Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, GLOBAL PUBLIC HEALTH (oxfordre.com/publichealth). (c) Oxford University Press USA, 2020. All Rights Reserved. Personal use only; commercial use is strictly prohibited (for details see Privacy Policy and Legal Notice).

date: 15 August 2020

Customer Assistance Programs and Affordability Issues in Water Supply and Sanitation

Summary and Keywords

Concerns about water affordability have centered on access to networked services in low-income countries, but have grown in high-income countries as water, sewer, and stormwater tariffs, which fund replacement of aging infrastructure and management of demand, have risen. The political context includes a UN-recognized human right to water and a set of Sustainable Development Goals that explicitly reference affordable services in water, sanitation, and other sectors. Affordability has traditionally been measured as the ratio of combined water and sewer bills divided by total income or expenditures. Subjective decisions are then made about what constitutes an “affordable” ratio, and the fraction paying more than this is calculated. This measurement approach typically omits the coping costs associated with poor supply, notably the time costs of carrying water home. Three less commonly used approaches include calculating (a) the expenditure related to procuring a “lifeline” quantity of water as a percent of income or expenditures, (b) the amount of income left for other needs after water and sewer expenditures are subtracted, and (c) the number of hours of minimum wage work needed to purchase an essential quantity of water.

Lowering water rates for all customers does not necessarily help those in need in low- and middle-income countries. This includes tariff structures that subsidize the price of water in the lowest block or tier (i.e., lifeline blocks) for all customers, not just the poor. Affordability programs that do not operate through tariffs can be characterized by (a) how they are administered and funded, (b) how they target the poor, and (c) how they deliver subsidies to the poor. Common types of delivery mechanisms include subsidizing public taps for unconnected households, subsidizing or financing the fees associated with obtaining a connection to the piped network, and subsidizing monthly bills for poor households. Means-tested consumption subsidies are most common in industrialized countries, whereas subsidizing public taps and connection fees are more common in low- and middle-income countries.

A final challenge is directing subsidies to renters who are more likely to be poor and who do not have a direct relationship with a water utility because they pay for water through their landlord, either included as part of their rent or as a separate water payment. Based on data from the 2013 American Housing Survey, approximately 21% of all housing units in the United States are occupied by this type of “hard to reach” customer, although not all of them would be considered poor or eligible for an assistance program. This ratio is as high as 74% of all housing units in metropolitan areas like New York City. Because of data limitations, there are no similar estimates in low-income countries.

Instead of sector-by-sector affordability policies, governments might do better to think about the entire package of services a poor person has a perceived right to consume. Direct income support, calculated to cover a package of basic services, could then be delivered to the poor, preserving their autonomy to make spending decisions and preserving the appropriate signals about resource scarcity.

Keywords: Affordability, water rates, pro-poor policies, poverty, water scarcity

Introduction

Concerns about the affordability of water and sewer services have been growing for several reasons. In higher-income countries like the United States, the cost of services for water, sewer, and stormwater are rising faster than inflation to replace aging infrastructure (AWWA, 2012), adapt to changing hydrological conditions, and comply with water quality regulations, particularly those limiting combined sewer overflows (National Academy of Public Administration, 2017).1 Rising water scarcity in many parts of the world is also straining the ability of utilities to match supply and demand, as evidenced most recently in Cape Town, South Africa (2017–2018), and Chennai, India (June 27, 2019).2 Although economists typically argue for the use of prices to help ration demand in face of rising scarcity, there is concern that high prices may cause consumption to fall “too much” among the poorest or lead to poor customers falling into arrears and being disconnected from service.

A parallel but longer-standing concern is improving access to safe water supply and sanitation in low-income countries. The World Health Organization (WHO) and UNICEF Joint Monitoring Program (2019) estimated that 3 in 10 people globally and 7 in 10 in sub-Saharan Africa do not use safely managed drinking water. Only 4 in 10 people used safely managed sanitation services. These statistics have improved in recent decades, particularly since the announcement of the WHO’s Millennium Development Goals, but a key barrier to continuing progress is how to finance the improvement and maintenance of built infrastructure in both rural and urban areas (Hutton & Varughese, 2016). Although donor support will continue to play a role in this financing, there must be an increase in local financing, including asking users and ratepayers to contribute more. Tariffs are in fact quite low in most urban water utilities in these countries, far below the real average cost of supply. A recent report using the World Bank’s International Benchmarking Network for Water and Sanitation Utilities benchmarking data estimated that only 15% of utilities in low-income countries cover operations and maintenance costs, let alone the cost of capital and capital depreciation (World Bank Group & UNICEF, 2017). Water tariffs in these places need to increase substantially, and governments and water utilities need ways to protect their poorest citizens from price increases while at the same time improving the quality and reliability of the service they receive. Indeed, the UN’s Sustainable Development Goals (SDGs) explicitly reference “affordable” service in medicine, education, credit, internet access, housing, energy, drinking water (SDG 6.1), and transport, and efforts are underway in several sectors to decide how these calculations should be made so they can be compared globally.

The first section of this article briefly reviews the methods used to measure affordability and discusses some of the measurement challenges faced. The next section discusses what policy options are available to utilities when policymakers wish to help the poor and is followed by a section describing ways to help renters and other “hard-to-reach” poor customers. The section provides an original empirical estimate of the share of U.S. households that fit this description using data from the most recent U.S. Census Bureau’s American Housing Survey (AHS). The conclusion discusses coordination of affordability programs across sectors.

It is important to note three limitations: First, with some exceptions, this article focuses on urban water customers and utilities. Most of the literature around this issue has until recently neglected the issue of “affordability” for households in rural areas, in part because these households primarily “pay” for water through the time they invest in collecting it. However, the issue of how and whether to subsidize rural water systems and the value of “demand-led” planning approaches is beyond the scope of this article. Second, treatment of sanitation is similarly limited and focuses on the common setting in high-income countries where water and sewer services are bundled together in the bill a customer receives. The term “water bills” generally refers to combined water and sewer bills. However, sewer services are sometimes paid through property taxes; combined bills may also include stormwater services; and households in cities in low-income countries may lack a connection to the piped water or sewer networks and pay for non-network water and rely on public latrines, open defecation, or pour-flush toilets that must be emptied manually or with vacuum trucks. Third, the discussion of affordability in high-income countries focuses largely on experience in the United States. Interested readers might consult another encyclopedia chapter (“Water tariffs in Spain”) for a discussion of the affordability measures undertaken by major Spanish cities (often in relation to household size).

Measuring Affordability

A first question that arises is how to define an “affordable service.” A number of reports and research articles have proposed and debated various approaches to measuring affordability (Fankhauser & Tepic, 2007; Hutton, 2012; Mack & Wrase, 2017; National Academy of Public Administration, 2017; Raucher, Clements, Rothstein, Mastracchio, & Green, 2019; Smets, 2017; Teodoro, 2018; Vanhille, Goedemé, Penne, Van Thielen, & Storms, 2018). Three broad approaches have been most commonly proposed: (1) the ratio of water and sewer bill compared to income or expenditures, or the conventional affordability ratio; (2) the potential expenditures that would be needed for consuming a predesignated minimum, essential, or lifeline quantity of water and sewer services as a percent of income or expenditures, that is, the potential affordability approach; and (3) the amount of income that would be left for other needs after actual water and sewer expenditures are subtracted, or the residual income approach. A more recent proposal has been to use the number of hours worked at the local prevailing minimum wage to cover the combined water and sewer bill for an “essential” amount of water (Teodoro, 2018).

As one would expect from its name, the conventional affordability approach is by far the most common. In the simplest case, suppose all customers pay a flat monthly service charge that is unrelated to usage and receive a regular monthly paycheck from an employer, which is observed by the government. In this case, expenditure on the water and sewer service could be presented as a fraction of monthly income. The distribution of these fractions could be informative (in this simple case, it will simply reflect the distribution of income in that community). What percent of users are paying more than 3% of their income on the service? What percent are paying more than 5%? The average fraction (i.e., the average customer spends 2.5% of their income on the service) is less informative, as it may not reflect the situation facing the poorest quintiles of income.

In many countries, policymakers have explicitly adopted a specific affordability target as a political or rights-based judgment. Smets (2017) surveyed affordability benchmarks globally, finding that most countries judge water and sanitation services affordable if they are in the range of 2% to 5% of income. Kenya, for example, uses a benchmark of 5% for combined water and sanitation spending, whereas Argentina considers unaffordable spending as more than 3% on water services only.

In 1997, the US Environmental Protection Agency (EPA) developed guidance to assess whether communities could afford improvements in wastewater treatment and combined sewer overflow reduction programs that were required by the Clean Water Act. The guidance determined that required wastewater investments that would raise bills to more than 2% of median household income were unaffordable. Water bills were not included. The origins of the 2% figure, and the use of median income as an indicator, are unclear but appear to have come from the use of median income in the 1972 Farm Home Loan program and spread to EPA documents as early as 1984 (EPA, 1984). The EPA (1998) added a threshold for water system costs of 2.5%, giving rise to the combined metric of 4.5% of median income. Although this metric was intended to be a measure of the overall financial capacity of a community, it has been widely applied to household-level affordability. Mack and Wrase (2017) use median income data from the U.S. Census, an average water tariff (based on a 2015 AWWA survey of 318 utilities) of $0.01 per gallon ($2.60/m3), and a consumption level of 12,000 gallons (45 m3) per month to estimate that 11.9% of U.S. households face “unaffordable” water and wastewater bills.3

This metric has been repeatedly criticized for being arbitrary and badly misrepresenting community affordability problems because of its reliance on the median income. Teodoro (2018) suggests basing metrics on the 20th percentile of income, which is more likely to capture affordability problems in the lower part of the income distribution. The EPA is in the process of redefining how it measures affordability (National Academy of Public Administration, 2017; Raucher et al., 2019).

Because it relies on the distribution of actual water use, the conventional affordability ratio also includes spending on nonessential water uses (e.g., seasonal outdoor water use). In contrast, the potential affordability approach relies on an assumption about what constitutes a minimum, or lifeline, consumption level to calculate the necessary expenditures needed to purchase that amount. These minimum consumption levels have also been the subject of discussion, given that water use varies substantially within similar geographies, let alone between countries and climates (Gleick, 1996). How much water is “essential”? At the level of survival, the WHO (2013) estimated that households in emergency situations like refugee camps needed between 7.5 and 15 liters per capita (LCD) per day for basic survival, cooking, and hygiene needs. Howard and Bartram (2003) classified households using 20 LCD as having “basic” access, 50 LCD “intermediate access,” and 100 LCD or more “optimal” access. Gleick (1996) also suggested 50 LCD (13.2 gallons) as a minimum. For a family of four, this corresponds to 6 m3 per month, which is also the amount guaranteed in South Africa’s Free Basic Water legislation. In many communities, however, what is deemed essential may be based not on basic survival needs but on what is considered a normal amount of water that most households would consume given the community’s culture and climate. A number of recent studies have attempted to estimate minimum consumption levels econometrically (García-Valiñas, Martínez-Espiñeira, & González-Gómez, 2010; Garcia-Valiñas, Athukorala, Wilson, Torgler, & Gifford, 2014; Sebri, 2015), although the approach used simply identifies the level of water consumption that is not sensitive to price changes and not a basic survival volume.4

Several other issues arise in calculating affordability using the conventional approach in the context of water and sanitation services in low-income countries (see Hutton, 2012, for a more detailed exploration). First, income data may be less reliably reported, so researchers rely more often on expenditures. Second, urban customers may supplement piped water with non-network sources like tanker trucks, and the utility will not observe these prices or quantities. Although the World Bank’s Living Standards Measurement Survey contains one question about combined water expenditures, most other national or international surveys do not (e.g., UNICEF’s Multiple Indicator Cluster Surveys). This also complicates the calculation of essential consumption. Using only utility billing data, it might be assumed that households are being forced to choose very small amounts because of unaffordable tariffs, and this may be the case. But the availability of groundwater, surface sources, or tanker water provides substitutes for municipal piped water, so households may in fact be consuming a sufficient amount of water in aggregate to meet their basic needs. Third, households often face unreliable service and therefore invest capital in storage tanks and treatment devices to cope (Cook, Kimuyu, & Whittington, 2016; Pattanayak, Yang, Whittington, & Kumar, 2005). Again, these decisions are not captured in regular survey efforts. Finally, including the economic value of time is critical in uncovering affordability problems, especially in rural areas where households spend significant time collecting water (Graham, Hirai, & Kim, 2016; Sorenson, Morssink, & Campos, 2011). In theory, placing a shadow value on this nonmarket time is straightforward, and a recent review supports using a “rule of thumb” in low-income countries of 50% of prevailing wages (Whittington & Cook, 2019). In the context of an affordability ratio, however, it is important to not only monetize water collection times in the numerator; all time available for labor also should be monetized in the denominator (i.e., “full income,” Becker, 1965).

What to Do about Affordability Concerns

Lower Water Rates for All: Logic and Evidence

For many, the first logical response to affordability concerns is to subsidize water prices for all customers to ensure it is affordable for the poorest. Water is by definition essential for survival, and is perceived as what economists call a “merit” good—one that all households should be provided regardless of their ability to pay. Indeed, a human right to water and sanitation was adopted by the UN General Assembly in 2010, and many countries such as South Africa have incorporated such a right into their constitutions (see Langford & Russell, 2017, for a history and introduction). There may also be a deep psychological aversion or “ancient instinct” to using prices and market-based logic to ration water supply (Whittington, 2016). Subsidy or welfare programs in which everyone benefits equally typically have more robust political support than those that target only the poor.

Lowering water rates for all could be accomplished by lowering the costs of supply through efficiency improvements such as reducing nonrevenue water, improving billing and collections, and improving labor productivity. Indeed, water sector professionals have largely focused on these types of efficiency improvements in the past. Utility benchmarking activities (Danilenko, van den Berg, Macheve, & Moffitt, 2014) would suggest that many utilities, particularly in low-income countries, have considerable scope for improvement along these dimensions.

Where efficiency in water production is already high, as in many utilities in high-income countries, prices are rising because of regulatory mandates and the need to replace aging infrastructure. Another approach to keeping costs low for all customers is to use large capital subsidies from the federal government or donors to replace, expand, and upgrade infrastructure facilities. Such an approach discourages utilities from doing careful, locally sensitive capital investment planning that uses ratepayer dollars wisely, fails to send the proper water scarcity signal to consumers, and directs a substantial amount of financial resources to nonpoor customers. This last point may be of less concern when these capital subsidies are raised progressively from national revenue (i.e., not external donor funding).

But this logic of protecting the poor through low rates for everyone is most commonly deployed to defend increasing block tariff structures, where (like tax brackets) marginal prices increase with increasing consumption blocks. As discussed in Whittington and Nauges (2020), a utility might set the amount of water in the first, “lifeline” block to be a minimum consumption level and set the volumetric price for that block at a very low or even zero price. In theory, higher prices in higher-consumption blocks cross-subsidize the low lifeline blocks. It is assumed that low-income households will only consume within the lifeline block, and mainly nonpoor households will consume into the higher tariff blocks. These tariffs structures are by far the most common approach that utilities use to try to manage affordability problems.

In theory, these programs can do a good job of helping the poor when the subsidized lifeline block rate is in fact targeted only to the poor (i.e., nonpoor households pay the full cost for consumption in the lifeline block). It may also be the case that untargeted lifeline rates do a good job of directing subsidies to the poor in high-income countries where most households are connected to a piped network, but evaluations of targeting performance (i.e., leakage, errors of inclusion, etc.) in that context are not available.

In low-income countries, there is ample evidence that this approach does not help the poor better than distributing subsidies at random (Angel-Urdinola & Wodon, 2012; Banerjee & Morella, 2011; Bardasi & Wodon, 2008; Barde & Lehmann, 2014; Diakité, Semenov, & Thomas, 2009; Fankhauser & Tepic, 2007; Foster, Gomez-Lobo, & Halpern, 2000; Fuente et al., 2016; Ruijs, 2009; Gomez-Lobo & Contreras, 2003; Nauges & Whittington, 2017; Whittington, Nauges, Fuente, & Wu, 2015). There are several reasons why this is so (see Whittington and Nauges (2020).

Customer Assistance Programs

There are several alternatives to subsidizing water rates for all customers, including innovative programs that attempt to help the poor pay their water and sewer bills. In the United States, these programs are often called customer assistance programs (CAPs). In a review of global CAP policies, Cook, Fuente, Matichich, & Whittington, 2020 describe four characteristics of these programs: (1) how subsidies are administered, (2) how they are funded, (3) how they target the poor, and (4) the policies or programs used to deliver the subsidies.

Program administration includes determining eligibility criteria, verifying and periodically reverifying eligibility, and designing how subsidies will be delivered. Cook et al. (2020) found that the vast majority of CAPs are administered by the utility or service provider. There are exceptions in low-income countries where donors’ preferences determine the design of the subsidy policy (e.g., subsidy connections through the Global Partnership on Output-Based Aid) or the rare programs that are funded at the federal level (i.e., Chile). There are also examples in the United States where eligibility verification is done by nonprofit organizations in coordination with income verification for home heating assistance (e.g., Cleveland, Ohio; see EPA, 2016).

Programs can be funded from the utility’s existing rate structure, the government, or voluntary/donor support. Many CAPs are described as being funded from the utility rate base, with nonpoor customers paying somewhat higher rates to support subsidy programs for the poor. However, such cross-subsidy programs are expressly forbidden by law in many U.S. states (Berahzer et al., 2017), and utilities typically rely on voluntary “round up your bill” programs to fund programs. Furthermore, such cross-subsidies are an illusion in the common situation where all customers are paying rates that are lower than the real average costs of supply. Utilities instead rely on periodic capital subsidies from donors or higher levels of government.

A different type of government support involves directly financing a utility’s assistance program. In Chile, for example, federal subsidies are distributed to states and then municipalities according to the percentage of the population who (a) would spend more than 3% of income on a combined water and sewer bill for 15 m3, (b) are elderly households in the bottom 40% of income, and (c) any households who qualify for Chile’s comprehensive welfare program Chile Solidario (Gomez-Lobo & Contreras, 2003; Contreras, Gomez-Lobo, & Palma, 2018). Households apply to the municipalities for eligibility determination. Municipalities notify the service provider, who then begins crediting the eligible customer’s account. The service provider bills the municipality for any subsidies awarded. The U-Save program in Singapore is another example of government funding (Chang & Fang, 2017).

How is eligibility for programs determined? In middle- and high-income countries with reliable income reporting, targeting is typically done with direct means-testing. This can be based on information reported to a tax agency or require the applicant to show paystubs or other documentation of insufficient income. In countries with less reliable income reporting, the rare means-testing targeting approaches use surveys and proxy measures for income such as dwelling type (e.g., Singapore) or durable assets (Coady, Grosh, & Hoddinott, 2004). Another targeting approach used in these settings is geographic, where all households living in a specific region are eligible for the program. These will obviously perform better when poverty is spatially concentrated. “Self-targeting” relies on variation in service-level quality and an assumption that only the poor would choose to use the cheaper (or free) lower-quality service. If given the choice between a piped connection and a public water point, for example, the assumption is that only the poor would choose the public water point. This assumption commonly breaks down when non-network water is in fact more expensive than a bill for the same amount of water from a piped connection and households would like to connect but cannot because of credit constraints, arduous application procedures (Devoto et al., 2012), corruption (Connors, 2007; Davis, 2004; Fass, 1988; Lovei & Whittington, 1993), or problems with land tenure that dissuade the utility from offering piped connections (Castro, 2009). Finally, demographic targeting is commonly used throughout the world (e.g., senior programs, disability status, veterans).

The final facet of programs is how the poor are assisted. Different authors and reports have used different typologies to describe programs (Banerjee & Morella, 2011; Komives, Halpern, Foster, & Wodon, 2006; EPA, 2016). Following the typology in Cook et al. (2020), program types include: (a) connection subsidies, which also includes preferential financing of connection loans; (b) payment flexibility programs, including allowing more or less frequent billing cycles to fit income patterns, the use of prepaid meters, and bill “leveling” policies; (c) conservation assistance, including rebates for low-flow appliances, irrigation timers, or conservation audits; (d) temporary crisis assistance, including one-time credits for those episodically in need; and (e) consumption subsidies, or subsidies that directly reduce the bill.

The type of consumption subsidy can in theory impact households’ incentives to conserve water depending on its effect on marginal prices.5 This may be important in settings where water scarcity and affordability overlap.6 On one end of that spectrum, fixed bill programs provide no financial incentive to use water wisely, for example by fixing leaking toilets. Philadelphia’s Tiered Assistance program limits the water bills of eligible customers to a fixed percentage of their income, completely decoupling the bill from volumetric water use (Nadolny, 2017). At the other end of the spectrum, programs like Singapore’s U-Save fixed rebate (Chang & Fang, 2017) should preserve incentives to use water wisely. The program deposits a set amount into the customer’s utility account each quarter and allows that credit to be rolled over and used in the future to pay water or electricity bills. The affordability program is therefore similar to providing lump-sum cash assistance. Although this might increase demand for water through an income effect, poor customers still face the same marginal price as nonpoor customers because the assistance is decoupled from the household’s water usage. In other words, a poor household might be incentivized to fix a leaking appliance because it will lower their monthly water bill without reducing the lump-sum rebate they receive each quarter. This is also similar to a policy advocated 20 years ago, which involves a simple tariff structure with a uniform volumetric price set and a fixed rebate (Boland & Whittington, 2000). Programs that have incentive properties between these two extremes include those that reduce bills by a set fraction (i.e., 50% off your total bill) or that provide a free or discounted lifeline block of water to poor customers only. Chile’s program reduced eligible customers’ combined water and sewer bills for consumption up to 15 m3 per household.

An EPA (2016) survey of 795 utilities in the United States found that a third were offering one, and often several, types of CAPs, including various types of programs for low-income customers, seniors, the disabled, and veterans. Consumption subsidies (“bill discounts” in the EPA report) were the most common (42% of programs), followed by payment flexibility programs, temporary assistance, and conservation assistance. The use of targeted lifeline block rates was used in only five of 365 CAPs, including those in Washington DC, Los Angeles, California, and Norman, Oklahoma (EPA, 2016). Similarly, WAREG, a network of European Water Regulators, surveyed utilities in 17 member countries (WAREG, 2017). Targeted lifeline rates were relatively rare, but fixed bill programs were somewhat more common than in the United States. European governments have also been more active than Americans in regulating the procedure for disconnecting customers for nonpayment, or even forbidding it outright (Aqua Publica Europea, 2016; Cook et al., 2020).

In low- and middle-income countries, Cook et al. (2020) found that the most common policies are connection subsidies (Jimenez-Redal, Parker, & Jeffrey, 2014; Virjee, 2009) and subsidies targeted to public taps (Banerjee & Morella, 2011; Bardasi & Wodon, 2008; Komives, Foster, Halpern, & Wodon, 2005) or shared private connections (Lauria, Hopkins, & Debomy, 2005; Mwangi, Otiego, & Ndakorerwa, 2015). The latter two use service-level targeting and are assumed to be used primarily by the poor. Fractional bills are used in several post-Soviet countries (Davis & Whittington, 2004; OECD, 2003), Argentina (Vagliasindi, 2013), and Panama City (Foster et al., 2000). Free or discounted “lifeline” allowances are also widely used, particularly in South Africa where the free basic water program subsidizes monthly consumption up to 6 m3, although it is up to municipalities to provide the allowance for all households or only for the poor (Calfucoy, Cibulka, Davison, Hinds, & Park, 2009; Department of Water Affairs and Forestry, 2002; Smith, 2010; Szabo, 2015). Payment flexibility programs are uncommon, with the exception of increasing use of prepaid meters in Africa (Heymans, Eales, & Franceys, 2014a, 2014b; Nhema & Zinyama, 2016). Cook, Fuente, and Whittington (2019) found no examples of conservation assistance programs in low-income countries.

Renters and the “Hard to Reach”

Another challenge is how to direct subsidies to households who are not customers of the utility, known as the “hard to reach” (Clements et al., 2017). Who are these households? In high-income countries where nearly all households have piped connections or private wells, they are renters who pay for water through their rent, rather than paying a water bill. This is most common in multifamily buildings where water is not submetered at each apartment. Although the Public Utilities Regulatory Policies Act of 1978 required that newly built apartments in the United States be individually metered for electricity, no such requirement exists for water. Because renters tend to have lower average incomes and wealth than owners and older buildings with lower rents that serve the poor are less likely to be submetered, there is good reason to believe that the poorest customers in most need of bill assistance are the hardest to reach. In low-income countries, there may be higher fractions of urban customers who are unconnected, collecting water from public water points, vendors, tanker trucks, or neighbors. It is also more common for households, particularly renters, to share access to a communal water tap located on the compound (Whittington, 1992).

What fraction of customers are hard-to-reach? Water utilities cannot produce a list of these households in their service area unless there is a centralized, up-to-date list of rental agreements and an inventory of multifamily buildings without submeters. If the utility knew only the latter, it could periodically ask the owner of a multifamily building to provide the names of tenants, but this not standard practice. Estimates for low-income countries are also not available, likely because of the complications already discussed and the lack of survey data that asks directly about whether water and sewer costs are included in the rent.

In the United States, however, one can estimate the fraction of hard-to-reach households using the (AHS, 2013). The AHS is a longitudinal survey that reports on housing units biennially. The national sample of 47,000 housing units was drawn in 1985. The data presented here are from the 2013 wave. The survey asks respondents living in a housing unit to report whether they own or rent the unit and whether they pay for utilities directly or through their rent. Clements et al. (2017) complete a similar exercise for nine counties using the Public Use Microdata Sample. Unlike the AHS, the Public Use Microdata Sample does not ask if households have a connection to a public or private utility. Table 1 provides the first national-wide estimates.

Nationwide, 29% of housing units are renter-occupied. Renters have on average lower household incomes ($55,264) than owners ($100,144). Of the renter-occupied units, 10% have electricity included in rent, 16% have natural gas included in rent, and 71% have water included in rent .7 The portion of all U.S. housing units occupied by a hard-to-reach water customer is therefore 21% (71% water-included of 29% rented units). The corresponding fractions for electricity and natural gas are much lower; 4% and 7%, respectively. These percentages vary substantially from city to city. Table 1 shows the 10 cities with the lowest percentage of hard-to-reach households and the 10 cities with the highest percentage. This varies from zero to 74% of all housing units. The 10 lowest cities, on average, have newer housing stock and higher rates of ownership. The two highest areas are in metropolitan New York City, where most people rent units in older multifamily buildings without individual submeters.

Table 1. Hard-to-Reach Housing Units in the United States, by Metropolitan Area

SMSA

Average building age (years)

% of units that are renter-occupied

% of renter-occupied units that do not have a separate fee for water

% Hard to Reach: units that are renter-occupied and do not have a separate fee for water

Panel A Greenville-Spartanburg, SC

49.2

24.1

0.0

0.0

Chattanooga, TN-GA

50.0

18.8

33.3

6.3

Birmingham, AL

57.4

26.9

28.6

7.7

Columbia, SC

48.6

21.9

42.9

9.4

Baton Rouge, LA

47.8

26.7

37.5

10.0

Fort Wayne, IN

50.1

20.0

50.0

10.0

Tucson, AZ

39.1

24.3

50.0

12.2

Austin, TX

32.9

46.8

27.6

12.9

Peoria, IL

62.1

45.5

30.0

13.6

Memphis, TN-AR-MS

48.4

35.0

39.3

13.8

Panel B Akron, OH

64.3

60.0

71.4

42.9

Fresno, CA

43.0

46.4

92.3

42.9

Lexington-Fayette, KY

58.5

54.3

78.9

42.9

Tulsa, OK

47.6

51.4

83.3

42.9

Los Angeles-Long Beach, CA

56.6

51.2

86.9

44.5

Sacramento, CA

44.3

53.2

87.9

46.8

Denver, CO

58.7

56.9

83.8

47.7

Oklahoma City, OK

44.6

51.9

92.9

48.1

New York City, NY

69.2

58.5

98.1

57.4

Jersey City, NJ

78.2

76.8

96.2

73.9

Average

50.4

32.9

74.5

24.5

Notes: All data from the 2013 AHS. SMSA = Standard Metropolitan Statistical Area, as defined by the AHS. Panel A shows the 10 SMSAs with the highest proportion of hard-to-reach housing units and panel B shows the 10 SMSAs with the lowest proportion. All analysis is restricted to housing units that report their main source of water is either a private or public water system. National averages vary slightly from the last row in the column because the table is restricted to housing units within a defined SMSA. National averages reported in the text are correct.

To reach these customers, Clements et al. (2017) suggest partnering with community organizations that are already serving the poor and working with landlords and housing groups to provide assistance. It is not clear, however, that subsidies delivered to landlords would not be simply captured by those landlords either directly or indirectly through higher rents. Such programs may be more successful in public housing units. The authors also suggest that many low-income households have had prior negative experiences trying to access social services, making them less likely to apply or enroll for other types of CAPs.

Conclusion

A dearth of careful empirical studies of existing affordability programs makes the characterization of any particular policy as the “best” or preferred policy difficult. Furthermore, context is undoubtedly important in crafting an effective strategy for a particular location. To help decision-makers understand program options, tradeoffs, and mistakes to avoid, Cook, Fuente, and Whittington (2019) address this by providing a set of scoping questions to identify important contextual factors, grouped by those that affect access to services and those that affect resources available for affordability programs, and present a simplified typology to map affordability policies to three illustrative cases. These cases vary in terms of the percentage of the population considered poor and the percent of the population connected to a piped network. In an urban setting where a large fraction of households would be considered poor and most are unconnected, most water subsidies should likely be directed toward a system of high-quality, accessible public taps. As economic growth lifts incomes and more people are connected to the system, connection subsidies are likely to be the more important tool in helping the poor. Finally, when nearly all households are connected and only a minority of households would be considered poor, means-tested consumption subsidies should pay a larger role, although targeting may require substantial administrative resources that should be coordinated with means-tested programs in other sectors. Utilities should also consider programs that allow flexible bill payment schedules, non-means-tested episodic bill forgiveness, and conservation assistance, particularly where water scarcity is a concern. Mistakes to avoid include subsidizing water bills for all customers in the hopes of reaching the poor and imposing administrative barriers for households who want a connection to get one. Similarly, means-tested programs should simplify application procedures or even use automatic enrollment. Finally, a key challenge is to convince central governments to finance affordability programs out of tax revenue rather than rely on locally generated sources, particularly cross-subsidies through the tariff.

The issue of affordability for water and sewer services should not be viewed in isolation from affordability for housing, food, energy, health care, and transportation, and may even be a misleading goal to pursue. Because of the UN’s Sustainable Development Goals, there are now sector-by-sector efforts to define affordability in each and to advance programs to protect the poor from paying too much of their income for each sector’s services. Discussions will center on how to properly measure affordability, and whether 3% or 5%, for example, is the right share for water. If 5% is chosen, then this decision impacts the available budget for healthcare or transportation (this logic is also reflected in the “residual income” approach to measuring affordability described in the section on “Measuring Affordability”).

Intelligently managing these interlocking value judgments is challenging. Furthermore, these judgments are inherently political, and there seems to be little consensus about what the appropriate cutoff should be. The affordability thresholds are largely determined by government officials (and often donors) without consultation with households. CAP programs might even be largely symbolic to show the public that something is being done about the affordability problem, but may in fact have relatively little effect on a poor household’s quality of life. Instead, governments might do better to think about the entire package of services a poor person has a perceived right to consume and ways to support the poor with direct income support, calculated to cover a package of basic services. Such direct income support would skirt the problem of hard-to-reach customers. In places where targeting the poor is a major challenge, investments could be made in centralized and reliable identification systems that can be used to direct bill support to water and energy utility accounts or mobile phone accounts like Kenya’s M-Pesa, a mobile phone–based money transfer, financing, and microfinancing service (Heymans, Eberhard, Ehrhardt, & Riley, 2016). This type of investment will seem far too expensive to help the poor in any one sector, but collectively will likely be more cost-effective in actually improving the lives of the poor, preserving their autonomy to make spending decisions, and preserving the appropriate signals about resource scarcity.

Acknowledgment

The author would like to thank Jake Wagner for research assistance in calculating the U.S. hard-to-reach estimates.

References

American Housing Survey (AHS). (2013). U.S. Census Bureau.Find this resource:

American Water Works Association (AWWA). (2012). Buried no longer: Confronting America’s water infrastructure challenge. Denver, CO, USA: American Water Works Association.Find this resource:

Angel-Urdinola, D. F., & Wodon, Q. (2012). Does increasing access to infrastructure services improve the targeting performance of water subsidies? Journal of International Development, 24(1), 88–101.Find this resource:

Aqua Publica Europea. (2016). Water affordability: Public operators’ views and approaches on tackling water poverty. Brussels, Belgium: Aqua Publica Europea.Find this resource:

Banerjee, S. G., & Morella, E. (2011). Africa’s water and sanitation infrastructure. Washington, DC: World Bank.Find this resource:

Bardasi, E., & Wodon, Q. (2008). Who pays the most for water? Alternative providers and service costs in Niger. Economics Bulletin, 9(20).Find this resource:

Barde, J. A., & Lehmann, P. (2014). Distributional effects of water tariff reforms: An empirical study for Lima, Peru. Water Resources and Economics, 6, 30–57.Find this resource:

Becker, G. (1965). A theory of the allocation of time. Economic Journal, 75(299), 493–517.Find this resource:

Berahzer, S. I., Clements, J., Raucher, R., Advisor, S., Associates, A., Giangola, L., Duckworth, M., Rubin, S., & Colton, R. (2017). Navigating legal pathways to rate-funded customer assistance programs: A guide for water and wastewater utilities.Find this resource:

Boland, J. J., & Whittington, D. (2000). The political economy of water tariff design in developing countries: Increasing block tariffs versus uniform price with rebate. In A. Dinar (Ed.), The political economy of water pricing reforms (pp. 105–120). Oxford, UK: Oxford University Press.Find this resource:

Calfucoy, P., Cibulka, J., Davison, J., Hinds, T., & Park, M. (2009). Improving free basic water provision in South Africa. Midrand, South Africa: Financial and Fiscal Commission, Republic of South Africa.Find this resource:

Castro, V. (2009). Setting up pro-poor units to improve service delivery: Lessons from water utilities in Kenya, Tanzania, Uganda and Zambia. Water and Sanitation Program. Washington, DC: World Bank.Find this resource:

Chang, Y., & Fang, Z. (2017). Efficient, equitable and sustainable energy policy in a small open economy: Concepts and assessments. Energy Policy, 105(March), 493–501.Find this resource:

Clements, J., Giangola, L., Issac Berahzer, S., Hughes, J., Rubin, S., & Colton, R. (2017). Customer assistance programs for multi-family residential and other hard-to-reach customers. Water Research Foundation.Find this resource:

Coady, D., Grosh, M. M., & Hoddinott, J. (2004). Targeting of transfers in developing countries. Washington, DC: World Bank.Find this resource:

Connors, G. (2007). Watering the slums: How a utility and its street-level bureaucrats connected the poor in Bangalore (Doctoral dissertation, Massachusetts Institute of Technology).Find this resource:

Contreras, D., Gomez-Lobo, A., & Palma, I. (2018). Revisiting the distributional impacts of water subsidy policy in Chile: A historical analysis from 1998–2015. Water Policy, 20(6), 1208–1226.Find this resource:

Cook, J., Fuente, D., Matichich, M., & Whittington, D. (2020). A global assessment of non-tariff customer assistance programs in water supply and sanitation. In Z. Chen, W. M. Bowen, and D. Whittington (Eds.), Development studies in regional science: Essays in honor of Kingsley E. Haynes. Singapore: Springer Nature.Find this resource:

Cook, J., Fuente, D., & Whittington, D. (2019). Choosing among pro-poor policy options in water supply and sanitation. Water Economics and Policy. Singapore: World Scientific.Find this resource:

Cook, J., Kimuyu, P., & Whittington, D. (2016). The costs of coping with poor water supply in rural Kenya. Water Resources Research, 52, 841–859.Find this resource:

Danilenko, A., van den Berg, C., Macheve, B., & Moffitt, L. J. (2014). The IBNET water supply and sanitation blue book 2014: The international benchmarking network for water and sanitation utilities databook. Washington, DC: World Bank.Find this resource:

Davis, J. (2004). Corruption in public service delivery: Experience from South Asia’s water and sanitation sector. World Development, 32(1), 53–71.Find this resource:

Davis, J., & Whittington, D. (2004). Challenges for water sector reform in transition economies. Water Policy, 6(5), 381–395.Find this resource:

Department of Water Affairs and Forestry. (2002). Free basic water implementation strategy. Pretoria, South Africa: Department of Water Affairs and Forestry.Find this resource:

Devoto, F., Duflo, E., Dupas, P., Parienté, W., Pons, V., Pariente, W., & Pons, V. (2012). Happiness on tap: Piped water adoption in urban Morocco. American Economic Journal: Economic Policy, 4(4), 68–99.Find this resource:

Diakité, D., Semenov, A., & Thomas, A. (2009). A proposal for social pricing of water supply in Côte d’Ivoire. Journal of Development Economics, 88(2), 258–268.Find this resource:

Fankhauser, S., & Tepic, S. (2007). Can poor consumers pay for energy and water? An affordability analysis for transition countries. Energy Policy, 35(2), 1038–1049.Find this resource:

Fass, S. M. (1988). Political economy in Haiti: The drama of survival. New Brunswick, NJ: Transaction.Find this resource:

Foster, V., Gomez-Lobo, A., & Halpern, J. (2000). Designing direct subsidies for water and sanitation services–Panama: A case study. Policy Research Working Paper Series 211 (May). Washington, DC: World Bank.Find this resource:

Fuente, D., Gakii Gatua, J., Ikiara, M., Kabubo-Mariara, J., Whittington, D., Mwaura, M., & Whittington, D. (2016). Water and sanitation service delivery, pricing, and the poor: An empirical estimate of subsidy incidence in Nairobi, Kenya. Water Resources Research, 52, 4845–4862.Find this resource:

Garcia-Valiñas, M. A., Athukorala, W., Wilson, C., Torgler, B., & Gifford, R. (2014). Nondiscretionary residential water use: The impact of habits and water-efficient technologies. Australian Journal of Agricultural and Resource Economics, 58(2), 185–204.Find this resource:

García-Valiñas, M. A., Martínez-Espiñeira, R., & González-Gómez, F. (2010). Affordability of residential water tariffs: Alternative measurement and explanatory factors in southern Spain. Journal of Environmental Management, 91(12), 2696–2706.Find this resource:

Gaudin, S. (2006). Effect of price information on residential water demand. Applied Economics 38(4), 383–393.Find this resource:

Gaudin, S., Griffin, R. C., & Sickles, R. C. (2001). Demand specification for municipal water management: Evaluation of the Stone-Geary form. Land Economics, 77(3), 399–422.Find this resource:

Gleick, P. H. (1996). Basic water requirements for human activities: Meeting basic needs. Water International, 21, 83–92.Find this resource:

Gomez-Lobo, A., & Contreras, D. (2003). Water subsidy policies: A comparison of the Chilean and Colombian schemes. World Bank Economic Review, 17(3), 391–407.Find this resource:

Graham, J. P., Hirai, M., & Kim, S.-S. (2016). An analysis of water collection labor among women and children in 24 sub-Saharan African countries. PLoS One, 11(6), e0155981.Find this resource:

Heymans, C., Eales, K., & Franceys, R. (2014a). The limits and possibilities of prepaid water in urban Africa: Lessons from the field. Washington, DC: World Bank.Find this resource:

Heymans, C., Eales, K., & Franceys, R. (2014b). Prepaid water in Mogale City: Case study. In The limits and possibilities of prepaid water in urban Africa: Lessons from the field (pp. 1–9). Washington, DC: World Bank.Find this resource:

Heymans, C., Eberhard, R., Ehrhardt, D., & Riley, S. (2016). Providing Water to Poor People in African Cities Effectively: Lessons from Utility Reforms.Find this resource:

Howard, G., & Bartram, J. (2003). Domestic water quantity, service level and health. Geneva, Switzerland: World Health Organization.Find this resource:

Hutton, G. (2012). Monitoring “affordability” of water and sanitation services after 2015: Review of global indicator options. United Nations Office of the High Commission for Human Rights.Find this resource:

Hutton, G., & Varughese, M. (2016). The costs of me eting the 2030 sustainable development goal targets on drinking water, sanitation, and hygiene: Summary report. Washington, DC: World Bank.Find this resource:

Ito, K. (2014). Do consumers respond to marginal or average price? Evidence from nonlinear electricity pricing. American Economic Review, 104(2), 537–563.Find this resource:

Jimenez-Redal, R., Parker, A., & Jeffrey, P. (2014). Factors influencing the uptake of household water connections in peri-urban Maputo, Mozambique. Utilities Policy, 28, 22–27.Find this resource:

Komives, K., Foster, V., Halpern, J., & Wodon, Q. (2005). Water, electricity, and the poor: Who benefits from utility subsidies? Washington, DC: World Bank.Find this resource:

Komives, K., Halpern, J., Foster, V., & Wodon, Q. (2006). The distributional incidence of residential water and electricity subsidies. Policy Research Working Paper No. 3878, World Bank, Washington, DC.Find this resource:

Langford, M., & Russell, A. (2017). Introduction: The right to water in context. In M. Langford & A. Russell (Eds.), The human right to water: Theory, practice and prospects. Cambridge, UK: Cambridge University Press.Find this resource:

Lauria, D. T., Hopkins, O. S., & Debomy, S. (2005). Pro-poor subsidies for water connections in West Africa. Water Supply and Sanitation Working Notes 3. Washington, DC: World Bank.Find this resource:

Lovei, L., & Whittington, D. (1993). Rent-extracting behavior by multiple agents in the provision of municipal water supply: A study of Jakarta, Indonesia. Water Resources Research, 29(7), 1965–1974.Find this resource:

Luby, I. H., Polasky, S., & Swackhamer, D. L. (2018). U.S. urban water prices: Cheaper when drier. Water Resources Research, 54(9), 6126–6132.Find this resource:

Mack, E. A., & Wrase, S. (2017). A burgeoning crisis? A nationwide assessment of the geography of water affordability in the United States. PLoS One, 12(1), 1–19.Find this resource:

Mwangi, P., Otiego, L., & Ndakorerwa, C. (2015). Innovation in scaling up access to water and sanitation services in Kenya. Water and Sanitation Program. Washington, DC: World Bank.Find this resource:

National Association of Clean Water Agencies (NACWA). (2005). Financial capability and affordability in wet weather negotiations. White Paper, Washington DC.Find this resource:

Nadolny, T. (2017, June 20). For low-income residents, Philadelphia unveiling income-based water bills. Philadelphia Inquirer.Find this resource:

Nataraj, S., & Hanemann, W. M. (2011). Does marginal price matter? A regression discontinuity approach to estimating water demand. Journal of Environmental Economics and Management, 61(2), 198–212.Find this resource:

National Academy of Public Administration. (2017). Developing a new framework for community affordability of clean water services. Washington DC.Find this resource:

Nauges, C., & Whittington, D. (2017). Evaluating the performance of alternative municipal water tariff designs: Quantifying the trade-offs between cost recovery, equity, and economic efficiency. World Development, 91(January), 125–143.Find this resource:

Nhema, A. G., & Zinyama, T. (2016). Prepaid water meters in Zimbabwe: The quest for an efficient water delivery and cost recovery system. Global Business and Economics Research Journal, 5(1), 1–30.Find this resource:

Organisation for Economic Co-operation and Development. (2003). Improving water management: Recent OECD experience. Paris, France: Organisation for Economic Co-operation and Development.Find this resource:

Pattanayak, S. K., Yang, J., Whittington, D., & Kumar, K. C. B. (2005). Coping with unreliable public water supplies: Averting expenditures by households in Kathmandu, Nepal. Water Resources Research, 41(1), 1–11.Find this resource:

Raucher, R., Clements, J., Rothstein, E., Mastracchio, J., & Green, Z. (2019). Developing a new framework for household affordability and financial capability assessment in the water sector. The American Water Works Association, National Association of Clean Water Agencies, and Water Environment Federation.Find this resource:

Ruijs, A. (2009). Welfare and distribution effects of water pricing policies. Environmental and resource economics, 43(2), 161–182.Find this resource:

Sebri, M. (2015). Water affordability and social equity in Tunisian governorates: A distributive approach. Water Policy, 17(1), 26–45.Find this resource:

Smets, H. (2017). Quantifying the affordability standard. In M. Langford & A. F. S. Russell (Eds.), The human right to water: Theory, practice and prospects. Cambridge UK: Cambridge University Press.Find this resource:

Smith, J. A. (2010). How much water is enough? Domestic metered water consumption and free basic water volumes—The case of Eastwood, Pietermaritzburg. Water SA, 36(5), 595–606.Find this resource:

Sorenson, S. B., Morssink, C., & Campos, P. A. (2011). Safe access to safe water in low income countries: Water fetching in current times. Social Science and Medicine, 72(9), 1522–1526.Find this resource:

Szabo, A. (2015). The value of free water: Analyzing South Africa’s free basic water policy. Econometrica, 83(5), 1913–1961.Find this resource:

Teodoro, M. P. (2018). Measuring household affordability for water and sewer utilities. Journal of the American Water Works Association, 110, 1–28.Find this resource:

U.S. Environmental Protection Agency. (1984). Financial capability guidebook. Washington, DC: Environmental Protection Agency.Find this resource:

U.S. Environmental Protection Agency. (1997). Combined sewer overflows: Guidance for financial capability assessment and schedule development. Washington, DC: Environmental Protection Agency.Find this resource:

U.S. Environmental Protection Agency. (1998, September). Variance technology findings for contaminants regulated before 1996. Washington, DC: Environmental Protection Agency.Find this resource:

U.S. Environmental Protection Agency. (2016). Drinking water and wastewater utility customer assistance programs. Washington, DC: Environmental Protection Agency.Find this resource:

Vagliasindi, M. (2013). Implementing energy subsidy reforms: Evidence from developing countries. Washington, DC: World Bank.Find this resource:

Vanhille, J., Goedemé, T., Penne, T., Van Thielen, L., & Storms, B. (2018). Measuring water affordability in developed economies: The added value of a needs-based approach. Journal of Environmental Management, 217, 611–620.Find this resource:

Virjee, K. (2009). Leveraging private sector finance for rural piped water infrastructure in Kenya: The use of output-based aid. OBApproaches 30(30).Find this resource:

WAREG. (2017). Affordability in European water systems.Find this resource:

Water Research Foundation. (2016). Residential end uses of water, version 2: Executive report. Denver, CO: Water Research Foundation.Find this resource:

Whittington, D. (1992). Possible adverse effects of increasing block water tariffs in developing countries. Economic Development and Cultural Change, 41(1), 75–87.Find this resource:

Whittington, D. (2016). Policy nook: Ancient instincts: Implications for water policy in the 21st century. Water Economics and Policy, 2(02), 1671002.Find this resource:

Whittington, D., & Cook, J. (2019). Valuing changes in time use in low- and middle-income countries. Journal of Benefit-Cost Analysis, 10, 51–72.Find this resource:

Whittington, D., & Nauges, C. (2020). An Assessment of the Widespread Use of Increasing Block Tariffs in the Municipal Water Supply Sector.Find this resource:

Whittington, D., Nauges, C., Fuente, D., & Wu, X. (2015). A diagnostic tool for estimating the incidence of subsidies delivered by water utilities in low- and medium-income countries, with illustrative simulations. Utilities Policy, 34, 70–81.Find this resource:

Wichman, C. J. (2014). Perceived price in residential water demand: Evidence from a natural experiment. Journal of Economic Behavior and Organization, 107(A), 308–323.Find this resource:

Wilkes, W. (2018). Make water affordable again? Journal of the American Water Works Association, 110(6).Find this resource:

World Bank Group and UNICEF. (2017). Sanitation and water for all: How can the financing gap be filled? World Bank, Washington, DC: World Bank.Find this resource:

World Health Organization. (2013). How much water is needed in emergencies? Geneva, Switzerland: World Health Organization.Find this resource:

World Health Organization and UNICEF Joint Monitoring Program. (2019). Progress on household drinking water, sanitation and hygiene 2000–2017: Special focus on inequalities. Geneva, Switzerland: World Health Organization and UNICEF.Find this resource:

Zetland, D., & Gasson, C. (2013). A global survey of urban water tariffs: Are they sustainable, efficient and fair? International Journal of Water Resources Development, 29(3), 327–342.Find this resource:

Notes:

(1.) The website Circle of Blue reports water rates for 30 U.S. cities. Water rates increased approximately 6%–8% per year from 2010 to 2015, with increases slowing somewhat in 2015–2018.

(2.) See Divya Karthikeyan’s article in The Guardian about historic droughts in Chennai, India.

(3.) This consumption level is likely an overestimate for the United States. Mack and Wrase (2017) cite an EPA website for this figure (“How we use water”) that, as of January 2020, reported that U.S. households use an average of 300 gallons per day. This corresponds to 34 m3 per month, or 283 LCD for a family with four members. The Water Research Foundation (2016) reported that an average U.S. household used 28 m3 per month.

(4.) These papers all build on Gaudin, Griffin, and Sickles (2001), who proposed a Stone-Geary function to estimate price elasticities. This form allows the researcher to estimate an intercept (γ‎) for water consumption that is empirically unresponsive to prices. Although researchers citing this paper have interpreted γ‎ as an “essential” amount, there are a number of reasons that consumers may be unresponsive to prices, including inattention and low prices overall. Indeed, Gaudin et al. (2001) caution against this interpretation and prefer to call the parameter the “conditional water use threshold” (p. 404): “Indeed the term ‘subsistence level’ is misleading for water demand analysis. The γ‎ parameter does not indicate how much water is needed to survive, but the amount of water that may not be responsive to prices.”

(5.) Do customers react to marginal prices in the tariff, or simply pay attention to their average bill? Tariff structures are often complex and hard to understand, tariff data is often not provided on the bill (Gaudin, 2006), billing is infrequent and after consumption occurs, and total bills may be too low to be worth the effort to comprehend. These all point to customers reacting to average prices, borne out by recent research (Ito, 2014; Wichman, 2014). Consumers may react to marginal prices, however, when the jump between price blocks is substantial enough to be salient (Nataraj & Hanemann, 2011).

(6.) It appears in the U.S. at least, however, most utilities may not be following economists’ standard prescriptions to use prices to ration long-term water demand. Luby, Polasky, & Swackhamer (2018) survey rates in 35 metropolitan areas in the U.S. and find that water rates are actually lower in cities facing more water stress. Similarly, Zetland and Gasson (2013) examine tariffs in 308 cities in 102 countries, and find that higher water tariffs are correlated with a lower risk of shortage but also lower water availability. Higher rates are also correlated with lower per capita consumption, smaller local populations, and higher demand.

(7.) Sixty-five percent of renter-occupied units have natural gas connections. Among all renter-occupied units, 16% do not have a separate fee for natural gas. Among the renter-occupied units with natural gas connections, 24% of them do not have a separate fee for natural gas.