energy affordability at a lower net present cost than bill assistance alone. This holds true, to varying degrees, across regions. Weatherization can decrease the solar installation needed; and both together can decrease or eliminate ongoing reliance on bill assistance. Fully rebated weatherization and solar rebated at $ 0.62/Watt, when combined with bill assistance, yield the same modeled net present cost as bill assistance alone and reduce the share of low-income households with high energy burdens from 66% to 17%, compared to bill assistance alone (to 34%). Absent tax credits, weatherization still reduces energy burden and program cost of bill assistance, whereas solar may not compete due to cost. Scientific community and society/Energy and society/Energy efficiency Scientific community and society/Energy and society/Energy justice Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Introduction Increasing prices for electricity and other fuel have raised concerns about energy affordability. States promote energy affordability via state appropriations, utility and ratepayer contributions, or voluntary mechanisms. 1 In addition to those broader efforts, targeted programs can provide direct relief to those most impacted by rising pricing, especially lower income households, who often have disproportionately higher energy burden, defined as the share of gross household annual income that goes towards electricity and non-electric heating costs. 2 One study estimated that low-income households’ median energy burden was triple that of non- low- or moderate-income households, 3 and some states have created parallel policy goals focused on reducing energy burden across households. 4 Historically, bill assistance is a common strategy to decrease energy burden. Some U.S. states have Percent of Income Payment Plan (PiPP) programs, whereby eligible customers receive a discount on their utility bill sufficient to reduce their energy costs to some specified percent of their income. That percentage is often set at 6%, the level often defined as equating to a “high” energy burden (see Table SI.2 for a summary of state PiPP programs). 3 PiPP can immediate relief, and one survey illustrates that bill assistance was effective in maintaining a comfortable temperature for participants and avoiding utility disconnection. 5 However, that relief is short-term and does not address the underlying causes of high energy burden or provide co-benefits. These programs can therefore become expensive over the long term if households re-apply over multiple years, which increases ongoing programmatic costs and applicant burden. 6 In addition, PiPP programs may provide only a partial solution for many participants, as the bill discounts are typically capped at some maximum dollar amount per customer, leaving many participating customers still facing high energy burdens above 6%. Alternatively, or in addition, to direct bill assistance, many states offer programs for income-qualifying customers to adopt home energy upgrades. Historically these programs have focused primarily on weatherization, but more recently some states have expanded their programs to include rooftop solar. 1 , 4 Compared to PiPP, installed measures have higher upfront costs, but offer long-term savings, as well as potential non-energy benefits such as improved home comfort, increased property value, reduced energy insecurity, and ability to afford other necessities. 7 These programs can be combined with PiPP to either “close the gap” for customers where the maximum dollar limit is reached, or to reduce the amount of direct bill relief needed to reduce energy burdens to the targeted level. This decreases bill assistance program cost directly, but also creates indirect avoided costs from reduced disconnections and arrearages or non-payment. 7 In tandem with income-qualified programs, some states and utilities have leveraged private and public dollars to introduce alternative financing via community development financial institutions or state investment banks and/or utility on-bill financing programs. These programs can mitigate high upfront measure costs and often offer lower interest rates through credit enhancements such as a loan-loss reserve or a guarantee to reduce financier risk. 8 When repayment is lower than bill savings throughout the loan term, households can pursue energy upgrades that will reduce their energy burden. Prior studies have demonstrated the savings potential from installed measures for low-income households. For example, Blasnik et al. (2014) found that participants in the federal weatherization assistance program saw an average 12.4% decrease in bills, or $ 283 per year, 9 and Tonn et al. (2015) found that the share of participants who struggled to pay energy bills declined from 74% before the program to 59% one year after. 10 , 11 Another income-qualified pilot program of a fully-incentivized 3.5 kW system produced $ 587 in savings, or a 45% bill reduction, for participants. 4 Analyses of rooftop solar adoption by low-income households have found that, in the U.S., low-income rooftop solar has reduced the incidence of high energy burden from an estimated 67% to 52% and have reduced the incidence of energy insecurity. 11 , 12 Kerby et al. (2024) studied storage, solar-plus-storage, weatherization, and appliance efficiency upgrades together and found that weatherization produced the highest savings, but that results varied widely between households. 13 As states grapple with growing energy affordability challenges, it will be important to consider a broad range of programmatic options and understand their complementarity. Our study aims to address that by evaluating the performance of a PiPP program compared to income-qualifying weatherization and rooftop solar programs, as well as all combinations thereof while considering varying combinations of loan-financing versus upfront rebates. We perform this analysis on a modeled dataset of 111,410 low-income households across the U.S., used widely in other published research studies. 14 The performance of each program model in evaluated in terms of two primary metrics: efficacy in reducing the percentage of low income households with “high” energy burdens (defined as exceeding 6% of income) and the program’s net present cost, assuming only cost-effective measures are installed. The results illustrate key tradeoffs and show that income-qualifying weatherization and rooftop solar programs can often be coupled with PiPP to improve both efficacy and cost compared to PiPP alone. Results In the following results, we begin by presenting outcomes for a PiPP program designed to reduce the incidence of high (>6%) energy burdens among low-income households (incomes less than or equal to 80% of their area median income). Outcomes are presented in terms of efficacy in reducing the incidence of low-income households with high energy burdens, as well as the net present cost (NPC) per low-income household. We then evaluate two installed measure programs—weatherization and rooftop solar—as alternative programmatic models for reducing energy burden, first considering each individually and in combination with each other, and then later in combination with PiPP. For these installed measures, we consider varying levels of upfront rebates and the remaining customer cost financed through loans with lower interest rates funded under a loan-loss reserve program. Those two financing models represent the most common approaches used in low-income technology upgrade programs. Both PiPP and the installed measures are constrained in their scale of implementation, though in different ways. In the case of PiPP, we assume all eligible low-income households receive assistance but limit the total annual bill discount to the lesser of: (a) the amount required to reduce energy burden to 6% or (b) a total annual bill discount of $1,800. That cap is based roughly on current program designs in the United States (see Methods and Table SI.2). The installed measures instead are limited to households where two cost-effectiveness criteria are both met: (1) the measure package has a “savings to investment” (SIR) ratio greater than or equal to 1 from the program perspective—that is, the total cost of the program (rebates and any capital for a loan-loss reserve) cannot exceed the gross value of the total lifetime energy savings—and (2) the upgrade is immediately cash-flow positive for the household—that is, the bill savings exceed any loan repayment costs. The first of those criteria is a standard screening metric used, for example, in the federal Weatherization Assistance Program for installed energy conservation measures. 15 The second is intended to represent the bare minimum financial return necessary to motivate a household to participate, recognizing that actual household decision-making is considerably more complex. 16–18 Within our analysis, those criteria are evaluated on a household-by-household basis, based on modeled energy and net bill savings for each individual household and the associated programmatic costs. Additional details on the application of these criteria are described in the Methods. The results presented below focus on national and regional outcomes. Bill assistance: Percent of Income Payment Plan Within our sample of modeled low-income households, 66% have an initial (pre-intervention) energy burden above 6% and are therefore eligible for PiPP. The $1,800 cap on annual bill discounts would be binding for roughly 37% of those participating households, thus reducing the incidence of high energy burdens from 66% to 34% of total low-income households. Provided continuously over 25 years, this level of bill assistance would incur a NPC of roughly $18,600 per participant, on average. That 25-year period is used to correspond roughly to the lifetime of the alternative installed measures considered. As shown in Fig. 1, outcomes vary across regions in terms of both the efficacy of the PiPP program in reducing energy burdens to below 6% and its cost, reflecting underlying regional differences in energy costs and incomes. The baseline incidence of high energy burden is considerably lower in the West than other regions, due in part to generally milder climates and thus lower overall energy costs. Consequently, PiPP leads to the lowest incidence of high energy burdens there, albeit starting from a much lower baseline. Among the other three regions, which all have a similar baseline incidence of high energy burdens, PiPP efficacy is greatest in the South, largely as a result of lower energy prices and costs. Conversely, efficacy is lowest in the Northeast, due to relatively high energy prices and thus more frequently binding caps on the annual bill discount. The NPC of PiPP is inversely related to its efficacy, with higher costs in regions where the funding cap is more frequently an active constraint. Of all PiPP-eligible participants, 52% of Northeast households require more than the maximum level of assistance whereas 41% do in the Midwest, 32% in the South, and 25% in the West. Installed Measure Programs: Participation Potential The efficacy of technology-based programs in reducing the incidence of high energy burden among low-income households is a function of participation rates and the net savings among participating households. Focusing first on participation rates, these depend on what fraction of eligible households pass the requisite set of cost-effectiveness thresholds. That fraction, in turn, depends on the funding model: a higher upfront rebate reduces cost-effectiveness to the program administrator (due to lower SIR), while a lower upfront rebate and correspondingly higher reliance on financing reduces cost-effectiveness to the household (due to lower net customer savings). In addition to those factors, participation may also be constrained by eligibility rules. In our analysis, the weatherization measures do not apply to the 1% of low income households with high energy burdens whose homes are already well-sealed (Table SI.3), and the rooftop solar program does not apply to the 3% in multi-family dwellings (see Table SI.3 and Table SI.4). Figure 2 illustrates how the program financing model (rebate vs. loan) impacts the set of households that pass the two cost-effectiveness screens. As shown, the solar-only program is most sensitive to the choice of financing, due to higher measure cost. Substituting even just a portion of the loan support with an upfront rebate (e.g. at the $0.5 or $1/W level) substantially increases the portion of households for which the measure is cost-effective, while still satisfying the program administrator SIR constraint. At a roughly $1.5/W rebate, the solar measure is cost effective for almost all eligible households, and any further increases in rebate level yield rapidly diminishing returns due to the program SIR constraint. By comparison, the cost effectiveness of the weatherization (Wx) measures is much less sensitive to the choice of financing, due to the significantly lower measure costs. 1 To illustrate the potential complementarity, we consider a program that combines fully rebated weatherization measures with solar under a varying loan/rebate mix. Notably, this combined program design generally improves cost-effectiveness for both the customer and the program administrator, leading to an upward shift in the curves for the combined measures (relative to the rightmost points on the corresponding weatherization-only curves). The combined program design also allows for households that could not benefit from one technology alone to benefit from the other, increasing potential further. Depending on the rebate level, either the customer cost-effectiveness or the program administrator SIR constraint may be binding. Combining these two, Figure 3 shows the percentage of the target population that passes both cost-effectiveness thresholds (we call this the “participation potential”) across the range of rebate/loan combinations for each program design and region. Not surprisingly, participation potential for the weatherization-only program is less impacted by the form of financial support and ranges from roughly 89% to 93% of the target population across the four regions, when fully rebated. Participation potential is somewhat lower in the West and South compared to the other regions, due to milder climates in the West and low energy prices in the South. For the solar program, participation potential is generally highest around a rebate of $1.5/W, reaching 90% or more of the target population in each region. When combined, weatherization provides more savings per dollar up to a limit while solar provides deeper savings, but at a higher cost. Weatherization can reduce the solar size in some cases, avoiding further cost while providing the same level of bill savings. Results show that, when weatherization is fully rebated, this allows for high participation potential for the combination across all regions regardless of solar rebate level. Installed Measure Programs: Net Bill Savings Net bill savings are the difference between the direct savings on the customer’s utility bills (inclusive of electricity, natural gas, and other heating fuels) and any loan repayment costs. The modeled weatherization program includes a standard set of insulation and air sealing measures, which generally yield relatively modest bill savings, but at a modest cost (see Methods and Table SI.5). Compared to the weatherization measures, the rooftop solar program offers greater bill savings potential, but at a substantially greater cost. The solar program assumes a maximum 5 kW installation, subject to each home’s usable roof space and limited in size such that annual solar production is no greater than the customer’s regional average annual ratio of solar to electricity consumption (see Methods and Table SI.6). These size limitations bind differently across regions, depending mostly on the intensity of heating/cooling loads and the prevalence of electric space heating, and are also impacted when solar is combined with weatherization (see Table SI.4). Figures 4 and 5 compare the net bill savings for each technology-based program to PiPP—focusing on the full national sample of low-income households in Figure 4 and the regional results in Figure 5. While Figures 2 and 3 illustrate how weatherization and solar complement one another to increase cost-effective participation, Figure 4 demonstrates that the complementarity extends to deeper participant savings that compete with PiPP levels. Alone, net annual bill savings from weatherization are $546 if loan-financed or $634 if financed entirely via upfront rebate. By comparison, median bill savings under the PiPP program are more than double that amount, at $1,373 per year. Weatherization bill savings vary to some extent by region (Figure 5)—with the greatest savings in Northeast, due to relatively high energy prices and cold winter temperatures—though in all cases are well below the median bill savings under PiPP. The weatherization measures considered in this analysis are rather modest. As such, deeper investments and savings could compete better with PiPP levels. Alone, solar provides a deeper investment and is considerably more expensive than the weatherization measures considered. Thus, net savings are more sensitive to the form of financing provided. If financed entirely through a loan, the median net bill savings across all participants is $205 annually, but grows to $1,042 with a full upfront rebate. However, even at that higher level, the savings are still less than PiPP for the full national sample. The West offers an exception when fully rebated, owing to greater solar production levels and a milder winter climate throughout much of the region. The combination of solar and weatherization alleviates the issue of deep potential savings for weatherization alone as well as the issue of high sensitivity to incentive levels for solar alone and can reach net savings comparable to that of PiPP. If financed fully through upfront rebates, median annual bill savings reach $1,620 across all participants nationally, which exceeds the benefit provided under PiPP. Notably, the same dynamic holds across all regions. Of course, as the financing model shifts away from upfront rebates and toward low-cost loans, the net savings diminish; with a fully loan-financed design, the median net bill savings in all regions falls below PiPP. Among the regions, the Northeast and West are closest to retaining parity with PiPP in terms of net bill savings, owing largely to the set of factors previously noted for those two regions. Comparing Efficacy and Cost across Alternate Program Models Based on the preceding participation rates and participant net bill savings, Figure 6 compares the efficacy and cost of each of the technology-based program models to PiPP (regional results are shown in Figure SI.1). As shown, the combination of weatherization measures and rooftop solar funded fully through upfront rebates compete directly with PiPP, providing better efficacy (reducing high burden incidence to 30% vs. 34%) at a lower program net present cost (25-year NPC of $11,454 vs. $12,288). When combined with PiPP, cost-effective installed measures reduce ongoing reliance on PiPP. When weatherization with solar are combined with PiPP, there is some “Pareto improvement” compared to PiPP alone (i.e., greater efficacy at the same or lower cost, or lower cost at the same or greater efficacy). This occurs either by allowing deeper energy burden reductions than can occur through bill assistance alone (due to the per-participant annual limits) and/or by reducing the amount of bill assistance needed to reduce energy burdens to the 6% threshold. These Pareto improvements are shown in Fig. 6 by outcomes in the lower left-hand quadrant. In the case of weatherization, Pareto improvements occur whenever cost-effective measures are combined with PiPP, regardless of the financing model. Instead, for rooftop solar, Pareto improvements are more likely to occur when the larger portion of the cost is loan-financed, rather than relying solely on upfront incentives. When fully-rebated weatherization and rooftop solar rebated at $0.62/W are both combined with PiPP, they can substantially reduce the incidence of high energy burden to 17% of low-income households (compared to 34% with PiPP alone), at the same NPC. Sensitivity to Tax Credit Availability The preceding results assume a 30% tax credit for the modeled technology upgrades, similar to the Investment Tax Credit for rooftop solar and the Energy Efficient Home Improvement Credit for weatherization. We also assume state solar tax credits and the Home Efficiency Rebates administered at the state level (see Methods). Since a subset of these tax credits will soon expire, we consider here the extreme case of how elimination of all tax credits (including state-administered credits) would impact the performance of the technology-based programs relative to PiPP. The loss of tax credits reduces cost-effective adoption of installed measures, and increases the cost for the remaining cost-effective participants (see Figure SI.2 for participation potential). This shifts the performance curves in Figure 7 upward (lower efficacy) and to the right (higher cost). This may require either lower rebates to achieve better program economics, which would reduce household savings; or higher rebates to improve affordability, which may lead to fewer households served. Without tax credits, installed measures cannot compete with PiPP on their own. Even with full rebates, efficacy of PiPP is superior to the combined installed measures. When combined with PiPP, providing access to low-cost capital still provides marginal improvements both in efficacy and reduced cost. For example, access to financing for weatherization and solar, separately, reduces NPC $1,546 and $98 per low-income household and reduces the incidence of high energy burden 5.7 and 0.7 percentage points, respectively, compared to PiPP alone. Close to Pareto improvement, fully-rebated weatherization with PiPP would have a NPC similar to PiPP alone, but reduce high energy burden an additional 11.8 percentage points. Discussion Energy affordability programs have expanded in recent years beyond direct bill assistance to include various cost-effective installed measures that can provide ongoing bill reduction such as weatherization and rooftop solar. Program goals may have multiple, sometimes competing, objectives such as those for households in reducing incidence of high energy burden, maximizing participation, or capturing non-energy benefits; and those for the program administrators such as controlling upfront and ongoing program costs, achieving state or utility targets or goals for affordability or DER deployment. There is no “best” program design and programs must balance these tradeoffs. On its own, direct bill assistance is flexible, does not require long installation time or large upfront investment, and can be applied to any household that pays their own utility bill regardless of building type or ownership status. However, bill assistance does not provide non-energy benefits or a long-term solution. The pros and cons for installed measures such as weatherization or solar are just the opposite. As such, programs have begun to explore the complementary relationship between direct bill assistance and installed measures. For example, an income-qualified solar program demonstrated economic benefits to the utility due to reduced assistance spending, reduced defaults and delinquencies, and reduced administrative costs. 19 Another program that used weatherization funds for solar acknowledged that solar unlocked the potential for much higher savings, however, at a higher cost that required supplemental funds and could limit the number of households served by a fixed program budget. 4 Our results find that bill assistance, weatherization, and solar—each with their respective tradeoffs—can complement one another. Together, weatherization reduces the size of solar required and improves the benefit-to-cost ratio while solar deepens potential savings. If further combined with PiPP, ongoing bill savings from weatherization and solar decreases or eliminates bill assistance that a household may require and serves more households. Results are optimized when rebates are set high enough to produce meaningful household savings, but low enough for a program to accommodate a larger number of participants. Indeed, our results show that the combination of cost-effective weatherization with a full rebate and solar with a $1/W rebate decreases incidence of high energy burden among low-income households more than PiPP would, and at a lower NPC. Combined with PiPP, all three interventions achieve the lowest incidence of high energy burden compared to any other scenario. These results illustrate a spectrum of options for affordability program design. For example, if a program instead is not aiming to optimize a PiPP program, but to encourage cost-effective weatherization and solar without a large level of funding, our results show that low-cost loans with customer protections could improve affordability, even without PiPP or upfront rebates. Indeed, weatherization with solar reduced the incidence of high energy burden 11 points with access to low-interest loans alone (Fig. 4). In another scenario, if a program solely aimed to maximize deployment of weatherization and solar without PiPP, it may focus on higher rebate levels instead. For a program with constrained upfront time and capital, PiPP could provide immediate affordability to any qualified household while allowing time for complementary program design. In these cases, pairing PiPP alone with financing and low-cost capital to allow for households to adopt cost-effective measures on their own improves both cost and efficacy. Our study includes several limitations. First, it does not include other strategies that could impact home energy burden such as demand response, community solar, energy storage, or electrification. It also only includes a set portfolio of weatherization measures without other energy efficient upgrades such as appliance replacement. Due to the focus on weatherization and rooftop solar, we do not consider renter-occupied households. We also use state-specific tariff assumptions to apply our findings across the United States that may not reflect nuances found in individual utility territories or programs. We also do not consider the cost of net energy metering or impacts on non-participating households of these programs. Another limitation is that we do not account for programmatic funding constraints that are binding in almost all actual cases. In reality, any assistance program may face funding limitations, and the level of allocated dollars may fluctuate greatly from year to year. 1,20 Finally, for participation, we assume a potential as a high bookend, but do not incorporate assumptions that would decrease actual participation, which would be much lower. For example, a study found that unsecured loans administered through similar financing programs had much lower default rates than unsecured consumer loans, potentially due to net bill savings and safeguards against predatory lending. 21 While this can expand access and reduce upfront costs for households, in cases where energy costs remain unaffordable, financing may expose vulnerable households that remain at risk for non-payment and deter participation. This study builds on a body of literature to better understand how states or utilities could employ multiple energy affordability strategies and how they may complement one another. Thoughtfully combining access to a series of interventions could help achieve broader affordability goals, decrease risk of customer shut off or non-payment, improve home comfort and health, and reduce reliance on bill assistance. Methods Filtering household models For this study, we specifically focused on households that would be candidates for bill assistance, weatherization, and/or low-income solar incentives. As such, we narrowed NREL’s full set of End Use Load Profiles (EULP) building models to those that were owner-occupied and with area median incomes (AMI) at or below 80%. 22 Data for each building model include building characteristics (e.g., units in structure, building envelope), household characteristics (e.g., tenure, number of occupants, income range), location (e.g., climate zone, municipal area, county), and energy use (e.g., fuel and electricity demand by end use). We used EULP data on each household’s county, number of occupants, and income range to determine AMI. Choosing a randomly selected value within the range and using the respective county and number of occupants, we assigned a percent AMI value to each household by comparing to the Housing and Urban Development specifications. 23 From a total set of 548,916 total building models, 307,480 were owner-occupied. Of those, 111,410 had an area median income at or below 80%, which make up the base of our study (see Table SI.1 for sample size of each step and other summary metrics by state). All monetary pricing and income data are taken from 2022 for consistency. Electricity prices Building models in EUSS include end-use electricity, natural gas, propane, and heating oil usage. Using the Energy Information Administration’s Form 861 electricity data, each utility has data on residential revenue, residential electricity sales, and residential customer counts. 24 Since these data do not include whether, or at what level, fixed charges or minimum bills are in place, we assume a monthly fixed charge of $10 for each residential customer. As a result, each utility’s volumetric rate is established by subtracting all fixed charges collected (equivalent to $120 annually by the total number of residential customers) from the total revenue collected and dividing that by the number of residential sales in kWh. For each county associated with more than one utility, the utility with the most customers for each respective state was selected as the county’s assigned utility. With this, each building model was linked to a specific utility and corresponding volumetric electricity rate. Flat rates were assumed as the baseline with net metering-- i.e., compensation for solar generations, including exports, at the retail rate value. Time sensitive rates were not considered the base case because only 10% of residential customers were enrolled in time-sensitive rates in 2023. 24 Only nine states had above-average shares of customers on time sensitive rates while the rest had near-zero shares. The states with the above-average shares include Delaware (57% of customers), Maryland (47%), Missouri (44%), Colorado (37%), Arizona (37%), Michigan (37%), Oklahoma (33%), California (30%), and Montana (15%). 24 Other energy prices For non-electric energy prices, we used EIA 2022 state data (concurrent with electricity prices) for residential natural gas, propane, and heating oil. 25,26 All EULP energy consumption is in kWh. So, all non-electric prices were converted to kWh with: 1,039 British thermal unit (BTU) to 1 cubic foot of natural gas; 91,452 BTU to gallon of propane; 138,500 BTU to gallon of fuel oil; and 3,412 BTU to kWh. 27 Bill assistance: Percent of income payment plan (PiPP) design Percent of income payment plans (PiPP) are offered in specific states or utility territories and are structured such that income-qualified customers are given assistance on their energy bills that drives their energy burden down to a specified level (most often 6%, but sometimes 10% if bills include more than solely electricity). Often these programs have a maximum annual discount of $150 per month, or $1,800 per year. Table SI.2 summarizes existing programs across the U.S. We use values modeled after existing programs to structure our PiPP bill assistance for this study, offering the smaller between a discount that would reduce household energy burden to 6% or $1,800 per year. As such, customers with a baseline energy burden below 6% would not receive additional bill assistance. The level of bill assistance changes based on how installed weatherization or solar may impact costs. This includes both bill savings as well as on-bill loan repayment, where applicable. Weatherization measures, costs, and incentives NREL’s EULP building models have corresponding upgrade scenarios (“End-Use Savings Shapes (EUSS)”). For the weatherization case, we consider the “Enhanced Enclosure” scenario, which includes attic, wall, and basement insulation as well as air and duct sealing (see Table SI.5). 28 NREL includes detailed logic on which buildings got what upgrades, based on baseline measures and housing condition as well as climate. 28 As such, each building is prescribed a different level of upgrades. For example, a building with no or very poor insulation in a region with many heating degree days may be prescribed more insulation in the weatherization scenario whereas a building with sufficient insulation in a mild climate may not. Additionally, some measures may not apply to certain buildings. For example, insulation for finished attics will not be performed for housing without a finished attic. To best pair with EUSS measures, we drew costs from NREL’s “National Residential Efficiency Measures Database”, which represents the total cost to implement the retrofit measure. The database presents one primary value as well as a range, depending on the pre-measure condition and measure selected. For each category, the database provides additional information on cost drivers. See Table SI.5 for the cost (and range) of each relevant measure. We use this pricing for each building’s characteristics (e.g., square footage of various conditioned spaces) to determine the total cost of the weatherization upgrade for each home. 29 NREL’s EUSS database then provides the consequent changes in energy consumption as a result of the upgrades, which we can use to quantify savings from weatherization for each building. Where federal tax credits are included, we consider two separate incentives. The first is modeled after the energy efficiency home improvement credit, which allows for 30% of the installed price up to a maximum limit of $1,200. 30 We also consider the Home Efficiency Rebates (HOMES) program, which is a rebate for home weatherization that is allowed to be stacked on top of other incentives. This only applies to households with 80% AMI or lower, which includes all households in our sample. The rebate provides differing levels of incentives based on the amount of modeled savings from the proposed upgrade. An 80% rebate is offered for households with AMI at or below 80% and energy savings more than 20% above the baseline. For measures leading to 20%-35% of savings, the maximum limit is $4,000 while, for savings above 35%, the maximum limit is $8,000. Measures that produce savings less than 20% are not offered this incentive. Rooftop solar sizing, cost, and incentives Generation is determined using the centroid of each household’s county and inputting that locational information into NREL’s System Advisor Model. We assume a South-facing array, tilt at latitude, fixed array with a 1.2 inverter loading ratio and 14% system losses. For system degradation, we assume 0.5% degradation per year. We then determine each household’s solar array sizing, taking into account restrictions created by either rooftop square footage or total annual load. Using Energy Sage data, we find the median ratio of household annual solar generation to annual load from each state (see Table SI.6). We apply that ratio to buildings based on their state as one constraint. We also consider rooftop square footage, where we assume that 70% may be suitable to host a solar array, 31 and that each square foot could host 15 Watts as the second constraint. The final rooftop solar array size is selected as the minimum between 5 kW, the solar generation to load ratio, or the available rooftop size restriction (see Table SI.4 for share of binding constraints). Rooftop solar costs are taken for each state, where available, or region. These data originate from Berkeley Lab’s Tracking the Sun. These include both material and soft costs in addition to any upfront or performance based incentives, but excludes tax credits and solar renewable energy credits. Per the authors’ recommendation, system costs were filtered to exclude third-party owned arrays and then discounted by 20% to account for high loan origination fees. For each region, 2022 mean and standard deviation of solar system costs were taken (see Table SI.7). For each of the 1,000 iterations, each customer was assigned a cost per Watt based on a normal distribution that was specific to their state (or region). Since tax credits and renewable energy credits are not included in the Tracking the Sun values, we apply them separately. Where federal tax credits are considered, we assume 30%. Some states offer additional incentives which are considered (see Table SI.8). Since we assume host-ownership, we do not include any additional adders applicable for third parties such as energy community or low-income adders. Upfront rebates and loan assumptions We consider an income-qualified rebate program that provides capital required for low-income households with high or severe energy burdens to adopt weatherization and/or rooftop solar. Rebates may represent the full amount, intermediate levels (only for solar), or none at all. If provided in full, the program administrator covers all costs of adoption after relevant state and/or federal incentives. Intermediate levels are in various $/W levels for solar and it is assumed that the household would pay for the rest via a loan. Consistent with programs that allow use of bill assistance and/or weatherization to go towards solar installations, 4 we assume that programs will only provide solar incentives where the savings-to-investment ratio is greater than 1 (i.e., savings over lifetime of solar asset are more than upfront cost). As described in the Introduction, there are various programs that provide low-cost capital with consumer protections to households that seek to install energy technology such as weatherization measures or rooftop solar. These often include credit enhancement such as loan-loss reserves to buy down interest rates and zero/low levels of loan origination fees. Funds for loan-loss reserves may be supplied by a one-time grant or influx of money from public or ratepayer dollars and the capital itself may be supplied by financiers such as community development financial institutions or other impact oriented or local institutions. Where they exist, regional community development financial institutions or state investment banks may play a role in facilitating these partnerships. We assume that a program administrator could create a loan-loss reserve with 5% of the capital lent to install these weatherization and/or solar measures. We then assume that this could reduce loan origination fees to negligible levels and provide interest rates with a mean of 7.5% and 1.25% standard deviation, determined randomly via standard distribution for each customer. We assume a household will use the loan to cover the full upfront cost that they are responsible for (i.e., after all rebates and tax credits) such that no money is required upfront. We take into consideration the loan repayment on the customer’s bill. As such, the resulting energy burden post-intervention will include both the consequent bill savings in addition to repayment of both loan principal and interest. Consequently, we assume that a household will not adopt if the investment causes their net energy costs to increase. It is important to note that the level of participation considered for this study is much higher than is realistic because, on the program side, there may be limits to spending either in terms of available capital for rebates or ongoing money for bill assistance. The customer side, there may be households unwilling to adopt, that may not qualify for loans, or who may want to avoid debt. Rather, this study is meant to illustrate and compare how different interventions could improve affordability and their subsequent programmatic costs. Declarations Competing Interests The authors declare no competing interests. Author Contributions SPF conceived and designed the analysis, collected the data, contributed data or analysis tools, performed the analysis, wrote the paper. EO provided feedback on scope, provided edits. GB provided feedback on scope, assisted with writing, provided edits. Acknowledgements This work was supported by the U.S. Department of Energy’s Office of Energy Efficiency and Renewable Energy under Solar Energy Technologies Office Agreement Number 38444 and Contract No. DE-AC02-05CH11231 in Fiscal Year 2024. References Shields, L. Bolstering Federal Energy Assistance and Weatherization With State Clean Energy Programs . https://www.ncsl.org/energy/bolstering-federal-energy-assistance-and-weatherization-with-state-clean-energy-programs (2020). Brown, M. A., Soni, A., Lapsa, M. V, Southworth, K. & Cox, M. High energy burden and low-income energy affordability: conclusions from a literature review. Prog. Energy 2 , (2020). Drehobl, A., Ross, L. & Ayala, R. How High Are Household Energy Burdens? An Assessment of National and Metropolitan Energy Burden across the United States . ACEEE (2020). Carrera, A. Incorporating Renewable Energy Technology into the Minnesota Weatherization Assistance Program: Reducing Energy Burden Among Low-Income Households. (University of Minnesota, 2023). APPRISE. 2018 National Energy Assistance Survey Final Report . https://neada.org/wp-content/uploads/2015/03/liheapsurvey2018.pdf (2018). Graff, M. Addressing energy insecurity: Policy Considerations for enhancing energy assistance programs. Heliyon 10 , (2024). Tonn, B., Rose, E., Hawkins, B. & Conlon, B. Health and Household-Related Benefits Attributable to the Weatherization Assistance Program . (2014) doi:ORNL/TM-2014/345. Bell, C. J., Nadel, S. & Hayes, S. On-Bill Financing for Energy Efficiency Improvements: A Review of Current Program Challenges, Opportunities, and Best Practices ACEEE E118 . https://www.aceee.org/sites/default/files/publications/researchreports/e118.pdf (2011) doi:E118. Blasnik, M. et al. National Weatherization Assistance Program Impact Evaluation: Energy Impacts for Single Family Homes . https://weatherization.ornl.gov/wp-content/uploads/pdf/WAPRetroEvalFinalReports/ORNL_TM-2015_13.pdf (2014) doi:ORNL/TM-2015/13. Tonn, B., Rose, E. & Hawkins, B. Survey of Recipients of Weatherization Assistance Program Services: Assessment of Household Budget and Energy Behavior Pre- to Post- Weatherization . https://www.osti.gov/servlets/purl/1223652 (2015) doi:ORNL/TM-2015/64. Yozwiak, M. et al. The effect of residential solar on energy insecurity among low- to moderate-income households. Nat. Energy 10 , (2025). Forrester, S. P., Montañés, C. C., O’Shaughnessy, E. & Barbose, G. Modeling the potential effects of rooftop solar on household energy burden in the United States. Nat. Commun. 15 , (2024). Kerby, J., Hardy, T., Twitchell, J., O’Neil, R. & Tarekegne, B. A targeted approach to energy burden reduction measures: Comparing the effects of energy storage, rooftop solar, weatherization, and energy efficiency upgrades. Energy Policy2 184 , (2024). NREL. ResStock Publications. https://resstock.nrel.gov/page/publications. DOE. Weatherization Assistance Program Notice 22-7 Table of Issues. at https://www.energy.gov/scep/wap/articles/weatherization-program-notice-22-7-weatherization-health-and-safety (2021). McKenna, C., Gronlund, C., Hernández, D. & Vaishnav, P. When homeowners lose momentum after an energy audit: Barriers to completing weatherization in the United States Midwest. Energy Res. Soc. Sci. 122 , (2025). Wolske, K. S., Stern, P. C. & Dietz, T. Explaining interest in adopting residential solar photovoltaic systems in the United States: Toward an integration of behavioral theories. Energy Res. Soc. Sci. 25 , (2017). Legault, L., Bird, S. & Heintzelman, M. D. Pro-environmental, prosocial, pro-self, or does it depend? A more nuanced understanding of the motivations underlying residential solar panel adoption. Energy Res. Soc. Sci. 111 , (2024). Navigant. California Solar Initiative—Biennial Evaluation Studies for the Single ‐ Family Affordable Solar Homes (SASH) and Multifamily Affordable Solar Housing (MASH) Low ‐ Income Programs: Impact and Cost-Benefit Analysis Program Years 2011-2013 . https://www.cpuc.ca.gov/-/media/cpuc-website/files/legacyfiles/n/9323-navigant-csi-sash-mash-impact-and-cost-benefit-analysis-2011-2013.pdf (2015). Adams, J. A., Carley, S. & Konisky, D. M. Utility assistance and pricing structures for energy impoverished households: A review of the literature. Electr. J. 37 , (2024). SEEAction. Long-Term Performance of Energy Efficiency Loan Portfolios . https://emp.lbl.gov/publications/long-term-performance-energy (2022). NREL. End-Use Load Profiles for the U.S. Building Stock. https://www.nrel.gov/buildings/end-use-load-profiles.html (2022). HUD. Neighborhood stabilization program data. https://www.huduser.gov/portal/datasets/NSP.html (2022). U.S. Energy Information Administration. Annual Electric Power Industry Report, Form EIA-861 detailed data files . https://www.eia.gov/electricity/data/eia861/ https://www.eia.gov/electricity/data/eia861/ (2023). EIA. Natural Gas Prices. https://www.eia.gov/dnav/ng/ng_pri_sum_a_EPG0_FWA_DMcf_a.htm (2023). U.S. EIA. Weekly Heating Oil and Propane Prices (October-March). https://www.eia.gov/dnav/pet/pet_pri_wfr_a_EPLLPA_PRS_dpgal_w.htm. U.S. EIA. Units and calculators explained: British thernal units (Btu). https://www.eia.gov/energyexplained/units-and-calculators/british-thermal-units.php. NREL. End-Use Savings Shapes: Public Dataset Release . https://docs.nrel.gov/docs/fy23osti/84931.pdf (2022). NREL. National Residential Efficiency Measures Database. https://remdb.nrel.gov/group_listing. IRS. Energy Efficient Home Improvement Credit. Energy Efficient Home Improvement Credit (2025). https://www.irs.gov/credits-deductions/energy-efficient-home-improvement-credit Gagnon, P., Margolis, R., Melius, J., Phillips, C. & Elmore, R. Photovoltaic Technical Potential in the United States . https://www.nrel.gov/docs/fy16osti/65586.pdf (2016). Footnotes It is partly for this reason that we model only the full-loan and full-rebate financing models for weatherization, rather than also modeling intermediate scenarios with varying combinations of loan- and rebate-financing as we do for solar. Additional Declarations There is NO Competing Interest. Supplementary Files SupplementalInformation.docx Supplementary information Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8725239","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":583478264,"identity":"456f46e8-7cb3-468a-af9a-698b7cc249e8","order_by":0,"name":"Sydney Forrester","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA00lEQVRIiWNgGAWjYFACxgcI1sMGC2K0MBuAKR4QK7FBgjQtbBJEaZGf3cz4uHKPjb09e++zisQdEgzy7j0GeLUY3DnMbHjmWVpiD89xsxuJZyQYDM+cIaBFIv+YZMOBwwk8EmlsNxLbgFpmpCXgd9iMZDaglv/2PPLP2AqI0sJwA6zlAGOPBBsbA0iLvETyAcJ+aTiQnNhzJo1ZAugXHgOew/i1gELsYcMBO3v29mOMHz7usJGTb29swO8w9IjgMcBvBxYtDPIE7BgFo2AUjIKRBwDwjkM+FNb/WQAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0001-8430-843X","institution":"Lawrence Berkeley National Laboratory","correspondingAuthor":true,"prefix":"","firstName":"Sydney","middleName":"","lastName":"Forrester","suffix":""},{"id":583478265,"identity":"0dede55a-a057-4c0f-a364-93eaef92bbba","order_by":1,"name":"Eric O’Shaughnessy","email":"","orcid":"","institution":"Lawrence Berkeley National Laboratory","correspondingAuthor":false,"prefix":"","firstName":"Eric","middleName":"","lastName":"O’Shaughnessy","suffix":""},{"id":583478266,"identity":"00cda0c9-ebdd-4263-867f-8d683ba5a25e","order_by":2,"name":"Galen Barbose","email":"","orcid":"https://orcid.org/0000-0003-4687-2309","institution":"Lawrence Berkeley National Laboratory","correspondingAuthor":false,"prefix":"","firstName":"Galen","middleName":"","lastName":"Barbose","suffix":""}],"badges":[],"createdAt":"2026-01-28 21:45:28","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8725239/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8725239/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":101762836,"identity":"9e7a395c-3cb5-46a9-8540-dd1cee24702f","added_by":"auto","created_at":"2026-02-03 11:29:04","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":72917,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eCost and efficacy of modeled percent-of-income payment plans: Assistance is bound to a $1,800 annual maximum. Results are presented across regions, with the top of each line segment (diamond) corresponding to the baseline pre-intervention share of low-income households with high energy burdens and the bottom of the line segment (circle) showing the share after receiving bill assistance.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-8725239/v1/43706cdf4737e41acaef9e9a.png"},{"id":101762842,"identity":"edf7e98c-e147-4ae6-a837-00a26557e2ae","added_by":"auto","created_at":"2026-02-03 11:29:04","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":99320,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eCost effective share of eligible households as a function of incentive level: Dotted line follows share of household investments that yield SIR over 1 (cost effective from program perspective) while solid line follows share of household investments that yield positive net savings (cost effective from household perspective). The rebate levels for weatherization go from zero to full, while that for solar include intermediate $/W levels. For the combination of solar and weatherization, weatherization is fully rebated while solar is incentivized at varying levels.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-8725239/v1/79fa474b51b571c58477af2b.png"},{"id":101942884,"identity":"c0be3ce2-62fc-4843-b1f8-4ee0703352b5","added_by":"auto","created_at":"2026-02-05 09:39:10","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":165698,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eParticipation as a function of incentive level: Levels represent the share of households that yield both household net savings and program SIR \u0026gt; 1. Results are shown for (a) weatherization, (b) rooftop solar, and (c) weatherization + rooftop solar. Note that in (c), weatherization is fully rebated and rebates only vary for solar.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-8725239/v1/b07277940f7d2fef62d833bf.png"},{"id":102294840,"identity":"961ed30a-330c-45ae-8ce0-84bc9506d865","added_by":"auto","created_at":"2026-02-10 09:59:57","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":108441,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eNet bill savings across participant households from installed measures compared to that of PiPP: Participant households include only those that receive net savings and have a SIR\u0026gt;1. The percentage shown is the percent of low-income, energy burdened (i.e., PiPP-eligible) households that the participants represent. Boxes indicate median and 25\u003c/em\u003e\u003csup\u003e\u003cem\u003eth\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e/75\u003c/em\u003e\u003csup\u003e\u003cem\u003eth\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e percentiles while whiskers indicate 5\u003c/em\u003e\u003csup\u003e\u003cem\u003eth\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e and 95\u003c/em\u003e\u003csup\u003e\u003cem\u003eth\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e percentiles.\u003c/em\u003e \u003cem\u003eWeatherization measures are fully subsidized in the solar + weatherization case.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-8725239/v1/ccc79421d6fe1524f7a07c71.png"},{"id":101762844,"identity":"d54c370b-c98c-417e-8584-4e5ae817ace1","added_by":"auto","created_at":"2026-02-03 11:29:04","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":132007,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eRegional net bill savings across participant households from installed measures compared to that of PiPP: Participant households include only those that receive net savings and have a SIR\u0026gt;1. The percentage shown is the percent of low-income, energy burdened (i.e., PiPP-eligible) households that the participants represent. Boxes indicate median and 25th/75th percentiles while whiskers indicate 5th and 95th percentiles. Weatherization measures are fully subsidized in the solar + weatherization case.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-8725239/v1/4e2719c37ceab78f38fd7615.png"},{"id":101762841,"identity":"319b204c-177c-4bb8-8365-b6eb04485a4e","added_by":"auto","created_at":"2026-02-03 11:29:04","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":134033,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eComparison of various program designs: Dotted lines include the installed measures with bill assistance (percent of income payment plan). Line segments represent changes moving (left to right) from fully loan-financed (empty circle) to a full upfront rebate. For the weatherization + solar case, weatherization is fully rebated. Dotted axes show efficacy and NPC of PiPP. Thus, points in the upper-left quadrant are cheaper, but less effective; lower-left are both cheaper and more effective; lower-right are more expensive and more effective when compared to PiPP alone.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-8725239/v1/dc127f1dee644691f7c6a2c0.png"},{"id":101762843,"identity":"96adb6b9-a4f7-48bb-a4c5-a3b9166876e1","added_by":"auto","created_at":"2026-02-03 11:29:04","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":129143,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eComparison of various program designs without state or federal tax credits: Dotted lines include the installed measures with bill assistance (percent of income payment plan). Line segments represent changes moving (left to right) from fully loan-financed (empty circle) to a full upfront rebate. For the weatherization + solar case, weatherization is fully rebated. Dotted axes show efficacy and NPC of PiPP. Thus, points in the upper-left quadrant are cheaper, but less effective; lower-left are both cheaper and more effective; lower-right are more expensive and more effective when compared to PiPP alone.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-8725239/v1/813f6b97ce48eb34490170c5.png"},{"id":102299688,"identity":"fd9975f3-f685-47e3-b4ec-fcdcdba77adb","added_by":"auto","created_at":"2026-02-10 11:08:00","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1132407,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8725239/v1/5331a128-a0d8-4759-8fed-e95eb8e922be.pdf"},{"id":101762839,"identity":"ae6f2c8f-5387-4544-b21a-f86cf7e82b92","added_by":"auto","created_at":"2026-02-03 11:29:04","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":107248,"visible":true,"origin":"","legend":"Supplementary information","description":"","filename":"SupplementalInformation.docx","url":"https://assets-eu.researchsquare.com/files/rs-8725239/v1/209eef839681d7b758218baf.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Evaluating technology upgrades as a complement to traditional bill assistance programs","fulltext":[{"header":"Introduction","content":"\u003cp\u003eIncreasing prices for electricity and other fuel have raised concerns about energy affordability. States promote energy affordability via state appropriations, utility and ratepayer contributions, or voluntary mechanisms.\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e In addition to those broader efforts, targeted programs can provide direct relief to those most impacted by rising pricing, especially lower income households, who often have disproportionately higher energy burden, defined as the share of gross household annual income that goes towards electricity and non-electric heating costs.\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e One study estimated that low-income households\u0026rsquo; median energy burden was triple that of non- low- or moderate-income households,\u003csup\u003e3\u003c/sup\u003e and some states have created parallel policy goals focused on reducing energy burden across households.\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eHistorically, bill assistance is a common strategy to decrease energy burden. Some U.S. states have Percent of Income Payment Plan (PiPP) programs, whereby eligible customers receive a discount on their utility bill sufficient to reduce their energy costs to some specified percent of their income. That percentage is often set at 6%, the level often defined as equating to a \u0026ldquo;high\u0026rdquo; energy burden (see Table SI.2 for a summary of state PiPP programs).\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e PiPP can immediate relief, and one survey illustrates that bill assistance was effective in maintaining a comfortable temperature for participants and avoiding utility disconnection.\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e However, that relief is short-term and does not address the underlying causes of high energy burden or provide co-benefits. These programs can therefore become expensive over the long term if households re-apply over multiple years, which increases ongoing programmatic costs and applicant burden.\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e In addition, PiPP programs may provide only a partial solution for many participants, as the bill discounts are typically capped at some maximum dollar amount per customer, leaving many participating customers still facing high energy burdens above 6%.\u003c/p\u003e \u003cp\u003eAlternatively, or in addition, to direct bill assistance, many states offer programs for income-qualifying customers to adopt home energy upgrades. Historically these programs have focused primarily on weatherization, but more recently some states have expanded their programs to include rooftop solar.\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e Compared to PiPP, installed measures have higher upfront costs, but offer long-term savings, as well as potential non-energy benefits such as improved home comfort, increased property value, reduced energy insecurity, and ability to afford other necessities.\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e These programs can be combined with PiPP to either \u0026ldquo;close the gap\u0026rdquo; for customers where the maximum dollar limit is reached, or to reduce the amount of direct bill relief needed to reduce energy burdens to the targeted level. This decreases bill assistance program cost directly, but also creates indirect avoided costs from reduced disconnections and arrearages or non-payment.\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e In tandem with income-qualified programs, some states and utilities have leveraged private and public dollars to introduce alternative financing via community development financial institutions or state investment banks and/or utility on-bill financing programs. These programs can mitigate high upfront measure costs and often offer lower interest rates through credit enhancements such as a loan-loss reserve or a guarantee to reduce financier risk.\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e When repayment is lower than bill savings throughout the loan term, households can pursue energy upgrades that will reduce their energy burden.\u003c/p\u003e \u003cp\u003ePrior studies have demonstrated the savings potential from installed measures for low-income households. For example, Blasnik et al. (2014) found that participants in the federal weatherization assistance program saw an average 12.4% decrease in bills, or \u003cspan\u003e$\u003c/span\u003e283 per year,\u003csup\u003e9\u003c/sup\u003e and Tonn et al. (2015) found that the share of participants who struggled to pay energy bills declined from 74% before the program to 59% one year after.\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e,\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e Another income-qualified pilot program of a fully-incentivized 3.5 kW system produced \u003cspan\u003e$\u003c/span\u003e587 in savings, or a 45% bill reduction, for participants.\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e Analyses of rooftop solar adoption by low-income households have found that, in the U.S., low-income rooftop solar has reduced the incidence of high energy burden from an estimated 67% to 52% and have reduced the incidence of energy insecurity.\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e Kerby et al. (2024) studied storage, solar-plus-storage, weatherization, and appliance efficiency upgrades together and found that weatherization produced the highest savings, but that results varied widely between households.\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eAs states grapple with growing energy affordability challenges, it will be important to consider a broad range of programmatic options and understand their complementarity. Our study aims to address that by evaluating the performance of a PiPP program compared to income-qualifying weatherization and rooftop solar programs, as well as all combinations thereof while considering varying combinations of loan-financing versus upfront rebates. We perform this analysis on a modeled dataset of 111,410 low-income households across the U.S., used widely in other published research studies.\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e The performance of each program model in evaluated in terms of two primary metrics: efficacy in reducing the percentage of low income households with \u0026ldquo;high\u0026rdquo; energy burdens (defined as exceeding 6% of income) and the program\u0026rsquo;s net present cost, assuming only cost-effective measures are installed. The results illustrate key tradeoffs and show that income-qualifying weatherization and rooftop solar programs can often be coupled with PiPP to improve both efficacy and cost compared to PiPP alone.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eIn the following results, we begin by presenting outcomes for a PiPP program designed to reduce the incidence of high (\u0026gt;6%) energy burdens among low-income households (incomes less than or equal to 80% of their area median income). Outcomes are presented in terms of efficacy in reducing the incidence of low-income households with high energy burdens, as well as the net present cost (NPC) per low-income household. We then evaluate two installed measure programs\u0026mdash;weatherization and rooftop solar\u0026mdash;as alternative programmatic models for reducing energy burden, first considering each individually and in combination with each other, and then later in combination with PiPP. For these installed measures, we consider varying levels of upfront rebates and the remaining customer cost financed through loans with lower interest rates funded under a loan-loss reserve program. Those two financing models represent the most common approaches used in low-income technology upgrade programs.\u003c/p\u003e\n\u003cp\u003eBoth PiPP and the installed measures are constrained in their scale of implementation, though in different ways. In the case of PiPP, we assume all eligible low-income households receive assistance but limit the total annual bill discount to the lesser of: (a) the amount required to reduce energy burden to 6% or (b) a total annual bill discount of $1,800. That cap is based roughly on current program designs in the United States (see Methods and Table SI.2). The installed measures instead are limited to households where two cost-effectiveness criteria are both met: (1) the measure package has a \u0026ldquo;savings to investment\u0026rdquo; (SIR) ratio greater than or equal to 1 from the program perspective\u0026mdash;that is, the total cost of the program (rebates and any capital for a loan-loss reserve) cannot exceed the gross value of the total lifetime energy savings\u0026mdash;and (2) the upgrade is immediately cash-flow positive for the household\u0026mdash;that is, the bill savings exceed any loan repayment costs. The first of those criteria is a standard screening metric used, for example, in the federal Weatherization Assistance Program for installed energy conservation measures.\u003csup\u003e15\u003c/sup\u003e The second is intended to represent the bare minimum financial return necessary to motivate a household to participate, recognizing that actual household decision-making is considerably more complex.\u003csup\u003e16\u0026ndash;18\u003c/sup\u003e Within our analysis, those criteria are evaluated on a household-by-household basis, based on modeled energy and net bill savings for each individual household and the associated programmatic costs. Additional details on the application of these criteria are described in the Methods. The results presented below focus on national and regional outcomes.\u003c/p\u003e\n\u003cp\u003eBill assistance: Percent of Income Payment Plan\u003c/p\u003e\n\u003cp\u003eWithin our sample of modeled low-income households, 66% have an initial (pre-intervention) energy burden above 6% and are therefore eligible for PiPP. The $1,800 cap on annual bill discounts would be binding for roughly 37% of those participating households, thus reducing the incidence of high energy burdens from 66% to 34% of total low-income households. Provided continuously over 25 years, this level of bill assistance would incur a NPC of roughly $18,600 per participant, on average. That 25-year period is used to correspond roughly to the lifetime of the alternative installed measures considered.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAs shown in Fig. 1, outcomes vary across regions in terms of both the efficacy of the PiPP program in reducing energy burdens to below 6% and its cost, reflecting underlying regional differences in energy costs and incomes. The baseline incidence of high energy burden is considerably lower in the West than other regions, due in part to generally milder climates and thus lower overall energy costs. Consequently, PiPP leads to the lowest incidence of high energy burdens there, albeit starting from a much lower baseline. Among the other three regions, which all have a similar baseline incidence of high energy burdens, PiPP efficacy is greatest in the South, largely as a result of lower energy prices and costs. Conversely, efficacy is lowest in the Northeast, due to relatively high energy prices and thus more frequently binding caps on the annual bill discount. The NPC of PiPP is inversely related to its efficacy, with higher costs in regions where the funding cap is more frequently an active constraint. Of all PiPP-eligible participants, 52% of Northeast households require more than the maximum level of assistance whereas 41% do in the Midwest, 32% in the South, and 25% in the West.\u003c/p\u003e\n\u003cp\u003eInstalled Measure Programs: Participation Potential\u003c/p\u003e\n\u003cp\u003eThe efficacy of technology-based programs in reducing the incidence of high energy burden among low-income households is a function of participation rates and the net savings among participating households. Focusing first on participation rates, these depend on what fraction of eligible households pass the requisite set of cost-effectiveness thresholds. That fraction, in turn, depends on the funding model: a higher upfront rebate reduces cost-effectiveness to the program administrator (due to lower SIR), while a lower upfront rebate and correspondingly higher reliance on financing reduces cost-effectiveness to the household (due to lower net customer savings). \u0026nbsp;In addition to those factors, participation may also be constrained by eligibility rules. In our analysis, the weatherization measures do not apply to the 1% of low income households with high energy burdens whose homes are already well-sealed (Table SI.3), and the rooftop solar program does not apply to the 3% in multi-family dwellings (see Table SI.3 and Table SI.4).\u003c/p\u003e\n\u003cp\u003eFigure 2 illustrates how the program financing model (rebate vs. loan) impacts the set of households that pass the two cost-effectiveness screens. As shown, the solar-only program is most sensitive to the choice of financing, due to higher measure cost. \u0026nbsp;Substituting even just a portion of the loan support with an upfront rebate (e.g. at the $0.5 or $1/W level) substantially increases the portion of households for which the measure is cost-effective, while still satisfying the program administrator SIR constraint. \u0026nbsp;At a roughly $1.5/W rebate, the solar measure is cost effective for almost all eligible households, and any further increases in rebate level yield rapidly diminishing returns due to the program SIR constraint. \u0026nbsp;By comparison, the cost effectiveness of the weatherization (Wx) measures is much less sensitive to the choice of financing, due to the significantly lower measure costs.\u003ca href=\"#_ftn1\" name=\"_ftnref1\" title=\"\"\u003e\u003c/a\u003e\u003csup\u003e1\u003c/sup\u003e To illustrate the potential complementarity, we consider a program that combines fully rebated weatherization measures with solar under a varying loan/rebate mix. Notably, this combined program design generally improves cost-effectiveness for both the customer and the program administrator, leading to an upward shift in the curves for the combined measures (relative to the rightmost points on the corresponding weatherization-only curves). The combined program design also allows for households that could not benefit from one technology alone to benefit from the other, increasing potential further.\u003c/p\u003e\n\u003cp\u003eDepending on the rebate level, either the customer cost-effectiveness or the program administrator SIR constraint may be binding. \u0026nbsp;Combining these two, Figure 3 shows the percentage of the target population that passes both cost-effectiveness thresholds (we call this the \u0026ldquo;participation potential\u0026rdquo;) across the range of rebate/loan combinations for each program design and region. \u0026nbsp;Not surprisingly, participation potential for the weatherization-only program is less impacted by the form of financial support and ranges from roughly 89% to 93% of the target population across the four regions, when fully rebated. Participation potential is somewhat lower in the West and South compared to the other regions, due to milder climates in the West and low energy prices in the South. \u0026nbsp; For the solar program, participation potential is generally highest around a rebate of $1.5/W, reaching 90% or more of the target population in each region. \u0026nbsp;When combined, weatherization provides more savings per dollar up to a limit while solar provides deeper savings, but at a higher cost. Weatherization can reduce the solar size in some cases, avoiding further cost while providing the same level of bill savings. Results show that, when weatherization is fully rebated, this allows for high participation potential for the combination across all regions regardless of solar rebate level.\u003c/p\u003e\n\u003cp\u003eInstalled Measure Programs: Net Bill Savings\u003c/p\u003e\n\u003cp\u003eNet bill savings are the difference between the direct savings on the customer\u0026rsquo;s utility bills (inclusive of electricity, natural gas, and other heating fuels) and any loan repayment costs. The modeled weatherization program includes a standard set of insulation and air sealing measures, which generally yield relatively modest bill savings, but at a modest cost (see Methods and Table SI.5). Compared to the weatherization measures, the rooftop solar program offers greater bill savings potential, but at a substantially greater cost. The solar program assumes a maximum 5 kW installation, subject to each home\u0026rsquo;s usable roof space and limited in size such that annual solar production is no greater than the customer\u0026rsquo;s regional average annual ratio of solar to electricity consumption (see Methods and Table SI.6). These size limitations bind differently across regions, depending mostly on the intensity of heating/cooling loads and the prevalence of electric space heating, and are also impacted when solar is combined with weatherization (see Table SI.4).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFigures 4 and 5 compare the net bill savings for each technology-based program to PiPP\u0026mdash;focusing on the full national sample of low-income households in Figure 4 and the regional results in Figure 5. While Figures 2 and 3 illustrate how weatherization and solar complement one another to increase cost-effective participation, Figure 4 demonstrates that the complementarity extends to deeper participant savings that compete with PiPP levels. Alone, net annual bill savings from weatherization are $546 if loan-financed or $634 if financed entirely via upfront rebate. By comparison, median bill savings under the PiPP program are more than double that amount, at $1,373 per year. \u0026nbsp;Weatherization bill savings vary to some extent by region (Figure 5)\u0026mdash;with the greatest savings in Northeast, due to relatively high energy prices and cold winter temperatures\u0026mdash;though in all cases are well below the median bill savings under PiPP. The weatherization measures considered in this analysis are rather modest. As such, deeper investments and savings could compete better with PiPP levels.\u003c/p\u003e\n\u003cp\u003eAlone, solar provides a deeper investment and is considerably more expensive than the weatherization measures considered. Thus, net savings are more sensitive to the form of financing provided. If financed entirely through a loan, the median net bill savings across all participants is $205 annually, but grows to $1,042 with a full upfront rebate. \u0026nbsp;However, even at that higher level, the savings are still less than PiPP for the full national sample. The West offers an exception when fully rebated, owing to greater solar production levels and a milder winter climate throughout much of the region.\u003c/p\u003e\n\u003cp\u003eThe combination of solar and weatherization alleviates the issue of deep potential savings for weatherization alone as well as the issue of high sensitivity to incentive levels for solar alone and can reach net savings comparable to that of PiPP. If financed fully through upfront rebates, median annual bill savings reach $1,620 across all participants nationally, which exceeds the benefit provided under PiPP. Notably, the same dynamic holds across all regions. Of course, as the financing model shifts away from upfront rebates and toward low-cost loans, the net savings diminish; with a fully loan-financed design, the median net bill savings in all regions falls below PiPP. Among the regions, the Northeast and West are closest to retaining parity with PiPP in terms of net bill savings, owing largely to the set of factors previously noted for those two regions.\u003c/p\u003e\n\u003cp\u003eComparing Efficacy and Cost across Alternate Program Models\u003c/p\u003e\n\u003cp\u003eBased on the preceding participation rates and participant net bill savings, Figure 6 compares the efficacy and cost of each of the technology-based program models to PiPP (regional results are shown in Figure SI.1).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAs shown, the combination of weatherization measures and rooftop solar funded fully through upfront rebates compete directly with PiPP, providing better efficacy (reducing high burden incidence to 30% vs. 34%) at a lower program net present cost (25-year NPC of $11,454 vs. $12,288). \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWhen combined with PiPP, cost-effective installed measures reduce ongoing reliance on PiPP. When weatherization with solar are combined with PiPP, there is some \u0026ldquo;Pareto improvement\u0026rdquo; compared to PiPP alone (i.e., greater efficacy at the same or lower cost, or lower cost at the same or greater efficacy). \u0026nbsp;This occurs either by allowing deeper energy burden reductions than can occur through bill assistance alone (due to the per-participant annual limits) and/or by reducing the amount of bill assistance needed to reduce energy burdens to the 6% threshold. \u0026nbsp;These Pareto improvements are shown in Fig. 6 by outcomes in the lower left-hand quadrant. \u0026nbsp;In the case of weatherization, Pareto improvements occur whenever cost-effective measures are combined with PiPP, regardless of the financing model. Instead, for rooftop solar, Pareto improvements are more likely to occur when the larger portion of the cost is loan-financed, rather than relying solely on upfront incentives. When fully-rebated weatherization and rooftop solar rebated at $0.62/W are both combined with PiPP, they can substantially reduce the incidence of high energy burden to 17% of low-income households (compared to 34% with PiPP alone), at the same NPC.\u003c/p\u003e\n\u003cp\u003eSensitivity to Tax Credit Availability\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe preceding results assume a 30% tax credit for the modeled technology upgrades, similar to the Investment Tax Credit for rooftop solar and the Energy Efficient Home Improvement Credit for weatherization. We also assume state solar tax credits and the Home Efficiency Rebates administered at the state level (see Methods). Since a subset of these tax credits will soon expire, we consider here the extreme case of how elimination of all tax credits (including state-administered credits) would impact the performance of the technology-based programs relative to PiPP.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe loss of tax credits reduces cost-effective adoption of installed measures, and increases the cost for the remaining cost-effective participants (see Figure SI.2 for participation potential). This shifts the performance curves in Figure 7 upward (lower efficacy) and to the right (higher cost). This may require either lower rebates to achieve better program economics, which would reduce household savings; or higher rebates to improve affordability, which may lead to fewer households served.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWithout tax credits, installed measures cannot compete with PiPP on their own. Even with full rebates, efficacy of PiPP is superior to the combined installed measures. When combined with PiPP, providing access to low-cost capital still provides marginal improvements both in efficacy and reduced cost. For example, access to financing for weatherization and solar, separately, reduces NPC $1,546 and $98 per low-income household and reduces the incidence of high energy burden 5.7 and 0.7 percentage points, respectively, compared to PiPP alone. Close to Pareto improvement, fully-rebated weatherization with PiPP would have a NPC similar to PiPP alone, but reduce high energy burden an additional 11.8 percentage points.\u0026nbsp;\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eEnergy affordability programs have expanded in recent years beyond direct bill assistance to include various cost-effective installed measures that can provide ongoing bill reduction such as weatherization and rooftop solar. Program goals may have multiple, sometimes competing, objectives such as those for households in reducing incidence of high energy burden, maximizing participation, or capturing non-energy benefits; and those for the program administrators such as controlling upfront and ongoing program costs, achieving state or utility targets or goals for affordability or DER deployment. There is no \u0026ldquo;best\u0026rdquo; program design and programs must balance these tradeoffs. On its own, direct bill assistance is flexible, does not require long installation time or large upfront investment, and can be applied to any household that pays their own utility bill regardless of building type or ownership status. However, bill assistance does not provide non-energy benefits or a long-term solution. The pros and cons for installed measures such as weatherization or solar are just the opposite. As such, programs have begun to explore the complementary relationship between direct bill assistance and installed measures. For example, an income-qualified solar program demonstrated economic benefits to the utility due to reduced assistance spending, reduced defaults and delinquencies, and reduced administrative costs.\u003csup\u003e19\u003c/sup\u003e Another program that used weatherization funds for solar acknowledged that solar unlocked the potential for much higher savings, however, at a higher cost that required supplemental funds and could limit the number of households served by a fixed program budget.\u003csup\u003e4\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003eOur results find that bill assistance, weatherization, and solar\u0026mdash;each with their respective tradeoffs\u0026mdash;can complement one another. Together, weatherization reduces the size of solar required and improves the benefit-to-cost ratio while solar deepens potential savings. If further combined with PiPP, ongoing bill savings from weatherization and solar decreases or eliminates bill assistance that a household may require and serves more households. Results are optimized when rebates are set high enough to produce meaningful household savings, but low enough for a program to accommodate a larger number of participants. Indeed, our results show that the combination of cost-effective weatherization with a full rebate and solar with a $1/W rebate decreases incidence of high energy burden among low-income households more than PiPP would, and at a lower NPC. Combined with PiPP, all three interventions achieve the lowest incidence of high energy burden compared to any other scenario.\u003c/p\u003e\n\u003cp\u003eThese results illustrate a spectrum of options for affordability program design. For example, if a program instead is not aiming to optimize a PiPP program, but to encourage cost-effective weatherization and solar without a large level of funding, our results show that low-cost loans with customer protections could improve affordability, even without PiPP or upfront rebates. Indeed, weatherization with solar reduced the incidence of high energy burden 11 points with access to low-interest loans alone (Fig. 4). In another scenario, if a program solely aimed to maximize deployment of weatherization and solar without PiPP, it may focus on higher rebate levels instead. For a program with constrained upfront time and capital, PiPP could provide immediate affordability to any qualified household while allowing time for complementary program design. In these cases, pairing PiPP alone with financing and low-cost capital to allow for households to adopt cost-effective measures on their own improves both cost and efficacy.\u003c/p\u003e\n\u003cp\u003eOur study includes several limitations. First, it does not include other strategies that could impact home energy burden such as demand response, community solar, energy storage, or electrification. It also only includes a set portfolio of weatherization measures without other energy efficient upgrades such as appliance replacement. Due to the focus on weatherization and rooftop solar, we do not consider renter-occupied households. We also use state-specific tariff assumptions to apply our findings across the United States that may not reflect nuances found in individual utility territories or programs. We also do not consider the cost of net energy metering or impacts on non-participating households of these programs. Another limitation is that we do not account for programmatic funding constraints that are binding in almost all actual cases. In reality, any assistance program may face funding limitations, and the level of allocated dollars may fluctuate greatly from year to year.\u003csup\u003e1,20\u003c/sup\u003e Finally, for participation, we assume a potential as a high bookend, but do not incorporate assumptions that would decrease actual participation, which would be much lower. For example, a study found that unsecured loans administered through similar financing programs had much lower default rates than unsecured consumer loans, potentially due to net bill savings and safeguards against predatory lending.\u003csup\u003e21\u003c/sup\u003e While this can expand access and reduce upfront costs for households, in cases where energy costs remain unaffordable, financing may expose vulnerable households that remain at risk for non-payment and deter participation.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThis study builds on a body of literature to better understand how states or utilities could employ multiple energy affordability strategies and how they may complement one another. Thoughtfully combining access to a series of interventions could help achieve broader affordability goals, decrease risk of customer shut off or non-payment, improve home comfort and health, and reduce reliance on bill assistance.\u0026nbsp;\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003eFiltering household models\u003c/p\u003e\n\u003cp\u003eFor this study, we specifically focused on households that would be candidates for bill assistance, weatherization, and/or low-income solar incentives. As such, we narrowed NREL\u0026rsquo;s full set of End Use Load Profiles (EULP) building models to those that were owner-occupied and with area median incomes (AMI) at or below 80%.\u003csup\u003e22\u003c/sup\u003e Data for each building model include building characteristics (e.g., units in structure, building envelope), household characteristics (e.g., tenure, number of occupants, income range), location (e.g., climate zone, municipal area, county), and energy use (e.g., fuel and electricity demand by end use). We used EULP data on each household\u0026rsquo;s county, number of occupants, and income range to determine AMI. Choosing a randomly selected value within the range and using the respective county and number of occupants, we assigned a percent AMI value to each household by comparing to the Housing and Urban Development specifications.\u003csup\u003e23\u003c/sup\u003e From a total set of 548,916 total building models, 307,480 were owner-occupied. Of those, 111,410 had an area median income at or below 80%, which make up the base of our study (see Table SI.1 for sample size of each step and other summary metrics by state). All monetary pricing and income data are taken from 2022 for consistency.\u003c/p\u003e\n\u003cp\u003eElectricity prices\u003c/p\u003e\n\u003cp\u003eBuilding models in EUSS include end-use electricity, natural gas, propane, and heating oil usage. Using the Energy Information Administration\u0026rsquo;s Form 861 electricity data, each utility has data on residential revenue, residential electricity sales, and residential customer counts.\u003csup\u003e24\u003c/sup\u003e Since these data do not include whether, or at what level, fixed charges or minimum bills are in place, we assume a monthly fixed charge of $10 for each residential customer. As a result, each utility\u0026rsquo;s volumetric rate is established by subtracting all fixed charges collected (equivalent to $120 annually by the total number of residential customers) from the total revenue collected and dividing that by the number of residential sales in kWh.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFor each county associated with more than one utility, the utility with the most customers for each respective state was selected as the county\u0026rsquo;s assigned utility. With this, each building model was linked to a specific utility and corresponding volumetric electricity rate. Flat rates were assumed as the baseline with net metering-- i.e., compensation for solar generations, including exports, at the retail rate value. Time sensitive rates were not considered the base case because only 10% of residential customers were enrolled in time-sensitive rates in 2023.\u003csup\u003e24\u003c/sup\u003e Only nine states had above-average shares of customers on time sensitive rates while the rest had near-zero shares. The states with the above-average shares include Delaware (57% of customers), Maryland (47%), Missouri (44%), Colorado (37%), Arizona (37%), Michigan (37%), Oklahoma (33%), California (30%), and Montana (15%).\u003csup\u003e24\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003eOther energy prices\u003c/p\u003e\n\u003cp\u003eFor non-electric energy prices, we used EIA 2022 state data (concurrent with electricity prices) for residential natural gas, propane, and heating oil.\u003csup\u003e25,26\u003c/sup\u003e All EULP energy consumption is in kWh. So, all non-electric prices were converted to kWh with: 1,039 British thermal unit (BTU) to 1 cubic foot of natural gas; 91,452 BTU to gallon of propane; 138,500 BTU to gallon of fuel oil; and 3,412 BTU to kWh.\u003csup\u003e27\u003c/sup\u003e \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eBill assistance: Percent of income payment plan (PiPP) design\u003c/p\u003e\n\u003cp\u003ePercent of income payment plans (PiPP) are offered in specific states or utility territories and are structured such that income-qualified customers are given assistance on their energy bills that drives their energy burden down to a specified level (most often 6%, but sometimes 10% if bills include more than solely electricity). Often these programs have a maximum annual discount of $150 per month, or $1,800 per year. Table SI.2 summarizes existing programs across the U.S. We use values modeled after existing programs to structure our PiPP bill assistance for this study, offering the smaller between a discount that would reduce household energy burden to 6% or $1,800 per year. As such, customers with a baseline energy burden below 6% would not receive additional bill assistance. The level of bill assistance changes based on how installed weatherization or solar may impact costs. This includes both bill savings as well as on-bill loan repayment, where applicable.\u003c/p\u003e\n\u003cp\u003eWeatherization measures, costs, and incentives\u003c/p\u003e\n\u003cp\u003eNREL\u0026rsquo;s EULP building models have corresponding upgrade scenarios (\u0026ldquo;End-Use Savings Shapes (EUSS)\u0026rdquo;). For the weatherization case, we consider the \u0026ldquo;Enhanced Enclosure\u0026rdquo; scenario, which includes attic, wall, and basement insulation as well as air and duct sealing (see Table SI.5).\u003csup\u003e28\u003c/sup\u003e NREL includes detailed logic on which buildings got what upgrades, based on baseline measures and housing condition as well as climate.\u003csup\u003e28\u003c/sup\u003e As such, each building is prescribed a different level of upgrades. For example, a building with no or very poor insulation in a region with many heating degree days may be prescribed more insulation in the weatherization scenario whereas a building with sufficient insulation in a mild climate may not. Additionally, some measures may not apply to certain buildings. For example, insulation for finished attics will not be performed for housing without a finished attic.\u003c/p\u003e\n\u003cp\u003eTo best pair with EUSS measures, we drew costs from NREL\u0026rsquo;s \u0026ldquo;National Residential Efficiency Measures Database\u0026rdquo;, which represents the total cost to implement the retrofit measure. The database presents one primary value as well as a range, depending on the pre-measure condition and measure selected. For each category, the database provides additional information on cost drivers. See Table SI.5 for the cost (and range) of each relevant measure. We use this pricing for each building\u0026rsquo;s characteristics (e.g., square footage of various conditioned spaces) to determine the total cost of the weatherization upgrade for each home.\u003csup\u003e29\u003c/sup\u003e NREL\u0026rsquo;s EUSS database then provides the consequent changes in energy consumption as a result of the upgrades, which we can use to quantify savings from weatherization for each building.\u003c/p\u003e\n\u003cp\u003eWhere federal tax credits are included, we consider two separate incentives. The first is modeled after the energy efficiency home improvement credit, which allows for 30% of the installed price up to a maximum limit of $1,200.\u003csup\u003e30\u003c/sup\u003e We also consider the Home Efficiency Rebates (HOMES) program, which is a rebate for home weatherization that is allowed to be stacked on top of other incentives. This only applies to households with 80% AMI or lower, which includes all households in our sample. The rebate provides differing levels of incentives based on the amount of modeled savings from the proposed upgrade. An 80% rebate is offered for households with AMI at or below 80% and energy savings more than 20% above the baseline. For measures leading to 20%-35% of savings, the maximum limit is $4,000 while, for savings above 35%, the maximum limit is $8,000. Measures that produce savings less than 20% are not offered this incentive.\u003c/p\u003e\n\u003cp\u003eRooftop solar sizing, cost, and incentives\u003c/p\u003e\n\u003cp\u003eGeneration is determined using the centroid of each household\u0026rsquo;s county and inputting that locational information into NREL\u0026rsquo;s System Advisor Model. We assume a South-facing array, tilt at latitude, fixed array with a 1.2 inverter loading ratio and 14% system losses. For system degradation, we assume 0.5% degradation per year. We then determine each household\u0026rsquo;s solar array sizing, taking into account restrictions created by either rooftop square footage or total annual load.\u003c/p\u003e\n\u003cp\u003eUsing Energy Sage data, we find the median ratio of household annual solar generation to annual load from each state (see Table SI.6). We apply that ratio to buildings based on their state as one constraint. We also consider rooftop square footage, where we assume that 70% may be suitable to host a solar array,\u003csup\u003e31\u003c/sup\u003e and that each square foot could host 15 Watts as the second constraint. The final rooftop solar array size is selected as the minimum between 5 kW, the solar generation to load ratio, or the available rooftop size restriction (see Table SI.4 for share of binding constraints).\u003c/p\u003e\n\u003cp\u003eRooftop solar costs are taken for each state, where available, or region. These data originate from Berkeley Lab\u0026rsquo;s Tracking the Sun. These include both material and soft costs in addition to any upfront or performance based incentives, but excludes tax credits and solar renewable energy credits. Per the authors\u0026rsquo; recommendation, system costs were filtered to exclude third-party owned arrays and then discounted by 20% to account for high loan origination fees. For each region, 2022 mean and standard deviation of solar system costs were taken (see Table SI.7). For each of the 1,000 iterations, each customer was assigned a cost per Watt based on a normal distribution that was specific to their state (or region).\u003c/p\u003e\n\u003cp\u003eSince tax credits and renewable energy credits are \u003cem\u003enot\u003c/em\u003e included in the Tracking the Sun values, we apply them separately. Where federal tax credits are considered, we assume 30%. Some states offer additional incentives which are considered (see Table SI.8). Since we assume host-ownership, we do not include any additional adders applicable for third parties such as energy community or low-income adders.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eUpfront rebates and loan assumptions\u003c/p\u003e\n\u003cp\u003eWe consider an income-qualified rebate program that provides capital required for low-income households with high or severe energy burdens to adopt weatherization and/or rooftop solar. Rebates may represent the full amount, intermediate levels (only for solar), or none at all. If provided in full, the program administrator covers all costs of adoption after relevant state and/or federal incentives. Intermediate levels are in various $/W levels for solar and it is assumed that the household would pay for the rest via a loan. Consistent with programs that allow use of bill assistance and/or weatherization to go towards solar installations,\u003csup\u003e4\u003c/sup\u003e we assume that programs will only provide solar incentives where the savings-to-investment ratio is greater than 1 (i.e., savings over lifetime of solar asset are more than upfront cost).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAs described in the Introduction, there are various programs that provide low-cost capital with consumer protections to households that seek to install energy technology such as weatherization measures or rooftop solar. These often include credit enhancement such as loan-loss reserves to buy down interest rates and zero/low levels of loan origination fees. Funds for loan-loss reserves may be supplied by a one-time grant or influx of money from public or ratepayer dollars and the capital itself may be supplied by financiers such as community development financial institutions or other impact oriented or local institutions. Where they exist, regional community development financial institutions or state investment banks may play a role in facilitating these partnerships.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe assume that a program administrator could create a loan-loss reserve with 5% of the capital lent to install these weatherization and/or solar measures. We then assume that this could reduce loan origination fees to negligible levels and provide interest rates with a mean of 7.5% and 1.25% standard deviation, determined randomly via standard distribution for each customer. We assume a household will use the loan to cover the full upfront cost that they are responsible for (i.e., after all rebates and tax credits) such that no money is required upfront. We take into consideration the loan repayment on the customer\u0026rsquo;s bill. As such, the resulting energy burden post-intervention will include both the consequent bill savings in addition to repayment of both loan principal and interest. Consequently, we assume that a household will not adopt if the investment causes their net energy costs to increase.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIt is important to note that the level of participation considered for this study is much higher than is realistic because, on the program side, there may be limits to spending either in terms of available capital for rebates or ongoing money for bill assistance. The customer side, there may be households unwilling to adopt, that may not qualify for loans, or who may want to avoid debt. Rather, this study is meant to illustrate and compare how different interventions could improve affordability and their subsequent programmatic costs.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e \u003ch2\u003eCompeting Interests\u003c/h2\u003e \u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eAuthor Contributions\u003c/h2\u003e \u003cp\u003eSPF conceived and designed the analysis, collected the data, contributed data or analysis tools, performed the analysis, wrote the paper. EO provided feedback on scope, provided edits. GB provided feedback on scope, assisted with writing, provided edits.\u003c/p\u003e\u003ch2\u003eAcknowledgements\u003c/h2\u003e \u003cp\u003eThis work was supported by the U.S. Department of Energy\u0026rsquo;s Office of Energy Efficiency and Renewable Energy under Solar Energy Technologies Office Agreement Number 38444 and Contract No. DE-AC02-05CH11231 in Fiscal Year 2024.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eShields, L. \u003cem\u003eBolstering Federal Energy Assistance and Weatherization With State Clean Energy Programs\u003c/em\u003e. https://www.ncsl.org/energy/bolstering-federal-energy-assistance-and-weatherization-with-state-clean-energy-programs (2020).\u003c/li\u003e\n\u003cli\u003eBrown, M. A., Soni, A., Lapsa, M. V, Southworth, K. \u0026amp; Cox, M. High energy burden and low-income energy affordability: conclusions from a literature review. \u003cem\u003eProg. Energy\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, (2020).\u003c/li\u003e\n\u003cli\u003eDrehobl, A., Ross, L. \u0026amp; Ayala, R. \u003cem\u003eHow High Are Household Energy Burdens? An Assessment of National and Metropolitan Energy Burden across the United States\u003c/em\u003e. \u003cem\u003eACEEE\u003c/em\u003e (2020).\u003c/li\u003e\n\u003cli\u003eCarrera, A. Incorporating Renewable Energy Technology into the Minnesota Weatherization Assistance Program: Reducing Energy Burden Among Low-Income Households. (University of Minnesota, 2023).\u003c/li\u003e\n\u003cli\u003eAPPRISE. \u003cem\u003e2018 National Energy Assistance Survey Final Report\u003c/em\u003e. https://neada.org/wp-content/uploads/2015/03/liheapsurvey2018.pdf (2018).\u003c/li\u003e\n\u003cli\u003eGraff, M. Addressing energy insecurity: Policy Considerations for enhancing energy assistance programs. \u003cem\u003eHeliyon\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, (2024).\u003c/li\u003e\n\u003cli\u003eTonn, B., Rose, E., Hawkins, B. \u0026amp; Conlon, B. \u003cem\u003eHealth and Household-Related Benefits Attributable to the Weatherization Assistance Program\u003c/em\u003e. (2014) doi:ORNL/TM-2014/345.\u003c/li\u003e\n\u003cli\u003eBell, C. J., Nadel, S. \u0026amp; Hayes, S. \u003cem\u003eOn-Bill Financing for Energy Efficiency Improvements: A Review of Current Program Challenges, Opportunities, and Best Practices ACEEE E118\u003c/em\u003e. https://www.aceee.org/sites/default/files/publications/researchreports/e118.pdf (2011) doi:E118.\u003c/li\u003e\n\u003cli\u003eBlasnik, M. \u003cem\u003eet al.\u003c/em\u003e \u003cem\u003eNational Weatherization Assistance Program Impact Evaluation: Energy Impacts for Single Family Homes\u003c/em\u003e. https://weatherization.ornl.gov/wp-content/uploads/pdf/WAPRetroEvalFinalReports/ORNL_TM-2015_13.pdf (2014) doi:ORNL/TM-2015/13.\u003c/li\u003e\n\u003cli\u003eTonn, B., Rose, E. \u0026amp; Hawkins, B. \u003cem\u003eSurvey of Recipients of Weatherization Assistance Program Services: Assessment of Household Budget and Energy Behavior Pre- to Post- Weatherization\u003c/em\u003e. https://www.osti.gov/servlets/purl/1223652 (2015) doi:ORNL/TM-2015/64.\u003c/li\u003e\n\u003cli\u003eYozwiak, M. \u003cem\u003eet al.\u003c/em\u003e The effect of residential solar on energy insecurity among low- to moderate-income households. \u003cem\u003eNat. Energy\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, (2025).\u003c/li\u003e\n\u003cli\u003eForrester, S. P., Monta\u0026ntilde;\u0026eacute;s, C. C., O\u0026rsquo;Shaughnessy, E. \u0026amp; Barbose, G. Modeling the potential effects of rooftop solar on household energy burden in the United States. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, (2024).\u003c/li\u003e\n\u003cli\u003eKerby, J., Hardy, T., Twitchell, J., O\u0026rsquo;Neil, R. \u0026amp; Tarekegne, B. A targeted approach to energy burden reduction measures: Comparing the effects of energy storage, rooftop solar, weatherization, and energy efficiency upgrades. \u003cem\u003eEnergy Policy2\u003c/em\u003e \u003cstrong\u003e184\u003c/strong\u003e, (2024).\u003c/li\u003e\n\u003cli\u003eNREL. ResStock Publications. https://resstock.nrel.gov/page/publications.\u003c/li\u003e\n\u003cli\u003eDOE. Weatherization Assistance Program Notice 22-7 Table of Issues. at https://www.energy.gov/scep/wap/articles/weatherization-program-notice-22-7-weatherization-health-and-safety (2021).\u003c/li\u003e\n\u003cli\u003eMcKenna, C., Gronlund, C., Hern\u0026aacute;ndez, D. \u0026amp; Vaishnav, P. When homeowners lose momentum after an energy audit: Barriers to completing weatherization in the United States Midwest. \u003cem\u003eEnergy Res. Soc. Sci.\u003c/em\u003e \u003cstrong\u003e122\u003c/strong\u003e, (2025).\u003c/li\u003e\n\u003cli\u003eWolske, K. S., Stern, P. C. \u0026amp; Dietz, T. Explaining interest in adopting residential solar photovoltaic systems in the United States: Toward an integration of behavioral theories. \u003cem\u003eEnergy Res. Soc. Sci.\u003c/em\u003e \u003cstrong\u003e25\u003c/strong\u003e, (2017).\u003c/li\u003e\n\u003cli\u003eLegault, L., Bird, S. \u0026amp; Heintzelman, M. D. Pro-environmental, prosocial, pro-self, or does it depend? A more nuanced understanding of the motivations underlying residential solar panel adoption. \u003cem\u003eEnergy Res. Soc. Sci.\u003c/em\u003e \u003cstrong\u003e111\u003c/strong\u003e, (2024).\u003c/li\u003e\n\u003cli\u003eNavigant. \u003cem\u003eCalifornia Solar Initiative\u0026mdash;Biennial Evaluation Studies for the Single\u003c/em\u003e\u003cem\u003e‐\u003c/em\u003e\u003cem\u003eFamily Affordable Solar Homes (SASH) and Multifamily Affordable Solar Housing (MASH) Low\u003c/em\u003e\u003cem\u003e‐\u003c/em\u003e\u003cem\u003eIncome Programs: Impact and Cost-Benefit Analysis Program Years 2011-2013\u003c/em\u003e. https://www.cpuc.ca.gov/-/media/cpuc-website/files/legacyfiles/n/9323-navigant-csi-sash-mash-impact-and-cost-benefit-analysis-2011-2013.pdf (2015).\u003c/li\u003e\n\u003cli\u003eAdams, J. A., Carley, S. \u0026amp; Konisky, D. M. Utility assistance and pricing structures for energy impoverished households: A review of the literature. \u003cem\u003eElectr. J.\u003c/em\u003e \u003cstrong\u003e37\u003c/strong\u003e, (2024).\u003c/li\u003e\n\u003cli\u003eSEEAction. \u003cem\u003eLong-Term Performance of Energy Efficiency Loan Portfolios\u003c/em\u003e. https://emp.lbl.gov/publications/long-term-performance-energy (2022).\u003c/li\u003e\n\u003cli\u003eNREL. End-Use Load Profiles for the U.S. Building Stock. https://www.nrel.gov/buildings/end-use-load-profiles.html (2022).\u003c/li\u003e\n\u003cli\u003eHUD. Neighborhood stabilization program data. https://www.huduser.gov/portal/datasets/NSP.html (2022).\u003c/li\u003e\n\u003cli\u003eU.S. Energy Information Administration. \u003cem\u003eAnnual Electric Power Industry Report, Form EIA-861 detailed data files\u003c/em\u003e. \u003cem\u003ehttps://www.eia.gov/electricity/data/eia861/\u003c/em\u003e https://www.eia.gov/electricity/data/eia861/ (2023).\u003c/li\u003e\n\u003cli\u003eEIA. Natural Gas Prices. https://www.eia.gov/dnav/ng/ng_pri_sum_a_EPG0_FWA_DMcf_a.htm (2023).\u003c/li\u003e\n\u003cli\u003eU.S. EIA. Weekly Heating Oil and Propane Prices (October-March). https://www.eia.gov/dnav/pet/pet_pri_wfr_a_EPLLPA_PRS_dpgal_w.htm.\u003c/li\u003e\n\u003cli\u003eU.S. EIA. Units and calculators explained: British thernal units (Btu). https://www.eia.gov/energyexplained/units-and-calculators/british-thermal-units.php.\u003c/li\u003e\n\u003cli\u003eNREL. \u003cem\u003eEnd-Use Savings Shapes: Public Dataset Release\u003c/em\u003e. https://docs.nrel.gov/docs/fy23osti/84931.pdf (2022).\u003c/li\u003e\n\u003cli\u003eNREL. National Residential Efficiency Measures Database. https://remdb.nrel.gov/group_listing.\u003c/li\u003e\n\u003cli\u003eIRS. Energy Efficient Home Improvement Credit. Energy Efficient Home Improvement Credit (2025). https://www.irs.gov/credits-deductions/energy-efficient-home-improvement-credit\u003c/li\u003e\n\u003cli\u003eGagnon, P., Margolis, R., Melius, J., Phillips, C. \u0026amp; Elmore, R. \u003cem\u003ePhotovoltaic Technical Potential in the United States\u003c/em\u003e. https://www.nrel.gov/docs/fy16osti/65586.pdf (2016).\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Footnotes","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003e It is partly for this reason that we model only the full-loan and full-rebate financing models for weatherization, rather than also modeling intermediate scenarios with varying combinations of loan- and rebate-financing as we do for solar.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-8725239/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8725239/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eIncreasing electricity prices have raised concerns about energy affordability. State programs to decrease energy burden (share of income spent on energy) could help mitigate these concerns. Many states offer some combination of bill assistance and rebates for installed measures such as weatherization and rooftop solar to income-qualifying households. Here, we analyze how these approaches may complement each other to improve energy affordability. Results illustrate tradeoffs between program costs and energy burden reduction as well as between a program\u0026rsquo;s upfront capital costs versus ongoing costs. Generally, the three strategies complement one another to improve energy affordability at a lower net present cost than bill assistance alone. This holds true, to varying degrees, across regions. Weatherization can decrease the solar installation needed; and both together can decrease or eliminate ongoing reliance on bill assistance. Fully rebated weatherization and solar rebated at \u003cspan\u003e$\u003c/span\u003e0.62/Watt, when combined with bill assistance, yield the same modeled net present cost as bill assistance alone and reduce the share of low-income households with high energy burdens from 66% to 17%, compared to bill assistance alone (to 34%). Absent tax credits, weatherization still reduces energy burden and program cost of bill assistance, whereas solar may not compete due to cost.\u003c/p\u003e","manuscriptTitle":"Evaluating technology upgrades as a complement to traditional bill assistance programs","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-03 11:28:59","doi":"10.21203/rs.3.rs-8725239/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"d10f7325-b1ac-4710-996f-cc3a67b6b16b","owner":[],"postedDate":"February 3rd, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":62080010,"name":"Scientific community and society/Energy and society/Energy efficiency"},{"id":62080011,"name":"Scientific community and society/Energy and society/Energy justice"}],"tags":[],"updatedAt":"2026-05-05T19:55:28+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-03 11:28:59","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8725239","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8725239","identity":"rs-8725239","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}