UC Davis

The scope of this study is to estimate the change in household energy consumption before and after the Low-Income Weatherization Program. In chapter 1, we econometrically estimated daily energy savings by detailed weatherization types and hourly energy savings. We use California ISO’s wholesale electric price to estimate the savings at electric market perspective. We conducted robustness tests and confirmed that our approach could be used to estimate the impact of weatherization. We compare average daily energy savings from treated households to control to evaluate inconsistencies. Calculated savings were 1.9 kWh to 2 kWh daily on average. We find that weatherized homes in the sample reduced electricity consumption during afternoon peak hours when the cost of supplying the energy is high. Electricity savings were also greater when ambient temperature was high.In chapter 2, we present three steps analysis to understand weatherization program participation and savings heterogeneity. Our approach is to use households’ prior energy consumption and home characteristics. We estimated household specific usage pattern from pre-weatherization energy use. We found that households with greater temperature sensitivity (i.e., had higher energy demand on hot days), was associated with greater reduction in energy consumption after the weatherization. This study contributes to understanding how households’ prior energy use and home characteristics can explain heterogeneity in household-level savings. Our study also shows not necessarily that households with greater saving potential are opting into the program. Although the direction of the effect is as expected, sensitivity to temperature was not a significant indicator for predicted probability of participation. Other home characteristics appears to explain participation more. There are also non-participating and non-responding households who would have saved energy. In chapter 3, we study how targeting, and the scale of energy efficiency programs affect the energy savings achieved by energy efficiency program. Our study tries to further close the gap between literatures on targeting and opt-in decisions to energy efficiency program. We find that households with highest predicted probability of participation is driven by variables such as home year built, but household with demand sensitive to high temperatures has most potential to save. Our results implicate that allowing the households to select into the program without targeting (opting-in) or relying on the attributes that drives the choice to participate does not achieve the highest per household energy savings. We find that energy savings from weatherization could be increased if policymakers and utilities could effectively target treatment towards households whose energy use is closely tied to outdoor temperature. Our findings also indicates that the difference between the savings achieved by targeting and the existing opt-in program design diminishes as the program scales up.