Post

The Potential Economic and Health Hardship Effects of Time-Of-Use Rates

Posted to AESP
AESP Contributor's picture
Communications AESP

Founded in 1990, the Association of Energy Services Professionals is the leading organization in the U.S. and Canada serving the networking and professional development needs of energy...

  • Member since 2011
  • 232 items added with 186,655 views
  • Aug 10, 2018
  • 1024 views

This item is part of the Special Issue - 2018-10 - AESP EE Day, click here for more

By Benjamin L. Messer

Introduction & Background
Time-of-Use (TOU) rates have the potential to help utilities manage and control their peak load by sending a price signal to customers that electricity rates are higher during peak hours and lower during off-peak hours. If customers respond to the price and can reduce their usage or shift usage from peak to off-peak hours, they may benefit from lower electricity bills and the utility may benefit from a lower peak load.
 
The customers who are unable to reduce or shift their usage during peak hours, however, may experience economic and/or health hardship resulting from high electricity bills. For example, if TOU rates resulted in higher electricity bills, customers might cut back on other important expenses, forgo paying some of their bills, or go into debt to meet their basic needs (economic hardship). Customers could also reduce their air conditioning on hot days or their heating on cold days to save money on their electricity bill, which could result in a household member experiencing a heat- or cold-induced medical event (health hardship).
 
In preparation for mandatory TOU rates, the public utility commission in California (CPUC) required the three largest investor-owned utilities (PG&E, SCE, and SDG&E) to implement a TOU Opt-In Pilot. A key question the Pilot sought to address was whether a TOU rate would cause any households to experience substantial financial or health hardships, particularly low-income and senior households in hot climate regions. This article presents a summary of economic and health hardship results from two customer surveys we designed and fielded during the TOU pilot.


Methods and Data
The Pilot began in June 2016 and ended in December 2017 and was designed as a randomized control trial (RCT). The IOUs recruited over 55,000 customers into the Pilot and randomly assigned them to the standard, tiered control rate (Control group) or an experimental TOU rate (Rate group)1. Customers who opted into the Pilot were provided a $100 bill credit and bill protection during the first year of the Pilot to ensure that a sufficient number of different customer segments participated.

Pilot participants were divided into three climate regions (hot, moderate, and cool) based on where they live, and up to five customer segments based on their income, household size, and age2. In this article, we limit the focus to two segments:

1) Low-income customers enrolled in the California Alternate Rates for Energy (CARE) program and Federal Electric Rates Assistance (FERA) program (CARE/FERA customers)

2) Moderate- and high-income customers not enrolled in or eligible for CARE or FERA.

We conducted two participant surveys, the first in the fall of 2016 to ask about customers’ experiences during the previous summer (Wave 1 survey) and the second in the summer of 2017 to ask about customers’ experiences during the previous winter and spring (Wave 2 survey). The surveys included questions about customers’ economic and health experiences, satisfaction with and understanding of their rate, actions taken to shift or reduce usage during peak hours, and other relevant topics.

We implemented the surveys to achieve a near census of responses using a web-mail-phone approach. Customers were first mailed an invitation letter with a survey URL and two follow-up letter or email reminders, followed by a paper booklet sent to nonrespondents, and concluding with phone calls targeted to nonrespondents in the segments with lower response rates. Respondents were provided a $50, $75, or $100 bill credit for completing each survey. Both Wave 1 and Wave 2 surveys achieved high response rates (82% and 81%, respectively). More than 44,500 customers responded to the Wave 1 survey and more than 38,500 responded to the Wave 2 survey, which was a sufficient number to achieve the statistical power to make the RCT comparisons by regions, segments and rate groups3.

We included multiple economic- and health-related questions in the surveys and used customers’ responses to these questions to create an economic index, a health index, and two health metrics. We created the economic hardship index using exploratory factor analysis (EFA) with four survey questions comprised of 18 items. The economic survey questions loaded strongly into an index that we standardized into an 11-point scale, in which 0 means very low hardship and 10 means very high hardship. The survey questions asked about:

  • How well each of five statements in the Consumer Financial Protection Bureau’s (CFPB) Financial Well-Being Index described their financial situation and outlook, using five-point scales.
  • How often they had problems paying their electricity bills and other important bills, using a scale from zero to three or more times.
  • How concerned they were about having enough money to pay their electricity bill in the future, using an 11-point scale.
  • What sources they used to pay their electricity bills, including their income and other non-income sources such as using their savings or a credit card they couldn’t pay off, borrowing the money, cutting back on expenses or not paying bills on time, or receiving public assistance.

We used EFA with several health-related questions included in the Wave 1 survey to create a health hardship index, but the questions did not load strongly into a single index. Instead, we calculated a ‘too hot’ health metric that measured the percentage of respondents with air conditioning who reported at least one heat-induced medical event due to too much heat in their home4. In the survey, respondents were asked how many times they needed medical attention due to their home being too hot, their type of cooling equipment, and if they had a disabled household member5.

In the Wave 2 survey, we included the ‘too hot’ metric questions, as well as additional health questions to provide more robust measures of health hardship. We added questions about cold-induced medical events, electric heating equipment, and disabled household members to create a ‘too cold’ metric, similar to the too hot metric. We calculated the average percentage of respondents in each segment who reported having electric heating equipment and who reported at least one medical event due to too much cold in their home6.

We also added two questions from the Center for Disease Control’s (CDC) Behavioral Risk Factor Surveillance System (BRFSS) survey to the Wave 2 survey that measured how often members of the household were in poor health and how often their poor health limited them from performing their usual activities, using five-point scales. The BRFSS questions strongly loaded into a general health hardship index that we standardized into an 11-point scale, in which 0 means very low hardship and 10 means very high hardship.

We calculated the average index scores for each index and each customer segment, and statistically compared the average scores between the Control and Rate groups using a two-tailed t-test. We also calculated the average percentages for the ‘too hot’ and ‘too cold’ metrics for each customer segment and compared the averages between Control and Rate groups using a two-tailed z-test.

Economic Hardship Results

The distribution of economic hardship index scores from the Wave 1 survey is nearly normal for CARE/FERA respondents, with an average around 4.0, whereas it is skewed toward lower hardship for non-CARE/FERA respondents, with an average around 2.0 (Figure 1). The distributions are nearly identical for the Wave 2 survey (not shown in a Figure).

Figure 1: Distributions of CARE/FERA and Non-CARE/FERA Economic Hardship Index Scores from the Wave 1 Survey

As shown in Figure 2, SCE Rate 3 hot region CARE/FERA respondents reported significantly higher economic hardship than corresponding Control group respondents in the Wave 1 survey. In the Wave 2 survey, PG&E Rate 3 hot region CARE/FERA respondents and SDG&E Rate 1 moderate region CARE/FERA respondents reported significantly higher economic hardship compared to corresponding Control group respondents. All other Wave 2 CARE/FERA Rate group respondents, and Wave 1 and 2 Non-CARE/FERA Rate group respondents (not shown in Figure), reported similar or significantly lower economic hardship index scores than corresponding Control group respondents.

 

Figure 2: Comparisons of CARE/FERA Economic Hardship Index Average Scores from Wave 1 and Wave 2 Survey Results*

 

* Patterned bars indicate a statistically significant difference vs. the Control group at p≤.05.

Health Hardship Results

As shown in Figure 3, significantly higher percentages of Wave 2 SDG&E Rate 1 and 2 moderate region CARE/FERA respondents reported experiencing one or more medical events due to too much heat in their home compared to corresponding Control group respondents. Similar or significantly lower percentages of Wave 1 CARE/FERA Rate group respondents with AC, and Wave 1 and 2 non-CARE/FERA Rate group respondents (not shown in Figure), reported experiencing one or more heat-induced medical events than corresponding Control group respondents.

Looking at respondents with AC and a disabled household member, significantly higher percentages of SCE Rate 1 and 3 hot region CARE/FERA respondents reported experiencing one or more medical events due to too much heat in their home compared to corresponding Control group respondents in the Wave 1 survey (Figure 3). Similar or significantly lower percentages of CARE/FERA Rate group respondents in the Wave 2 survey, and non-CARE/FERA respondents in the Wave 1 and 2 surveys (not shown in Figure), reported experiencing a heat-induced medical event compared to corresponding Control group respondents.

Figure 3: Comparisons of the Percentage of Respondents with AC and/or a Disabled Household Member Who Reported One or More Medical Events Due to Too Much Heat in Their Home*

* Patterned bars indicate a statistically significant difference vs. the Control group at p≤.05.

** Counts of respondents were too low to provide highly accurate or representative percentages or statistical comparisons.

In the Wave 2 survey, a significantly higher percentage of SCE Rate 1 hot region CARE/FERA respondents with electric heating equipment reported at least one medical event due to their home being too cold compared to the corresponding Control group (not shown in Figure). Similar or lower percentages of CARE/FERA Rate and Control group respondents with electric heating equipment and a disabled household member, and non-CARE/FERA Rate and Control group respondents with electric heating and with or without a disabled household member, reported a cold-induced medical event (not shown in Figure).

The distribution of general health hardship index scores from the Wave 2 survey is skewed toward lower hardship for both CARE/FERA and non-CARE/FERA respondents (Figure 4). However, CARE/FERA respondents reports slightly higher health hardship (2.8), on average, than non-CARE/FERA respondents (2.2).

Figure 4: Distributions of CARE/FERA and Non-CARE/FERA General Health Hardship Index Scores from the Wave 2 Survey

As shown in Figure 5, SDG&E Rate 2 moderate region CARE/FERA respondents reported significantly higher general health hardship compared to the corresponding Control group. All other CARE/FERA Rate group respondents, and the non-CARE/FERA Rate group respondents (not shown in Figure), reported similar or significantly lower general health hardship compared to corresponding Control group respondents.

Figure 5: Comparisons of CARE/FERA General Health Hardship Index Average Scores from Wave 2 Survey Results*

 

* Patterned bars indicate a statistically significant difference vs. the Control group at p≤.05.

Conclusions

Overall, TOU rates did not cause a significant increase in economic or health hardship for most customers either during the summer (Wave 1) or winter/spring (Wave 2) of the Pilot. However, a few CARE/FERA customer groups, especially in the hot regions, did report significantly greater economic and/or health hardship than Control groups due to their TOU rate. In addition, bill impact analysis results (conducted by Nexant) show that bills significantly increased for many of the TOU Rate groups compared to the Control groups, particularly in the summer. As a result, the CPUC administrative law judges ruled that the IOUs had to exclude CARE/FERA customers in hot climate regions from being defaulted onto TOU rates in the future.

Benjamin Messer is a Senior Consultant at Research Into Action and was project manager of the California TOU Opt-In Pilot customer surveys. He has managed several research projects involving the design and implementation of large-scale surveys for program evaluation and market research.

This article is contributed by the AESP Market Research and Evaluation Topic Committee.

PG&E and SCE tested three TOU rates and SDG&E tested two TOU rates. The rates varied in their peak hours and peak hour prices.

SDG&E did not have enough Pilot participants in the hot climate regions, and thus included only moderate and cool regions in the Pilot.

3 Due to attrition in the Pilot, there were fewer respondents to the Wave 2 survey than the Wave 1 survey, even though both achieved a similar response rate.

4 Pilot participants without AC equipment could not reduce their usage on hot days to save money and thus were excluded from the metric.

We also calculated the ‘too hot’ metric for respondents with AC and a disabled household member since participants with a disabled household member were more likely to be vulnerable to heat and less likely to afford an increase in their bill.

6 Pilot participants with non-electric heating equipment would not save money on their electricity bill by reducing their heating usage, and thus were excluded from the ‘too cold’ metric. We also calculated the ‘too cold’ metric for respondents with electric heating equipment since participants with a disabled household member were more likely to be vulnerable to coldness and less likely to afford an increase in their bill.

 

AESP
AESP leads a vibrant community of professionals dedicated to improving energy efficiency through learning, networking and knowledge sharing.
RECENT POSTS FROM THIS COMPANY
AESP Contributor's picture
Thank AESP for the Post!
Energy Central contributors share their experience and insights for the benefit of other Members (like you). Please show them your appreciation by leaving a comment, 'liking' this post, or following this Member.
More posts from this member

Discussions

Spell checking: Press the CTRL or COMMAND key then click on the underlined misspelled word.

No discussions yet. Start a discussion below.

Get Published - Build a Following

The Energy Central Power Industry Network is based on one core idea - power industry professionals helping each other and advancing the industry by sharing and learning from each other.

If you have an experience or insight to share or have learned something from a conference or seminar, your peers and colleagues on Energy Central want to hear about it. It's also easy to share a link to an article you've liked or an industry resource that you think would be helpful.

                 Learn more about posting on Energy Central »