Residential Energy Consumption Survey Research Articles

Multidimensional well-being of US households at a fine spatial scale using fused household surveys

Social science often relies on surveys of households and individuals. Dozens of such surveys are regularly administered by the U.S. government. However, they field independent, unconnected samples with specialized questions, limiting research questions to those that can be answered by a single survey. The presented data comprise the fusion onto the American Community Survey (ACS) microdata of select donor variables from the Residential Energy Consumption Survey (RECS) of 2015, the National Household Travel Survey (NHTS) of 2017, the American Housing Survey (AHS) of 2019, and the Consumer Expenditure Survey - Interview (CEI) for the years 2015–2019. This results in an integrated microdataset of household attributes and well-being dimensions that can be analyzed to address research questions in ways that are not currently possible. The underlying statistical techniques, designed under the fusionACS project, are included in an open-source R package, fusionModel, that provides generic tools for the creation, analysis, and validation of fused microdata.

Addressing energy insecurity: Policy Considerations for enhancing energy assistance programs

Household energy consumption is crucial for a productive and safe life. Despite its importance, 33.7 million U.S. households experienced energy insecurity in 2020. This paper examines the prevalence, correlates, and effects of energy bill assistance programs, which aim to alleviate the hardship. This analysis relies on logistic regressions and the 2020 Residential Energy Consumption Survey, a nationally representative survey administered by the Energy Information Administration. Results reveal 16 percent of the energy insecure population received assistance to pay its energy bill in 2020. Several socioeconomic and housing characteristics are associated with receipt of assistance; however, logistic regression estimates suggest prior participation in energy assistance and receipt of a disconnection notice from a utility provider are indicators substantively and significantly associated with energy assistance participation that warrant attention from scholars and practitioners. Lastly, outcomes generally indicate previous participation in energy assistance does not impact the odds a household will experience energy insecurity. Based on findings, I offer three policy recommendations: 1) increase resources spent on energy assistance to align with demand; 2) enhance communication between utility providers and low-income households regarding available assistance opportunities; and 3) prioritize engagement with populations that never participated in energy assistance to facilitate successful application processes.

Race, ethnicity, and the distribution of energy efficiency incentives

This paper examines the distribution of energy efficiency incentives by race and ethnicity among residential American households based on data from the Residential Energy Consumption Survey. Across a variety of types of incentives, non-Hispanic White households are the mostly likely to have received an incentive relative to American Indian or Alaskan Native; Asian; Black or African American; Hispanic White; Native Hawaiian or Other Pacific Islander; or multi-racial households. I compute concentration indices to compare the degree of inequality by type of incentive. Tax credits, utility rebates, and subsidies for home energy audits have similar levels of inequality by race and ethnicity. Subsidies for efficient light bulbs have a more equal rate of uptake. The driving mechanism for these disparities appears to be that non-Hispanic White households are more likely to be homeowners, as the differences in uptake disappear in models that control for homeowner/renter status.

Rising U.S. Income Inequality and Declining Residential Electricity Consumption: Is There a Link?

Abstract After growing steadily for decades, in the mid-2000s, average US household energy consumption began declining. Using household-level data from the Residential Energy Consumption Survey and Current Population Survey between 1990 and 2020, we decompose overall changes in per-household consumption into three components: a) average income; b) cross-household income distribution; and c) consumption habits, which include energy efficiency. Growth of average income caused consumption to increase by 11 percent, and rising income inequality reduced consumption by 8 percent, nearly entirely offsetting the effect of income growth. Changes in habits also reduced consumption. Back-of-the-envelope calculations indicate an unexpected effect of rising income inequality: climate and air quality improvements valued at $9 billion in 2020 due to lower electricity consumption. The results indicate the importance of coordinating policies that address inequality and pollution.

Data-driven analysis of influential factors on residential energy end-use in the US

Decarbonizing the residential buildings sector is important to realize a net-zero future. However, little research comprehensively explores how residential energy by end-uses, i.e., space heating, space cooling, water heating, and appliances, vary across different climate, building, and occupant characteristics. Therefore, based on Residential Energy Consumption Survey 2015, this study investigates how these energy end-uses correlated with different influential factors, i.e., climate, demographics, housing and appliance features. Recursively random forest regression and partial dependency plots were employed to analyze and quantify the impact of influencing factors on end-use energy usage. The developed models illustrate nonlinear and varying impacts of influential factors on different end-uses, with R-square from 0.481 (appliance) to 0.885 (space cooling). Specifically, housing size and climate are determinant factors of space heating/cooling, while family size and total appliance number determine water heating and appliance usage, respectively. Use frequency, size, and vintage of appliances (e.g., refrigerator, laundry) affect appliance energy consumption. Furthermore, one-way ANOVA confirms the statistically significant differences in energy consumption across varying household income groups, i.e., households with higher incomes consume more energy. Houses with older vintages use more total energy, especially in space heating, while less in cooling due to lower penetration of air conditioners. Finally, the analysis proves that residential end-use electrification, mainly space heating and water heating, is beneficial for energy efficiency. The findings contribute to understanding influential factors on different end-uses in households, which from a holistic level informs the potential pathway for building decarbonization, especially targeting at specific end-uses or occupant groups.

Analyzing electric vehicle uptake based on actual household distributions: A contribution to empirical policy formulation

This paper assesses the influence of policy-relevant factors for electric vehicle (EV) uptake, while considering the related solar-panel context, using the 2020 Residential Energy Consumption Survey for the U.S. Our results suggest that the link from household income to EV uptake is relatively weak; we find the link from education to EV uptake is stronger. Three policy reforms are suggested based on our analysis of actual household data. If income is used in determining subsidies, then increased targeting of policy support to lower income levels may be warranted. Policymakers may also consider alternative approaches, rather than relying on income thresholds. Policy bundling is also suggested, where households can use subsidies for their chosen energy investment, rather than separate schemes for solar panels and EVs. This flexible approach could be appropriate for renters, who were less likely to have at least one of an EV or solar panels by up to 3.5 percentage points.

Investigating the application of a commercial and residential energy consumption prediction model for urban Planning scenarios with Machine Learning and Shapley Additive explanation methods

Building energy forecasting methodologies utilized by municipal governments tend to be geared heavily towards depicting broader qualitative representations of regional change and are in need of complementary data-driven models that can produce quantitatively reliable depictions of future energy consumption at the neighborhood-level. The current research demonstrates an application of a Machine Learning (ML) model in the form of an Extreme Gradient Boosting (XGBoost) algorithm for forecasting the energy use of commercial and residential buildings. The methodology serves to improve on municipal scenario planning by providing a more spatially granular representation of future energy use. In this way, city government and urban planners can more accurately set carbon emission benchmarks and target specific locales for sustainability initiatives. The second major contribution of the study is to demonstrate how scenario planning approaches can utilize existing Machine Learning techniques to compensate for gaps in the data. This work is developed through a case study of Philadelphia. The study begins with the construction of residential and commercial energy models for the year 2015 and corresponding models for the year 2045. The forecast models integrate regional socioeconomic trends from the Delaware Valley Regional Planning Commission (DVRPC) scenario of Enduring Urbanism. The commercial energy use model is developed from the DVRPC’s open-source Geographic Information System (GIS) datasets, the Commercial Buildings Energy Consumption Survey (CBECS), and CoStar commercial real estate data. The residential model applies the Residential Energy Consumption Survey (RECS), the Public Use Microdata Sample (PUMS), and Census Bureau American Community Survey (ACS) estimates. A corresponding SHAP (SHapley Additive exPlanations) analysis is implemented to pinpoint feature contributions to the model’s energy estimates. By using the PopGen software, the model’s energy estimates could be analyzed at the household level, the smallest possible scale. To provide a useful resource for key stakeholders, the study aggregates model output by Traffic Analysis Zone (TAZ) and Public Use Microdata Area (PUMA) to display a detailed forecast of energy use. The results indicate that the DVRPC Enduring Urbanism trends in income and employment do not significantly affect energy consumption for the study area. However, features related to lower building intensity (e.g., lower square footage, fewer floors per building) were associated with reduced energy use in both models. Additionally, the study found residential buildings under the “single-family attached” zoning designation to correspond with higher energy estimates.

Residential responses to service-specific electricity demand: Case of China

Understanding the diversity of the residential demand for various electrical services is critical for utilities and policymakers in conducting effective demand side management and narrowing urban-rural inequality. Previous research has usually treated the household as a unit of analysis, and thus may have ignored the fact that household electricity consumption is derived demand driven by specific services, which fails to examine the heterogeneous behavioral responses. Therefore, this paper presents a new pattern of residential demand for various electrical services and quantifies the impacts of socioeconomic determinants in China. The conditional demand analysis is performed on the unique dataset of the Chinese Residential Energy Consumption Survey of 2014 to estimate the electricity demand distribution in eight types of services and to investigate the effect of socioeconomic variables on service-specific electricity consumption. The results show that, together, entertainment and food refrigeration account for about half of the total annual electricity consumption, followed by laundry, lighting, space cooling, and hot water. Rural households use about 7.2% of total electricity for cooking purposes, while urban counterparts hardly use electricity to cook at all. Electricity consumption for space heating is negligible for both urban and rural households. Heterogeneity in socioeconomic determinants is found not only among different electrical services but also between urban and rural households.

Visualizing the Spatial Distribution of Energy Insecurity in the United States

Prior to 2020, little was known about how energy insecurity varies across U.S. states. The recent release of energy insecurity data from the Energy Information Administration facilitates the exploration of energy insecurity at the state level using the Residential Energy Consumption Survey. In this visualization, the authors use choropleth maps to show spatial variation in (1) forgoing basic necessities such as food or medicine to pay an energy bill and (2) keeping the home at an unhealthy temperature. Both are strategies used by households to prevent energy insecurity. The visualization highlights the state-level prevalence of each indicator.

Characteristics and patterns of residential energy consumption for space cooling in China: Evidence from appliance-level data

The adoption of air conditioners (ACs) in households is rapidly growing in response to rising global temperatures, thereby posing significant sustainability challenges. It is vital to identify the characteristics and patterns of household demand for AC-induced electricity consumption. However, owing to huge costs and privacy concerns, appliance-level data sets are rare. Therefore, we conducted the Chinese Residential Energy Consumption Survey (CRECS) 2021, collected high-frequency operation data from 1140 active ACs from 673 households in 40 southern Chinese cities, and depicted the overall residential space-cooling electricity consumption profile. We found that, first, urban households in southern China showed a highly consistent pattern in AC time use. The peak usage in both living rooms and bedrooms occurred at noon (11 a.m.–3 p.m.) and night (9 p.m.–1 a.m.), when electricity consumption was 29%–38% higher than at other times of the day. Second, AC usage patterns showed great heterogeneity in terms of city-level and household-level characteristics. Households in large cities typically owned less energy-efficient equipment and consumed more AC-related electricity than did those in mega- and small-and-medium cities. Middle-income households, smaller room sizes, less-educated heads of households, and households with older adults were associated with lower AC-related electricity use.

Hyperparameter optimized classification pipeline for handling unbalanced urban and rural energy consumption patterns

Energy consumer locations are required for framing effective energy policies. However, due to privacy concerns, it is becoming increasingly difficult to obtain the locational data of the consumers. Machine learning (ML) based classification strategies can be used to find the locational information of the consumers based on their historical energy consumption patterns. The ML methods in this paper are applied to the Residential Energy Consumption Survey 2009 dataset. In this dataset, the number of consumers in the urban area is higher than the rural area, thus making the classification problem unbalanced. The unbalanced classification problem has been solved in original and transformed or reduced feature space using Monte Carlo based under-sampling of the majority class datapoints. The hyperparameters for each classification algorithm family is represented as an optimized pipeline, obtained using the genetic programming (GP) optimizer. The classification performance metrics are then obtained for different algorithm families on the original and transformed feature spaces. Performance comparisons have been reported using univariate and bivariate distributions of the classification metrics viz. accuracy, geometric mean score (GMS), F1 score, precision, area under the curve (AUC) of receiver operator characteristics (ROC). The energy policy aspects for the urban and rural residential consumers based on the classification results have also been discussed.

Why is it still too warm or cold in my house? Examining the relationships between energy efficient capital and household energy insecurity

This paper examines the relationships between energy efficient (EE) capital technology and household energy insecurity in the United States. The theoretical model of these relationships employs household production theory to capture the demand for and production of household energy services, and a stochastic production frontier approach to describe how having access to and the usage intensity of EE capital technology could help alleviate inefficiency in the production of household energy services. A working hypothesis formulated from the theoretical model posits that having EE capital technology in the home will reduce the level of household energy insecurity experienced. The extent of energy insecurity experienced is inferred from an energy insecurity index value assigned to each household, generated via the application of a dichotomous Rasch model to questions contained in the 2015 Residential Energy Consumption Survey. Noting the potential simultaneous relationship that exists between a household having access to and the usage intensity of EE capital technology and the experience of being energy insecure, an instrumental variables approach was employed to estimate a series of ordered logit models. Results suggest access to EE capital technology in the form of Energy Star® appliances, Energy Star® windows, or a SMART thermostat does not reduce the probability of experiencing a greater level of energy insecurity. Nor does the usage intensity of EE capital. Thus, policy instruments designed to alleviate household energy insecurity may need to go beyond simply helping households obtain EE capital technology.

The effect of carbon tax incidence on household energy demand and welfare in the U.S.

This study develops a model based on a general equilibrium framework to assess the excess burden of carbon taxes imposed on energy commodities (electricity and natural gas) among residential sector. The model takes into account labor market distortions from the tax and cross-price effects among energy commodities. Using data from the U.S. Residential Energy Consumption Survey, the own-price and cross-price elasticities of energy commodities are estimated. A substitution effect is found between electricity and natural gas, and omitting this effect would overestimate the excess burden of the carbon tax. The results show that the carbon tax behaves differently in affecting the excess burden for low-, middle-, and high-income households. The excess burden is lower for high-income households than for low-income households at lower pre-determined labor tax rates, but the effect is reversed at higher pre-determined labor tax rates. In addition, the empirical results show that the excess burden is different across the nine U.S. regions, while minor gas price changes have no significant effect on the excess burden.

Photovoltaics and the Solar Rebound

<h3>Abstract</h3> Recent research suggests that households increase their electricity consumption after installing photovoltaic (PV) panels, a behavioral change commonly referred to as the solar rebound. Using panel data originating from the German Residential Energy Consumption Survey (GRECS), we investigate the existence of a solar rebound effect. Our empirical results suggest that PV panel adoption does not change the amount of electricity that households take from the grid. As we derive theoretically, this implies a solar rebound that is bounded from above by about 50%, while back-of-the-envelope calculations suggest a lower bound of 12% and an average solar rebound of 35%.

How do machines predict energy use? Comparing machine learning approaches for modeling household energy demand in the United States

This paper illustrates the use of different machine learning techniques to estimate household energy demand. To demonstrate the performance of the techniques, we discuss how the different machine learning algorithms select a model or models of energy usage and we explore how well the models predict usage. Our study employs a high-dimensional dataset of housing, socioeconomic, and behavioral characteristics, provided by the U.S. Energy Information Administration's ongoing Residential Energy Consumption Survey. In addition to discussing the machine learning models, we estimate energy price elasticities, which are important indicators of how sensitive households are to changes in residential energy prices. Given the broad set of data in the survey, we compare and contrast various machine learning techniques to see which model provides the best overall fit to the data. We find that a random forest algorithm performs better than the other machine learning approaches, which include a step-wise Akaike Information Criterion, partial least squares, ordinary least squares, k-nearest neighbors, penalized regression, and gradient boosting methods. Finally, we discuss how machine learning can be used to inform residential energy policies and predict household energy consumption.

Exploring the factors that influence energy use intensity across low-, middle-, and high-income households in the United States

This study examines the relationships between energy use intensity (EUI), which is considered to be an indicator of energy efficiency, and dwelling or housing characteristics, technology (appliances), socio-demographic characteristics, geographic factors, and energy-related behavioral actions. Additionally, it explores whether these relationships vary across low-, medium-, and high-income households. The study is based on regression analyses conducted on a representative sample of households, the 2015 U.S. Residential Energy Consumption Survey. Overall, the analysis revealed two important findings. First, residential energy use intensity is shaped significantly by housing characteristics, socio-demographic factors, technology, and energy-related behavioral actions. Second, the relationships between the factors examined and energy use intensity vary quite substantially across income groups. Lower income households have a higher EUI than higher income households. The policy implications of these findings are that reducing EUI in the residential sector, which may help with addressing energy burdens and poverty among low-income households, will require paying careful attention to these factors and their dynamic impacts across income groups.

Carbon Emission Characteristics and Reduction Pathways of Urban Household in China

With China’s rapid urbanization and rising living standards, the household sector has become the second largest contributor to urban carbon emissions and important pathway to achieve China’s carbon reduction targets. Based on data of Chinese Residential Energy Consumption Survey in 2014, this paper establishes standard households in 54 cities to identify the carbon emission characteristics and explores the emission reduction paths of urban households. We present evidence that rich households tend to increase their use of clean energy, low-income households will not completely abandon inferior energy because of their usage habits and costs and they always live in suburban areas and urban villages. Cities with high household carbon emissions are almost located in north of Huai River/Qinling Mountains line, while cities with low carbon emissions are mostly located around the line and which are mainly on the south side. Monocentric urban development patterns and frigid climates are easy to increase higher household carbon emissions. In this regard, policy makers should help low-income households to escape from inferior energy dependence, encourage the use of new energy vehicles and green housing technologies, introduce a two-part pricing system for central heating services and maintain relative balance of residential and public resources in urban planning.

Extreme temperatures and residential electricity consumption: Evidence from Chinese households

We estimate the causal effect of extreme temperatures on household electricity consumption using the China Residential Energy Consumption Survey (CRECS), which contains validated data on electricity consumption from respondents' electricity bills. We find that for each additional day in which the mean temperature exceeds 32 °C, relative to a day in which temperature is in the range 10 °C–16 °C, there is an 8.9% increase in annual electricity consumption. Exploiting rich information on the use of major home appliances which is available in CRECS, we demonstrate that this large effect is due to households using air conditioners more, at both the extensive and intensive margins, to cope with heat stress. Combining our baseline estimates with daily temperature projections from eight recent climate models, we simulate changes in future electricity consumption due to global warming. We find that residential electricity consumption would increase by 9.59–30.09% in the medium term and by 9.77–47.70% in the long term, depending on whether measures are taken to curb greenhouse gas emissions.

A Comparison of Energy Consumption in American Homes by Climate Region

The present research analyzes the impact of nine factors related to household demographics, building equipment, and building characteristics towards a home’s total energy consumption while controlling for climate. To do this, we have surveyed single-family owned houses from the 2015 Residential Energy Consumption Survey (RECS) dataset and controlled the analysis by Building America climate zones. Our findings are based on descriptive statistics and multiple regression models, and show that for a median-sized home in three of the five climate zones, heating equipment is still the main contributor to a household’s total energy consumed, followed by home size. Social-economic factors and building age were found relevant for some regions, but often contributed less than size and heating equipment towards total energy consumption. Water heater and education were not found to be statistically relevant in any of the regions. Finally, solar power was only found to be a significant factor in one of the regions, positively contributing to a home’s total energy consumed. These findings are helpful for policymakers to evaluate the specificities of climate regions in their jurisdiction, especially guiding homeowners towards more energy-efficient heating equipment and home configurations, such as reduced size.

The Household Appliance Stock, Income, and Electricity Demand Elasticity

This paper estimates household electricity demand using data from the Energy Information Administration’s (EIA) 2015 Residential Energy Consumption Survey (RECS). Previous research has focused on estimating price elasticities, and we contribute to this literature by examining how price elasticity is impacted by household income and the appliance stock. Results show that as income increases, households rely less on electricity for space and water heating, but the number of electronic appliances increases with income. The changing stock suggests that the mechanism through which energy reduction occurs differs across income levels. The results can aid policymakers concerned about electricity demand, rising electricity rates, and the impact on low-income households. The results can also inform the design of demand response and demand side management programs.