-------
-------
Contents
Contents
Acknowledgements
Executive Summary
Trends and Key Findings
Survey Methodology
Intended Uses of CWS Survey Data
1. Introduction
1.1 Background
1.2 Data Presentation
1.3 Organization of the Report
2. National Projections Summary
2.1 Water System Profiles
2.2 Customer Profile
2.3 Operational Summary
2.4 Financial Summary
3. Trends and Key Findings
3.1 Trends
3.2 Key Findings
4. Intended Uses of CWSS Data
4.1 Regulatory Development Analyses
4.2 Policy Development Analyses
4.3 Regulatory Implementation Analyses
4.4 Compliance Analyses
II
• • •
III
iii
iv
v
1
1
3
4
5
5
8
9
12
17
17
21
29
29
32
34
34
-------
Acknowledgements
Many dedicated owners, operators, and managers of community water
systems made this survey possible. We would like to thank the almost 2,000
water systems that devoted valuable time to searching through records and
completing questionnaires.
The Community Water System Survey was managed by Mr. Brian C.
Rourke of EPA's Office of Ground Water and Drinking Water (OGWDW). He
was assisted by Mr. Benjamin P. Smith, also of OGWDW.
The Cadmus Group, Inc. served as prime contractor for this project.
Westat, Inc., a subcontractor, was responsible for survey design, data collection
and database development.
-------
Executive Summary
Trends and Key Findings
r I ^he U.S. Environmental Protec-
• tion Agency (EPA) conducted
JL the 1995 Community Water
System (CWS) Survey to obtain data
to support its development and evalua-
tion of drinking water regulations.
Most of the operating characteristics
of community water systems are
unchanged from 1976, when the first
CWS Survey was conducted. The vast
majority of systems are small and
privately owned, but most people are
customers of large publicly owned
systems. Nevertheless, there have been
some important changes since the first
CWS Survey. Trends and key findings
from the survey include:
• The percentage of systems that do
not treat their water steadily
declined from 1976 to 1995. This is
consistent with the emphasis on
water quality monitoring and
treatment since the Safe Drinking
Water Act (SDWA) was first passed
in 1974. For more information on
this trend, see section 3.1.1 of
Chapters.
Percentage of Systems
Not Providing Treatment
101-500 501-1,000 1,001-3,300 3,301-10,000
SYSTEM SIZE
E3 1976 • 1962 D 1986 B 1995
• In spite of this decrease in systems
not providing treatment, only 19.5
percent of the capital investment
made by community water sys-
tems is for water quality improve-
ments. Community water systems
reported spending $32.6 billion in
the 8 years following SDWA
reauthorization in 1986. The largest
category of capital investment was
expansion (49.9 percent of the total),
followed by repair and replacement
of infrastructure (30.6 percent). This
is discussed more fully in section
3.1.2.
Water Industry
Capital Investment Expenditures
Water Quality 19.5%
Repair/Replace. 30.6%
Expansion 49.9%
Standard financial ratios indicate
that many small community water
systems are, on average, not
financially healthy. The operating
ratio is calculated by dividing total
operating revenues by operation and
maintenance (O&M) expenses. If
the ratio is 1.0, a system can cover
its daily expenses, but little else. A
higher ratio means that funds are
available for non-operating func-
tions, such as servicing debt or
establishing a capital reserve fund.
As shown in the following table,
over 30 percent of all small systems
serving fewer than 500 people have
operating ratios of less than 1.0.
in
-------
1995 Community Water System Survey
Di'ilnliumm al Opor.itmij Ratio
fP«'i(:t!iiliKjt* ol System*.!
rubteWiurs
o
1(01,2
>1.2
rrtviti Wittr I
<1
I 101.2
>1.2
System SU>
80.000
rttona
34,8
17.8
47.4
yttxtu
32,3
19.0
48.1
19,0
23.6
67.4
17.2
18.7
64.1
14.4
10,8
74.8
6.O
16.8
77.2
8.3
9.5
82.2
6.4
5.1
88.5
Another measure of financial
condition is the debt service cover-
age ratio. It gauges the ability of the
system to cover debt service after all
other "cash" expenses have been
paid. The numerator of this ratio is
the annual net revenue available to
pay debt service costs, and the
denominator is the amount of debt to
be retired plus associated interest on
that debt. Generally, this ratio
should exceed 1.0. The table below
shows that approximately half of the
systems in the smallest size category
have ratios of less than 1.0.
Distribution al Debt Service Coverage Ratio
(Percentage of Systems)
SyMmStz*
BO,000
Pubic WKvEy)«m»
<1X>
t.Otol.5
>1,B
62.1
19.3
28.6
41.B
16.5
4ZO
25.7
21 ,0
53.3
15.7
21.2
63.1
Prhtt* WXir SyitMTD
<1O
1 0 to 1.6
>W
48.7
12.6
40.7
23.5
11.4
60.1
16.4
14,8
69.8
7.7
3.1
892
When assessing this financial
information about water systems,
please note the survey results
represent only one year of financial
performance. Using only this
"snapshot" of the industry in 1995,
EPA's ability to derive conclusions
about the overall financial health of
the water industry is limited. For a
more detailed interpretation of this
financial data, see section 3.1.2.
Over one-third of all community
water systems participate in some
type of source water protection
effort. These efforts can be a low-
cost method of preventing contami-
nation of water sources. Systems
serving populations greater than
3,300 reported a higher rate of
participation in source water protec-
tion efforts (45 percent) than did
systems that serve fewer than 3,300
people (32 percent).
Percentage of Systems
Participating in Source Water
Protection
100%
80%
60%
40%
20%
0%
45%
>3,300 <3,300
Population Served
The 1996 SDWA Amendments made
source water protection a national
priority. The survey results show that a
substantial portion of community
water systems already participate in
some type of contamination preven-
tion efforts. These efforts may or may
not meet requirements under state
source water protection programs. For
more information on source water
protection, see section 3.1.1.
Survey Methodology
EPA surveyed community water
systems in 1976, 1982, and 1986.
Consistent with the previous surveys,
the 1995 CWS Survey collected
information on the most important
operational and financial characteris-
tics of community water systems.
EPA started the 1995 CWS Survey in
the fall of 1994. During Phase I, EPA
developed a preliminary survey
instrument and sampling plan, then
conducted a pretest with nine water
systems to gauge respondents' reac-
tions to the draft questionnaire. Next,
EPA conducted a full-scalefpilot test
of the questionnaire. The Agency used
a computer-assisted telephone inter-
view questionnaire to identify eligible
iv
-------
Executive Summary
systems and appropriate respondents
for the pilot test and for the mail
questionnaire.
In Phase II, EPA revised the survey
instrument and sampling plan based on
results of the pilot test. The Agency
mailed the questionnaire to 3,700
eligible community water systems in
June 1995 and maintained a toll-free
helpline to answer questions from
respondents. EPA continued to receive
completed questionnaires until Febru-
ary 1996, at which time approximately
54 percent of the eligible participants
had completed questionnaires.
A more complete discussion of the
survey methodology can be found in
Chapter 1 of Volume I and in Volume
II, Part 3 (Methodology Report).
Intended Uses of CWS
Survey Data
EPA developed the CWS Survey
database to provide critical data to
support regulatory development and
implementation. The Agency plans to
use the data to support the types of
analyses discussed below.
Regulatory Development Analyses.
Before new regulations are established
under the SDWA, the Agency must
satisfy the requirements of various
statutes and regulations including:
Executive Order 12866, the Paperwork
Reduction Act, the Regulatory Flex-
ibility Act, the Small Business Regula-
tory Enforcement Fairness Act, and
the Unfunded Mandates Reform Act.
Data from the CWS Survey provide
baseline information that is critical to
the preparation of these analyses.
Policy Development Analyses. The
diversity of water systems in the CWS
Survey database provides sufficient
financial and operational data to
support a variety of Agency initiatives
to develop policies and guidance
documents for states and public water
systems concerning the implementa-
tion and enforcement of drinking
water regulations. The Agency also
receives periodic requests from
Congressional staff and committees,
other federal agencies, and the public
for information on the water supply
industry. The 1995 CWS Survey
provides current information on the
water industry to satisfy these re-
quests.
Regulatory Implementation Analy-
ses. A critical issue for EPA to address
under the 1996 SDWA Amendments
is whether the drinking water industry,
and small systems in particular, have
the technical and financial capacity to
comply with SDWA regulations over a
sustained period. Congress has pro-
vided money to assist the states and
EPA in building additional capacity
through State Revolving Loan Funds
for public water systems. CWS Survey
data, in conjunction with data from the
Drinking Water Infrastructure Needs
Survey, may be used to assess the
financial ability of the water industry
to finance infrastructure investment.
Compliance Analyses. EPA will use
the CWS Survey database in develop-
ing profiles of operational and finan-
cial characteristics for different types
of water systems that can be statisti-
cally correlated with the Agency's
database of compliance records
contained in the Safe Drinking Water
Information System (SDWIS). The
objective of this analysis is to identify
those operational and financial charac-
teristics that may lead to future
compliance problems.
-------
-------
1. Introduction
To support its regulatory
development initiatives, the
U.S. Environmental Protection
Agency (EPA) periodically collects
information on the financial and
operating characteristics of the public
water supply industry. EPA conducted
the 1995 Community Water System
(CWS) Survey as part of this effort.
EPA uses the information from this
survey to prepare regulatory impact
analyses (RIAs) in support of regula-
tory development and to analyze
economic and operating factors that
affect national drinking water quality.
A complete discussion of the uses of
the survey data will be found in
Chapter 4 of this volume.
This report presents the information
collected from the 1995 CWS Survey
in two volumes. Volume I, the Over-
view, provides perspective on these
details by extrapolating from the
survey data to present a national
picture of water systems. It does this
by grouping systems into fewer
categories than in Volume II in order
to observe patterns that characterize
the industry, and by comparing the
1995 data to similar data from previ-
ous CWS Surveys (1986, 1982, and
1976).
Volume II, the Detailed Report,
presents the survey findings in a series
of tables that break out water systems
according to detailed size, ownership,
and water source categories. Volume
II also provides a detailed methodol-
ogy.
1.1 Background
EPA began the 1995 CWS Survey in
the fall of 1994. The survey effort
proceeded in two phases. In Phase I,
EPA developed a preliminary survey
instrument and sampling plan, and
then conducted a pretest of nine water
systems to gauge respondents' reac-
tions to the draft questionnaire.
Following the pretest, EPA conducted
a full-scale pilot test using three
versions of the survey instrument: one
for publicly owned systems, one for
privately owned systems, and one for
ancillary systems (a system where
water supply is ancillary to its primary
business—e.g., mobile home park).
(The three versions of the question-
naire varied only slightly to reflect
minor variations in methods of ac-
counting and financial data.) The goal
of the pilot test was to evaluate the
design and procedures for the full
study. In order to identify eligible
systems and appropriate respondents
for the mail questionnaire, the survey
used a computer-assisted telephone
interview (CATI) questionnaire to
conduct a preliminary screening. The
CATI survey identified over 4,700
eligible water systems, of which EPA
selected 62 for the pilot test. Twenty-
one systems returned completed
questionnaires.
In Phase II, EPA prepared a revised
sampling plan and questionnaires
based on the results of the pilot test. In
June 1995, the surveys were distrib-
uted to a stratified random sample of
3,700 water systems nationwide. As
shown in the table on the following
page, the revised questionnaire com-
prised 24 operational questions and 13
financial questions. Three questions
related to general information also
were asked. Most questions requested
several line items of information.
Community water system respondents
had until February 1996 to return
completed questionnaires. Slightly
more than 54 percent of the systems
that received questionnaires responded
to the survey. From June 1995 to
-------
1995 Community Water System Survey
Question
Number
Summary of 1995 CWS Survey Questionnaire
Summary of Question
General Information
Contact information (e.g., name and telephone number of persons filling out questionnaire).
Year for which operating and financial information are provided.
Source of regulatory information, operator training, and technical assistance.
10
TT
13
W
W
19
sST
21"
IF
23~
24~
25~
28~
27~
28
29~
30~
IT"
32~
33"
34
35
36
37~
38
39~
40
Part I - Operating Characteristics
Number of gallons produced in the last year and number of intake points with disinfection.
Peak daily production and maximum dally treatment design capacity.
Factors determining maximum design capacity.
Presence of treated water storage.
Treated water storage (e.g., type of storage, number of tanks, and storage capacity).
Distribution system (e.g., type and length of existing pipe, pipe replacement, main repairs, and
months between flushes).
Length of pipe for expansion in last five years.
People served and active connections currently and five years ago.
ZIP codes covered by service area.
Presence of drinking water operators.
Number, employment status, and training level of operators.
Presence of system interconnection for emergency purposes (e.g., hot summers).
Solutions in case of permanent contamination of water source.
Indication and identification of connection to long term alternate water source.
Treatment facility information (e.g., name, location, number of wells, average and potential
flows, and treatment provided).
Indication of points in distribution system where disinfectant residuals are boosted.
Sources not receiving treatment (e.g., name, type, location, and average and potential flows).
Indication of source water or wellhead protection program.
Indication of source water or wellhead protection measure applied.
Indication of source water or wellhead protection management (e.g., local government).
Indication of how management area is delineated.
Indication of presence and type of contaminants located within 2 miles of water intakes.
Identification of laboratory analysis provider.
Laboratory analysis payment method.
Part II - Financial Information
Indication of application of Generally Accepted Accounting Principles for financial reports.
Water sales revenues and deliveries by customer category (e.g., residential, commerical, etc.).
Water related revenue sources (e.g., connection fees, general fund revenues, etc.).
Billing structure (e.g., customer category rate increase information, and active connections).
Uncompensated usage (e.g., free service to municipal buildings, uncollected bills, leaks, etc.).
Expanse summary (e.g., operating, debt service, and other).
Assets, liabilities, and total debt outstanding.
Indication of payment for capital improvements, repairs, or expansion since 1987.
Source of funds for capital improvements, repairs, or expansion since 1987.
Cancellation of plans for capital improvements, repairs or expansion due to borrowing limits.
Indication of whether bonds have been rated by a rating service.
Summary of bond rating and type of bond last issued.
Additional comments.
February 1996, EPA maintained a toll-
free helpline to answer questions and
provide guidance to water systems
participating in the survey.
To enhance the quality of the data
provided in the completed question-
naires, EPA conducted a manual
quality assurance (QA) review begin-
ning in December 1995. This QA
review focused on the eight survey
questions that EPA considered to be
the most critical to supporting its
regulatory initiatives. These critical
questions addressed:
1. Annual water production by water
source—ground water, surface
water, and purchased water. (Ques-
tion 4)
2. Service population and active
connections. (Question 11)
3. Characteristics of treatment facili-
ties—location, average and potential
daily flows, and treatments pro-
vided. (Question 18)
4. Characteristics of untreated
sources—name, location, and
average and potential daily flows.
(Question 20)
-------
Chapter 1—Introduction
5. Water sales revenues and deliveries
by customer category. (Question 29)
6. Water-related revenues—connection
fees, inspection fees, interest earn-
ings, etc. (Question 30)
7. Water system expenses—operating
expenses, debt service expenditures,
and other expenses. (Question 33)
8. Water system assets, liabilities, and
debt. (Question 34)
The first step of the QA review
process was to determine if responses
to any of the eight critical questions
appeared inconsistent with information
provided in other survey questions, or
with expert knowledge of the water
industry. In cases where respondents
made obvious mistakes and solutions
were apparent (e.g., reporting thou-
sands of gallons instead of millions of
gallons), responses were corrected. In
cases where mistakes were suspected,
but solutions were not apparent, the
water systems in question were
contacted. Reviewers used both
approaches to obtain answers for blank
or incomplete critical questions.
EPA identified mistakes or inconsis-
tencies in responses to critical ques-
tions on about half of the completed
questionnaires—approximately 1,000
questionnaires. EPA telephoned
approximately 500 water systems to
clarify the mistakes or inconsistencies.
Ultimately, all identified mistakes and
inconsistencies were corrected.
Because approximately half of the
systems responding to the survey
provided inconsistent responses to at
least one of the eight critical ques-
tions, EPA was concerned that incon-
sistencies may be present in the 32
noncritical questions that did not
undergo manual QA review. EPA
conducted automated data validation
checks on most of the noncritical
questions to identify and remove data
anomalies and outliers, but it was not
within EPA's budget or the scope of
this project to conduct detailed QA
reviews of the noncritical questions.
To ensure high data quality, EPA
relied on expert peer review and on the
assistance of known experts in all
phases of the CWS Survey project. A
complete discussion of the peer review
process and of the quality assurance
efforts can be found in the methodol-
ogy chapter of Volume II.
1.2 Data Presentation
Volumes I and II of the CWS Survey
Report present tabulations of the data
collected in the CWS Survey. In
Volume I, the data are generally
presented according to four service
population categories:
• Very Small—From 25 to 500 served
• Small—501 to 3,300 served
• Medium—3,301 to 50,000 served
• Large—More than 50,000 served
In Volume II, results are presented
according to eight service population
categories:
• From 25 to 100 served
• 101 to 500 served
• 501 to 1,000 served
• 1,001 to 3,300 served
• 3,301 to 10,000 served
• 10,001 to 50,000 served
• 50,001 to 100,000 served
• More than 100,000 served
-------
1995 Community Water System Survey
These different size categories support
the different analytic purposes men-
tioned earlier.
Data tabulations also are presented
according to ownership (e.g., public,
private, or ancillary) and primary
water source (e.g., primarily ground
water, primarily surface water, and
primarily purchased water). The most
detailed level of data disaggregation
presented in this report is by owner-
ship and primary water source (e.g.,
the number of publicly owned systems
with primarily surface water sources).
1.3 Organization of the
Report
This report comprises two volumes:
Volume I: The Overview. Volume I
presents the most important and
interesting findings of the survey. It is
composed of an Executive Summary,
which summarizes the key findings
and highlights of the survey results,
and four chapters:
• Chapter 1. Introduction. Chapter 1
describes the background, purpose,
survey methodology, and the organi-
zation of the overall report.
Chapter 2. National Projection
Summary. Chapter 2 provides an
aggregate perspective on basic water
industry demographics and opera-
tional and financial characteristics of
the industry. It presents a national
profile of water systems, their
customers, and their operating and
financial characteristics.
Chapter 3. Key Trends and
Survey Findings. This chapter
provides a discussion, supported by
graphics, of the principal findings of
the CWS Survey. Chapter 3 provides
a summary of operational and
financial survey findings, as well as
a comparison of findings in the 1995
survey to those of the 1986, 1982,
and 1976 surveys.
Chapter 4. Uses of Data. This
chapter describes the intended uses
of the CWS Survey data. It shows
how the survey questions relate to
the data requirements of drinking
water RIAs and Regulatory Flexibil-
ity Analyses (RFAs).
Volume II: Detailed Survey Result
Tables and Methodology Report.
Volume II presents a detailed sum-
mary of data collected in the CWS
Survey. There are no narrative de-
scriptions accompanying these tabula-
tions. The results are divided between
operating and financial characteristics.
The order of presentation generally
corresponds to the order and organiza-
tion of the survey questionnaire.
Volume II also provides a detailed
description of the survey methodol-
ogy. It provides information on sample
design and weighting, the telephone
screener survey, the mail survey, and
quality assurance. Copies of the
survey questionnaires are supplied as
appendices.
-------
2. National Projections Summary
r"lP|he 1995 CWS Survey data-
• base contains financial and
-M. operational data for a wide
variety of public water supplies. As an
introduction to the database, this
section of the report describes the
current operating and financial charac-
teristics of the water industry. The data
are presented, in general, as industry
totals based on projections from the
survey sample statistics. The objective
of this presentation is to establish the
preliminary themes and patterns in the
CWS Survey results that are more
fully explored in Chapter 3.
A Note on the Data
Because not every respondent answered
every survey question, some variables dis-
cussed in Chapters 2 and 3 may have more
or fewer available observations than other
variables. This can reduce consistency
across variables, since different groups of
systems may be represented in different
calculations. For example, the survey es-
timated water produced, water delivered,
and water losses due to unaccounted for
use. In theory, one might assume that
water produced minus water delivered
would equal water loss. In fact, if the re-
spondents are different for each variable,
this simple calculation does not work. EPA
and CWS Survey analysts decided that the
increased accuracy for each variable (e.g.,
water produced) outweighed any reduction
in consistency. When inconsistency seems
significant, we explain it in a footnote, or a
note in the relevant table.
2.1 Water System
Profiles
Altogether, there are more than 180
thousand water systems in the United
States serving over 250 million
people. Included in this estimate of
systems are community water systems,
transient noncommunity water sys-
tems, and nontransient noncommunity
water systems.
Number of Public Water
Systems
System Type
Community Water
Systems in Survey (1,2)
Nontransient
Noncommunity Water
Systems (3)
Transient Noncommunity
Systems (3)
Number
50,289
23,639
106,436
%of
Total
28%
13%
59%
(1) Certain types of community water systems were
excluded from the estimate. See Vo ume II .for more
information.
(2) The number of community water systems is an
estimate from the CWS Survey.
(3) Data on the number of noncommunity systems come
from "The National Public Water System Supervision
Program" (EPA-81 2-R-95-001 , July 1995), which relies on
a national inventory of public water systems.
The vast majority of the U.S. popula-
tion served by water systems are
served by community systems. While
people also drink water from
noncommunity systems, community
water systems provide the most
exposure to risks from contaminants.
Therefore, they are the focus of this
report.
Population Served by Types
of Public Water Systems
System Type
Community Water
Systems in Survey
Nontransient
Noncommunity Water
Systems
Transient
Noncommunity Systems
Population
(Millions)
243. 0
6.O
13.6
%of
Total
93%
2%
5%
Data on the population served by types of public water
systems come from "The National Public Water System
Supervision Program" (EPA-812-R-95-001, July 1995).
2.1.1 Source and Ownership
The water industry consists of a
diverse group of water suppliers. They
range from large municipal systems
that use primarily surface water
sources and serve several million
persons, to mobile home parks,
schools, and institutions that use
primarily ground water sources and
serve fixed populations of as few as 25
persons at least 6 months of the year.
The results of the 1995 CWS Survey
confirm this diversity. Almost 80
-------
1995 Community Water System Survey
percent of community water systems
use primarily ground water (i.e.,
ground water makes up the largest
portion of the system's total produc-
tion). Less than 10 percent of systems
surveyed primarily use surface water
sources, and systems that primarily
use purchased water, either raw or
finished, account for approximately 11
percent of the national total.
In the table below, we show commu-
nity water systems by source of water
and by ownership type (publicly
owned, privately owned, or ancillary1).
While systems that use primarily
ground water are rather evenly divided
by source, publicly owned systems are
predominant in both primarily surface
and primarily purchased sources. (See
Community Water Systems By Source and Ownership Type
Total Drinking Water Systems
Primarily
Ground Water
Primarily
Surface Water
Primarily
Purchased
40,123
4,832
5,334
79.8%
9.6%
10.6%
Primarily Ground Water
Public
Private
Ancillary
14,321
14,168
11,634
35.7%
35.3%
29.0%
Primarily Surface Water
Public
Private
Ancillary
3,641
957
234
75.4%
19.8%
4.8%
Primarily Purchased
Public
Private
Ancillary
3,827
1,415
92
71.8%
26.5%
1.7%
Ancillary 29.0%
Ancillary 1.7%
Volume II, Table 1-3, for more
detailed data on this subject.)
Source water characteristics are
incorporated into EPA's regulatory
analyses to account for operational
configurations, regulatory require-
ments, and costs that are associated
with different types of water quality
conditions. For example, approxi-
mately 21 percent of systems use
mixed sources (i.e., they do not rely
100 percent on any single source).
In EPA regulatory analyses, water
systems are categorized based on the
size of their service populations to
account for differences in operational
characteristics such as production
requirements, treatment processes in-
place, operators' skills and experience,
as well as technical and financial
capability. Water production typically
demonstrates economies of scale (i.e.,
declining average costs of production
as volume increases) because of the
large capital investments required.
Water system size is an increasingly
'In an "ancillary" system, providing potable water
is not the entity's primary business. Instead, these
systems provide water as an ancillary function of
their principal business or enterprise. Examples
include mobile home parks, schools, hospitals,
prisons, and commercial businesses.
-------
Chapter 2—National Projections Summary
important characteristic of the industry
for the Agency to consider in its
development of regulations.
Recognizing differences among
systems based on size, Congress, in
the reauthorized Safe Drinking Water
Act (SDWA), directed the Agency to
provide regulatory flexibility to
"small" community water systems,
i.e., those serving populations of 3,300
or fewer persons. Eighty-five percent
of community water systems fit this
small system designation. Only 2
percent of community water systems
are considered large systems, i.e.,
serving populations of more than
50,000 persons.
Community Water Systems
by Size
Medium 13%
targe 2%
Small 85%
In addition to size, issues related to
how water system ownership is
structured are important when EPA
estimates the financial impact that
drinking water regulations may have
on the water industry. Publicly and
privately owned systems differ in rate
structures, accounting practices, and
their ability to raise capital.
The results of the CWS Survey
indicate that 43 percent of community
water systems are publicly owned.
This group comprises water systems
that are owned by municipalities,
townships, counties, water districts,
and water authorities. The survey also
shows that 33 percent of all commu-
nity water systems are privately
owned. Private ownership encom-
passes a broad range of owners, from
homeowners associations to investor-
owned water companies.
Number & Percentage of Systems
by Type of Ownership
Privately Owned 33%
(16,540)
Publicly Owned 43%
(21,789)
Ancillary Systems 24%
(11,960)
Privately Owned Systems
Investor-Owned
Dependent on
Parent Company
41.2%
Homeowner's Association
34.6%
Investor-Owned
Not Dependent on
Parent Company
8.4%
Other
15.8%
The remaining 24 percent of commu-
nity water systems are classified by
EPA as ancillary2 systems, all of
which serve populations of 3,300 or
fewer persons. These systems, as
explained above, are not typical water
utilities; they provide water as an
ancillary function of their principal
business or enterprise. Often they
provide water as a convenience to
their patrons, employees, or residents.
Compared to publicly and privately
owned systems, ancillary systems
serve smaller populations, produce
smaller flows, have limited operator
capability (i.e., no full-time, certified
operators), and do not bill customers
or users directly for the water service.
2In EPA's data system on public water systems,
ancillary systems are a sub-set of privately owned
systems. For purposes of this report, we have
defined them as a separate category.
-------
1995 Community Water System Survey
The largest category of ancillary
systems is mobile home parks.
Ancillary Systems
Mot** Horn* P«*»xn
87S Jlffl
The vast majority of water system
customers—about 86 percent—get
their water from publicly owned
systems.
Percent of Population Served
by Type of Ownership
Public 86%
Prtv««« 13% AndHiiy 1%
2.2 Customer Profile
Community water systems serve more
than 75 million customer connections,
representing a service population of
approximately 245 million persons.
Residential connections comprise 87
percent of total connections (of those
systems that were classified by type of
connection in the survey).
CWS Connections
Nonresidential 13%
Residential 87%
The remaining 13 percent of connec-
tions are classified as nonresidential—
primarily commercial and industrial
(including multi-family dwellings and
apartment complexes), governmental,
wholesale, and agricultural customers.3
Large systems (primarily publicly
owned) provide potable water to over
half of all customers served by com-
munity water systems. In fact, the
1995 CWS Survey confirms previous
industry surveys' conclusions that
large systems account for a tiny
'Nonresidential customers, for this summary, also
include some residential connections since some
respondents were not able to provide detail on
connections by customer class.
percentage of systems, but more than
50 percent of the connections served.
Community Water Systems
Connections vs. Systems
Conversely, small and very small
systems account for roughly 85
percent of systems, but only 12
percent of customer connections
served.
Total water consumption (i.e., deliver-
ies) in community water systems is
estimated to be approximately 37
billion gallons per day.4 Deliveries to
residential and nonresidential custom-
ers represent 47 percent and 53
percent, respectively, of total con-
sumption.
'This may under-estimate total national deliveries
because of item nonresponse on this survey
question.
8
-------
Chapter 2—National Projections Summary
Water Deliveries by
Customer Type
Residential 47%
Nonresidential 53%
Of the nonresidential deliveries where
respondents could specify the type of
customer, the largest customer cat-
egory was commercial (43.4 percent),
followed by wholesale (26.5 percent),
and industrial (20.8 percent), as shown
in the pie chart below. Wholesale
customers often are other water
systems that may supply residential
customers.
Water Deliveries to Nonresidential
Customers (Excluding Other)
Commercial 43.4%
Industrial 20.8%
Agricultural 1.2%
Other Gov't 3.3%
Local Qov't 4.8%
Wholesale 26.5%
Deliveries per connection for all
community water systems totaled
approximately 547 gallons per day
(approximately 200 thousand gallons
annually). Residential deliveries per
connection totaled 295 gallons per
day, compared to 2,174 gallons per
day to nonresidential connections.
Even though the gallons delivered to
each residential connection are fewer
than the gallons delivered to nonresi-
dential connection, residential connec-
tions make up the majority of all
connections.
Residential consumption per connec-
tion translates to 107 thousand gallons
annually, which confirms assumptions
in EPA RIAs that the average house-
hold consumes approximately 100
thousand gallons per year. Annual
consumption per nonresidential
connection totaled 793 thousand
gallons. (See Volume II, Table 1-14,
for more detailed data on this subject.)
1.000,000
Average Annual Water Use
by Consumer Type
Nonresidential
2.3 Operational Summary
The 1995 CWS Survey database
enables the Agency to identify differ-
ences in operational conditions among
water systems and to develop an up-
to-date characterization of baseline
conditions at water treatment facilities
throughout the industry. In particular,
data were collected on facility opera-
tional parameters such as sources/
intakes, treatment processes in-place,
production capacity, storage and
distribution composition, and operator
skills and training. Specification of
these operational conditions provides
the Agency with a baseline from
which to estimate the incremental
impact of facility compliance with
SDWA regulations. A brief summary
of operational characteristics is
provided below.
-------
1995 Community Water System Survey
2.3.1 Industry Production
Total water production from all
community water system facilities
averaged 43 billion gallons per day.
Systems using primarily surface water
sources accounted for 48 percent of
industry production, or, about 21
billion gallons per day. Publicly
owned systems accounted for the
majority of water production (88
percent); most large surface water
systems are publicly owned.
Water Production by Ownership
ftfvattty Owned 12%
(1.9 Trillion Giftoni/yaw)
Publicly Owned 88%
(13.9 Trillion Gallons/Year)
To process these volumes of water,
community water systems employed
an estimated 91,944 operators,
approximately 70 percent of whom
were employed full time by water
systems. Thirty-eight percent of
systems employed at least one full-
time operator. (See Volume II, Table
10
1-15, for more detailed data on this
subject.)
Drinking Water System
Operators
Part Time 30%
Full Time 70%
Seventy-six percent of all operators
were state certified in water system
operations, and an additional 10
percent had received some formal
training through a national or state
program, but were not fully state
certified. Seventy percent of part-time
operators were employed in small
systems. (See Volume II, Table 1-16,
for more detailed data on this subject.)
System Operator Background
Some Training 10%
Untrained 14%
State Certified 76%
Total industry production per em-
ployed operator (full and part time)
was 484 thousand gallons per day for
all community water systems. Produc-
tion per operator per day by very small
systems (i.e., serving fewer than 500
persons) totaled 23 thousand gallons,
compared to 2.4 million gallons per
operator per day for large systems.
2.3.2 Treatment and
Distribution
Eighty-one percent of community
water systems reported performing
some treatment on all or a portion of
their water sources. To put this result
in perspective, the survey estimates
that over 99 percent of systems using
surface water sources performed some
treatment of their source water. Of
those systems reporting no treatment,
80 percent rely on ground water as
their only source.
The types of treatments applied vary
according to type of water source. For
example, of systems that answered the
survey questions on types of treat-
ment, more than twice as many
exclusively ground water systems
apply iron and manganese removal
and aeration as do exclusively surface
-------
Chapter 2—National Projections Summary
water systems. The exclusively
surface water systems that replied to
the questionnaire apply flocculation/
coagulation and filtration more than
twice as often as do the ground water
only systems.
The reader should use caution in
interpreting some of the percentages in
the following table. It shows the
percentage of systems reporting each
type of treatment at any treatment
facility. If one looked at these data by
facility, instead of by system, the
percentages would be lower.
Percentage of Systems
Applying Various Treatments
at One or More Treatment
Facilities
Aeration
Disinfection/
Oxidation
Iron and
Manganese
Removal
Flocculation/
Coagulation
Filtration
Organics Removal
Corrosion Control
Other
100%
Ground
Water
33%
92%
34%
33%
39%
27%
36%
28%
100%
Surface
Water
12%
99%
14%
84%
89%
23%
62%
12%
In addition to the SDWA source water
treatment requirements, water systems
are confronting increasing costs for
replacement and repair of existing
infrastructure, such as storage facili-
ties and distribution systems. For
many water systems, distribution
system replacement and repair are
long overdue. In other systems, repairs
are being accelerated in conjunction
with treatment installations or up-
grades. In either situation, as the cost
of providing high-quality water
increases, water systems and custom-
ers may be more concerned about the
significant portion of production lost
or adversely affected by poor distribu-
tion systems.
To achieve greater insight into the
condition of distribution systems, the
1995 CWS Survey collected detailed
information on the current size and
composition of distribution piping and
maintenance. Community water
systems maintain approximately 29
million miles of pipe. (See Volume II,
Tables 1-10,1-11, and 1-12, for more
detailed data on this subject.)
Community water systems reported
expanding their distribution systems
by less than 1 percent (113,265 miles)
from 1990 through 1994. Community
water systems also replaced 1.0
million miles of pipe and performed
over 360 thousand main repairs.
2.3.3 Source Water
Protection
Source water protection efforts can be
a low-cost option for many water
systems to protect sources from
contamination. Water systems whose
sources are vulnerable to contamina-
tion can reduce future capital expendi-
tures for treatment plants and equip-
ment by adopting "best management
practices" and land use controls to
prevent or reduce the probability of
contamination.
Survey results indicate that over one-
third of all community water systems
participate in some type of source
water protection effort. The three
primary methods of source protection
identified by these respondents were
zoning or land use controls, best
management practices, and education
on land use impacts.
About 80 percent of the systems that
participated in source water protection
efforts reported using these three
11
-------
1995 Community Water System Survey
methods. These efforts, however, may
or may not meet the requirements for
state source water protection pro-
grams. A greater proportion of large
systems (45 percent) reported partici-
pation in source protection programs
than did small systems (32 percent).
Source and ownership appear to be
less significant in determining partici-
pation in source water protection
efforts. An equal percentage of sys-
tems also participate in source water
protection efforts managed by a local
or state governmental agency, but
small systems report greater involve-
ment with source water protection
efforts managed by a state agency.
(See Volume II, Table 1-24, for more
detailed data on this subject.)
Percentage of Systems
Participating in Source Water
Protection
Snufl Large
System Size
Water systems were asked to identify
potential sources of contamination
within 2 miles of their water supply
intakes or wells. Overall, the most
frequently identified sources of
contamination were septic systems
(cited by 79 percent of systems) and
agricultural runoff (reported by 55
percent of systems). Other potential
sources of contamination that were
identified included petroleum products
(e.g., fuel and heating oil tanks) in 38
percent of systems; urban runoff in 31
percent of systems; and sewage
discharge in 27 percent of systems.
(See Volume II, Table 1-28, for more
detailed data on this subject.)
Percentage of Systems Citing
Water Contamination Sources
^Septic Systems Petroleum Products Sewage Discharge
Agricultural Runoff Urban Runoff
Water systems were asked to indicate
their laboratory analysis provider.
Overall, approximately 40 to 50
percent of systems identified the state
or a private firm as their primary
laboratory analysis provider. A higher
percentage of large water systems
reported using in-house laboratory
services to analyze water for metals,
inorganic chemicals, and microbial
contaminants.
2.4 Financial Summary
Water systems were asked to provide
basic financial information from their
income statements and balance sheets.
EPA requires an accurate baseline of
the financial characteristics of commu-
nity water systems to forecast the
ability of these systems to make the
technical and capital investments
required for sustainable water opera-
tions.
The summary below provides an
aggregate profile of the water industry
in 1995. Given that the survey results
represent only one year of financial
performance, EPA's ability to derive
conclusions about the financial health
of the water industry is limited.
Instead, the data are presented as a
"snapshot" of the industry in 1995. A
variety of financial characteristics
must be examined over time to fully
assess the ability of water systems to
sustain an adequate and safe supply of
water.
12
-------
Chapter 2—National Projections Summary
2.4.1 Summary of Revenues
and Expenses
Most water industry revenues are
generated directly through the sale of
water. Water rates are the primary
mechanism by which customers are
charged for service. Systems may also
generate revenues from fees (e.g.,
connection or inspection fees), fines
and penalties, and other non-consump-
tion-based charges. Publicly owned
water systems may receive contribu-
tions from a municipal general fund.
Water industry revenues from all
sources for publicly and privately
owned systems were estimated to total
$25.9 billion, most of which was
derived from water sales. Publicly
owned systems accounted for 86
percent ($22.2 billion) of total industry
revenues, compared to 14 percent
($3.7 billion) for privately owned
systems.
Water System Annual
Revenues and Expenses
Publicly Owned
Water Systems
Privately Owned
Water Systems
Total
Annual
Revenue
(§Billion)
22.2
3.7
25.9
Annual
Expenses
(^Billion)
19.O
3.1
22.1
The vast majority of water-related
revenues come directly from water
sales. As shown in the pie charts that
follow, this is particularly true for
privately owned systems, which derive
92 percent of their revenues from
water sales. Publicly owned systems,
obtain 86 percent of their revenues
from water sales. The remainder
consists of other types of water-related
revenues (e.g., connection fees,
inspection fees, and interest earnings).
(See Volume II, Table 1-35, for more
detailed data on this subject.)
Water-Related Revenues of
Privately Owned Systems
Other Water-Related 8%
Water Sales 92%
Water-Related Revenues of
Publicly Owned Systems
Other Water-Related 14%
Water Sales 86%
Across all system size categories, sales
to residential customers accounted for
54.8 percent of total water sales.
Revenues from water sales to commer-
cial and industrial customers totaled
18.3 percent of total water sales, while
wholesale customers and other cus-
tomer categories accounted for 4.1
percent and 22.8 percent respectively.
The percentage of water sales rev-
enues from residential customers is
highest in small systems, as shown in
the table below. The category "other"
included governmental customers,
13
-------
1995 Community Water System Survey
Sources of System Water Sales Revenue by Type of Customer
(Excluding Ancillary Systems)
Customer Type
Residential
Commercial/Industrial
Wholesale
Other
Total
System Size
<600
70.5%
8.1%
5.4%
16.0%
100.0%
601-3,300
61.7%
12.3%
1.8%
24.2%
100.0%
3,301-50,000
52.4%
19.9%
3.0%
24.7%
100.0%
agricultural customers, categories not
defined in the survey, and total sales
revenues for systems that did not
disaggregate by customer type.
Nonresidential connections generated
about 5 times more revenue per
connection, but delivered 7 times more
water than did residential connections.
(See the graph, "Average Annual
Water Use by Consumer Type," on
page 9.) Specifically, nonresidential
water sales per connection totalled
$1,177, compared to sales revenues
per residential connection of $218.
Annual Water Sales per
Connection, by Customer Type
11,400
11,000
UM
MOO
WOO
«CO
so
ResManUal
NonrosWenUal
Residential customers pay a higher
rate, however, $2.94 per thousand
gallons, compared with $2.51 for
>so,ooo
54.5%
20.2%
5.4%
19.9%
100.0%
Water Sales Revenue by
Customer Type
commercial custom-
ers and $1.70 for
wholesale custom-
ers.5 Rates generally
are higher for all
customer categories
in small systems, as
shown in the chart
"Water Sales Revenue by Customer
Type."
Total expenses for all community
water systems were $22.1 billion. (See
the table "Water System Annual
Revenues and Expenses" on page 13.)
Operations and maintenance (O&M)
expenses, which include all direct
To calculate the residential customer rate from the
preceding paragraph, one cannot simply divide the
national estimate of average annual water sales per
residential connection by the national estimate of
residential consumption per connection. While data
for all three variables came from the first line of
survey question 29, each was calculated indepen-
dently to provide the greatest precision in each
estimate. The estimate of average annual water
sales came from all respondents who provided
"water sales revenue" data on line 1 of question 29.
The estimate of residential consumption came from
respondents who provided "gallons delivered" on
line 1 of question 29. The average residential
customer rate is derived from information provided
by all respondents who provided both "water sales
revenue" and "gallons delivered."
1 Very Small
Residential* 305
Comm/lnd n 289
Wholesale H 160
Small
301
262
156
Medium
270
217
190
Large
205
174
132
costs of production such as labor,
materials, chemicals, electricity, taxes
and payments in lieu of taxes, ac-
counted for $16.9 billion, or 76.5
percent of total industry expenses. Of
the water industry's remaining operat-
ing expenses, depreciation expenses
were $2.4 billion, interest expenses
were $2.7 billion, and "other" ex-
penses were $0.1 billion.
Water Industry Expenses
Depreciation 10.8%
Other 0.5%
Interest Expense 12.2%
O&M 76.5%
14
-------
Chapter 2—National Projections Summary
As shown in the graph below, ex-
penses for publicly owned systems
totaled $19 billion, (86 percent of total
industry expenses). Private system
expenses totaled $3.1 billion (14
percent).
Water System Expenses
by Type of Ownership
Privately Owned 14% ($3.1 B)
Publicly Owned 86% ($19.08)
As previously mentioned, water rates
are the method by which most water
systems attempt to recover operating
expenses. Meters allow water systems
to monitor customer demand and to
establish charges based on usage. The
CWS Survey estimates that 95 percent
of residential and 98 percent of
nonresidential connections were
metered industry-wide.
Metered Connections by
Customer Type
Residential Connections
Nonresidential Connections
95%
98%
Of all the rate structures that water
systems use for their residential
customers, the uniform block rate is
the most common (49 percent).
Declining block rates and separate flat
fees are next (16 percent and 15
percent, respectively). Note that rate
structures other than those explicitly
included in the CWS survey accounted
for about 8 percent of all the rate
structures applied. (See Volume II,
Table 1-43, for more detailed data on
this subject.)
Percentage Use of Various
Residential Rate Structures
Uniform Rate
Declining Block
Increasing Block
Peak Period
Separate Flat Fee
Combined Flat Fee
Other
49.0%
16.0%
1 1 .0%
0.9%
15.3%
10.0%
8.2%
Note: The percentages in this table do not
total 1 00 percent because some systems
used more than one rate structure.
2.4.2 Industry Balance Sheet
Community water systems reported
assets totaling approximately $132
billion in 1995. Total liabilities were
$62 billion. Survey respondents were
not asked to provide detail on assets
by major system component (e.g.,
treatment plant or distribution sys-
tems).
Water System Annua|l
Revenues and Expenses
Publicly Owned
Water Systems
Privately Owned
Water Systems
Total
Assets
($ Billion)
117.8
14.1
131.9
Liabilities
($Biilion)
56.2
5.7
61.9
Consistent with previous surveys of
water systems, the water industry can
be characterized as very capital
intensive. The results of the 1995
survey indicate that water systems
overall maintain about $5 of gross
assets for every $1 of revenue. This
ratio is highest in the smallest system
size category. (See Volume II, Tables
1-56, 1-57, 1-58, and 1-59, for more
detailed data on this subject.)
15
-------
1995 Community Water System Survey
Asset-to-Revenue Ratio
(Excluding Ancillary Systems)
System Size Category
<500
B01 -3,300
3,301-50.000
> 50,000
Ratio
6.3
4.5
4.9
5.3
This high asset-to-revenue ratio
reflects, among other things, the high
capital investment of water utilities.
The asset-to-revenue ratio for inves-
tor-owned electric utilities and gas
utilities, by comparison, was about $3
to $1 in 1994, according to the Energy
Information Administration and
American Gas Association.
16
-------
3. Trends and Key Findings
Y • Ihis chapter summarizes the
I most important trends and
-*- findings from the responses to
the 1995 CWS Survey. The chapter is
organized into two parts. The first part
covers a few trends derived from
comparing this survey with previous
surveys. The second part of this
chapter describes key findings. In both
parts, we present the operational
characteristics first, followed by
financial characteristics.
There were three previous CWS
surveys (1976,1982, and 1986),
providing almost 20 years of experi-
ence for comparative purposes. We are
particularly interested in changes since
the 1986 survey because results from
that survey are used as baseline
characteristics in EPA's cost models.
3.1 Trends
3.1.1 Operating
Characteristics
As Chapter 2 showed, most of the
operating characteristics of commu-
nity water systems are the same as
those in 1976. Most systems are small.
Most small systems are privately
Percentage of Systems
Not Providing Treatment
SYSTEM SIZE
1976 1
1982 •
1986 C
1995 H
0-100
67%
56%
56%
31%
101-500
58%
50%
40%
16%
501-1,000
45%
33%
28%
11%
1,001-3,300
34%
33%
12%
3,301-10,000
23%
20%
5%
owned or ancillary systems. Most
people, however, are customers of
large publicly owned systems. Never-
theless, there has been one important
change.
The percentage of systems that do
not treat their water has steadily
declined from 1976 to 1995. Differ-
ences in questionnaires preclude trend
analysis for all size categories, but we
have comparable data for the five
smallest sizes. These categories are of
interest, however, because most large
systems provide some treatment. As
shown in the graph above, the percent-
age of small and medium systems not
providing treatment has fallen steadily
since the SDWA was enacted in 1974.
This is consistent with the SDWA's
emphasis on water quality monitoring
and treatment.
The importance of this change is that
treatment in place is a significant
variable in calculating the cost of
compliance for RIAs. In the models
used to estimate compliance costs,
systems with treatment in place do not
need to invest in additional treatment
(provided, of course, that the treatment
in place is capable of meeting regula-
tory objectives). (See Volume II,
Table 1-18, for more detailed data on
this subject.)
3.12 Financial
Characteristics
Water sales revenues have increased
since 1986 (in constant dollars) for
systems in most size categories.
These increases reflect increased
investment in fixed assets and in-
creased costs of operation and mainte-
17
-------
1995 Community Water System Survey
nance (O&M). (See the table "Trends
in Water Sales Revenues.") Although
we cannot do a one-to-one comparison
of the larger size categories, it is clear
that there have been significant
increases in all systems serving more
than 50,000 people.1 For example,
water sales revenues for systems
serving 50,001 to 100,000 people are
220 cents per thousand gallons of
water delivered, significantly higher
than the two corresponding size
categories in the 1986 survey. Simi-
larly, systems serving more than
100,000 people in 1995 posted higher
water sales revenues than did all
comparable size categories for 1986.
(See Volume II, Table 1-39, for more
detailed data on this subject.)
1 In the size categories where we could not do a
one-to-one comparison, the percent change for a
given size category in the 1995 survey was
calculated by comparing the 1995 value with the
mean value of the corresponding two or three
categories from the 1986 survey.
18
Trends in Water Sales Revenues
(Cents per 1,000 Gallons Sold)
System Size
<100
101-500
501-1,000
1,001-3,300
3,301-10,000
10,001-50,000
50,001-100,000
> 100,000
1995
345
342
356
295
293
240
220
189
Percentage
Change
from 1986
(in 1995$)
25.09%
1 .30%
38.95%
3.87%
40.87%
17.33%
49.25%
31.80%
During the past decade, water rates
have increased faster than the
Consumer Price Index (CPI). The
CWS Survey examined rate increases
by system size. These data show that
large systems increase rates more
frequently than do small systems. This
is a well-established pattern in the
literature on the water industry. Large
systems have substantial resources to
plan and implement regular rate
increases. Small systems tend to wait
longer between increases, but seek
larger increases when they do so.
The lag between rate increases may be
an important factor explaining the
weaknesses in many small systems'
financial conditions, as measured by
financial ratios. For any given year,
the financial health of many small
systems may not be good because
these systems have not yet increased
rates.
The table below shows the size of the
most recent rate increases for systems
in the survey. Rate increases were
annualized by dividing the size of the
increase by the number of years
between increases. The data show
annual increases ranging from 4.9
percent to 14.8 percent. This rate of
increase is larger than the CPFs,
which has been approximately 3
percent for the past several years. (See
Volume II, Table 1-44, for more
detailed data on this subject.)
Residential Rate Increase
Profile
System Size
<100
101-500
501-1,000
1,001-3,300
3,301-10,000
10,001-50,000
50,001-100,000
> 100,000
Annualized
Percentage
Increase
14.80%
5.79%
8.62%
7.37%
4.85%
5.52%
7.22%
7.50%
-------
Chapter 3—Trends and Key Findings
The CWS Survey results are consistent
with analyses conducted by Professor
Janice A. Beecher of Indiana Univer-
sity on the pattern of rate increases for
water and sewer systems over the past
20 years.
The graph below, prepared by Profes-
sor Beecher, shows that increases in
water and sewer prices were slightly
below the increase in the CPI from
1970 to 1984, but then began to
steadily outstrip the CPI increase.
These increases are even more signifi-
cant when compared with those of
other utilities (telephone, piped gas,
electricity) which had substantially
lower rates of increase.
In spite of substantial rate increases,
it appears that many water systems
in all size categories still are not
raising enough revenue. Indeed, the
1995 CWS Survey confirms the
findings of the three previous
surveys: many systems have costs
that exceed revenues. A substantial
percentage of systems participating in
Consumer Price Indexes for Utilities
1970 to 1995
(1982-1984=100)
1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994
Cable television Water & Sewer CPI
Electricity Piped gas Interstate telephone
—-X-—
the 1995 CWS Survey reported a
revenue deficit (if publicly owned) or
a loss (if privately owned).
Percentage of Systems With
Deficit (Loss)
501-3,300 3,301-60,000 > 50,000
System size
As shown above, the percentage of
systems in the survey reporting
deficits (or losses) decreases with
system size. The overall percentage
is about the same as it was in the
1976 survey (the last time this
characteristic was measured). The
percentage of systems with deficits
(or losses) was highest in the small-
est system size categories, where it
was approximately 40 percent. But
the percentage of systems with
deficits (or losses) declined as the
system size increased. A higher
percentage of publicly owned
systems had deficits than privately
owned systems had losses.
19
-------
1995 Community Water System Survey
In theory, privately owned systems
with repeated losses cannot stay in
business, and both publicly owned
systems with deficits and privately
owned systems with losses should
have more difficulty in securing
capital for infrastructure investments.
Additional information on the finan-
cial health of water systems, including
their ability to borrow, is presented in
section 3.2.2.
Before drawing any conclusions from
these data about the overall financial
health of the industry, we offer some
caveats:
• The survey's estimates of deficit or
surplus come from a single year's
financial data. As noted above,
water utilities often face temporary
deficits while waiting for the imple-
mentation of higher rates. During the
QA process on the survey, analysts
telephoned all large systems show-
ing a deficit.2 All data were con-
firmed, and several respondents
indicated that lagging rate increases
accounted for the temporary deficit.
Other respondents said that the
current year's loss (or deficit)
reflected the need to reduce large
reserve fund balances.
20
Combined systems (e.g., water and
sewer) may have had difficulty
disaggregating their operating
expenses. Many combined utilities
track sales revenue for each opera-
tion separately, but combine operat-
ing expenses. Telephone queries to
survey respondents indicated that
many used simple decision rules to
disaggregate their operating ex-
penses (e.g., applying the proportion
of total sales represented by water
sales to operating expenditures).
Other respondents indicated that
they disaggregated "big ticket"
items, but did not remove non-water
expenditures from less significant
line items. The cumulative effect of
these factors likely may be to
overestimate the percentage of
systems in deficit.
Our questionnaire asked respondents
to identify "water sales" and "water-
related" revenues. The second
category was defined broadly, but
some systems may not have reported
revenues that should have been
classified as "water related." If
2 Analysts telephoned all publicly owned systems
serving more than 50,000 persons and all privately
owned systems serving more than 10,000 persons to
confirm survey responses.
reported, these revenues would have
improved their financial position.
The comparatively small percentage
of large, privately owned systems
with losses may reflect the reliance
of these systems on equity capital.
Profits are needed to pay dividends
to stake holders.
It is important to note that some
systems, particularly small systems,
are technically insolvent. This is
consistent with the findings of other
studies of such systems.3 For very
small systems, there is a thin line
between solvency and insolvency. In
a homeowners association serving
100 people, for example, temporary
insolvency can be resolved by a
small assessment on all customers.
Observers have noted that water
systems may have negative net
income on their income statements,
but positive cash flow. Depreciation
3 See, e.g., Dreese, G. R. and Beecher, J.A.
"Financial Distress Models for Small Water
Utilities," Proceedings of the Eighth NARUC
Biennial Regulatory Information Conference, IV:
175-95 (Columbus, OH 1992). See also Cromwell,
J. E. and Rubin, S. J., "Development of Benchmark
Measures for Viability Assessment," Report for the
Pennsylvania Department of Environmental
Protection, 1996.
-------
Chapter 3—Trends and Key Findings
is counted as an expense but requires
no cash outlay.4
3.2 Key Findings
3.2.1 Operating
Characteristics
Production and distribution of drink-
ing water are essential functions of
most community water systems. As
with all characteristics of water
systems, of course, there is much
diversity in both production and
distribution. Some systems purchase
finished water and have little or no
production function. Other systems
produce finished water and act solely
as wholesale distributors with few (if
any) retail customers. A summary of
production and storage data from the
1995 survey appears on this page.
One interesting item in this table is the
ratio of maximum daily treatment to
peak daily production. For both
ground water and surface water
4 This issue is discussed in a recent article by
Professor Jeffrey L. Jordan, using data from
financial statements of all Georgia water utilities.
See "Do You Use Your Depreciation Funds
Wisely," Opflow, Vol. 21, No. 12 (December
1995), p. 1.
Summary of Production and Storage
Primarily Ground Water
System Size
<500
501-3.300 3,301-50,000 > 50,000
Average Daily Production (Gallons)
Peak Daily Production (Gallons)
Maximum Daily Treatment Capacity (Gallons)
Ratio of Maximum Daily Treatment to Peak Daily Production
Storage Capacity (Million Gallons)
Average Daily Deliveries (Gallons)
Ratio of Average Daily Deliveries to Average Daily Production
17,148
46,320
152,383
4.66
0.03
15,170
0.97
198,011
403,979
880,078
2.59
0.26
164,804
0.89
1,892,761
3,597,039
5,560,389
1.62
2.62
1,525,106
0.87
25.238,287
47,813,370
58,971,239
1.40
20.92
21,123,817
0.89
Primarily Surface Water
Average Daily Production (Gallons)
Peak Daily Production (Gallons)
Maximum Daily Treatment Capacity (Gallons)
Ratio of Maximum Daily Treatment to Peak Daily Production
Storage Capacity (Million Gallons)
Average Daily Deliveries (Gallons)
Ratio of Average Daily Deliveries to Average Daily Production
45,963
160,758
304,872
2.28
0.23
18,871
0.91
278,479
624,576
1,527,963
1.83
0.60
229,110
0.86
2,677,060
4,389,824
6,409,019
1.57
4.00
2,228,237
0.87
38,099,428
56,606,417
79,023,967
1.38
149.83
32,402,153
0.87
Readers should note that the ratios in this table—the ratio of maximum daily treatment to peak daily production and the
ratio of average daily deliveries to average daily production—cannot be computed from other data in the table. For .
example, dividing maximum daily treatment (as shown in the table) by peak daily production (as shown in the table) will
not yield the ratio shown in the table. Each mean value in the table was calculated independently to provide the
greatest precision in each estimate. The mean maximum daily treatment comes from all respondents who answered
that question; the peak daily production comes from all respondents who answered that question; and the ratio is
derived from information provided by all respondents who answered both questions.
systems, the ratio declines with system
size. Previous CWS Surveys have
concluded that this trend is related
exclusively to system size—large
systems have more efficient opera-
tions. The data show, however, that
the ratio of maximum daily treat-
ment to peak daily production is
related not only to system size, but
also to the source of unfinished
water. Size matters, but so does
source.
The ratio of maximum daily treatment
to peak daily production appears to
reflect the treatment and storage
conditions associated with different
sources of water. Ground water
systems generally rely on, and can
afford, additional pumping and
treatment capacity to meet peak
demands. Surface water systems, in
contrast, generally use more capital-
intensive treatment techniques and
tend to rely on storage facilities to
meet peak momentary and hourly
demands.
As expected, the ratios for ground
water systems and surface water
systems tend to converge as system
size increases. The smallest ground
water systems often use hydro-
pneumatic tanks which, practically
speaking, are designed to meet mo-
mentary and hourly demands through
source pumping rather than storage.
Large ground water systems, however,
tend to have more sophisticated and
capital-intensive treatment processes.
Like their surface water counterparts,
they rely more on storage to meet peak
demands.
21
-------
1995 Community Water System Survey
Notwithstanding the relationship
between these ratios and source, the
table also shows that the ratios for
both surface water and ground water
systems are inversely related to system
size. This suggests that, to some
extent, large systems have a more
sophisticated understanding of the
fluctuations in demand and have sized
production and storage facilities to
account for them.
The other ratio in the table "Summary
of Production and Storage" compares
average daily deliveries with average
daily production. It shows that ap-
proximately 10 percent of the water
produced does not get delivered and
therefore results in "uncompensated
usage." There are many reasons for
this phenomenon, but one of the most
important is losses from the distribu-
tion system because of leaks. (For
more information on production and
storage, see Volume II, Tables 1-2,1-
4, and 1-8.)
The relationship between size and
production efficiency is examined
again in the table "Ratio of Peak Daily
Production to Average Daily Produc-
tion." This table compares the average
ratio of peak daily production to
22
average daily production, and shows
that the ratios are inversely related to
system size.
System Size
<100
101-600
501-1,000
1,001-3,300
3,301-10,000
1 0,001 -50.000
B0.001 -100.000
> 100,000
Publicly Owned Systems
Mem
2.36
2.47
3.18
1.98
2.33
1.66
1.69
1.60
Median
1.59
1.82
1.89
1.61
1.64
1.53
1.66
1.61
Privately Owned Systems
Mean
3.56
2.17
1.69
2.34
1.90
1.74
1.47
1.61
Median
1.44
1.72
1.55
1.57
1.62
1.48
1.44
1.54
Small systems, on average, have a
higher ratio of peak daily production
to average daily production. This may
reflect the comparatively large fluctua-
tions in demand that occur in small
systems. In a small system, changes in
consumption by a few households can
have a significant impact on demand.
In large systems, with larger and more
predictable commercial and industrial
customers, there may be less variance
in demand. (See Volume II, Tables 1-6
and 1-7, for more detailed data on this
subject.)
System Size
501-1,000
1,001-3.300
3,301-10,000
10,001-50,000
50,001-100,000
> 100,000
83%
61%
Mean
95%
50%
The table above shows that the small-
est systems typically serve primarily
residential customers. For example,
the median values show that at least
half of all privately owned systems
serving fewer than 1,000 people serve
residential customers exclusively. As
system size increases, other customers
become more significant. These
include commercial, industrial,
governmental, wholesale, and agricul-
tural customers. The ratio of residen-
tial deliveries to total deliveries is
highest for systems serving fewer than
100 persons (almost 100 percent), and
it declines to almost 50 percent for
systems serving more than 100,000
persons. (See Volume II, Table 1-14,
for more detailed data on this subject.)
-------
Chapter 3—Trends and Key Findings
Average Annual Water Delivered per Connection (Thousand Gallons)
System Size
Customer Category
Comm erica)/
Residential Industrial Wholesale Governmental Agricultural
Publidy Owned Systems
<100
1,061
7.703
9.919
22,483
System Size
<100
101-500
501-1,000
1,001-3,300
3,301-10,000
10,001-50,000
50,001-100,000
> 100,000
PubHdy Owned Systems
Mean
95%
68%
84%
75%
73%
65%
67%
59%
Median
100%
91%
89%
79%
75.%
67%
67%
59%
Privately Owned Systems
Mean
99%
93%
92%
88%
84%
74%
59%
60%
Median
100%
100%
100%
96%
91%
71%
61%
62%
When residential sales are compared
with total sales, the same general
pattern emerges. As shown in the table
above, the percentage of total sales
that are residential declines as the size
of system increases. When we com-
pare sales with deliveries, a slightly
different pattern emerges. In the
smallest systems, residential sales are
a lower percentage of total sales than
residential deliveries are a percentage
of total deliveries. At the largest size
categories, however, this pattern is
reversed. For public systems serving
more than 100,000 people, residential
sales account for 59 percent of total
sales, but residential deliveries ac-
count for only 53 percent of total
deliveries, suggesting a small subsidy
in the other direction. (See Volume II,
Table 1-36, for more detailed data on
this subject.)
3.301-10,000
10,001-50,000
50,001-100,000
> 100,000
Privately Owned Systems
1,001-3,300
3,301-10,000
10,001-50,000
50,001-100,000
> 100,000
The table titled "Average
Annual Water Delivered per
Connection (Thousands of
Gallons)" shows the annual
distribution of water
delivered per connection.
The range for residential
connections is from 81,000
gallons to 127,000 gallons.
This is consistent with the
traditional assumption of 100,000
gallons per year per connection. The
largest deliveries, as one might expect,
are to nonresidential users. Average
annual deliveries to commercial and
industrial users average as high as 8.1
million gallons. While water use
among many commercial and indus-
trial customers is similar to that of
residential users, there are others (e.g.,
laundries and some manufacturing
operations) where water use is very
high. The average annual deliveries to
governmental customers are even
larger, as high as 34 million gallons
for publicly owned systems serving
50,001 to 100,000 persons. These may
reflect large public entities, some of
which (e.g., hospitals) are intensive
users of water. The largest category is
wholesale customers. Many large
utilities sell finished water to other
water systems, and the average deliv-
126,889
34,482
24,566
3,030
1,488
8,114
25,413
9,264
31,140
343,625
194,286
7,015
1,479
2,026
1,763
1.113
eries can be substantial—632 million
gallons annually for the largest cat-
egory of publicly owned systems. (See
Volume II, Table 1-14, for more
detailed data on this subject.)
Miles of Pipe Replaced as a Percentage of Total WJiles of
Existing Pipe, by Ownership ]
' !
System Size
<100
101-500
501-1,000
1,001-3,300
3,301-10,000
10,001-50,000
50,001-100,000
> 100,000
Publicly Owned Systems
Mean
0.61%
6.27%
3.93%
3.09%
3.12%
1.62%
0.87%
1.16%
Median
0.50%
0.61%
0.63%
0.70%
1.14%
0.67%
0.41%
0.34%
Privately Owned Systems
Mean
1.26%
4.59%
2.94%
0.77%
1.33%
0.52%
0.38%
0.34%
Median
0.25%
1.21%
2.36%
0.19%
0.28%
0.36%
0.18%
0.11%
The table above shows the percentage
of distribution piping replaced per year
as a percent of the total miles of pipe
in the system. The mean values show
substantial variance, particularly for
small systems. The median values
generally show an inverse relationship
23
-------
1995 Community Water System Survey
between the percentage of pipe
replaced and system size. The data for
systems serving more than 10,000
people are consistent with results from
the 1992 survey by the American
Water Works Association (AWWA).5
(See Volume II, Tables 1-10 and 1-11,
for more detailed data on this subject.)
Another characteristic of distribution
systems is the percentage of new pipe
installed each year for expansion. This
is calculated by dividing the miles of
new pipe installed by the total miles of
pipe already in place. The table below
shows that the mean values generally
are between 1 and 2 percent, consis-
tent with the 1992 AWWA survey.6
Miles ol New Pipe for Expansion as a Porcontagc of
Total Miles of ExiMiruj Pipe, by Ownership
System Size
100.000
•ufclcty Owned Systems
Mem
2,33%
1.48%
2.28%
2.18%
1.86%
1.69%
1.25%
1.70%
Median
0.00%
0.00%
0.44%
1.25%
1.07%
1.08%
0.80%
0.85%
Privately Owned Systems
Mean
0.71%
2.13%
1.33%
1.86%
1.50%
6.01%
1.35%
1.16%
Medten
0.00%
0.00%
0.19%
0.58%
0.84%
0.68%
1.09%
1.01%
'American Water Works Association, Water
Industry Data Base, 1992, p. 91.
«/Wrf.,page92.
24
A final statistic on distribution sys-
tems is the population served per mile
of existing pipe. The data are pre-
sented in the next table. Both the mean
and median values show that the
largest systems have substantially
higher populations per mile of pipe
than smaller systems. This is consis-
tent with the fact that large systems,
particularly publicly owned systems,
are in densely populated areas.
Population Served per Mile of Existing Pipe, by Ownership
System Size
<100
101-500
501-1.000
1.001-3,300
3,301-10,000
10.001-50,000
60.001 -100.000
> 100,000
PubBdy Owned Systems
Mean
100
117
172
122
161
255
246
90S
Median
36
70
IIS
96
136
189
237
288
Privately Owned Systems
Mean
151
160
101
102
97
229
187
317
Median
72
52
59
40
56
178
177
260
As we explained during the discussion
of trends in section 3.1, the percentage
of systems providing no additional
treatment has decreased steadily since
1976. The table on the next page
shows the percentage of systems
applying various treatments at one or
more treatment facilities, by water
source, and by system size. (We have
limited the analysis to 100-percent
ground water systems and 100-percent
surface water systems to explore the
effects of source as well as size. Using
data from primarily ground water
systems, for example, could represent
a mixture of surface and ground
sources.)
The data show that treatment configu-
rations become more complex as size
of system increases. In part, this
reflects the fact that very small sys-
tems usually cannot afford the capital
investment or maintenance costs
associated with complex treatment
processes.
For systems with 100 percent ground
water, for example, the most common
treatment was simple disinfection. For
large ground water systems, however,
there is a high percentage of systems
that also install filtration for treat-
ments such as iron and manganese
removal or softening of hard water
using the lime-soda ash process. Large
ground water systems also have more
sources, and only a few of these may
have water quality problems. For
example, the table shows that 77
percent of these systems have organics I
removal, but this may require only one |
treatment facility and may serve one
(or a few) wells.
-------
Chapter 3—Trends and Key Findings
Percentage of Systems Applying Various Treatments at
One or More Treatment Facilities
riean Number of Treatment
Facilities
Treatment Categories
<500
501-3,300 3,301-50,000 > 50.000
100% Ground Water
Percent of Systems Not
'roviding Treatment
1.2
26%
1.6
13%
2.7
5%
6.2
0%
Aeration
Disinfection/Oxidation
Iron and Manganese Removal
=locculation/Coagulation
Filtration
Organics Removal
Corrosion Control
Other
15%
89%
21%
20%
25%
14%
20%
15%
48%
95%
45%
41%
48%
35%
46%
38%
74%
72%
82%
62%
77%
60%
87%
99%
74%
92%
95%
77%
91%
86%
100% Surface Water
Mean Number of Treatment
Facilities
Percent of Systems Not
Providing Treatment
2%
0%
1.1
0%
1.8
0%
Treatment Categories
Aeration
Disinfection/Oxidation
Iron and Manganese Removal
Flocculation/Coagulation
Filtration
Organics Removal
Corrosion Control
Other
1%
98%
1%
48%
75%
3%
23%
10%
100%
10%
85%
82%
10%
49%
11%
97%
15%
99%
93%
37%
70%
12%
59%
100%
63%
100%
96%
79%
87%
62%
Note: The percentage of systems not providing treatment in this table is slightly different
from the percentages reported in the table on page 17. This table reports only on systems
that are either 100 percent ground water or 100 percent surface water. The table on page
17 reports on all systems, regardless of source.
The percentage of surface water systems disinfecting should be 100 percent, at least for
all systems serving more than 501 people. Failure to include disinfection appears to have
been an error on the part of a few respondents.
For surface water systems, a similar
pattern emerges. Small surface water
systems use simple disinfection and
filtration. In large surface water
systems, other types of treatment (e.g.,
organics removal) become more
common.
The reader should note the constraints
we imposed on this analysis. The table
shows the percentage of systems
reporting each type of treatment at any
treatment facility. Thus, if a ground
water system had eight facilities, and
only one
reported
filtration, this
still was
counted as a
report of
filtration at that
system. If one
looked at these
data by facility,
instead of by
system, the
percentages
would be lower.
3.2.2 Financial
Characteristics
As the previous section demonstrated,
size of system matters for several of
the most important operating charac-
teristics. Size of system is even more
important for financial characteris-
tics. There are economies of scale in
the water industry. The industry is
characterized by substantial invest-
ment in fixed assets. This investment
can be spread across a broader cus-
tomer base in large systems, leading to
lower per capita costs and lower per
capita revenues. Also, large systems
can afford professional management.
As we demonstrate in the following
section, using several financial ratios,
some small systems do not appear to
be financially healthy.
The following financial ratios group
systems into three or four categories
based on commonly applied thresh-
olds. In general, these thresholds
indicate a level of financial health. For
example, an operating ratio of less
than 1 generally indicates a weak
financial condition. A ratio of 1 to 1.2
represents marginal to acceptable
performance, and a ratio greater than
1.2 represents a generally strong
financial condition. While the ratio
thresholds are intended to characterize
the financial condition of CWSs in
general, they do not characterize the
financial position of a particular water
system. For example, it is entirely
possibly, even likely, that some public
water systems have operating ratios of
less than 1.0 for reasons that are
consistent with good planning and
management (e.g., because they are
drawing down large reserves). Such
water systems would not accurately be
characterized as financially weak,
25
-------
1995 Community Water System Survey
even though they have a low operating
ratio.
One final note of caution about
interpretation of these financial ratios.
As explained in Section 5.8 of Volume
II, financial data are recorded and
reported in different ways by privately
owned and publicly owned systems.
Furthermore, within the category of
publicly owned systems, some use
enterprise fund accounting, and others
do not. Comparing these data involved
making assumptions about and adjust-
ments to the data as they were initially
reported in the questionnaires. The
objective of these adjustments was to
define various revenue and expense
items as consistently as possible in
order to provide comparable measures
of financial condition. Section 5.8 of
Volume II provides details on how the
ratios were calculated and guidelines
to their interpretation.
The operating ratio is defined as the
ratio of operating revenues to O&M
expense.7 The ratio is calculated by
dividing total operating revenues by
7 The use of the term "operating ratio" here is an
accounting measure used by analysts to assess the
financial condition of water utilities. It should not
be confused with the operating ratio as used by
public utility commissions to calculate a rate base.
26
O&M expenses. Items such as depre-
ciation charges, interest, or other debt
service payments are excluded from
expenses when creating the ratio.
If the ratio is less than 1.0, the system
is either running an operating deficit
(or loss) that year, or is relying on
non-operating revenues to finance its
operations. A higher ratio means that
funds are available from operations for
non-operating functions such as
servicing debt.
Distribution of Operating Rntio
(Percentage of Systems)
Public Water S
<1
1 to 1.2
>1.2
Private Water S
<1
1 to 1.2
>1.2
System Size
<600 | 501-3,300 1 3,301 -60,000 | > 50,000
^sterns
34.8
17.8
47.4
ystems
32.3
19.6
48.1
19.O
23.6
67.4
17.2
18.7
64.1
14.4
10.8
74.8
6.0
16.8
77.2
8.3
9.5
82.2
6.4
5.1
88.5
The table above shows the distribution
of operating ratios by size of system.
Notably, over 30 percent of all small
systems (serving 500 or fewer people)
have operating ratios of less than 1.0.
Furthermore, less than half of systems
in this size category have ratios of
more than 1.2, a standard of good
financial health. As system size
increases, the ratios get better. In the
largest size category, more than 80
percent of systems (almost 90 percent
of privately owned systems) have
ratios greater than 1.2.
Another indication of a water system's
financial health is the ratio of total
debt to total annual revenue. This ratio
provides a measure of a water
system's ability to incur new debt. The
lower the ratio, the better. As shown in
the following table, there is a slight
downward trend in both means and
medians as system size increases.
Total Debt as a Ratio of Total Revenue
System Size
<100
101-500
601-1,000
1,001-3,300
3,301-10,000
10,001-50,000
60,001-100,000
> 100,000
PubUcly Own«d Systems
Mean
3.21
2.98
2.63
2.84
2.62
2.08
3.09
2.64
Median
3.34
2.38
1.82
2.47
2.02
1.86
1.69
1.94
Privately Owned Systems
Mean
3.18
2.64
2.70
2.43
3.93
1.48
1.S6
1.22
Median
2.23
1.76
1.94
1.89
1.86
1.32
1.60
1.26
Closely related to total debt is the debt
service coverage ratio. This measures
the ability of water systems to cover
their debt service after all operating
expenses (excluding depreciation,
interest, and other debt service) have
been paid. Debt service coverage is
computed by dividing net available
revenue by annual principal and
interest (i.e., debt service) charges.
Net available revenue is the cash
-------
Chapter 3—Trends and Key Findings
available to pay debt service expenses
after current O&M expenses have
been paid. It is calculated by subtract-
ing only O&M expenses (excluding
interest, other debt service payments,
and any depreciation charges) from
gross revenues (i.e., operating plus
non-operating revenues).
The numerator in the debt service
coverage ratio represents annual net
revenues available to pay debt service,
and the denominator is the amount of
debt to be retired and the interest on
that debt for one year. It is a critical
ratio used by lenders and bond rating
services. This ratio should exceed 1.0,
and analysts consider a range of 1.0 to
1.5 as acceptable. The debt service
coverage ratio may be the subject of
bond issue requirements for purposes
of setting rates and for meeting tests
before additional bonds may be issued.
Distribution of Debt Service Coverage Ratio
(Percentage of Systems)
System Size
<500
501-3,300
3,301-50,000
> 50,000
Pubic Water Systems
<1.0
1.0to1.5
>1.5
52.1
19.3
28.6
41.5
16.5
42,0
25.7
21 .0
53.3
15.7
21.2
63.1
Private Water Systems
1.5
46.7
12.6
40.7
28.5
11.4
60.1
15.4
14.8
69.8
7.7
3.1
89.2
The table above shows that approxi-
mately half of the systems in the
smallest size category have ratios of
less than 1.0. Privately owned systems
are somewhat better in all size catego-
ries. For the largest systems, over 80
percent of publicly owned systems,
and over 90 percent of privately
owned systems, have ratios greater
than 1.0.
Another characteristic, called the net
takedown ratio, is an indicator of
water system profitability (or surplus).
It is determined by dividing net
available revenue (defined above) by
total gross revenue. Total gross
revenues are the sum of both operating
and non-operating revenues.
Distribution of Net Takedown Ratio
(Percentage of Systems)
Public Water Sy
<0%
0% to 20%
>20% to 40%
>40%
Private Water S
<0%
0% to 20%
> 20% to 40%
>40%
System Size
<500 | 501-3.300 {3,301-50.000 | > 50,000
items
29.6
11.2
26.8
32.4
stems
19.7
9.6
32.2
38.5
13.O
22.9
28.2
35.9
10.2
18.6
31. 0
40.2
1O.O
13.5
35.5
41 .0
5.6
14.O
42.O
38.4
8.O
8.O
32.2
51.8
7.7
3.O
78.5
10.8
As a general rule, lenders like to see a
net takedown ratio greater than 20
percent. The data in the table above
show that about 40 percent of the
smallest publicly owned systems and
about 30 percent of the smallest
privately owned systems fall below the
20 percent threshold. As with all of the
financial ratios, the larger systems
look much better. Over 80 percent of
the publicly owned systems and
almost 90 percent of the privately
owned systems have debt service
coverage ratios of more than 20
percent.
As a final comment on financial ratios,
we note that publicly owned systems
frequently appeared to have financial
ratios that were slightly worse than
those of privately owned systems in
most size categories. This is due in
part to the ratios chosen for analysis,
most of which emphasize net rev-
enues. Privately owned systems often
raise capital through sales of equity,
reducing their reliance on debt capital.
To pay dividends to their shareholders,
privately owned systems would need
to generate higher net revenues than
publicly owned systems.
The general theme of this section—
that larger systems have financial
characteristics that are different from
small systems—also appears to be true
when one examines capital expendi-
tures. Across all size categories, the
27
-------
1995 Community Water System Survey
largest category of investment was
system expansion, followed by repair/
replacement and water quality im-
provements.
Water Industry Capital
Expenditures
Water Quality
Improvements
19.5%
Large systems, however, generally
invest a slightly greater percentage of
their capital investment budgets in
water quality improvements than do
small systems. (See Volume II, Table
1-66, for more detailed data on this
subject.)
PfrcQnlAge of Capital Expenditures by Piunose,
by Ownership
ruMdr Owixd S«50,000
83
36
Six percent or less of systems in the
smallest two size categories were rated
"A" or better, but 83 percent of the
largest publicly owned systems were
rated "A" or better. The data also show
that publicly owned systems generally
have a higher percentage of systems
rated "A" or better. This reflects the
fact that most publicly owned systems
serving more than 10,000 people have
bonds that are rated; the majority of
privately owned systems are not rated.
Also, most small system bonds are not
rated. (See Volume II, Table 1-69, for
more detailed data on this subject.)
-------
4. Intended Uses of CWSS Data
rTl116 1995 CWS Survey database
I was developed primarily to
_M_ provide the Agency with
critical data to support its regulatory
development and implementation
efforts. The Agency last undertook
this effort in 1986, to coincide with the
1986 Amendments to the SDWA.
Since 1986, the Agency has developed
regulations covering 84 contaminants
in public drinking water supplies, the
filtration of surface water supplies, the
use of lead plumbing, and the control
of underground injection of wastes.
EPA undertook the 1995 CWS Survey
to determine a current baseline of
operational and financial characteris-
tics of the water supply industry. By
comparing the results of this survey
with the 1986 CWS Survey, changes
in water industry operations and
expenses resulting from the 1986
SDWA Amendments—and from
customer demands for improvements
in water quality and service—can be
measured.
EPA plans to use the 1995 CWS
Survey data to support the following
types of analyses:
• Regulatory development analyses
• Policy development
• Regulatory implementation
• Compliance analyses
The CWS Survey data elements that
can be used as inputs to each of these
types of analyses are described below.
4.1 Regulatory
Development
Analyses
The 1996 Amendments to the SDWA
extend EPA's mandate to establish
regulations (i.e., set maximum con-
taminant levels) for contaminants in
public drinking water supplies includ-
ing arsenic, sulfate, and at least five
additional contaminants every 5 years.
Before any new regulations are
established, however, the Agency must
satisfy the analytic requirements of
various statutes and regulations
including:
• Executive Order 12866.
• Paperwork Reduction Act.
• Regulatory Flexibility Act.
• Small Business Regulatory Enforce-
ment Fairness Act.
• Unfunded Mandates Reform Act.
The 1996 Amendments reinforce
EPA's current SDWA requirement to
specify best available technologies
(BATs) for the removal of drinking
water contaminants to consider those
technologies that are affordable to
different classes (i.e., sizes) of water
systems. The Agency traditionally has
conducted analyses of affordability in
the context of implementing the
SDWA's provisions regarding vari-
ances and exemptions. The 1996
Amendments to the SDWA add a new
section, "Small System Variances," .
which provides new procedures for
variances for small systems. This
section also directs the Agency to
publish information to assist states in
developing affordability criteria. Data
from the CWS survey will be useful
when implementing these require-
ments.
In addition, the 1996 SDWA Amend-
ments formalize the cost-benefit
analysis requirements of Executive
Order 12866. Under this order, the
Agency must prepare Regulatory
29
-------
1995 Community Water System Survey
Impact Analyses (RIAs) that detail the
national costs and benefits of all
proposed regulatory actions and
alternatives under consideration. The
RIAs are reviewed by the Office of
Management and Budget to determine
whether a proposed regulation can be
justified from an economic perspective
(i.e., whether the public health and
other benefits achieved are sufficient
to justify the costs imposed on the
nation).
The RIAs prepared in support of
proposed drinking water regulations
estimate the economic and financial
impacts of these regulations on the
nation as a whole, on individual water
utilities, and on individual households.
The national-level economic impact
analyses focus on estimating the net
benefits of each regulatory alternative
under consideration. The financial
analyses examine the impact of
additional capital requirements on
water utility operating expenses and
revenues. They also estimate how
capital and O&M expenditures that are
necessary to achieve compliance will
affect household water rates and
expenses. These effects are also
known as distributional impacts.
30
As shown in the table on the next
page, a drinking water RIA is a
compilation of separate analyses and
data collection efforts that provide
data for the Agency's national cost
and benefit estimation models. One
example is the Office of Ground
Water and Drinking Water's "Safe-
Water" Model, which is an updated
version of the old "What-If' Model.
These analyses include:
• System Characterization (Source,
Size, Ownership): Describe the
type, size, and number of water
treatment processes public water
systems would have to construct or
install to remove the regulated
contaminant.
• Contaminant Occurrence Analysis:
Estimate the current levels of a
particular contaminant in drinking
water supplies and the number of
water utilities that potentially exceed
various regulatory alternatives (i.e.,
Maximum Contaminant Levels) in
violation of the proposed standard.
• Exposure/Health Effects: Describe
the potential health risks associated
with particular drinking water
contaminants and assess the number
of persons at risk from exposure.
• Decision Trees/Treatment Profiles:
Describe the current treatment
profile (i.e., treatment in-place) of
community water systems and
forecast the likelihood that they will
select one of EPA's proposed BATs.
• Entry Points, Distribution of Treat-
ment Facilities: Examine entry
points into the distribution system,
their average and potential flows,
extent to which they are treated, and
the types of treatments applied.
• Unit Engineering Costs: Examine
in detail the capital and O&M
requirements of a particular treat-
ment technology that has been
demonstrated to be effective in
removing the specific contaminant.
• Economic and Financial Input
Estimates: Evaluate baseline
economic and financial conditions in
the water industry and estimate the
impact of new regulations on
financial condition.
The table on the next page compares
RIA and Information Collection
-------
Chapter 4—Intended Uses of CWSS Data
1 Comparison of RIA and ICR Data Requirements to CWS Survey Data Elements 1
Survey Question Number
0.1 : Contact Information
Q2: Year Data
Q3: Information Source
Q4: Source Detail
Q5: Peak/Maximum Designs
Q6: Design Factors
Q7/8: Finished Storage
Q9: Pipe Detail
Q1 0: New Pipe Length
Q1 1 : Population/Connections
Q1 2: Zip Codes
Q13/14: Operators
Q15: Interconnections
Q1 6: Contaminant Options
Q1 7: Distance to Alternative
Source
Q18: Treatment Facilities
Q1 9: Disinfection Residuals
Q20: Untreated Facilities
Q21-24: Surface/Wellhead
Protection
Q25: Contaminant Sources
Q26: Lab Analysis Provider
Q27: Lab Payment Method
Q28: Generally Accepted
Accounting Principles
Q29: Revenues From Water
Q30: Other Revenues
Q31: Rates/Meters
Q32: Uncompensated Usage
Q33: Expenses.
Q34: Balance Sheet
Q35-37: Capital Improvement
Detail
Q38/39: Bond Ratings
Q40: Comments
RIA Data Requirements
System
Characterization
(Source, Size,
Ownership)
•
•
•
•
•
Contaminant
Occurence
Analysis
•
•
•
•
•
•
•
•
•
•
•
•
•
Request (ICR) data requirements to
CWS Survey data elements. In gen-
eral, the CWS Survey data elements
provide baseline information that is
Exposure/
Health
Effects
•
•
•
•
Decision Trees/
Treatment
Profiles
•
•
•
•
•
•
•
•
•
Entry Points,
Distribution of
Treatment
Facilities
' •
•
•
•
•
•
critical to the preparation of the
analyses discussed above. Without an
accurate baseline, changes imposed by
regulations cannot be measured
Unit
Engineering
Costs
•
•
•
•
•
•
•
•
•
•
•
. •
Economic and
Financial Impact
Estimates
•
•
ICR
Compliance
Monitoring
Burden
•
•
•
•
•
•
•
•
accurately. For example, basic data to
characterize the industry, such as the
number of water systems categorized
by their type of source, ownership,
31
-------
1995 Community Water System Survey
population served, and production
volume are collected in response to
CWS Survey questions 4, 5, and 11.
Contaminant occurrence analyses and
exposure assessments use such indus-
try characterization data; they also rely
on information about source and
facility characteristics that is provided
in response to questions 18 and 20.
Analyses such as these support EPA's
estimates of the cost of complying
with new regulations. The cost of
compliance includes installing or
upgrading treatment facilities; in-
creases in O&M expenses (mainly
labor, chemicals, and power) associ-
ated with new treatment processes;
and the cost of collecting and analyz-
ing drinking water samples to monitor
compliance.
The Paperwork Reduction Act requires
the Agency to identify the reporting
and recordkeeping burden imposed on
regulated industries and on federal and
state governmental agencies that
manage the public water supply
supervision program and the compli-
ance monitoring program specified in
SDWA regulations. The estimated
burden of these recordkeeping and
32
reporting requirements is detailed in
ICRs. The preceding table also shows
the CWS Survey data elements that
support the development of SDWA
burden estimates for use in ICRs.
The Regulatory Flexibility Act (RFA)
and the newly authorized Small
Business Regulatory Enforcement
Fairness Act (SBREFA) require the
Agency to demonstrate that SDWA
regulations do not impose an unrea-
sonable economic and financial
burden on small businesses or govern-
ments. The analyses required by the
RFA and SBREFA can be supported
by many of the same CWS Survey
data elements as the RIA and ICR
analyses. The table on the opposite
page compares RFA data requirements
and CWS Survey data elements. The
financial section of the CWS Survey
database provides a number of critical
data elements for input into EPA's
small business impact analyses.
4.2 Policy Development
Analyses
The diversity of water systems con-
tained in the CWS Survey database
provides the Agency with a sufficient
set of financial and operational data
that can be used to support a variety of |
Agency initiatives to develop policies
and guidance to states and public
water systems concerning the imple-
mentation and enforcement of drink-
ing water regulations. These policy
initiatives can involve, for example,
defining financial affordability criteria
for granting variances and assessing
community-level affordability (i.e.,
ability to pay). Issues of affordability
can then be examined by merging
CWS Survey data with current popula-1
tion Census tract data to compare
financial and operational performance
measures to median income levels in
service areas. Further, the source water |
protection and operator training and
certification data contained in the
CWS Survey can be used by EPA and
the states to refine guidance for state
programs.
The Agency is continually engaged in
efforts to provide summary informa-
tion and reports on the status of
regulatory development, implementa-
tion, and enforcement activities. For
example, the Agency has periodically
prepared comprehensive drinking
water program-level studies that
describe the total cost and benefits of
-------
Chapter 4—Intended Uses of CWSS Data
Regulatory Flexibility/Small System Impact Analysis 1
Survey Question Number
Q1 : Contact Information
Q2: Year Data
Q3: Information Source
Q4: Source Detail
Q5: Peak/Maximum Design
Q6: Design Factors
Q7/8: Finished Storage
Q9: Pipe Detail
Q1 0: New Pipe Length
Q1 1 : Population/Connections
Q1 2: Zip Codes
Q13/14: Operators
Q15: Interconnections
Q16: Contaminant Options
Q1 7: Distance to Alternative Source
Q1 8: Treatment Facilities
Q19: Disinfection Residuals
Q20: Untreated Facilities
O2 1-24: Source/Wellhead
3rotection
Q25: Contaminant Sources
Q26: Lab Analysis Provider
Q27: Lab Payment Method
Q28: Generally Accepted
Accounting Principles
Q29: Revenues From Water
Q30: Other Revenues
Q31: Rates/Meters
Q32: Uncompensated Usage
Q33: Expenses
Q34: Balance Sheet
Q35-37: Capital Improvement Deta
0.38/39: Bond Ratings
Q40: Comments
Defining Small
Entities (Source, Size,
Ownership, etc.)
•
•
•
•
•
•
•
•
•
•
•
•
Determine
Health Risk
Posed by
Small Entities
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
Determine
Reporting,
tecordkeeping.
Requirements
(Monitoring)
•
•
•
•
•
•
•
•
•
•
•
Determine Small Entity's
Ability to Absorb/Pass on
Cost Increase
Financial
Analysis
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
Socio-Economic
Analysis
•
•
•
•
•
•
•
•
•
•
•
Alternatives to
Regulation
•
•
•
•
•
all SDWA regulations. Most recently, financial requirements of SDWA program-level ICR to document the
the Agency was required by Congress regulations. Data from previous CWS burden imposed on states, the water
to prepare an overall assessment of the Surveys were used extensively in these industry, and federal agencies to
capacity of states and public water reports. Further, the Agency is re- implement SDWA regulations. The
suppliers to meet the technical and quired periodically to prepare a Agency also receives periodic requests
33
-------
1995 Community Water System Survey
from Congressional staff and commit-
tees, other federal agencies, and the
public for information on the water
supply industry. The 1995 CWS
Survey provides current information
on the water industry to satisfy these
requests.
Finally, data from the CWS Survey
can be used in conjunction with data
from prospective drinking water RIAs
to develop estimates or profiles of the
net needs of public water systems.
Specifically, the CWS Survey data-
base contains information on the
current treatment and distribution
configuration of water systems that
can be used as a baseline for project-
ing future capital and O&M require-
ments for water systems.
4.3 Regulatory
Implementation
Analyses
A critical issue for EPA to address
under the 1996 SDWA Amendments
is whether the drinking water industry,
and small systems in particular, have
the technical and financial capacity to
comply with SDWA regulations over a
sustained period. In addition to the
34
financial problems faced by small
water systems, larger systems have
potentially serious financial concerns
as the combined effects of regulatory
compliance and infrastructure repair
and replacement drive operating costs
higher. As a result, the Agency is
engaged in efforts to assist states and
water suppliers in building the neces-
sary technical and financial capacity.
Congress has provided money to assist
the states and EPA in building addi-
tional capacity through State Revolv-
ing Loan Funds for public water
systems. CWS Survey data, in con-
junction with data from the Drinking
Water Infrastructure Needs Survey,
may be used to assess the ability of the
water industry to finance infrastructure
investment.
Another potential use of the CWS
Survey database is for the develop-
ment of operational and financial
performance measures for individual
water systems to gauge their relative
technical and financial performance.
Data on specific CWS Survey vari-
ables or ratios of variables can be
expressed as industry mean or median
values for comparison by individual
utilities against their particular opera-
tional situation. The data can also be
used in statistical (e.g., regression
analysis) models to develop relation-
ships between variables for use in
predicting potential operational or
financial problems. Water system
analysts could use these models to
identify systems in need of additional
technical and financial assistance
before critical problems develop.
4.4 Compliance
Analyses
The Agency is engaged in several
efforts to upgrade and expand its water
industry databases. One intended use
of the CWS Survey database is to
support the development of profiles of
operational and financial characteris-
tics for different types of water sys-
tems that can be statistically correlated
with the Agency's database of compli-
ance records contained in the Safe
Drinking Water Information System
(SDWIS). The objective of this analy-
sis is to identify those operational and
financial characteristics that can
potentially result in future compliance
problems. EPA can then develop
guidance to target systems that may
exhibit these characteristics.
------- |