We also included data from the Bureau of Labor Statistics at the county level regarding unemployment to see how unemployment levels have changed in the counties served by utilities.

The amount of water then depends a lot on how many people are living in your house and the number of days in the month to start with. Since we are looking at affordability, there is concern that people have "enough" water to live on (not so worried about swimming pools and green lawns). A one person household could use between 1,500 gallons to 3,000 gallons in a month while a four person household would use 6,000 to 12,000 gallons in a month. And this does not include considerations for the types of water fixtures, any medical needs requiring water, and so on.

Currently, we only have the service area boundary for drinking water providers. We calculate the average bill when there are multiple wastewater or stormwater providers within the service area. This was especially common in Pennsylvania where drinking water and wastewater are often provided by different utilities. We provide a table with all wastewater and stormwater provider used to calculate the average bill for a selected utility in the How much do water services cost tab in the dashboard. For example, the North Penn Water Authority provided drinking water services to all or portions of 10 municipalities; yet each municipality had its own wastewater utility with rates ranging from as low as $15.93 per month in Lansdale to as much as $57.30 per month in Soulderton at 4,000 gallons per month. No wastewater rates were identified for two townships, although these townships may receive services from nearby townships. In these instances, given missing data and lack of spatial wastewater service areas, we calculated the mean wastewater bills within the drinking water utility service.

We substitute an estimated monthly amount that homeowners will spend to maintain a septic system based on average costs to maintain and replace septic systems. We recognize there is a tremendous range in septic system costs depending on the soil and drainage characteristics of the landscape and may not accurately reflect your conditions. We are open to suggestions for improving these estimates to tailor to state or other conditions:

• The median cost to replace septic tanks was estimated at $6,733 with a lifetime of 25 to 30 years.
• The median costs for routine plumbing were $391 every 3 to 5 years.
• The median cost to address problems requiring a plumber were $520 and occur once every 10 years.

This results in an estimated cost to homeowners of $32.89 per month.

We provide a table with all wastewater providers in the How much do water services cost tab in the dashboard to identify when we believed the community used septic/on-site lots.

Utilities may charge different rates to different regions within their service area, often based on distribution type (pump or gravity), location (closer or farther away), elevation (more expensive to pump water higher), and so on. Here, we estimate the average bill because we have no way of knowing where these different rates are applied. There is one exception, and that is when there are “inside-outside” rates, which was very common in some states. Here, utilities charge different rates depending on if a customer was located inside or outside of a utility’s political jurisdiction or municipal boundary.

We estimated inside-outside rates in this project because: (1) inside-outside rates were common, (2) the rates were often quite different (outside rates were often 1.5 to 2 times higher than inside rates), and (3) we have current municipal boundaries. We assigned households to inside rates if they were located within the municipal boundary and outside rates if households were located outside of the municipal boundary (but inside the service area). For example, Greensboro, NC provides water to those living inside the Greensboro municipal limits as well as outlying areas. Those living inside city limits were assigned inside rates (an estimated monthly bill of $46 at 4,000 gallons), while those living outside of the city were assigned outside rates (with an estimated monthly bill of $110 for 4,000).

The big assumption is that current municipal boundaries accurately reflect inside and outside charges. For utilities with inside and outside rates, and where there were known stormwater services, stormwater rates were only applied inside the municipal boundary because stormwater services are often provided by the public works department of municipalities.

• Compare different affordability measures for the same utility and across utilities
  • Each metric tells a different part of the affordability story, how similar or different are they?
  • What do these metrics tell me about a utility of interest?
  • How does my selected utility compare with other utilities selected (e.g. state, size, or ownership)?
  • How does affordability change when different amounts of water are used?
• Gain more understanding about the costs of different water services.
  • How do estimated bills for a selected utility compare with its peers?
  • How do rate structures compare, particularly the amount of the bill that is fixed and the cost for each 1,000 gallons of water?
  • Go to the links to where rates information were originally found.
  • See in table format how many utilities may provide services in an area and were used to estimate the average bill.
• Learn more about the characteristics of those living in the service area in terms or population, age, race, building age, and unemployment. Each of these attributes may provide insight about the ability of the community to pay for water services.
• Create more transparency and understanding about water affordability in the U.S. by making the data and code publicly available and creating a process that allows more data to be added and updated over time.

The dashboard is best viewed on a desktop. It was not designed for use on mobile devices. There are two main components to the dashboard.

(1) To get started, select a utility in the map. You can do this by:

• Clicking on a utility in the map
• Using the drop down menu for Select Utility on the right
• Type in an address to zoom into an area. Select a utility on the map.

(2) Next, decide what utilities you want to compare with the selected utility. You can do this by using the selection boxes to the right of the map. You may select peers based on:

• Location: state
• Size of the system: population served by the drinking water utility
• Ownership type: public (local government), private, or other for the drinking water utility
• Volume of water: select the volume of water used to estimate bills and affordability

(3)Once you have made your selections, scroll down below the map and select the tab of interest to learn more about the utility you selected and how it compares with its peers.

If at any point you are not sure what to do, click on the little, red information icons.

How the dashboard is organized

The map shows you the service area boundaries of utilities with data in the dashboard. The color of the utilities reflects the burden level based on poverty prevalence and household burden at the selected volume of water. If you place your mouse on top of a utility, you will see information about that utility appear in the bottom left corner. You will also see a chart showing the percent of block groups inside the service area by burden level. If most of the block groups are at one level, the income and rates of the utility are similar throughout. If the block groups are spread between several burden levels then there is greater diversity in the income and/or rates in the service area.

You may find a utility by typing an address in the box on the top right. You can also turn county and municipal layers on and off to help find familiar areas. The buttons turning those layers on and off are above the legend on the right side of the map.

The map colors reflect the affordability burden of the utility

Once you click on a utility (or select a utility from the drop down menu), the map will zoom in. The block groups that were used to estimate affordability metrics will be shown and colored based on the affordability burden of that block group. You may place your mouse over the block groups to see information about the block group including number of households, median income, low income, and affordability metrics. The block group you hover over will be highlighted in white. There is no further action if you click on the block group. You may click on other utilities on the map to zoom in and see their block groups. The block groups largely reflect the distribution of income in the service area.

Zoom into see block groups of selected utility

The charts are organized into 4 sections:

• Who lives in my utility? The charts in this section provides basic demographic information about the selected utility.
• How much do water services cost? The charts in this section provide information about the cost of water services and the different components of the bill.
• How affordable are water services? The charts in this section provide information about how affordable water services are for the community.
• How does affordability change with water usage?The charts in this section summarizes how affordability changes for utilities as the volume of water used increases from 0 to 16,000 gallons.

Each chart highlights the selected utility. The other utilities selected using the selection choices by the map are drawn in gray to provide context and comparison. The title of several of the chart changes to provide some basic takeaways.

The charts are interactive. If you hover over the chart, the tooltip will provide some text to help you interpret what the chart is showing. There are also red information buttons in the dashboard that you may click on for guidance.

Below the map you will see the definition of metrics, your selection of utilities, and tabs to learn more information.

+ How much data to show in line charts?

The amount of data displayed is growing over time. The line charts in the dashboard go through a loop to draw each line individually. This takes a long time when looping through thousands of utilities. We have added a drop down menu to allow users to select the percent of the data they want to display - with a maximum allowance of 800 utilities. The data drawn are randomly selected so the charts may vary over time. If you filter the map by state or system size, all or most of the data can be drawn if you select 100% in the drop down menu. You may notice in the 'Who lives in my utility' tab and the IDWS data that less than 100% of the data are drawn. This is because some very small and small utilities are too small for this approach (using block groups) to ensure good accuracy of the results.

The place to change the percent of data is located between the map and chart tabs.

+ Who lives in my utility?

This tab provides some basic demographic information about the utility that might be relevant to affordability and equity. Note that very small and small utilities may be much smaller than block groups, creating problems for estimating who lives in the utility. More refined census data or survey methods are needed to have a good sense of who lives in those utilities.

• Population. The estimated population in the service area. We use the Environmental Protection Agency's Safe Drinking Water Information System (EPA SDWIS) data to set the current population. We adjust the estimated population for previous years according to the difference between our estimated population for today and the SDWIS estimate. For example, if we estimated 100 persons today, but SDWIS reports 110 persons (10% higher), then we increase the estimated population in 1990, 2000, and 2010 by 10%. Service areas with a growing population have a growing revenue base, while those with a shrinking population have a shrinking revenue base.

• Working Age. The percent of the population to be in different age groups. We assume those under 18 are not contributing towards paying water bills. We assume those between 18 to 59 are more likely to be employed and paying for water services. We assume those over 65 may be retired and on a fixed income where it becomes harder to pay for services as those costs increase. We assume those nearing retirement are still working, but the utility may be interested in the size of the population nearing retirement (especially if they have Customer Assistance Programs for senior citizens).

• Racial Diversity. The percent of the population identifying with different racial and ethnic groups. The census treats race and ethnicity as separate categories. Racial categories included: Asian, Black, Native, White, and Other (includes those identifying as multiple races, Pacific Islander, or other). Ethnic categories included: Hispanic or Not Hispanic. The racial categories add up to 100% and the ethnic categories add up to 100%. There are many conversations around the inequities of water services an affordability issues that are tied to racial inequities.

Example of charts providing information about the selected utility's population, age, and racial diversity

• Income Diversity. The estimated percent of households based on their income brackets. We added all the households in the block groups for the service area. We divided the number of households in each census bracket by the total number of households. This chart shows the distribution of income in the service area (how diverse are incomes?). The title highlights the percent of households earning less than $75,000 (the middle census bracket).

• Unemployment over time. The average annual unemployment rate for counties that intersect the utility service area. We assume higher unemployment would result in more difficulty for some households to afford water services. It is more likely these customers would benefit from some type of customer assistance program.

• Unemployment and the COVID-19 Pandemic. The COVID-19 Pandemic resulted in an economic slow down. The timing and number of jobs lost and later recovered has varied across the U.S. At a time when the importance of water for public health became critical, many struggled to pay for water services because of job losses.

• Household Age. The estimated percent of houses built each decade. This chart provides an approximation for the age of the pipe infrastructure in the service area. We assume that those service areas with the majority of houses built prior to 1950 might have higher replacement costs as those pipes are at the end of their expected life. Additionally, those with older houses built prior to 1986 when Congress banned the use of lead in pipes may have more lead in their pipes that need replacement. The title provides information about how many households are more than 70 years old (built before 1950) and how many are more than 30 years old (built before 1990).

Example of charts providing information about the selected utility's income, housing age, and unemployment

+ How much do water services cost?

This tab provides some basic information about the monthly bill of the utility selected in the map for a particular volume of water, a table with information about the rates used to estimate the bill, and some charts showing the different components of the bill and how those components compare with their peers.

• Total Monthly Bill shows how much of your monthly bill is spent for drinking water, wastewater, and stormwater each month for the selected volume of water used. The monthly bill is the average of all utilities in the service area (i.e. if there are multiple wastewater or stormwater providers) and all class categories (i.e. if the utility charges different rates based on location - such as elevation zones, neighborhoods, and so on). The exception is for inside and outside classes, in which case we show the estimated monthly bill for those living inside the municipality and those located outside of the municipality but within the service area.

• Find Utility Websites is a table providing information about all utilities within the selected service area that were used to estimate the monthly bill:
• pwsid: is the unique identifier for the drinking water utility and the service area boundaries used here.
• service area: is the name of the drinking water utility and service area.
• city name: some service areas include several cities and towns that may have different wastewater and stormwater providers. When this happens, we include each city/town and list the drinking water, wastewater, and potentially stormwater utility associated with that city/town.
• utility name: the wastewater and stormwater services may be provided by different utilities from the drinking water utility. When that occurs, we list the name of the utility. We list the different utility names associated with each city and town when there are more than one wastewater or stormwater providers in the drinking water service area. We estimate the average bill for each service provided from each unique combination of city name and utility name.
• service type:the type of service provided: drinking water, wastewater, or stormwater. If the wastewater system is thought to be septic, we list the service as septic/on-site.
• year rates started:the year this rate was established. If there was no data on the website, we listed it as current year.
• rates found here: a url link taking you to the website where rates were identified. These links are owned by hundreds of utilities and may no longer be correct.
• data last updated: the year the data were entered. Eventually, we hope to continually update the underlying rates database.

• Explore Rate Structures compares the rate structure (e.g. percent of bill that is fixed, the amount charged per 1,000 gallons of water used) of the selected utility with their peers. Select whether you want to explore drinking water or wastewater rates.

Usage cost per 1,000 gallons of water shows the cost per 1,000 gallons of water based on the commodity charge (i.e. not including any fixed charges). Many utilities include a minimum amount of water in the fixed cost, which is why many lines start at zero when there are 1,000 or 2,000 or 3,000 gallons used each month. Costs may decrease if the utility caps the volume of water charged (i.e. a wastewater utility may only charge for the first 12,000 gallons of water used). Lines with odd shapes reflect situations where the average usage cost of several utilities with different rate structures or class categories were combined.

Rate components drinking water or wastewater services shows the different bill components for the selected utilities with its peers (based on selections made in the map). The radio button allows users to select "drinking water" or "wastewater". The monthly bill for drinking water or wastewater is provided as well as how much of that bill came from fixed charges (i.e. service charges), usage charges (i.e. commodity charges), and surcharges (extra charges that are temporary are administered by a separate entity). The bottom boxplot shows the percent of the monthly bill that is fixed and independent of how much water is used. The boxplots allow you to see what is "normal" relative to selected peers with the small dots representing all utilities so you can see the distribution.

+ How affordable are water services?

We provide the results for five affordability metrics in this tab.

• Income Dedicated to Water Services (IDWS) is the metric that is initially displayed. IDWS shows the prevalence of shared affordability burdens by showing how many households spend a similar percent of their income on water services. For context, each 4.6% of income represents a day of labor each month. The light color lines show all selected peers while the highlighted lines show the IDWS curve for the selected utility. Note, utilities that cover a very small proportion of a block group are not shown here because when weighted less than 1 or 2 households may be present. For example, if a utility covers 0.2% of a block group, we have no way of knowing if the population served by the utility is representative of the remainder of the population in this block group. Additionally, users may click on the checkbox to see a table showing the percent of households in a shared income bracket and the proportion of the income in that bracket that is spent on water services.

• Compare Metrics shows four metrics together. Minimum Wage Hours, Traditional, and Household Burden metrics are shown within the context of how many days of labor are needed to pay for water services (with each day of labor in a month equivalent to 4.6% of income earned).The Poverty Prevalence indicator is shown within the context of low, moderate, or high levels of poverty (defined as the percent of the community below 200% of the Federal Poverty Level)... although remember that these thresholds are somewhat arbitrary and there is little difference between 19% ("low") and 21% ("moderate"). Lastly, remember that Minimum Wage Hours is the only metric truly concerned with the affordability of water services for a specific household (i.e. that of the minimum wage earner). The other three metrics are concerned with the affordability of a representative population in the community (the median and lowest quintile incomes). Metrics are displayed as boxplots to allow users to compare a selected utility with the selected peers.

Household Affordability Metrics

Community Financial Capability Metrics

• Affordability Burden provides an affordability burden level for the utility after combining two metrics: Household Burden and Poverty prevalence per recommendations from a 2019 report by the American Water Works Association, National Association of Clean Water Agencies, and the Water Environment Federation. The colors of utilities and block groups in the map are based on this chart and assign burden levels from low to very high. The farther to the right a utility is located on the chart, the greater the importance of widespread poverty to creating affordability challenges. The higher up in the chart a utility is located, the greater the burden of rates on low-income (20th percentile) households to afford water services. The selected utility will be highlighted in the chart among its peers. The categories from low to very high were created by intersecting the recommended thresholds for poverty prevalence with each day of labor needed to pay for water services in a month. Notice, that the categories do not change once more than 2 days of labor are spent; however, each day is clearly demarcated to show the increasing affordability burden for utilities.

Affordability Burden

Lastly, we provide a chart that only shows how affordability changes for the selected utility from zero to 16,000 gallons. The result is a vertical line since the poverty prevalence is not contingent on water usage. The amount of water used changes the bill, which changes the percent of income spent on water services. Notice that some utilities appear start at 2,000 or 3,000 gallons - this simply reflects that there is no charge for some minimum amount of water used. The length of the lines is sensitive to both (1) rate structure and (2) the incomes of the lowest quintile. For example, with rate structures, the cost will increase much more for a utility charging $10.00 per 1,000 gallons than a utility charging $4.00. For example, with the lowest quintile income, a $100 increase in a bill reflects a 5% increase in the household burden for a household earning $2,000 a month compared with an 8.3% increase in a household earning $1,200 a month.

Change in Household Burden with Water Usage

Burden Table

In addition to the affordability chart, we also provide a table that summarizes the number of utilities in the database and the number of peer utilities based on map selections that fall within each burden level. Additionally, for the selected utility we provide the number of block groups in each burden level. The number of block groups is also reflected in the map when you hover over a utility.

+ How does affordability change with water usage?

This tab summarizes how many utilities share a similar affordability challenge for all volumes of water used (0 to 16,000 gallons a month). The charts only show those utilities selected in the map. The y-axis is the percent of utilities that share a similar burden while the x-axis is the volume of water. Charts are drawn for:

• Burden Level: see the percent of selected utilities that share similar affordability burdens at different volumes of water usage. These percentages reflect the number of utilities in the map showing a particular affordability burden.
• Minimum Wage Hours: see the percent of utilities requiring minimum wage earners to spend less than half a day, 0.5 to 1 day, 1.1 to 2 days, 2.1 to 3 days, or more than 3 days of labor each month to pay for water services depending on the volume of water used. Note, that the minimum wage is reflective of the state minimum wage and does not capture local ordinances requiring higher wages.
• Poverty Prevalence: see the percent of utilities sharing a similar poverty prevalence (defined as the percent of the community below 200% of the Federal Poverty Level). Notice the percents do not change with volume! That is because this metric is not based on water usage of bills, simply the composition of the community.
• Traditional: see the percent of the median household income in a utility going to pay for water services. Utilities are classified by the number of days spent each month paying for water services with 4.6% reflecting 1 day of labor.
• Household Burden: see the percent of the 20th percentile household income in a utility going to pay for water services. Utilities are classified by the number of days spent each month paying for water services with 4.6% reflecting 1 day of labor.

See the change in the percent of utilities requiring similar amounts of time working to pay for water services when using metrics of Minimum Wage Hours, Household Burden, and Traditional metrics.

General overview on how to navigate through the dashboard

The Poverty Prevalence metric indicates that the many communities are experiencing widespread poverty. Over 3-quarters of utilities served communities where more than 20% of the population was below 200% of the federal poverty level. The average utility served a community with a poverty prevalence of 30%. Regardless of water rates or usage, poverty can make affording water services a challenge for households, and simultaneously create financial challenges for a utility: if a large portion of the population served by a utility has low income, then the revenue potential for the utility is constrained.

No utilities with a Poverty Prevalence below 20% were classified with high or very high affordability burdens at all volumes of water use (although minimum wage earners in the community would still struggle to pay high bills).

We applied the state minimum wage to utilities in this study, recognizing that local governments may have higher minimum wages that would change our results. North Carolina uses the federal minimum wage of $7.25. At 6,000 gallons per month, minimum wage earners would spend 15 hours (almost two days) each month to pay for water services in the average utility. In contrast, California had a minimum wage of $12 in 2019, with minimum wage earners spending a little over 8 hours (1 day) of labor to pay for water services each month in the average utility. Higher minimum wages would help alleviate household affordability challenges and increase the financial capability of utilities.

For example, let's assume a day of labor is affordable. A minimum wage of $7.25 means water services (drinking, wastewater, and stormwater combined) would have to be less than $58 a month ($7.25 * 8 hours). Now, if the minimum wage was $12 then the combined water services could charge up to $96 a month and be considered affordable. At $15, the combined water services could charge $120 a month, and so on. In short, higher minimum wages benefit households while increasing the financial capability of utilities.

The Income Dedicated to Water Services (IDWS) metric provides a new way to explore affordability. While the other metrics in this study look at a single population (Minimum Wage Earners) or a representational slice of a community (the median or low-income household), this approach combines both the prevalence of affordability challenges (how many households) with the depth of affordability challenges (what percent of income is spent on water services). This metric allows us to consider what constitutes undue hardship, how many households does this influence, and the volume of water that is considered enough to meet basic needs. All of these are value based decisions that may differ between communities.

For example, in our study exploring 1,800 utilities we looked at 5% of income, or a little more than a day of labor a month, as representative of reasonable hardship. We found that 10% of households spent more than 4% of their income water services when using 2,000 gallons per month (a relatively small amount of water). This number increased to 18% of households by 6,000 gallons per month and 30% of households by 12,000 gallons per month. The percent of households in "undue" hardship changes as the definition of undue hardship changes. At this definition, between a tenth to a third of households spend more than a day of labor (4.6% of income) each month to pay for water services. Is that acceptable? The IDWS allows conversations of undue hardship to take place while understanding how many households may be affected.

• To learn more about the method visit here
• To learn more about how to interpret the chart visit here and click on What do the charts below the map tell me? and then How affordable are water services?

Affordability is more than cost. It combines the cost of services with the income of individuals living in that service area. We looked at all estimated bills costing between $70 to $80 and found when we looked at the affordability burden for a $10 range that:

Due to poverty prevalence alone:

• 18% classified as having a low affordability burden
• 27% classified as having a low-moderate affordability burden

Due to poverty prevalence and household burden:

• 28% classified as having a moderate-high affordability burden
• 25% classified as having a high affordability burden (had to work more than 1 day to pay for services)
• 1% classified as having a very high affordability burden (had to work more than 2 days to pay for services)

Even at nearly the same cost, the ability for households in a community to pay those costs can vary tremendously.

Within a utility, the affordability burden can be similar or change depending on the range of incomes in the community. For example, some utilities have similar incomes throughout while other utilities may have a wide range of incomes. In utilities with a wide range of incomes throughout the service area, some have incomes grouped in certain regions while others have incomes mixed throughout. Since the estimated bill is applied throughout the service area, the differences in affordability reflect differences in incomes.

Examples of how affordability and incomes vary within a single utility's service area

Several metrics were developed to understand different aspects of water affordability in the U.S. in terms of the financial capability of utilities and household affordability. These metrics are intended to identify what constitutes enough water (volume of water used) and undue hardship to pay for basic water services (learn more). Rather than advocating for one metric, we calculate several metrics across a range of water volumes and provide context for hardship in terms of the number of days of labor needed to pay water bills each month. We found each metric provided different insights, and in combination, can provide a more comprehensive understanding of water affordability challenges for utilities and households.

For example:

• The Poverty Prevalence showed that many of the communities served by utilities are experiencing widespread poverty. More than 75% of utilities served communities where more than 20% of the population was below 200% of the federal poverty level. The average utility served a community with a poverty prevalence of 30%. Regardless of water rates or usage, poverty can make affording water services a challenge for households, and simultaneously create financial capability challenges for a utility: if a large portion of the population served by a utility is low income, then the revenue potential for the utility will be constrained.
• Minimum Wage Hours was not influenced by the community composition but on the cost of water and was greatly influenced by how much water was considered "enough". This metric showed that in a typical utility, minimum wage earners must work 1 to 2 days per month to pay for water services, even when households use relatively low amounts of water (4,000 gallons per month).

This approach allows our understanding of affordability to expand because we can include more data, update the data, and repeat the metric calculations using the same methods. We also developed an interactive data visualization tool to allow users to explore some demographic characteristics of utilities, some aspects of their rates, and some affordability metrics. The COVID-19 pandemic has reminded us of the immense importance of water for public health and the Biden Administration has recently launched a relief program to improve access to water affordability for low-income households through the Low-Income Household Water Assistance Program (LIHWAP). It is important for us to better understand the scale and dimensions of water affordability challenges to best inform how new policies and programs are implemented.

There are many more questions that can be asked of the data and our hope is that by making the underlying data and code publicly available, many more will do so. You can find the data and code at the water-affordability github repository.

Many essential services have recommendations about how much of a household's income should go to various essential services:

• shelter/housing: less than 30% of your income
• energy services: less than 6% of your income
• water services: less than 4.5% of your income
• insurance (car/health): less than 6.5% of your income
• taxes (assuming low income): around 11% of your income
• food: depends on household size, but let's estimate $400 a month ($4,800 annually)
• wifi/internet: lets assume $50 a month for basic speeds ($600 annually)

A household would need to make at least $12,855 to break even assuming no assistance. The lowest-quintile income (20th percentile) is higher than $12,855 for all states, ranging from less than $5,000 higher (Mississippi) to more than $22,000 higher (Maryland). However, between 5.5% (New Hampshire) and 13.6% (Mississippi) households are estimated to earn less than $12,855. The problem with individualized thresholds for different essential services is that they are all applied to the same low-income households. In other words, a household struggling to make ends meet for one service is likely struggling to make ends meet in all essential services.

Our approach is less robust for utilities with a very small service area. Some utilities represent less than 1% of a single block group. Our approach estimates affordability based on the weighted income composition of the block group. For utilities with small service areas the weighted household may be a few houses and it is unknown how accurately the composition of such small utilities represents the composition of the overall block group. As such, for small and very small utilities the results should be viewed with caution. We highlight utilities that may be too geographically small in the dashboard when looking at the IDWS metric results.

The filter criteria were based on data obtained from the Environmental Protection Agency’s (EPA) Safe Drinking Water Information System. Note that these characteristics only reflect the drinking water utility and may not be reflective of separate wastewater utilities or stormwater utilities located within the drinking water service area. We used the size of the state, system size, and ownership type because these attributes are commonly thought to influence affordability. By filtering the map, you are able to select peers for comparison.

• State: based on the physical location of the utility


• Size: based on the reported population served. Very small to medium sizes were based on EPA classifications. We created additional categories for medium-large to very large sizes as follows:
• Very Small: 500 persons
• Small: 501 to 3,300 persons
• Medium: 3,301 to 10,000 persons
• Medium-Large: 10,001 to 75,000 persons
• Large: 75,001 to 500,000 persons
• Very Large: > 500,000 persons


• Owner Type: based on who own the utility. The majority are owned by local governments (cities, towns, districts, authorities, etc) or private companies. A few systems were served by a blend of public/private utilities, state government, or the federal government. We combined those into a single category for now. This may change if other states have a greater number of utilities with different ownership.


• Volume: the volume of water used by households in a month. This volume is what determines the estimated household bill. We calculated bills for every 1,000 gallons (134 cubic feet) from 0 to 16,000 gallons (2,139 cubic feet). This range represents the reasonable amount of water that is likely to be considered essential for basic uses.

You may notice some errors in the size or owner classification. Please reach out to EPA to correct the underlying data. You may also reach out to us: water-affordability@duke.edu to temporarily correct errors.

If there are additional filters that may be beneficial, please email and we will try to incorporate those options in future versions of the dashboard. The list of potential data that EPA collects can be found here.

Block groups may not contain residential households, in which case there is no income data. We also flagged those block groups with more than eight persons per household as these often reflect communal living situations, such as universities or prisons. We display the income data when you hover over the block group but did not include those block groups when we calculated affordability metrics. If you zoom into the map you may see a map labeling a university, airport, etc.

The information box on the bottom left of the map expands when you hover over a utility. If the size of the box covers the location of your mouse, the information box will not appear. Move the map to the right so you can hover over the utilities of interest. Similarly, the map may begin to shake if you are trying to click on a utility behind the information box or the map is only partially on the screen.

The dashboard was designed for use on a desktop. It does not work well for a mobile device. If you must use a mobile device, you will need to use the drop down menu to select a utility. The dashboard has been tested with Chrome, Edge, Firefox, and Safari. It does not work with Internet Explorer.

There are few reasons you might not be able to find your utility.

1. We do not yet have the data for the utility (either the spatial boundary or the rates). If you have these data and want to include the data in the dashboard, please email with the data and we will add the utility.
2. You may have turned on some filters that do not match the criteria of your utility. Turn off all filters to see if your utility is present. There may be a data error in terms of population or owner size. Email and we will update the data. However, it would be best for EPA to update the underlying data.

We manually collected rates and are focused on adding new data to the dashboard rather than updating data. We hope to be able to work with others to develop a process to allow others to share and update the data to make data entry more sustainable. For now, please email to describe the error and/or with a link to where the new rates are located. Thank you for your help!

Here is what the icons do in order from left to right.

• [camera]: select this icon to save and download an image of the plot
• [magnifying glass]: select the icon, click and drag over a portion of the chart to zoom in
• [cross arrows]: select the icon, click and drag to move the chart around (pan)
• [+ sign]: select the icon to zoom in
• [- sign]: select the icon to zoom out
• [rotated cross arrows]: select the icon to zoom out
• [house]: select the icon to reset the chart
• [chart icon]: select the icon to go to plotly, which provided the libraries used to create these charts

Please visit our HOW TO USE THE DASHBOARD tab to learn about the maps and charts. There are YouTube videos available that walk through navigating the different elements of the dashboard.

• The raw data, processed data and code to create the processed data (written in R) are provided in our gitHub repository.
  The repository also provides a pdf detailing the work flow from collecting data to calculating metrics to developing a dashboard.


• The data and code are licensed under the creative commons clause and we only ask that you use the citation provided in GitHub if you use the data or code in your research or applications as described in the terms of use. This helps us to know whether to continue pursuing long-term maintenance and development of the underlying rates database.


• If you are publishing a paper, we ask that you use the citation to our paper (once accpeted); in the interim, please use the citation to the static data repository used to create the results in our paper:
  
  Patterson, L. & Doyle, M. (2021). Data and scripts from: Exploring the affordability of water services within and across utilities. Duke Research Data Repository. https://doi.org/10.7924/r4862k514


• The paper describing our approach and results.