United States
Environmental
Protection Agency
Office of Policy
(1807T)
March 2011
EPA-100-R-11-006

 Evaluation of the Drinking
 Water and Clean Water
 Infrastructure Tribal Set-
 Aside Grant Programs

 Final Report
 Promoting Environmental Results
 <                        >
 Through Evaluation

-------
    EVALUATION OF Til	IE               AND             INFRASTRUCTURE TRIBAL SET-ASIDE
ACKNOWLEDGEMENTS
Ross & Associates Environmental Consulting, Ltd. (Ross & Associates) and Industrial Economics, Incorporated (lEc)
conducted this evaluation under  Contract EP-W-07-028 between lEc and  EPA's  Office  of  Policy  (OP).  The
evaluation team consisted of Anna Williams, Amy Wheeless, and Darcy Peth of Ross & Associates  and  Angela
Helman and  Lindsay Ludwig  of lEc.  Planetary Engineers, staffed by Crispin Kinney, a former IMS manager who
assisted with the development of the IMS STARS database, also provided assistance in this evaluation.

The Evaluation Advisory Team that supported this evaluation included David Harvey and Kyle Carey of EPA's Office
Ground Water and Drinking Water (OGWDW), Matthew Richardson of EPA's Office of Wastewater Management
(OWM), and  Michelle Mandolia of EPA's OP's Evaluation Support Division. An Evaluation Review Group held two
calls to provide feedback on evaluation products: this group included the Evaluation Advisory Group members,
Michael Mason (OWM), Gerald McKenna (Region 2), Brian Smith (Region 4), Stephen Poloncsik (Region 5),  Minnie
Adams (Region 8),  Erskine Benjamin (Region 9), Linda Reeves (Region 9), Loretta Vanegas (Region 9),  and  Dennis
Wagner (Region 10).

EPA's Program Evaluation Competition, sponsored by OP, provided support for this evaluation. To access copies of
this  or  other  EPA  program evaluations, please  visit  EPA's Evaluation  Support  Division's  website at
http://www.epa.gov/evaluate.
Evaluation Report-March 2011

-------
    EVALUATION OF Til	IE               AND CLEAN       INFRASTRUCTURE TRIBAL SET-ASIDE

Table of Contents
Executive Summary	ES-1
Conclusions	ES-2
Recommendations	ES-3
Introduction	1
Evaluation Purpose and Audience	1
DWIG-TSAand CWISA Program Background	2
History and Funding	2
EPA Infrastructure Grant Program Goals and Measures	3
Projects  Eligible for DWIG-TSA and CWISA Funding	5
Role of IMS in EPA Infrastructure Grant Programs	6
Methods	8
Evaluation Approach	8
Information Collection Approach	8
Findings: Program Implementation	12
EPA Headquarters Implementation of Tribal Infrastructure Programs	12
EPA Regional Implementation: DWIG-TSA Program	16
EPA Regional Implementation: CWISA Program	19
Regional Communication with IMS, Tribes, and Other Parties	23
Role of IMS in Program Implementation	25
ARRA Implementation	27
Findings: Program Goals and Measures	30
The Compliance Strategic Measure: Progress and Factors Influencing Progress	30
The Access Strategic Measures: Progress and Factors Influencing Progress	36
Conclusions and Recommendations	43
Conclusions	43
Recommendations	45
Additional Considerations	54


Appendices (provided separately)

Appendix A.      DWIG-TSA and  CWISA Program Evaluation Logic Model
Appendix B      Evaluation Interview Guides
Appendix C.      DWIG-TSA Regional Implementation Table
Appendix D.      Results of Correlation Analyses
Appendix E.      Detailed Data Tables
Appendix F.      Information Sources
Evaluation Report-March 2011

-------
    EVALUATION OF THE DRINKING       AND CLEAN       INFRASTRUCTURE TRIBAL SET-ASIDE
ACRONYMS
ACS            Annual Commitment System
AI/AN          American Indian and Alaska Native
ARRA          American Recovery and Reinvestment Act of 2009
CDP            Community Deficiency Profile (a component of STARS)
CWA           Clean Water Act
CWISA         Clean Water Infrastructure Set-Aside
CWS           Community Water System
CY             Calendar Year
DG             Direct Grant
DL             Deficiency Level (a STARS designation)
DWIG-TSA      Drinking Water Infrastructure Grant Tribal Set-Aside
DWINS         Drinking Water Infrastructure Needs Survey
EPA            United States Environmental Protection Agency
FDL            Final Deficiency Level
FTE            Full Time Equivalent
FY             Fiscal Year
HUD           United States Department of Housing and Urban Development
IA             Interagency Agreement
IDL            Initial  Deficiency Level
IHS            United States Department of Health and Human Services, Indian Health Service
NPDWR        National Primary Drinking Water Regulation
NPM           National Program  Manager
O&M          Operations and Maintenance
OGWDW       EPA Office of Ground Water & Drinking Water (part of the EPA Office of Water)
OP             EPA Office of Policy
OW            EPA Office of Water
OWM          EPA Office of Wastewater Management (part of the EPA Office of Water)
PDS            Project Data System (a component of STARS)
PWS           Public Water System
PWSID         Public Water System Identification Number
SDS            Sanitary  Deficiency System (a  component of STARS)
SDWA          Safe Drinking Water Act of 1974, amended in 1986 and 1996
SDWIS          Safe Drinking Water Information System
SFC            IHS Division of Sanitation Facilities Construction
STARS          Sanitation Tracking and Reporting System
USDA          United States Department of Agriculture
Evaluation Report-March 2011

-------
    EVALUATION OF Til	IE               AND             INFRASTRUCTURE TRIBAL SET-ASIDE
EXECUTIVE SUMMARY
The Environmental Protection Agency (EPA) Clean Water Indian Set-Aside (CWISA) Grant Program and the Drinking
Water Infrastructure Grant Tribal Set-Aside (DWIG-TSA)  Program provide funding for wastewater infrastructure
and drinking water infrastructure to American Indian tribes and Alaska Native (AI/AN) Villages. EPA coordinates
with Indian Health Service (IMS) on these programs.

Through this program evaluation, EPA sought to determine the extent to which the combined efforts of the EPA
DWIG-TSA program and the EPA CWISA program have resulted  in increased access to safe drinking water and
sanitation  facilities and in increased compliance with the Safe Drinking Water Act  (SDWA) in Indian country.  EPA
also sought  to  better understand and  explain  how  these programs are  implemented, how implementation
influences program performance  toward their goals, as tracked by each program's strategic  measures, and
whether these measures are accurate  indicators of each  program's  progress.   In  addition, because  IMS  often
oversees and directly implements the infrastructure projects, EPA hoped to gain a deeper understanding  of its own
scope of influence and if and how EPA could better demonstrate program results. The evaluation looked at DWIG-
TSA and  CWISA program  activities and influence since 2003.  The evaluation team  did  not focus  on IMS
implementation of the programs, except as it relates to EPA program implementation and accomplishments.

Ross & Associates Environmental Consulting, Ltd.  (Ross & Associates) and Industrial Economics, Incorporated (lEc)
conducted this evaluation with support  through the EPA Program Evaluation Competition, sponsored by the EPA
Office of Policy (OP).

The evaluation focused on responding to the following questions:
   I.    How does EPA implement the DWIG-TSA and CWISA programs at headquarters and in the regions?
  II.   To what extent are the DWIG-TSA and CWISA programs achieving their stated goals?
       What factors influence whether the programs achieve their stated goals?
       Are the current DWIG-TSA and CWISA program performance measures accurate indicators of EPA's
        progress?
  V.   What implementation improvements or innovations could be made by EPA to make the DWIG-TSA and
       CWISA programs more effective in meeting the water and wastewater infrastructure needs of tribes?

To answer these questions, the evaluators collected  and  reviewed documents,  conducted interviews with 39
people, and analyzed data on over 650  DWIG-TSA and CWISA  projects that  EPA contributed funding to between
2003 and 2009.

EPA tracks progress made  by the DWIG-TSA and  CWISA programs toward the  EPA Strategic Plan Goal 2—
protecting America's waters—according to performance on three Strategic Plan measures.
      Compliance Strategic Measure:  By 2015,  88 percent  of the population in  Indian country served  by
      community water systems will receive drinking water that meets all applicable health-based drinking water
      standards. (2005 baseline: 86 percent. Status as of FY 2009: 81 percent.)1
1 In the FY2011-2015 EPA Strategic Plan, this measure is under Goal 2: Protecting America's Waters, Objective 2.1: Protect Human Health
Evaluation Report - March 2011                                                                       Page ES-1

-------
    EVALUATION OF Til	IE                AND CLEAN       INFRASTRUCTURE TRIBAL SET-ASIDE

      Drinking Water Access Strategic Measure:  By 2015,  in coordination with other federal agencies, provide
      access to safe drinking water for 136,100  American Indian and Alaska Native  homes. (FY 2009 baseline:
      80,900 homes. Universe: 360,000 homes.)2
      Sanitation Access Strategic Measure: By 2015, in coordination with other federal  agencies, provide access to
      basic sanitation for 67,900 American  Indian and Alaska Native homes. (FY 2009 baseline:  43,600 homes.
      Universe: 360,000 homes.)3
The access strategic measures relate to a US commitment made in 2002 at the United Nations World Summit on
Sustainable Development, held  in Johannesburg, South Africa.4 Specifically, the US committed to reducing by
half the proportion of people that did not have access to safe drinking water or basic sanitation.5

Conclusions

    The goals and priorities for these programs could be clearer and more focused.   The goals and priorities for
    both programs lack precision and clarity: EPA staff have different ideas of what the term "goal" means for
    these programs and how the drinking water program  priorities of compliance and access relate to each  other.
    EPA has the opportunity to more clearly articulate and focus the goals of both programs, and then to focus
    program design  and implementation  accordingly in  a  set of "cascading" decisions and  actions that  would
    follow clearer goals and priorities.
    Funding allocation for  both programs could be improved to be more in  line with EPA's priorities.  The
    method used to allocate funds for each program has an understandable basis and history; however, neither
    allocation process is an ideal match for the current program priorities and strategic  measures.
    For both programs, project selection  could be more clearly and  consistently tied to EPA's priorities, while
    still maintaining regional discretion and flexibility.  Project selection under the  DWIG-TSA program  differs
    from region to region,  and for both programs the relationship between project selection and  EPA's priorities
    and scope of influence is not as clear or consistent  as it could be.  For the CWISA  program, using the  IMS
    Sanitary Deficiency System (SDS) list is  not ideal for CWISA's priorities and ability to  fund projects.
    Despite improvement in recent years, communication within  EPA and between EPA and IMS is inconsistent
    and not optimal for strategic program management or strategic learning and improvement.
    IMS is a vital partner for EPA to implement these programs.  IMS has the in-house expertise, field presence,
    and relationships with  tribes necessary  to  do much of the work  of these grant programs.  In addition, the
    STARS database is also generally considered to be the best source of information on tribal water infrastructure
    information and is thus an invaluable tool  for these  EPA programs.  The STARS database is an inventory of
    wastewater and water infrastructure need and ongoing projects in  Indian country, as tracked by IMS.
    There  are opportunities for  improved  communication with tribes about these  programs.  EPA could
    strengthen its direct communication with the  tribes to enhance knowledge of tribal contexts and needs, more
    consistently provide tribes  with  information  about funding options, understand program accomplishments,
    and identify areas for program improvements.
2 Ibid.
3 In the FY2011-2015 EPA Strategic Plan, this measure is under Goal 2: Protecting America's Waters, Objective 2.1: Protect and Restore
Watersheds and Aquatic Ecosystems.
4 United Nations. Plan of Implementation of the World Summit on Sustainable Development. Section II. September 4, 2002. Available:
http://un-documents.net/iburgpln.htmtfll.
5 Specific commitment language was as follows: "The provision of clean drinking water and adequate sanitation is necessary to protect human
health and the environment. In this respect, we agree to halve, by the year 2015, the proportion of people who are unable to reach or to afford
safe drinking water (as outlined in the Millennium Declaration) and the proportion of people who do not have access to basic sanitation..."
Evaluation Report - March 2011                                                                           Page ES-2

-------
    EVALUATION OF THE DRINKING       AND CLEAN       INFRASTRUCTURE TRIBAL SET-ASIDE

    EPA has limited ability to make progress on the current strategic measures focused on providing access to
    drinking water and clean water to tribal homes.  EPA's limitations include:
            Lower program funding relative to the funding needed to provide for all identified access needs;
            A history of funding a substantial proportion of projects that do  not provide access (as defined by
            moving from a higher deficiency level (DL) to a lower one) due to project selection processes;
            Restrictions on EPA's authority to fund some  projects that would  provide access, such as funding
            individual drinking water wells or providing sanitation infrastructure services to homes that are not
            on reservation land;
            Reliance on a metric which assumes changes in DLs (from a DL4-5 to a lower DL) that are immediate
            and permanent, when  in reality, DLs may not change in this manner as a result of funding (e.g.,
            because a DL-4 project only partially funds the infrastructure needed to change the associated homes
            to  a  DL-3 or lower) and there  is sometimes  "backsliding" of DL designations due to compliance
            violations, lack of infrastructure maintenance, or other reasons;
            Substantial reliance on other federal agencies,  particularly IMS, to  contribute toward progress on the
            access measure; and
            Disconnects between reported progress on providing access (reported to be immediate upon project
            funding) and  actual progress on the ground, which typically occurs  years later, after construction is
            completed and the new or upgraded infrastructure is operating.
    EPA appears to be quite limited in its ability to  make progress toward the current compliance strategic
    measure, though the extent of the limitations is not clear due to a lack of measurable data.  The drivers of
    some if not most compliance problems may related to operations and maintenance (O&M) at tribal systems or
    the introduction of new drinking water  rules, rather  than to  infrastructure problems that the DWIG-TSA
    program  can  address.   In addition, the allocation of DWIG-TSA funding has not been directly tied to
    compliance needs  or  opportunities,  nor has project  selection  clearly  and consistently been  linked to
    compliance support on a national scale.
    EPA's meaningful contributions to meeting tribal infrastructure needs are  not well reflected by the reported
    program performance.  Despite the identified challenges  surrounding making progress toward the current
    strategic measures, both  programs have made significant contributions  over the years to Indian country
    through these programs by providing funding for vital water infrastructure needs.  These contributions—and
    the public and environmental health benefits they have provided—have not been clearly recognized in a
    public or formal way.  As one EPA regional coordinator said, "I  don't think we hear enough about what the
    achievements are—we know what the goals are, but I don't know what the  achievements have been."

Recommendations

In response to the evaluation  question about possible program improvements (evaluation question V), we offer
eight  recommendations  and  three  additional  items  for  consideration.     We   consider the  first  four
recommendations to be policy recommendations as they would require approval from senior  EPA management
and, in the case of at least recommendations 1-3, approval from the US Office of Management and Budget as well.
Recommendations 5-8 are more  program-oriented and could  be considered  and implemented at the program
level.  Additional discussion on each  recommendation  is  provided in the main body of the report (beginning on
page 45).
Evaluation Report - March 2011                                                                        Page ES-3

-------
    EVALUATION OF THE                AND             INFRASTRUCTURE TRIBAL SET-ASIDE

1.  Clarify the goals and priorities of both programs.

To  reduce confusion and  increase program focus and  management, we recommend that EPA identify clear,
specific, and measurable goals and priorities for both the DWIG-TSA and CWISA programs.  A set of cascading
changes to both programs' processes (e.g., allocation of funds, project selection) and performance measures could
then logically follow as described in the following recommendations.

2.  Consider changing the access strategic measures by which EPA tracks its performance to measures that
    more directly reflect EPA's mandate, authority, and scope of influence; take into consideration data
    availability and quality; and reflect changes or clarifications to the programmatic goals made in response to
    recommendation 1.

In the Recommendations section of the full report, we offer several options for EPA's consideration regarding the
current  access strategic measures,  which we do not think serve EPA or reflect  progress toward meeting tribal
needs as well as other options would.

3.  Consider changing the DWIG-TSA compliance strategic measure to a measure that more directly reflects the
    drivers of compliance problems and EPA's scope of influence over these problems.

A revised compliance measure targeted specifically to infrastructure-solvable compliance  problems  that EPA has
the ability to  address  would  more directly and clearly reflect the influence  of the DWIG-TSA program  on
compliance.  In the Recommendations section of the full report, we  offer three options for such a measure.

4.  Reassess the national annual budget allocation for both programs to be more clearly tied to EPA's goals and
    priorities.

Once EPA has determined  whether and how to update each program's goals, priorities, and measures,  deciding
whether and how to update the  allocation formulas for one or  both  programs will be easier.   However, we
recommend  updating the allocation formulas to more directly and clearly support EPA's  current strategic measures
even if EPA makes no changes in response to recommendations 1-3.

5.  For both programs, update and clarify expectations for project selection to be more clearly in line with the
    program priorities,  allow for regional flexibility and discretion, and promote increased consistency and
    transparency.

The project selection processes for both programs have their bases and merits, but neither process is ideal.  For
the DWIG-TSA program, consideration of both compliance needs  and provision  of access varies from region to
region; regions often choose projects based on other considerations, making it difficult for EPA to make  progress
on its strategic measures.  For the CWISA program, exclusive use of the SDS list is limiting and does not recognize
EPA's distinct mission, authority to fund, or priorities, including provision of access. Updating the project selection
guidances to focus on a simple set of core program elements while allowing for regional flexibility and discretion
would provide more coherence across the regions, particularly for the DWIG-TSA program, and more direct linkage
to program priorities for both programs.

6.  For both programs, routinely collect and analyze data to enhance transparency and strategic coordination
    and improve EPA's ability to report on, advocate, and improve both programs through learning and
    adaptation.
Evaluation Report - March 2011                                                                        Page ES-4

-------
    EVALUATION OF THE                AND             INFRASTRUCTURE TRIBAL SET-ASIDE

Improved information availability will enable EPA to better understand its contributions through these programs
and support learning over time to improve program effectiveness. Given there is currently very little information
collected by EPA nationally other than  basic funding statistics, even a modest increase in data collection would
provide a markedly improved ability to understand, explain, and advocate for these infrastructure programs at the
regional and national levels. The type of information that  could be beneficial would  not be complicated or
extensive.

7.   Update the national guidelines for both programs.

Following the  consideration of the above recommendations and any resulting changes, we recommend that EPA
headquarters work with EPA regions to update the guidelines for both programs. In addition, EPA headquarters
should work with the EPA regions to update any communication materials about these programs.

8.   Continue to enhance and improve communication within EPA, between EPA and IMS,  and between EPA and
    the tribes.

Communication between  EPA headquarters and EPA  regions, between  EPA and IMS, and between EPA and the
tribes  is  important to  effectively implement  and measure  the impact that  these programs have on tribal
communities.  Although the evaluators heard from interviewees many positive reflections on communication, we
also heard that communication improvements in several areas would be useful.

Additional Considerations
We offer three additional considerations which reflect  our support for efforts that are already underway or, in one
case, previous work done to reflect on many of the same issues that this evaluation has covered:
1.   Continue to focus on the critical issues surrounding infrastructure sustainability, including tribal O&M capacity
    issues;
2.   Continue to communicate about interagency agreement (IA) requirements, expectations, and inter-agency
    roles between EPA and IMS;  and
3.   Reflect on other reports and recommendations prepared by other workgroups that pertain to these programs.
Evaluation Report - March 2011                                                                        Page ES-5

-------
    EVALUATION OF Til	IE               AND             INFRASTRUCTURE TRIBAL SET-ASIDE
INTRODUCTION
The Environmental Protection Agency (EPA) Clean Water Indian Set-Aside (CWISA) Grant Program and the Drinking
Water Infrastructure Grant Tribal Set-Aside (DWIG-TSA) Program contributes funding for wastewater infrastructure
and drinking water infrastructure to American Indian tribes and Alaska Native (AI/AN) Villages. EPA coordinates
with Indian Health Service (IMS) on these programs.  This evaluation is intended to assess these tribal set-aside
programs,  including their implementation, program  measures and outcomes, and  identify opportunities for
program improvements.

Evaluation Purpose and  Audience

Through this program evaluation, EPA sought to determine the extent to which the combined efforts of the DWIG-
TSA and CWISA programs have  resulted in the provision of access to safe drinking water and sanitary facilities in
Indian country, and in increased compliance with the Safe Drinking Water Act (SDWA) in Indian country. EPA also
sought to better understand and explain how these programs are implemented, how  implementation influences
program performance toward their goals, as tracked by each program's strategic measures, and whether these
measures are accurate indicators of each program's progress. In addition, because IMS often oversees and directly
implements the  infrastructure  projects, EPA hoped to better understand its own scope of influence  and, in  so
doing, better understand if and how EPA could better demonstrate program results.  The evaluation looked at
DWIG-TSA and CWISA program activities and influence since 2003.  The evaluation team did not focus  on IMS
implementation of the programs, except as it relates to EPA program  implementation and accomplishments.

The primary audiences for this evaluation are  the EPA managers and staff who oversee and implement these
programs in the Office  of Ground Water and Drinking Water (OGWDW),  Office of  Wastewater Management
(OWM), and EPA regions. IMS managers and staff may also be interested in the results of this evaluation,  as may
AI/AN representatives who are eligible for funding from the CWISA or DWIG-TSA programs.
Evaluation Report - March 2011                                                                        Page 1

-------
               OF THE                AND
DWIG-TSA AND CWISA  PROGRAM BACKGROUND
Historically, as compared non-tribal populations, those living in Indian country and in Alaska Native Villages have
had disproportionately lower access to safe drinking water and sanitary facilities, which indicates that there  is a
lack of a safe water supply and/or a system for sewage disposal. For example, in 2009, 43,437 or 12.1 percent of
the tribal homes tracked by IMS lacked access to safe drinking water in 2009, compared to the 0.6 percent of non-
tribal homes that lacked such infrastructure (as measured by the US Census Bureau in 2005).6

The CWISA and DWIG-TSA programs  are intended to provide funding to infrastructure  projects that will help
overcome these AI/AN access challenges. The EPA regions, often through federal interagency agreements (lAs)
with the IMS Division of Sanitation Facilities Construction  (SFC)7, administer the programs.  Grants are available to
any federally recognized Indian Tribe, as well as to Alaska  Native Villages working with the State of Alaska. Funding
recipients can receive funds either through a direct grant from EPA or through IMS, which  administers an IA with
EPA to receive and disperse the funds.

History and Funding

The DWIG-TSA Program was established under section 1452(i) of the 1996 amendments to the Safe Drinking Water
Act (SDWA) and is administered under the EPA Office of Ground Water and Drinking Water (OGWDW). Through
the 2009 federal fiscal year (FY2009), this program was funded from a 1.5 percent set-aside from the annual
Drinking Water State Revolving Fund  (DWSRF); as of FY2010, this set-aside amount  increased  to  2.0 percent.
DWIG-TSA funds are allocated to the EPA regions based on a combination of: (1) drinking water needs identified in
the IMS Sanitary Deficiency Survey (SDS)  list, an annual inventory of water infrastructure need in Indian country,
and (2) the tribal portion of the  EPA Drinking Water Infrastructure Needs Survey (DWINS), a survey of drinking
water infrastructure need  undertaken approximately every four years.  Annual funding since  the program's
inception until FY2010 has ranged from $10.9 million to $27.7 million. Funding fluctuations have been a function
of variances in DWSRF funding amounts and, in FY2010, the increased funding percentage from the DWSRF.

The CWISA Program  was  established in 1987 under section  518(c)  of the Clean Water Act  (CWA) and is
administered  under the EPA Office of  Wastewater Management (OWM). From  FY2001 to FY2009, this program
was funded from a 1.5 percent set-aside from the annual Clean Water State Revolving Fund  (CWSRF); as of FY2010,
this set-aside amount increased to 2.0 percent.8 These funds are allocated to the  EPA regions based on the level of
wastewater needs indicated in the IMS area SDS lists. Annual funding since the program's inception  until FY2010
has ranged  from  $4.7  million to  $42 million, due to different set-aside percent allocations and  CWSRF funding
amounts.

The American Recovery and Reinvestment Act (ARRA) of 2009 provided the CWISA and DWIG-TSA programs with
significant additional  funding  ($60 million for CWISA and $30 million for DWIG-TSA, or two  percent of  the
6 US EPA. FY2011 National Water Program Guidance (April 2010). Available:
http://www.epa.gov/ocfo/npmguidance/2011/owater/nwp program guidanceSOS 050510.pdf.
7 IMS is a Federal agency underthe US Department of Health and Human Services.
8 Between FY1987 and FY1990, CWISA was funded by a 0.5 percent set-aside from the Construction Grants Program. In FY1991, no funding was
appropriated. In FY1992, Congress gave EPA the authority to take a 0.5 percent set-aside from the CW State Revolving Fund. In FY2001, the
set-aside was increased to 1.5 percent.
Evaluation Report - March 2011                                                                          Page 2

-------
               OF THE                AND

respective State Revolving Fund).  Guidance associated with the ARRA funding directed that funds be allocated to
projects in a similar manner as the annual allocations, with a few management changes; these are described in the
section of this report that discusses ARRA implementation.

EPA Infrastructure Grant Program Goals and Measures

The DWIG-TSA and CWISA programs are guided by the EPA Strategic  Plan.  In September 2010, EPA issued its
FY2011-2015 EPA Strategic Plan, which includes strategic Goal 2: Protecting America's Waters.  The Strategic
Plan defines Goal 2 as, "Protect and restore our waters to ensure that drinking water is safe, and that aquatic
ecosystems sustain fish, plants and wildlife, and economic, recreational, and subsistence activities."9

The DWIG-TSA and CWISA programs provide funding for water infrastructure in Indian country and are aimed at
realizing Goal 2 specifically through  the Strategic Plan's Objective 2.1: Protect Human Health and Objective 2.2:
Protect and Restore Aquatic Watersheds and Ecosystems.10  Infrastructure funded by the DWIG-TSA program
protects public health  through exposure reduction by helping systems in  tribal  communities provide water in
compliance with the SDWA National Primary Drinking Water Regulations (NPDWRs). Infrastructure funded by the
CWISA program protects aquatic watershed and ecosystems by helping tribal communities ensure that wastewater
introduced to aquatic environments has been properly treated and in compliance with CWA regulations.  With
these and other programs and regulations within the Office of Water,  EPA works with other federal agencies and
state, local, and tribal governments to make progress on this high-level  strategic goal.

EPA tracks progress made by the DWIG-TSA and CWISA programs toward the Strategic Plan's Goal 2 and related
Objectives according to performance on three Strategic Plan  "measures."  These measures are reiterated as
strategic annual  commitment measures in the Office of Water National  Program Manager (NPM) Guidances.
We list  these strategic measures below  with the EPA annualized  measure designation  from the Annual
Commitment System (ACS).
  >    By 2015, 88 percent of the population  in Indian  country served by community water systems will  receive
      drinking water that meets all applicable health-based drinking water standards. (2005 baseline: 86 percent.
      Status as of FY 2009: 81 percent.) (ACS Measure SP-3)11
  >    By 2015, in coordination  with other federal agencies, provide  access to safe drinking water for 136,100
      American  Indian and Alaska Native homes.  (FY  2009 baseline:  80,900  homes. Universe: 360,000 homes.)
      (ACS Measure SOW-18)12
  >    By 2015, in coordination with other federal agencies, provide access to basic sanitation for 67,900 American
      Indian and Alaska Native homes. (FY 2009 baseline: 43,600 homes. Universe: 360,000 homes.) (ACS Measure
      WQ-24)13

These FY2011-2015 EPA strategic  measures are updates from the FY2006-2011 EPA Strategic Plan, which listed
three similar "strategic targets" related to the DWIG-TSA and CWISA programs.  '
9 US EPA. FY2011-2015 EPA Strategic Plan. Page 46. Available: http://www.epa.gov/planandbudget/strategicplan.html.
10 Ibid.
11 In the FY2011-2015 EPA Strategic Plan, this measure is under Goal 2: Protecting America's Waters, Objective 2.1: Protect Human Health
12 Ibid.
13 In the FY2011-2015 EPA Strategic Plan, this measure is under Goal 2: Protecting America's Waters, Objective 2.1: Protect and Restore
Watersheds and Aquatic Ecosystems
14 US EPA. FY2006-2011 EPA Strategic Plan. Available: http://www.epa.gov/planandbudget/archive.html.
Evaluation Report - March 2011                                                                            Page 3

-------
    EVALUATION OF THE DRINKING       AND CLEAN        INFRASTRUCTURE TRIBAL SET-ASIDE

      By 2011,  86 percent of the population in Indian country served by  community water systems will receive
      drinking water that meets all applicable health-based drinking water standards. (2005 Baseline: 86 percent)
      (ACS Measure: SP-3)16
      By 2015,  in coordination with other federal agencies, reduce by 50 percent the number of homes on tribal
      lands lacking  access to safe drinking water. (2003 baseline: Indian Health  Service data indicate that 12
      percent of homes on tribal lands lack access to  safe  drinking water (38,637  homes lack access).)  (ACS
      Measure: SP-5)17
      By 2015,  in coordination with other federal partners, reduce by 50 percent the number of homes on tribal
      lands lacking access to basic sanitation (cumulative). (2003 baseline: Indian Health Service data indicate that
      8.4 percent of homes on tribal lands lack access to basic sanitation (26,777 homes of an estimated 319,070
      homes).)  (ACS Measure: SP-15)18

This evaluation  has focused on the progress made toward the three FY2011-2015 EPA Strategic Plan DWIG-
TSA and CWISA strategic "measures" listed above. Where we refer to the higher-level Strategic Plan Goal 2,
"Protecting America's Waters," we note so explicitly in the text.

Throughout the  course of the evaluation, when discussing the DWIG-TSA and CWISA programs, EPA and IHS staff
also discussed four related "goals":

    1.  A long-term drinking water program compliance "goal" of having all public water systems in compliance
        with the Safe Drinking Water Act.  Although  the current Strategic Plan does not specifically reference
        this goal, EPA program managers routinely consider this programmatic goal when assessing progress on
        its programs. In this report, we refer to this as the "long-term drinking water compliance goal."

    2.  The CWISA program has a long-term access "goal" of all serviceable tribal homes having access to basic
        sanitation. In this report, we refer to this as the "long-term clean water access goal."

    3.  The  US government commitment  to the  Plan of Implementation of the 2002  United  Nations World
        Summit on  Sustainable Development, held in Johannesburg, South  Africa.19   Specifically, the US
        committed to reducing  by half the  proportion of people that did not have access to safe drinking water
        or basic sanitation.20  In 2007, a group of federal agencies assembled an Infrastructure Task Force to
        help achieve this commitment as it relates to tribal homes.  This Infrastructure Task Force is comprised
        of EPA, IHS,  the US Department of Agriculture (USDA), and the US Department of Housing and Urban
  SP-5 and SP-15, the former access strategic targets, were kept as "indicators" in the FY2011 Office of Water NPM Guidance, meaning that
EPA will continue to report on the number of tribal homes lacking access to safe drinking water and basic sanitation, but not against an annual
or long-term target.
16 In the FY2006-2011 EPA Strategic Plan, this target is under Goal 2: Clean and Safe Water, Objective 2.1: Protect Human Health, Sub-objective
2.1.1: Water Safe to Drink.
"ibid.
18 In the FY2006-2011 EPA Strategic Plan, this target is under Goal 2: Clean and Safe Water, Objective 2.2: Protect Water Quality, Sub-objective
2.2.1: Improve Water Quality on a Watershed Basis.
19 United Nations. Plan of Implementation of the World Summit on Sustainable Development. Section II. September 4, 2002. Available:
http://un-documents.net/iburgpln.htmtfll.
20 Specific commitment language was as follows: "The provision of clean drinking water and adequate sanitation is necessary to protect human
health and the environment. In this respect, we agree to halve, by the year 2015, the proportion of people who are unable to reach or to afford
safe drinking water (as outlined in the Millennium Declaration) and the proportion of people who do not have access to basic sanitation..."
Evaluation Report - March 2011                                                                               Page 4

-------
               OF THE                 AND

        Development (HUD).21 Stakeholders often refer to this cross-agency federal goal as the "access goal."
        In this report, we refer to this commitment as the "US access goal."

    4.  The logic model developed at the beginning of the evaluation (Appendix A) identified a general "goal"
        for both the DWIG-TSA and CWISA programs: To improve access to drinking water and basic sanitation
        in Indian country and to improve compliance of community water systems with applicable health-based
        standards.  In this report, unless noted otherwise, we are not referring to this more generalized goal,
        which supports and reiterates all of the aforementioned goals and measures.

Projects Eligible for DWIG-TSA and CWISA Funding

The  DWIG-TSA and  CWISA programs fund infrastructure projects  in tribal communities.   For the DWIG-TSA
program, most projects  that improve a tribal drinking water system are eligible for funding.  For example, eligible
projects could  be  those that rehabilitate  or develop sources of drinking  water;  install treatment  or  storage
facilities; and replace aging water system infrastructure. CWISA funding can be used for the planning, design, and
construction of wastewater collection and treatment systems.  For  example, on-site  septic systems and major
sewer rehabilitation are projects that are eligible for funding.  However, there are some limitations in what  each
program is able to fund, based the statutes that established the programs and interpretations of their respective
guidances.

The DWIG-TSA program cannot fund the following kinds of projects (per the DWIG-TSA national guidance, except
where footnoted):
  >    Projects that are not primarily or exclusively serving a tribal population;
  >    Individual home plumbing except when retrofitting for water efficiency;22
  >    New systems, except when an existing water source for a population is  contaminated or a public health
      concern, and a new system is a cost effective solution for addressing the public health risk;
  >    Individual home wells;
  >    Projects for systems that are for-profit non-community water systems; and
  >    Other ineligible projects listed in DWIG-TSA national guidance (e.g., reservoirs, dams, new growth, O&M).

The CWISA program cannot fund the following kinds of projects (per the CWISA national guidance, except where
footnoted):
  >    Indoor plumbing in individual homes;23
  >    Connection of individual homes to laterals (i.e., the connection from the home to the property line where a
      lateral connection exists);24
21 For more information, see http://www.epa.gov/indian/trprograms/infra-water.htm.
22 Mehan, G.Tracy. US EPA. Memorandum: "Use of Drinking Water State Revolving Fund (DWSRF) Program Funds for Water Efficiency
Measures." July 25, 2003. Available: http://www.epa.gov/ogwdw/dwsrf/pdfs/memo dwsrf policy  2003-07-25.pdf.
23 US EPA Office of Water. Clean Water Infrastructure Set-Aside Grant Program: Answers to Frequently Asked Questions. March 2007.
Available: http://water.epa.gov/type/watersheds/wastewater/upload/CWISA-tribal-faq-highres.pdf.
24 Frace, Sheila, US EPA. Memorandum: Eligibility of Lateral Connections under the Clean Water Indian Set-Aside Grant Program. September
18, 2007.
Evaluation Report - March 2011                                                                             Page 5

-------
    EVALUATION OF Til	IE                AND             INFRASTRUCTURE TRIBAL SET-ASIDE

      Projects not located within the jurisdiction  of the tribal recipient  (i.e., tribal reservation land) with the
      exception of Alaska Native Villages and tribes in Oklahoma on former reservations;
      Land purchases other than land used for the integral part of the wastewater treatment processes or used
      for the ultimate disposal of treatment residues; and
      Other ineligible projects listed in CWISA national guidance (e.g., new growth, economic development, O&M,
      industrial wastewater treatment systems, asset management).
Role of IMS in EPA Infrastructure Grant Programs
To implement these  infrastructure grant programs, EPA works in conjunction  with  IHS's Division of Sanitation
Facilities Construction (SFC).  The mission of the IMS is to work with AI/AN peoples to raise their health status to
the highest possible level. To support IHS's mission, IMS SFC works with tribal governments to eliminate sanitation
facilities deficiencies. Through SFC, IMS provides the  following services related to the EPA tribal infrastructure
programs.
      Sanitation Deficiency System (SDS) List. IMS develops and maintains an inventory of sanitation deficiencies
      in tribal communities, also  called the SDS  List,  which is a part of the Sanitation Tracking and Reporting
      System (STARS).  EPA uses this list to make funding allocation decisions for both the DWIG-TSA and CWISA
      programs. This list also guides project selection for the CWISA program and is used by some EPA regions for
      DWIG-TSA project selection. Exhibit 2 provides definitions for deficiency level designations that IMS uses.
      Interagency Agreements (lAs).  In  cases  where a  tribe chooses to have  funds awarded  through an
      interagency agreement (IA), EPA works with IMS  to set up the IA, and then IMS establishes a Memoranda of
      Agreement (MOA) with the tribal  grant recipients.   Under the ARRA funding of 2009, all projects  were
      funded through two national-level lAs with IMS: one for the DWIG-TSA program, and one for the CWISA
      program.
      Coordinated Project Funding. Often, IMS provides a proportion of funding for projects that EPA chooses to
      fund.
      Project Implementation.  Whether or not IMS administers the project funds through an IA, the tribes can
      use IMS to  implement projects. IMS has staff with  technical  expertise for sanitation and drinking water
      systems,  and can provide assistance with  feasibility studies, design, and construction of infrastructure
      projects.
      Program Reporting.  IMS annually provides data from the STARS system to EPA on progress made for EPA to
      report on its strategic measures.  IMS also responds to ad hoc requests for additional information  from
      STARS to answer questions on performance and implementation that arise.
Evaluation Report - March 2011                                                                          Page 6

-------
               OF THE                AND
                                Exhibit 2. IMS Deficiency Level Designations
    A home  that lacks access to safe drinking water or  basic sanitation is one that is defined as having a
    deficiency level (DL) ranking of 4 or 5 (on a 1-5 scale) by IMS in the STARS system.  Deficiency levels have
    the following general definitions:
    >   DL-5: An Indian tribe or community that lacks a safe water supply and a sewage disposal systems.
    s   DL-4: An Indian tribe or community with a sanitation system that lacks either a safe water supply or a
        sewage disposal system.
    s   DL-3: An Indian tribe or community with a sanitation system which has an inadequate or partial water
        supply  and  sewage disposal facility that does  not comply with all applicable  water supply  and
        pollution control laws, or has no solid waste disposal facility.
    >   DL-2: An Indian tribe or community with a sanitation system which complies with all applicable water
        supply and pollution control laws, and in  which the deficiencies relate to capital improvements  that
        are necessary to  improve the facilities in order to meet the needs of such tribe or community for
        domestic sanitation facilities.
    s   DL-1: An Indian tribe or community with a sanitation system which complies with all applicable water
        supply and pollution control laws, and in which the deficiencies relate to routine replacement, repair,
        or maintenance needs. A minimum level of technical assistance is required from the IMS.
    s   DL-0: No deficiencies to correct.
Evaluation Report - March 2011                                                                            Page 7

-------
              OF THE               AND
METHODS
Evaluation Approach

As a first step in the evaluation, the evaluation  team refined the initial evaluation questions that  had been
submitted  as  a part of OP's evaluation competition, identified the information needed to answer the refined
questions, and updated the draft program logic model (Appendix A) that had been previously developed by EPA.

Identification of Evaluation Questions
The evaluation team, with input from a small EPA evaluation advisory group, decided to focus on the following five
core evaluation questions:
                                  Exhibit 1. Core Evaluation Questions
   I.    How does EPA implement the DWIG-TSA and CWISA programs at headquarters and in the regions?
   II.    To what extent are the DWIG-TSA and CWISA programs achieving their stated goals?
  III.    What factors influence whether the programs achieve their stated goals?
  IV.    Are the current DWIG-TSA and CWISA program performance measures accurate indicators of EPA's
        progress?
  V.    What implementation improvements or innovations could be made by EPA to make the DWIG-TSA and
        CWISA programs more effective in meeting the water and wastewater infrastructure needs of tribes?
The core questions focused on  how the EPA headquarters and  regions implement the programs, whether the
programs were and are meeting their stated goals as articulated through the program strategic measures (and if
not, why not), and whether EPA could make program adjustments to maximize program effectiveness and success.
These core evaluation questions informed data collection and analysis approaches.

Information Collection Approach

The evaluation team employed a mixed-method approach to information collection, including document review,
interviews with stakeholders, and new data collection and synthesis.

Document Review
The evaluation team reviewed relevant EPA strategic reports and guidances, EPA OW NPM guidances, EPA regional
guidances and grant requests for proposals, IMS annual reports and other documentation, and Infrastructure Task
Force reports. A full list of information sources reviewed is included as Appendix F.

Interviews
Interviews were a crucial component of the qualitative data collection  for the evaluation, providing information,
insights, and perspectives that complemented and reinforced  the other information. The evaluation team first
conducted an initial set of scoping calls with 13  EPA regional contacts during the development of the evaluation
methodology to inform the evaluation approach. Over the course of the evaluation, the evaluation team gathered
Evaluation Report - March 2011                                                                        Page 8

-------
    EVALUATION OF THE                AND             INFRASTRUCTURE TRIBAL SET-ASIDE

information from 39 people during 28 separate interviews; these interviewees were from EPA headquarters and
regions, IMS headquarters and area offices, and tribal governments.  The evaluation team interviewed almost all of
the EPA regional coordinators across the nine regions that receive funding for these programs25 and a few other
regional staff members who were also familiar with these programs or had a related role in their regions.  EPA
headquarters staff interviewees consisted of DWIG-TSA  and CWISA program  managers and individuals with
oversight roles or knowledge of program measures and reported performance. IMS interviewees consisted of two
managers at IMS headquarters and three IMS area office staff members (IMS headquarters staff suggested who to
interview from the  IMS area offices). Tribal interviewees were tribal government staff familiar with these programs
(EPA regional coordinators suggested the tribal interviewees).

The roles of the individuals interviewed break down as follows:
      EPA regional staff: 20
      EPA headquarters staff: 10
      IMS staff: 5
      Tribal government staff: 4

Please see Appendix B for the interview guides used to structure the interviews.

Quantitative Data Synthesis and Analyses
The evaluation team collected data from a variety of sources:
      Historic DWIG-TSA and CWISA data.  EPA provided  historic program data in  Microsoft Excel spreadsheets
      that contained  data,  where available, on EPA funding amounts, funding years,  IA numbers, and  in some
      cases,  IMS Sanitation  Tracking  and Reporting System  (STARS) database identifiers  such  as  Sanitation
      Deficiency System (SDS) and Project Data System (PDS) numbers.
      Safe Drinking Water Information System (SDWIS) compliance data. This data included SDWA compliance
      violation information for tribal public water systems (PWS) for violation  years (calendar years (CY)) 2004 to
      2008, in Excel spreadsheets.  The evaluation team also received  summary SDWIS information in Excel pivot
      tables that is used for EPA reporting on its strategic measures  related to compliance.  These pivot  tables
      include data  as far back as 1998, but do not include specific violation information at the system level.
      IMS STARS data. IMS provided STARS data (in Microsoft Access format) that covered data from three areas
      of STARS:
            Community deficiency profile (CDP)  'snapshots' for 2004-2009:  Each  community served by  IMS  is
            evaluated annually by IMS engineers.  Each  profile for a community includes groups  of homes
            designated with a deficiency level for solid  waste, drinking water, and  sewer.  This home and
            deficiency level data was used to report on EPA's access strategic  targets in the FY2006-2011 EPA
            Strategic Plan.
            SDS List 'snapshots' for 2004-2009: Each year's 'snapshot' shows the SDS projects listed as  needs
            within a given  IMS area,  its area priority within the list in that year, the estimated initial deficiency
  EPA Region 3 has no federally recognized tribes within its boundaries and therefore does not receive funding or otherwise implement these
programs.
Evaluation Report - March 2011                                                                           Page 9

-------
    EVALUATION OF THE                AND             INFRASTRUCTURE TRIBAL SET-ASIDE

            level (IDL)  of the project, the estimated final deficiency level (FDL) of the project, and the scores in
            eight areas that contributed to the project's priority level within the area.
            PDS information: This information includes data on all projects where funding is directed through IMS,
            including IDL  and FDL of the project, number of homes served, funding  amount, and milestone
            information (e.g., date when construction  started). This is likely to be the data source for reporting on
            EPA's access strategic measures in  the FY2011-2015 EPA Strategic Plan. This  data was a 'snapshot'
            from 2009.

To best use this information, the evaluation team worked with EPA headquarters and EPA regional coordinators to
review the EPA project  data and fill in gaps (e.g., to add PWS identification numbers  (PWSIDs) and PDS numbers),
where possible. After the  evaluation team completed data cleanup, lEc developed an Access database ("project
database") that linked, as possible, EPA project data from both programs with IMS STARS information. The project
database also linked EPA  DWIG-TSA  project information  to  the detailed SDWIS violation data, by PWSID, as
possible.  Once connections between the datasets were made, the evaluation team  was able to  conduct data
analyses to help answer certain evaluation questions.  The evaluation team is  providing  this project database to
EPA at the end of this evaluation, along with a memorandum describing the database and how EPA could update
the database with new project data, STARS data, and/or SDWIS data.

The evaluation team conducted a  series of iterative data analyses, including statistical correlations (Appendix D)
and analyses of project performance information (e.g., project duration,  IDLs  and  FDLs of projects funded)
(Appendix E).  In the evaluation methodology, the evaluation team had  considered a regression analysis, with the
dependent  variable  being  the STARS  CDP deficiency level in  a set of community homes,  and the independent
variables including EPA  region,  IMS area, project duration, project funding, available  compliance information, and
other factors that may influence a community's  access deficiency  designation.  However, the correlation analyses
conducted  indicated a  lack of evidence  to suggest that a more sophisticated regression analysis would yield
meaningful results.  This conclusion was further supported by the evaluation team's increased understanding of
the lack of available data on many of the independent variables of interest.

Data analysis spanned projects from both programs that were funded between calendar years 2003 and 2009.  In
this evaluation, we refer to "funding years", which we  define as the calendar year when  a project was funded by
EPA, and except where noted, this unit is the timescale for analyses.  This designation is different from when
regions received funding from headquarters, which occurs on  a federal  fiscal year  basis.  It should be noted that
projects funded under the American Recovery and Reinvestment Act of 2009 (ARRA)  are analyzed separately from
the "routine" program implementation captured in the  2003-2009 time scope.

Data Analysis Limitations

The data available for  this evaluation had several gaps, and, as is typical of most  quantitative  analyses, the
evaluation team made assumptions in order to proceed with particular data analyses.  The evaluation team does
not think that either the gaps in the data or and the assumptions made have undermined or weakened the findings
and recommendations for this evaluation, though we do provide caveats and other explanations where needed.  A
few notes about data gaps and assumptions are provided below.
Evaluation Report - March 2011                                                                          Page 10

-------
    EVALUATION OF THE                AND             INFRASTRUCTURE TRIBAL SET-ASIDE

Lack of Linkages to STARS orSDWIS Data
      Even though the evaluation team, EPA headquarters staff, and EPA regional coordinators reviewed the EPA
      DWIG-TSA and CWISA project data, some gaps still persisted in  the  EPA project data, such as missing
      PWSIDs, IA numbers, and PDS or SDS numbers that would allow linkages between the EPA and IMS STARS
      information or SDWIS information. In particular, 51 of the 330 DWIG-TSA projects (15 percent) provided
      with funding between 2003 and 2009 (including ARRA projects) either lacked a PWSID or the PWSID could
      not be linked to provided SDWIS data.
      For CWISA projects, information on when funds were awarded to a tribe was not provided.  Therefore the
      evaluation team, in consultation with EPA headquarters staff, assumed that the fiscal year in which funds
      were appropriated was the same as the year in which a project was funded.
      Under both programs, projects funded by direct grant  (13.5 percent of  DWIG-TSA projects and 4.9 percent
      of CWISA projects, between 2003 and 2009) rather than through an IA with IMS often did not include PDS
      information available from IMS (e.g., IDL, FDL, number  of homes served, milestone information), unless IMS
      was involved in implementation of the tribe's grant or otherwise tracking the  project. The data analyses
      that reflect progress toward the access measures therefore did not cover all projects funded through the
      programs.
  Limited SDWIS Data
      The SDWIS dataset provided to the evaluators contained only violations that occurred for the first time
      between 2004 and 2008.  It did not contain ongoing violations that occurred prior to 2004, so the data may
      be missing some information relevant to  systems provided funding under the DWIG-TSA program.  In
      addition, as only 2004 to 2008 violation data were included, this evaluation did not have information on pre-
      2004 compliance history for EPA-funded systems or more recent compliance information.
      For analyses related to the SDWIS data, if a  project was funded the same year as when a violation occurred,
      the analyses assumed that violation occurred before funding.
Evaluation Report - March 2011                                                                         Page 11

-------
              OF THE               AND
FINDINGS:  PROGRAM  IMPLEMENTATION
Core Evaluation Question I: How does EPA implement the DWIG-TSA and CWISA programs
at headquarters and in the regions?

This section explores core evaluation question I, including how the programs allocate funds, how communication is
working within EPA and with tribes and IMS, and project selection under both programs. In addition, this section
highlights differences in implementation under the 2009 one-time ARRA funding.

EPA Headquarters Implementation of Tribal Infrastructure Programs

EPA  headquarters administers these tribal  infrastructure programs in two offices within  EPA Office of Water:
OGWDW  for the DWIG-TSA program and OWM for the CWISA program.  In addition to creating guidance for
implementation,  EPA  headquarters is responsible for  program oversight and ensuring that the programs are
implemented  effectively.    EPA  headquarters  manages  the allocation  of annual  funding to the  regions,
communicates with EPA regions regarding  program  implementation, monitors program implementation in the
context of the programs'  strategic measures, and coordinates with  IMS headquarters as needed on program
coordination and oversight. For ARRA-funded projects, EPA headquarters staff are also the project officers for the
lAs with IMS, whereas for the annual appropriations, EPA regions staff are the project officers.

DWIG-TSA Allocation of Funds

Each of the nine regions with federally recognized tribes (all EPA regions except Region 3) receives a two percent
base allocation from the annual DWSRF allotment, regardless of the extent of identified infrastructure needs in
each region. EPA headquarters allots the remaining 82 percent of the funds to the regions based on the identified
needs in the IMS area  SDS lists and the EPA DWINS. The national guidance for the DWIG-TSA program describes
why this allocation method was developed.
  >    The two percent base allocation was provided to, "allow for at least one viable project in each Region,  each
      year, and thus allow every Tribe to at least have the opportunity to receive a  grant from the DWIG TSA
      funds."26
  >    For using both the SDS list and DWINS, the guidelines acknowledge that neither tool is perfect, but that  both
      are used to balance out each tool's weaknesses.

Overall, the DWIG-TSA program stakeholders interviewed  indicated that they understood the reasons for the
design  of the  allocation formula, but many  shared some concerns with the formula.  In particular, interviewees
noted that EPA has not updated the DWINS for tribal systems since 1999 and thus the allocation is not based on
current need in Indian country. EPA is currently updating the tribal drinking water needs information through the
2011 DWINS,  but the survey results will  not  be available  until approximately 2013.2S   In addition, a  few
26 US EPA. Drinking Water Infrastructure Grants, Tribal Set-Aside Program: Final Guidelines (October 1998). Available:
http://www.epa.gov/safewater/dwsrf/allotments/tribes/pdf/guidelines dwsrf tribal.pdf.
"ibid.
28 Dougherty, Cynthia, US EPA. Letter to Tribal leaders on 2011 Drinking Water Infrastructure Needs Survey, September 15, 2010.
Evaluation Report - March 2011                                                                       Page 12

-------
    EVALUATION OF THE                AND             INFRASTRUCTURE TRIBAL SET-ASIDE

interviewees noted that the two percent base allocation to each region means that a region may receive funds but
not have any significant access needs or compliance issues to be addressed.

Interviewees also noted  that the IMS updates the SDS list annually, but that it may not be a complete list of all
drinking water infrastructure needs.  In particular, interviewees noted that IMS does not systematically track SDWA
compliance issues on the SDS list, and thus compliance issues that may contribute to public health impacts may
not be captured in the SDS list.  In addition, a few interviewees said that some tribes  do not work with IMS to get
their needs evaluated and listed, or that IMS may not be aware of all the specific needs  in some tribal communities.

The IMS SDS list also contains drinking water needs that are not eligible for DWIG-TSA funding, such as individual
drinking water wells.  However, projects that are not eligible for DWIG-TSA funding in the SDS list still contribute to
the overall drinking water need in tribal communities and thus the allocation of resources to the EPA regions.
Further, the SDS list includes projects that address  needs at all deficiency  levels.  Though EPA regions can fund
projects that address lower deficiency needs (i.e., DLs 1-3), the DWIG-TSA program progress is partly measured by
the number of homes provided access to safe  drinking water,  as defined from moving from a DL-4 or 5 designation
to a lower level. If an EPA region had many projects on the SDS list addressing lower deficiency needs that resulted
in a high dollar amount of need,  EPA would allocate proportionally more funding to this region, though these
needs do not align with meeting the DWIG-TSA strategic measure.

CWISA Allocation of Funds

EPA allocates  the annual CWISA  funding based on the  percent  of needs of individual IMS areas  as they  are
identified in the SDS list.  Each year, IMS provides the area SDS lists to EPA with all  of the sewer need infrastructure
costs summed in each IMS area; EPA headquarters allocates  the funds based on the  proportion of funding need
across  all IMS areas.  Allocation by IMS area need means that some EPA regions may not fund  any clean water
projects in a given year, depending on where  a region's projects fall on that year's IMS area SDS list. As some IMS
areas cover multiple EPA regions, one EPA region's clean water needs might all rank higher than  another region's
needs within the given funding allocation, and thus all the funding allocated to an IMS  area would go to projects in
the one EPA region.

For the CWISA allocation formula, some regional interviewees said that the allocation formula should be by EPA
region  rather than IMS area so that at least one clean water project could be funded in an EPA region every year.
However, other regional and EPA headquarters staff indicated that, as long as the funding is addressing the highest
priority needs throughout Indian country, the allocation formula does not need to  change.

Like the DWIG-TSA program, the SDS list includes projects that are not eligible for CWISA funding, such as projects
for individual  home  plumbing or projects that are not on  tribal reservation land.  However, projects  that are  not
eligible for CWISA funding in the SDS list still contribute to  the overall wastewater  infrastructure need and thus the
allocation  of funds from  this program.   Due to the inclusion of these non-EPA eligible projects, funding to a
particular IHS area would not align well with the CWISA program if many projects in the year's list were not eligible
for CWISA funding.

In addition, the SDS list includes projects that  address needs at all deficiency levels. Though EPA regions can fund
projects that address lower deficiency needs (i.e., DLs  1-3),  the CWISA program is measured by the number of
homes provided access, as defined by moving from a DL-4 or 5 designation to a lower level.  If an IHS area had
many projects addressing lower deficiency  needs that resulted in a high dollar amount of need, EPA would allocate
Evaluation Report - March 2011                                                                          Page 13

-------
    EVALUATION OF THE                AND             INFRASTRUCTURE TRIBAL SET-ASIDE

proportionally more funding to this area, though these needs do not align  with  meeting the  CWISA strategic
measure on providing access to basic sanitation.

Allocation Between Programs

A few regional coordinators  from both infrastructure programs and a tribal  government  interviewee indicated
interest in being able to shift funds between the CWISA and DWIG-TSA programs  to address the overall higher
priority infrastructure needs that may exist in a region within a given year.  For example, a region may have many
drinking water projects  that  have pressing public health concerns that  could be addressed by infrastructure
funding, but not enough DWIG-TSA funding to address the issues in that year.

Language that would allow this transfer of funds is in EPA's proposed FY2011 budget,  but as of this reporting
writing,  EPA does not yet know whether the US Congress will pass this language. If the US Congress approves the
language, EPA headquarters will work with the regions to establish how this transfer could be conducted.29

EPA Headquarters Communication and Coordination with EPA Regions

EPA headquarters has oversight responsibilities for the implementation  of the infrastructure  programs which
includes  summary  and  analysis  of fund  utilization  and annual impact  on strategic  measures, and  the
communication of these findings to interested senior leaders within EPA  and other federal managers (e.g., the
White  House's  Office of Management  and  Budget (OMB)). To fulfill these  duties,  EPA headquarters must
coordinate and communicate with EPA regional staff on project selection and implementation.

For both grant programs, interviews with  EPA staff indicate that, with the exception of ARRA funding, EPA regional
coordinators  and EPA headquarters staff communication about project selection and  progress is  ad hoc  and
project tracking is inconsistent. As a headquarters staff member said, "communication [about project selection] is
pretty superficial—we may hear about particular projects retrospectively on a case-by-case basis." A few regional
coordinators  said that the staff turnover at EPA headquarters had been  one contributing factor to the lack of
historic  communication  between headquarters and  the regions.  Some  of  these  coordinators  and others
mentioned  communications improvements in recent years due to more  engaged  and consistent headquarters
staff.

Previously the DWIG-TSA Grant Tracking Tool was used for national project tracking.  However, the DWIG-TSA
regional coordinators said that the tool was  difficult to  use, requested information  available in other  EPA
databases, and that headquarters did not seem to indicate to the regions that  its use was a  priority.  One regional
coordinator said, "Headquarters always talked about wanting updates on the projects and they developed  this
clunky grant tracking tool.  However, they never seemed serious about  wanting the information." (Headquarters
staff did request that regional coordinators use the  tool in annual funding memoranda to the regions.)   Both
CWISA and DWIG-TSA regional staff said  that  they provide  information on  which projects  are  selected to
headquarters, but do not provide other information, unless requested.

Regional coordinators and headquarters staff said that communication  and coordination around these programs
has increased and improved as a result of the ARRA processes (we discuss ARRA implementation in  more depth
later in this report).  However, with this increased communication, some regional coordinators  expressed concern
that EPA headquarters may  want to review regions'  selected projects and potentially  take  away the regional
  D. Harvey & M. Richardson, US EPA Office of Water. Personal communication. December 14, 2010.
Evaluation Report - March 2011                                                                         Page 14

-------
    EVALUATION OF THE                AND             INFRASTRUCTURE TRIBAL SET-ASIDE

autonomy in project selection.  One regional coordinator said, "We think headquarters is going toward the idea
that they want to decide what projects we fund and under what circumstances." Headquarters staff, for their part,
said that they want to know more about which projects EPA regions fund to better understand how EPA is meeting
or could better meet its strategic measures, but that they view the programs' regional coordinators as being in the
best position to understand tribal needs and determine project selection.

EPA headquarters staff that work directly on these programs said that they are often asked for reports on program
progress from the OMB, the EPA  Office  of the Chief Financial Officer (OCFO), the US Congress, and  EPA OW
management.  The  lack of consistent and  routine  tracking  and communication  about these  programs and
associated projects makes it difficult for EPA headquarters staff to respond in a timely and useful way to  these
ongoing requests. This dynamic also stymies the ability of headquarters staff to effectively advocate for the DWIG-
TSA and CWISA programs. A headquarters manager noted, "We hear a lot of anecdotal information about what's
going on... we don't have a lot of quantitative information to respond to specific questions about how we can best
protect that investment."

In addition to the issues described by  interviewees, during the course of the evaluation, the evaluators came to
recognize that there is a basic lack of information flowing between the EPA regions  and headquarters, as well as
between the regions. For example, headquarters DWIG-TSA staff did not have a strong understanding of how the
regions select projects, and the staff with both programs did not have a clear understanding of how many staff
resources the regions  devote to these programs.  The  regions also had an inconsistent understanding of each
others' work, including where they  could learn from each other's solutions to particular challenges that have arisen
in multiple regions.  Moreover,  for both programs, there is insufficient information for EPA at both regional and
headquarters levels to  learn about  what is working and not working well program-wide, and to provide a basis for
adapting the programs accordingly.

EPA Headquarters Communication and Coordination with IMS

As already described, IMS administers a large percentage of DWIG-TSA and  CWISA projects through lAs with EPA.
Further, EPA relies on the IMS STARS database to report to OMB, the US Congress, and the public on EPA's own
progress toward its strategic measures. Due to these implementation relationships  and reporting  dependencies,
as well as EPA's program oversight responsibilities, EPA headquarters staff periodically communicate with IMS
headquarters staff to, for example, request updated performance  information from STARS and to track dispersed
IA funds. Finally, IMS and EPA are both part of the Interagency Task Force related to the US access goal for water
infrastructure and work jointly, along with HUD and USDA, to help meet this commitment.

From  discussions with EPA  headquarters  and IMS  headquarters staff, it  appears that communication and
cooperation between the two agencies is working well overall. The ARRA process has led  to more coordination,
such as filling in  funding data  gaps and developing a standard template for  lAs.  Three IMS staff members
mentioned, however, that multiple and duplicative requests for information from EPA can be frustrating and time
consuming, but that this issue seems to be improving due to reduced staff turnover at EPA headquarters.

From interviews with IMS headquarters and area  staff members, there appears to be ongoing questions about the
expectations and  requirements around who is the  "lead" federal agency on  lAs and which agency's rules and
guidelines apply under lAs. One tribal government  interviewee who works closely with these programs said, "It
would be good if EPA would accept IMS as the lead agency. One federal entity needs to follow their [own] rules-
otherwise there are two sets of federal rules [to be followed]." As mentioned, a standard template with terms and
Evaluation Report - March 2011                                                                          Page 15

-------
    EVALUATION OF THE DRINKING WATER AND CLEAN WATER INFRASTRUCTURE TRIBAL SET-ASIDE GRANT PROGRAMS

conditions for all  EPA-IHS  DWIG-TSA and  CWISA  lAs was developed  as  a  result of the ARRA funding;  the
standardized IA may address these questions in the future, but it is too soon for this evaluation to say what effect
it may have both functionally on the implementation of these programs. One IMS staff person also mentioned that
having specific funds from EPA for IMS to administer projects through lAs would be useful.  EPA had previously
provided administrative funds to IMS through the CWISA  program,  but there are  currently no funds  for IA
administration for either infrastructure program.  These IA questions may be one area for continued discussion
and coordination between EPA and IMS headquarters.

Nationally, EPA does not consistently provide data on tribal community water systems' compliance history to IMS
for inclusion into the formula  that IMS uses  to rank projects on the SDS  list.  As discussed later in this report on
DWIG-TSA regional implementation, at a December 2010 meeting of EPA  headquarters and  regional staff, the
DWIG-TSA staff decided that, going forward, regions would more consistently provide compliance information to
IMS area offices to better inform the SDS list.

EPA Regional Implementation: DWIG-TSA Program

The evaluators examined DWIG-TSA regional project selection from three  perspectives: the processes used by each
region to select projects, the role of SDWA compliance history, and provision of access to drinking water through
DWIG-TSA funding.  We address our findings on  each of  these perspectives below, following a brief review of
projects funded and staffing levels in each region.

Table 1 provides the total number of DWIG-TSA projects funded by EPA between 2003  and 2009, excluding ARRA
projects. Of these 267 projects funded between 2003 and 2009, EPA administered 36 projects (13.5 percent of all
projects, representing 13.7 percent of funding) as  direct  grants to the tribes.  EPA has funded the remaining
projects through lAs with IMS.  Additional summary information about these projects, including total and average
funding amounts and number of lAs versus direct grants, is available in Appendix E (Tables 1-5).

                  Table 1. Number of DWIG-TSA Projects, by EPA Region and Funding Year

1
2
4
5
6
7
8
9
10
Total
0
0
1
4
2
2
2
19
16
46
0
1
1
2
3
4
4
16
14
45
4
1
1
4
2
1
6
11
10
40
0
0
1
8
2
3
4
13
7
38
4
2
1
4
3
3
8
6
8
39
3
0
1
3
0
3
7
7
4
28
0
1
1
4
2
2
4
9
8
31
11
5
7
29
14
18
35
81
67
267
Note: This table includes all projects funded by EPA from 2003-2009, excluding ARRA projects, but including direct grants and lAs with IMS.
Some projects are funded multiple times over multiple years under the same IMS STARS PDS number; we consider these separate projects for
this table to be consistent with EPA records. There are 257 unique projects within this timeframe.

As the data above shows, regions with larger tribal populations (e.g., Regions 9 and  10) have funded a higher
number of projects.
Evaluation Report-March 2011
Page 16

-------
    EVALUATION OF THE DRINKING WATER AND CLEAN WATER INFRASTRUCTURE TRIBAL SET-ASIDE GRANT PROGRAMS
                                                                              Table 2. Estimated Level of
                                                                             DWIG-TSA Regional Staffing
Region Staffing (FTE)
1
2
4
5
6
7
8
9
10
0.15
0.25
0.30
0.50
0.15
0.25
1.0
1.0
0.50
In terms  of administering programs,  we  asked  regional  coordinators  to
estimate the  level of staffing  (full-time equivalent,  or FTE) dedicated  to
coordinating the DWIG-TSA program in  their regions. Table 2 provides these
estimates of staffing.  Overall, regions with smaller tribal populations devote
fewer staff to the program.  In addition regional coordinators have other
responsibilities  beyond  the  DWIG-TSA  program,  though  no  regional
coordinators mentioned that having multiple responsibilities hinders their
ability  to  implement  the DWIG-TSA  program.  In Region  10, two regional
coordinators split the DWIG-TSA responsibilities between Alaska and the rest
of the Region  10 states; the one in Alaska works with  the State of Alaska on
compliance issues, as systems at Alaska Native Villages are state-regulated,
rather than EPA-regulated.

DWIG-TSA Regional Project Selection Processes

The national DWIG-TSA guidelines provide regions with  flexibility in most aspects of project solicitation, ranking,
and selection.  However, the national  DWIG-TSA guidelines do require the following of EPA regions in project
solicitation and selection:30
  >   A method for identifying water system projects and prioritizing those projects, and providing this method to
      tribes and other potentially interested  parties in  the region for review and comment;
  >   A quantifiable method of ranking all of the identified projects, with  priority given to projects that address
      the most serious risk to human health; are necessary to ensure compliance with the requirements of the
      SDWA; and  assist systems most in need on a per household basis;
  >   Coordination with local IMS  area office(s) on projects selected, even if the area  SDS list is  not used  for
      project ranking and selection; and,
  >   Determination of adequate capacity at a system  (beginning in FY2011, any system to be assisted with DWIG-
      TSA funds must have a certified operator).

In practice, project solicitation and selection varies widely by region: some regions rely completely on the IMS SDS
list for project selection while others use a  request for proposals (RFP) process for  soliciting projects from the
tribes and then rank those submissions. The table in  Appendix C provides a summary of the different processes
used in each of the regions. Most but  not all regional processes are in accordance with the requirements listed
above from the national guidelines.  From the  information the evaluation team received, it appears that three
regions are not explicit in how they quantify or consider compliance information or public health impacts in project
selection.31 However, the extent to which the regions believe they have discretion in developing processes that fit
individual regional conditions and circumstances, beyond what is outlined in the national guidance, is unclear.
30 US EPA. Drinking Water Infrastructure Grants, Tribal Set-Aside Program: Final Guidelines (October 1998). Available:
http://www.epa.gov/safewater/dwsrf/allotments/tribes/pdf/guidelines dwsrf tribal.pdf.
31 Regions 2, 4, and 7 did not provide any quantifiable guidance to the evaluators on how projects are selected and thus it is not transparent to
the evaluation how compliance or public health impacts are considered in project selection. Both Region 2 and Region 7 indicated that there
has not been a need to choose between projects in a given year during the program history. Region 4 uses a rotating method of awarding
projects to its six eligible tribes, working with the tribes to determine which project(s) to fund in a given year.
Evaluation Report-March 2011
                                                                                                     Page 17

-------
    EVALUATION OF THE                AND             INFRASTRUCTURE TRIBAL SET-ASIDE

DWIG-TSA Project Selection: Compliance History

Several EPA staff expressed a keen interest in improving compliance with the Safe Drinking Water Act (SDWA) and
achieving the associated safe drinking water access performance measure through improved system compliance
with health-based standards.  However,  the  role of  compliance history  and prevention of new  compliance
problems in project selection differs from region to region.  From the regional guidances reviewed, it appears that
all regions consider compliance and/or public health impacts when selecting projects to some extent, though, as
mentioned above, how  regions without quantifiable or written guidance weight these considerations  is not clear.
In addition, compliance  history or public health  impact is not the only factor regions consider in selecting projects;
other factors,  such  as  economic feasibility and tribal  engagement  (e.g.,  interest by the tribe in the project,
contributing funds) and capacity affect project  selection.  The table in Appendix C provides an overview of each
region's project selection process, including how compliance history or public health impacts of the project are
considered.

EPA does not provide data on tribal community water  systems'  compliance history to IMS for inclusion  into the
annual SDS process on  a nation-wide basis, though some EPA regions may share this information  with IMS area
offices.  Further, compliance history, even when  it is considered  in the generation of the SDS list, would not
necessarily result in a high rank for associated projects.  Other factors such economic feasibility and funding from
the tribe affect the ranking of projects and can outweigh any compliance issues. Those regions that select projects
based on their rank order  on the SDS list may therefore not be  selecting projects with the most pressing
compliance needs.  Moreover, information collected through this evaluation, described in more detail in the
section on Program Goals and Measures, suggests that many compliance problems are not driven by infrastructure
deficiencies that can be  addressed with DWIG-TSA funding.

At a December 2010 meeting of EPA headquarters staff and regional coordinators, the DWIG-TSA staff decided
that, going forward, regions would more consistently provide compliance information to IMS area offices to better
inform the SDS list. They also agreed to more routinely and thoroughly analyze their own regional compliance data
to understand the issues underlying  tribal SDWA compliance violations and, based on  this analysis, consider
funding eligible projects that would address these issues.  We support these actions for  increased analysis and
information sharing to help ensure that EPA and IMS are aware of  pressing compliance issues that would be eligible
for DWIG-TSA funding.

As described above, compliance issues at  tribal water systems may not necessarily be addressed  by DWIG-TSA
funding, due to other weighting factors in regions' project selection. To better understand the extent to which
DWIG-TSA projects addressed compliance issues, the evaluators examined funded systems' compliance versus the
universe of non-funded  tribal community water systems (CWSs).  From an analysis of available SDWIS information,
it appears that water systems associated with  DWIG-TSA projects have had more compliance issues than non-
funded tribal CWSs,  suggesting that existing compliance issues may contribute to EPA regional project selection.
As Table 3 shows, a higher percentage of EPA-funded  systems had at least one health-based violation between
2004 and 2008 when compared against non-funded systems.  In  addition, a  higher percentage of  EPA-funded
systems had repeat health-based  violations of the  same type and same  rule between  2004 and 2008,  when
compared to non-funded systems.  Note that with the data available, the evaluators cannot determine what how
many of these violations are infrastructure-related. Table 12 in Appendix E  provides more information about the
common types of violations at EPA-funded systems.
Evaluation Report - March 2011                                                                         Page 18

-------
    EVALUATION OF THE DRINKING WATER AND CLEAN WATER INFRASTRUCTURE TRIBAL SET-ASIDE GRANT PROGRAMS

           Table 3. Compliance Information for Systems in Evaluation Universe (SDWIS: 2004-2008)32
Universe
All EPA Tribal Systems (CWSs)
EPA-funded Systems (CWSs and 12 non-CWSs)
Non-Funded Systems (CWSs)
,of systems
733
249
496
# with at least 1 HB
314 (42.9%)
116 (46.6%)
198 (39.9%)
# with repeat HB
violations
195 (26.6%)
83 (33.3%)
112 (22.6%)
CWSs = Community Water Systems; HB= Health-Based

A requirement for certified operators at public water systems that receive EPA DWIG-TSA funding went into effect
for the FY2011 DWIG-TSA funding cycle.  This requirement may have consequences for which water systems are
eligible for funding, but it is not yet clear what the impacts may be on project selection and to what extent it will
affect non-compliant systems from being able to receive EPA infrastructure funding.

DWIG-TSA Project Selection: Access to Safe Drinking Water

Across the regions, analysis shows that the ability of a project to provide access to safe drinking water (as defined
as a  change from a DL-4 or -5 to  DL-3 (or lower) is one factor in determining if a project is selected, but not the
primary factor.  The evaluators analyzed available information  from STARS to determine the extent to which
DWIG-TSA funding has  addressed these  deficiencies to safe drinking water. From PDS  information, DWIG-TSA
projects addressed the needs of DL-2 and DL-3 homes more than those of DL-4 and DL-5 homes: only 36 percent of
projects addressed DL-4 or DL-5 needs, representing approximately 26 percent of homes served  by the programs.
However, these data do not explain why EPA regions selected certain projects.  For example, an EPA region  may
not have any DL-4 or DL-5 projects to choose from because access is not a  problem in the region or the drinking
water access needs that  do  exist are  not feasible to fund.   For reference, Appendix E (Tables 6-10)  provides
detailed  data  where it was available on the SDS priority level of DWIG-TSA  projects  funded, deficiency level
information from  the SDS list, and the IDLand FDL information of funded projects from PDS.33

EPA Regional  Implementation: CWISA Program

The evaluators examined  CWISA regional project selection from three perspectives: the processes used by each
region to select projects, the specific role of the SDS list in project selection, and whether use of the SDS list  to
select projects has maximized EPA's funding for projects that would  provide access to basic water sanitation for
tribal homes that currently  do not have access.  We share our findings on each  of these perspectives  below
following a brief review of CWISA projects funded by and staffing levels in each region.

Table 4 provides  the total number  of CWISA projects funded by EPA between 2003 and 2009, excluding ARRA
projects.  Of these 429 projects funded between 2003 and 2009, EPA administered 21 projects (4.9 percent of  all
projects; reflecting 2.9 percent of CWISA funding in this time period) as direct grants to the tribes.  EPA funded the
32 Non-funded systems analyses only looked at community water systems (CWSs) to be consistent with the EPA-reported numbers on
compliance. However, EPA DWIG-TSA projects have funded three non-community water systems and nine non-transient non-community water
systems between 2003 and 2009 and analyses on EPA DWIG-TSA-funded systems includes these systems. Unless otherwise specified, when
referring to "systems", we are referring to only CWSs for the non-funded universe and to all funded PWSs for the funded universe. These
analyses do not include the 51 DWIG-TSA projects that did not have PWSIDdata or had PWSIDs that could not be linked to provided SDWIS
data.
33 These analyses examine projects that could be linked to IHS STARS information. For PDS, this information includes 204 of the 257 unique EPA
projects (79 percent). For SDS, this information includes 72 of the 214 projects funded between 2004 and 2009 (34 percent). SDS 'snapshots'
were only available to this evaluation for 2004 to 2009.
Evaluation Report-March 2011
Page 19

-------
    EVALUATION OF THE DRINKING WATER AND CLEAN WATER INFRASTRUCTURE TRIBAL SET-ASIDE GRANT PROGRAMS

remaining projects as lAs with IMS.  Additional summary information about these projects, including total and
average funding amounts and number of lAs versus direct grants, is available in Appendix E (Tables 14-19).

                     Table 4. Number of CWISA Projects by EPA Region and Funding Year
Region 2003 2004 2005 2006 2007 2008 2009 Total
1
2
4
5
6
7
8
9
10
Total
2
0
9
1
5
2
7
52
17
95
2
1
3
5
3
1
7
23
12
57
1
1
0
7
3
0
6
34
9
61
0
1
4
2
3
1
4
20
10
45
1
1
1
1
5
1
4
42
9
65
0
0
1
3
2
1
3
26
11
47
1
1
3
5
3
0
6
35
5
59
7
5
14
24
24
4
36
224
65
429
Note: This table includes all projects funded by EPA from 2003-2009, excluding ARRA projects, but including direct grants and lAs with IMS.
Some projects are funded multiple times over multiple years under the same PDS number; we consider these as separate projects for this table
to be consistent with EPA records. There are 402 unique projects in thistimeframe.

By far,  EPA Region 9 has funded  the most CWISA projects,  followed  by EPA Region 10.  Given that Region 9
includes four IMS area offices  and that funding is  allocated by needs highlighted in  IMS areas, this result is not
surprising.
                                                                          Table 5. Estimated Level of
                                                                           DA/ISA regional Staffing
In terms of administering programs, we asked regional coordinators during
interviews  to  estimate  the level of  FTE  dedicated to coordinating  the
CWISA program in their regions. Table 5 provides these estimated staffing
levels.  Overall, regions with smaller tribal populations have fewer FTE and
staff generally have responsibilities  beyond the CWISA program.  Further,
for  some regions, fewer FTEs work on  the  CWISA program than on  the
DWIG-TSA program, as the staff do  not need to work with tribes to select
projects or there were no or very few projects funded from a region.

CWISA Regional Processes for Project Selection

When compared to the DWIG-TSA program, the national guidelines for the
CWISA program provide less flexibility to regions in selecting projects.  Per
the guidelines, regional  coordinators are expected to  select projects that
are on the SDS list, and fund them in priority order down the list. Regions may skip projects that have costs that
exceed available funding, have funding from another source such as IMS, or are ineligible for CWISA funding (e.g.,
drinking water projects, solid waste projects, or projects not on tribal  land).  Regions may also skip over projects
for other compelling reasons, but must provide an explanation to EPA headquarters for approval; according to EPA
headquarters,  there does not appear  to  have been such a  request within period of time  covered in the
evaluation.34
Region Staffing (FTE)
1
2
4
5
6
7
8
9
10
Negligible
Unknown (did
not interview)
0.15
0.5
0.15
0.10
1.0
1.0
0.50
  Richardson, M., US EPA Office of Water. Personal Communication. February 10, 2010.
Evaluation Report-March 2011
                                                                                                   Page 20

-------
    EVALUATION OF THE                AND             INFRASTRUCTURE TRIBAL SET-ASIDE

According to interviews with regional  coordinators, CWISA project selection at the regions is  conducted  in
accordance  with  the national guidelines, with the  exception  of one region.   The coordinator in this region
described how  the weighting factors that determine ranking on the SDS list mean that projects can rank highly
without  having access  problems, and that the  region  considers  the  EPA strategic  measure on access as a
compelling reason to skip projects on  the SDS  list.  The  coordinator said  the his  region  looks  for planned,
economically feasible  projects  that  will  address  EPA's  strategic  measure  for  access to  basic sanitation
infrastructure, and that, if EPA wanted the regions to  rank straight down the SDS list without using judgment, then
EPA should give the money directly to IMS for it to use.35

Given that, with the exceptions noted above, the CWISA program uses the SDS list in rank order to select projects
for funding, this evaluation looked at the  DLs of funded projects to determine the extent to which funded projects
would provide  access as it is defined as an immediate change in  DLs (from a DL-4 or DL-5 to a lower deficiency
level) upon project funding.  The analysis, summarized in Table 6, shows that the projects funded by CWISA range
widely across IMS  areas' SDS priority listing.  For example, between 2004 and 2009, the CWISA program has funded
projects  in the  Nashville Area offices with priority ranking between 1 and 13,  which represents projects falling
within the top 7 percent of needs in the area.  In the Navajo Area office, the CWISA program has funded projects
ranking between 1 and 546, with the latter falling in the bottom quartile of SDS projects ranked in that area in that
year.  The percentiles (listed in parentheses after the minimum and maximum ranking) are more illustrative than
are the absolute ranking of the projects  selected because some are SDS lists are substantially longer than others.
(Note that the EPA regional rankings are listed with each related IMS area office because each IMS area has its own
SDS list from which the EPA regions choose from.  In some cases, EPA regions are choosing from multiple IMS area
SDS lists because the regional office-area office boundaries overlap.)

This analysis shows that, on average, most of the EPA regions are selecting CWISA projects from  within  the top
quarter of the area SDS  priority lists. In  some cases,  a few regions chose projects that were not ranked highly on
the SDS lists; however, the data do not explain why the regions chose or skipped specific projects in any given IMS
area.  For example, an IMS area may already be funding highly ranked projects with IMS funds. Area office lists may
also have a large number of projects that are drinking water-related instead of sewer-related, are too expensive to
be funded with available EPA funds, or are otherwise ineligible for funding, forcing the EPA region  to skip many
projects on the  list to find an eligible one. Appendix E (Tables 21-22) provides additional data on information from
the SDS lists.36
  Poloncsik, S., US EPA Region 5. Interview and personal communication. August 26, 2010 and January 11, 2011.
36 These analyses examine projects that could be linked to the IMS SDS list, which is approximately 82 percent of the EPA, non-ARRA, projects
funded between 2004 and 2009 (255 out of 310). SDS 'snapshots' were only available to this evaluation for 2004 to 2009.
Evaluation Report - March 2011                                                                            Page 21

-------
    EVALUATION OF THE DRINKING WATER AND CLEAN WATER INFRASTRUCTURE TRIBAL SET-ASIDE GRANT PROGRAMS

 Table 6. IMS Area Priority Ranking of CWISA Projects for Funding Years 2004-2009, excluding ARRA projects, by
                                          IMS Area and EPA Region
IMS Area/ EPA Region # of Projects Minimum Maximum Average SDS Median SDS
Funded Rank/Percentile Rank/Percentile Rank Rank
Aberdeen
Region 7
Region 8
Alaska (RIO)
Albuquerque
Region 6
Region 8
Bemidji (R5)
Billings (R8)
California (R9)
Nashville
Region 1
Region 2
Region 4
Region 6
Navajo (R9)
Oklahoma
Region 6
Region 7
Phoenix
Region 8
Region 9
Portland (RIO)
Tucson (R9)
Overall
9
0
9
37
8
7
1
19
12
28
18
3
4
9
2
107
9
6
3
14
0
14
6
10
277
2 (1.0%)
n/a
2 (1.0%)
1 (0.2%)
2 (0.9%)
2 (0.9%)
8 (4.1%)
11 (6.1%)
1 (0.7%)
1 (0.4%)
1 (0.5%)
2 (1.3%)
1 (0.6%)
1 (0.5%)
1 (0.5%)
1 (0.1%)
4(1.2%)
4(1.2%)
15 (4.6%)
4(1.6%)
n/a
4(1.6%)
5 (2.5%)
1 (0.7%)
1
12 (5.7%)
n/a
12 (5.7%)
152 (27.2%)
21 (9.8%)
21 (9.8%)
8 (4.1%)
152 (66.4%)
34 (24.0%)
120 (40.4%)
13 (6.5%)
6 (3.0%)
9 (5.1%)
13 (6.5%)
9 (4.6%)
536 (76.4%)
36 (11.0%)
27 (8.3%)
36 (11.0%)
109 (42.7%)
n/a
109 (44.0%)
92 (46.9%)
40 (26.0%)
536
6.89
n/a
6.89
20.00
8.75
8.86
8.00
79.89
6.92
19.36
5.33
4.00
5.00
6.00
5.00
67.35
18.44
14.83
25.67
25.64
n/a
25.64
22.33
10.70
40.06
7
n/a
7
15
8
8
8
79
4
11
4.5
4
6
5
5
20
16
14
26
11.5
n/a
11.5
8
2
14
Note: The above is the minimum, maximum, average, and median priority rankings, by area, for all EPA projects funded between 2004 and
2009, excluding ARRA. Forthe minimum and maximum, the percentile ranks within that year's area SDS list is in parentheses. This information
is presented by area, as SDS lists are by IMS area, rather than in EPA regions. Where multiple EPA regions work with an IMS area, the region's
specific information is presented. Percentiles are not provided for median and average numbers, as these are a result of multiple years of data
and area SDS lists vary in length each year.

In short, choosing projects in rank order from the area office SDS lists is not a straight-forward exercise for EPA
regions; and based on this analysis alone, it is not clear whether this process supports EPA's access strategic
measure for basic sanitation.

CWISA Project Selection: Access to Basic Sanitation

The existing deficiency level (called "initial deficiency level" or IDL in STARS) is one of eight criteria that determines
a project's SDS ranking.  Given the EPA strategic  measure related to providing access to basic sanitation (moving
from a IDL-4 or IDL-5  to a lower final deficiency level (FDL)), the evaluators examined the IDLs and FDLs of selected
projects in the SDS list to understand if access to basic sanitation as EPA currently defines  it was supported in
practice. Analysis of SDS data showed that a majority of EPA CWISA projects funded (60.0 percent) were "access"
projects, with IDL-4 and DL-5 designations and FDL-3 or lower designations; however, the other 40.0  percent of
projects addressed IDL-2 and IDL-3 needs or would not  be expected to provide access (i.e., IDL and FDL of 4).  No
Evaluation Report-March 2011
Page 22

-------
    EVALUATION OF Til	IE                AND             INFRASTRUCTURE TRIBAL SET-ASIDE

DL-1 projects were  funded.37  A further analysis showed that this breakdown of IDLs is similar to the overall
deficiency designations of the top ranked projects across the IMS area SDS lists, which  indicates that top-ranked
projects within an SDS list are not necessarily addressing DL-4 or DL-5 needs. Appendix E provides tables related to
these analyses (Tables 21-22).

The Project Data System (PDS) is a part of STARS that contains information on funded and implemented projects,
and when looking at EPA-funded projects in PDS, we were able to find slightly more and different information than
when looking  at the SDS list information.  In PDS, of all homes in communities  that received CWISA funding
between 2003 and  2009, only 16 percent were DL-4  or DL-5 homes, even though a majority of projects  (56
percent) were addressing DL-4 and DL-5 needs. This finding suggests that many CWISA projects address the needs
of a small number of homes with high access needs. These analyses and related tables from PDS information  are
available in Appendix E (Tables 23-24).

We note that, until FY2008, the CWISA annual funding memos to the EPA regions directed the regions to consider
homes lacking access to basic sanitation as those with DLs of 3, 4, or 5,  rather than just DLs of 4 and 5.  EPA
headquarters changed this direction  in FY2008 to be consistent with the FY2006-2011 EPA strategic measure on
access  to basic sanitation.  Though  regions are expected to fund in order down the SDS lists for the CWISA
program, rather than by DL designation, it is possible that this change in how homes are defined  as lacking access
has changed project selection in terms of the deficiency levels of projects selected.  However, with only two years
of data since EPA changed the definition  of "access" to  exclude DL-3s, this evaluation cannot definitively say that
there has been a change in project selection related to access due to this change.

Regional Communication with IMS, Tribes, and Other Parties

EPA regional program staff coordinates with IMS and tribes in several respects, to promote awareness of available
funding, to manage project selection, and to track project implementation and progress.  The evaluators explored
how communication and coordination are working, and whether improvements in these areas would be helpful to
better meet the needs of the tribes.

Communication with IMS Area Offices

Based on the interviews, communication between the EPA regions and  IMS area offices on the whole appears to be
functioning fairly well, with only a modest number of issues raised by interviewees in both agencies. None of  the
EPA  regional coordinators from either infrastructure program  mentioned major issues associated with working
with their local IMS area offices when selecting projects or getting updates on projects funded.  However, during
both the interviews and  initial  evaluation scoping calls, some regional  coordinators indicated  that  IMS is
understaffed given the  level of work underway and ongoing need  in Indian country.  Regional coordinators
indicated that this understaffing can contribute to some delays, including lag time in closing out projects. During a
few  interviews, staff in both agencies  indicated that their working  relationships were  better with some  EPA
regions/IMS area offices than others, as is typical of most working relationships between different organizations
and their respective offices.  For the DWIG-TSA  program, EPA had not been consistently sharing data on SDWA
compliance with IMS for inclusion in the SDS list, but as  discussed earlier,  EPA DWIG-TSA regional coordinators
made a commitment in  December 2010 to communicate this information going forward.
  These analyses examine projects that could be linked to the IMS PDS list, which is approximately 85% of the unique EPA, non-ARRA, projects
funded between 2003 and 2009 (343 out of 402).
Evaluation Report - March 2011                                                                          Page 23

-------
    EVALUATION OF Til	IE                AND CLEAN       INFRASTRUCTURE TRIBAL SET-ASIDE

Communication with Tribes

For both programs, the level  and type  of communication with tribes varies across the regions, but is generally
more limited for the CWISA program.

For the DWIG-TSA program, interviews indicate that some regions with smaller tribal populations, such as Region
2, often rely on IMS to communicate more of the information about the program, while other regions often work
directly with tribes on a regular basis to communicate about this program. Region 4, for example, has a rotating
process with the tribes, which necessitates regular communication with its six  eligible tribes. Some regions, such
as Region 6, communicate with the tribes through notifications of funding or  RFPs.  In some cases, regions also
communicate information about the program and available funding through other tribal work or meetings within
the regions (e.g., regional meetings or announcements about all  EPA grant funding); regional coordinators often
have other communication avenues with tribes beyond these infrastructure grant programs.

Beyond the DWIG-TSA program, EPA's Public Water System Supervision (PWSS) grant program has funded "circuit
rider" positions where staff work for IMS and provide direct drinking water technical assistance to tribal systems.
As of FY2010, EPA had funded ten FTEs at IMS area offices, across six EPA regions to provide this support.  One
regional coordinator and an IMS area staff member noted that these positions are helpful for understanding tribal
drinking water needs and compliance issues, and for providing technical support to tribal communities that need it
the most.

For the CWISA program, communication with  the  tribes  is generally limited to discussions on the two available
funding mechanisms, typically after projects have been selected.  This difference may be due to the nature of the
project selection; as CWISA projects are selected directly from the SDS list, tribes must work with IMS to get needs
identified and listed on the SDS list, rather than discussing project ideas with  EPA regions  as they might through
the DWIG-TSA program.

Interviews from across all  the interview groups for this evaluation indicated  that direct, in-person  engagement
between tribes and EPA regions (as well as EPA headquarters) is important for understanding tribal context and
needs, as well  as progress on  projects underway and the benefits of completed projects. One tribal interviewee
noted that a recent field visit by the EPA Administrator and EPA staff to a  remote area showed a positive level of
engagement that they had not seen in other federal agencies. An  IMS area  staff member said that EPA staff should
be able to see the projects that they are funding to better understand progress  and needs. In regions with smaller
tribal populations or where the regional coordinator works on other tribal programs, the coordinators may be able
to regularly visit project sites.  However, with limited funding available for travel in most years, regions with larger
tribal populations or those with remote project sites may not have this opportunity.

Communication with Other Parties

Overall,  the regional staff have limited  or no  interaction with agencies or parties other than IMS and the tribes
when supporting these programs.  In some  cases, USDA or HUD may also be contributing funding to a project, but
any needed coordination happens with IMS, rather than with  EPA.  As discussed, the Tribal Water infrastructure
Task Force is composed of four federal partners (EPA, IMS, USDA, and HUD) and some strategic coordination takes
place within this group.38 We do not have sufficient information upon which to  base any particular findings on this
topic.
 ! For more information, see: http://www.epa.gov/indian/trprograms/infra-water.htm.
Evaluation Report - March 2011                                                                          Page 24

-------
    EVALUATION OF Til	IE                AND             INFRASTRUCTURE TRIBAL SET-ASIDE

Role of IMS in  Program Implementation

EPA relies heavily  on IMS for program implementation.  EPA uses STARS data to select CWISA projects and some
DWIG-TSA projects and to report on EPA's strategic access measures.  In addition, much of the programs' funds
have been provided to  IMS through lAs to administer projects on behalf of the tribes. Implementation of both
infrastructure programs is largely in IHS's realm, given that over 86 percent of DWIG-TSA projects and 95 percent
of CWISA projects funded between 2003 and 2009 were implemented under lAs with IMS.  In this section, we
explore IHS's role and how it influences EPA's own program implementation and influence.

IMS STARS Database
As already noted,  EPA uses the IMS STARS database to select CWISA projects, select DWIG-TSA projects in some
EPA regions, track performance on the EPA strategic measures related to access, and track progress for projects
funded through  lAs.  All of these functions are vital for the implementation of these programs.

STARS is considered by the Infrastructure Task Force and other stakeholders to be the best source of information
on tribal needs  related to access to drinking water and sanitation.39 However, interviewees who participated  in
this evaluation and the Infrastructure Task Force have noted issues with STARS, summarized below, in terms of its
use for EPA implementation of the DWIG-TSA and CWISA programs.
      STARS does not include all tribal homes or needs. STARS does not include all  tribal water infrastructure
      needs in  the country; it includes needs that  have  been identified by tribal or IMS  staff  and entered into
      STARS. Some tribes choose not to work with  IMS to have their infrastructure needs  on the list or they may
      not know how or have the capacity to do so; the extent to which this situation occurs is not clear to the
      evaluators.
      STARS does not track SDWA compliance history.
      STARS internal data linkages are limited. There is no automatic link in STARS between  information on a
      community's deficiency profile (CDP)  and projects proposed and/or funded.  Instead, IMS  engineers are
      expected  to keep the community deficiency profile up-to-date based on a holistic community-level view of
      identified needs and funded projects.  While there may be advantages and disadvantages for this approach
      to data tracking in STARS, the lack of an automatic link may  mean that information about a community's
      deficiency designation is neither current  nor a good  indicator of project accomplishments.   For EPA, the
      reported  progress on  its previous FY2006-2011 EPA Strategic Plan access "strategic targets"  was based on
      data from the CDP system, rather than from the Project Data System (PDS).  This evaluation has shown that
      progress  made through EPA-funded projects  was  not strongly related to changes in the community-level
      deficiencies (discussed further in the Program Goals and Measures section, and Appendix D, Correlation 6).
      IMS has different priorities  than  EPA. STARS was developed for  IHS's activities,  mission, and scope.
      Therefore, some  projects and homes within STARS are not within EPA's purview to fund or  do not reflect
      EPA's priorities (e.g., individual wells, sanitation plumbing). This difference can skew the allocation of funds
      under both  programs, as these ineligible projects are included when calculating total need nationally and  in
      the IMS areas and EPA regions.
  Federal Infrastructure Task Force Access Subgroup. Meeting the Access Goal: Strategies for Increasing Access to Safe Drinking Water and
Wastewater Treatment to American Indian and Alaska Native Homes. (March 2008)
Evaluation Report - March 2011                                                                         Page 25

-------
    EVALUATION OF THE                AND             INFRASTRUCTURE TRIBAL SET-ASIDE

      STARS does not uniquely identify homes. Individual homes are not uniquely identified in the system for
      projects, which can contribute to double counting for reporting on measures when homes are associated
      with multiple projects.

Interviews with IMS staff indicate that IMS is in the process of updating STARS to address some of the above issues.
In particular, the system will uniquely identify tribal homes, which will aim to address double counting and linkages
between a community's deficiency profile and projects in the same community.

IMS Project Implementation

For the  projects  implemented through  lAs  with EPA,  IMS  often  contributes critical design, planning,  and
construction management services to tribes, and in doing so,  helps tribes to receive needed infrastructure in a
timely manner. These and other implementation responsibilities require  a substantial amount of IMS staff time;
yet staff resources at IMS appear to be limited for the work underway.  In 2006, IMS estimated that the current
funding for program staff is less than 40 percent of what is needed to meet the  US access goal (i.e., at least $5.7
million annually for staffing).40 (Funding increased in FY2010 and under ARRA for EPA's programs, but not enough
to bridge this gap over time to meet all access needs or to meet all IMS staffing needs.) Interviewees from EPA
regions, EPA headquarters, IMS, and tribal governments all noted that IMS is understaffed for the current amount
of work underway. With the amount of work that there is yet to do in Indian country, this capacity issue at IMS is
not likely to improve without other changes such as additional staffing and  funding for the current as well as
increasing responsibilities.

For EPA-funded projects, some regional coordinators indicated that understaffing at IMS can manifest in longer lag
times for closing out projects.  An analysis of the universe of EPA projects (i.e.,  including projects funded before
2003 for which we had data) found that, according to the project milestones in STARS, the final project report had
been  completed in only one-quarter of CWISA construction-completed projects and just over half of DWIG-TSA
construction-completed projects  (see Table 7).   However, IMS provides a one-year warranty from  the time
construction is complete for projects to which IMS provides engineering services.  Given that warranty expenses
can affect financial information for this project, the  IMS cannot complete the final close-out report for these
projects at least until after the warranty expires.  Not having the final report for a project prevents the associated
IA between IMS and EPA from closing, but does not affect the provision of infrastructure to the tribal recipient.

Table 7 also provides project duration information for all EPA projects with available project milestone information
from IMS, excluding ARRA projects. This analysis includes information on projects funded by EPA before  2003, as
well as  during the evaluation time scope, to reduce any skewing of the results due to more-recently funded
projects that were implemented quickly.  Within the IMS areas and the EPA regions, there is a wide variation of
average  duration times  for  projects  in  the  two different programs. Level of IMS staffing,  different  water
infrastructure needs, type of facility construction, and  length of construction season are all significant factors that
could account for these differences.  For example, in Alaska, the construction season is very short in comparison to
other areas, which may force some projects to stretch over more construction seasons.  However, a correlation
analyses for this evaluation found that, generally, for both programs, there is no correlation between the duration
of projects being  constructed and  either  IMS areas or EPA regions.  More detailed information, including these
correlations, and average,  median, minimum, and maximum durations by EPA region and IMS area, is provided in
Appendix D (Correlations 1-4) and Appendix E (Tables 25-30).
 J Ibid.
Evaluation Report - March 2011                                                                          Page 26

-------
    EVALUATION OF THE DRINKING WATER AND CLEAN WATER INFRASTRUCTURE TRIBAL SET-ASIDE GRANT PROGRAMS

          Table 7. Project Duration Information, All EPA Project Information, excluding ARRA Projects
Duration Information DWIG-TSA CWISA
Number of Projects
Projects with Construction Completed
Projects with Final Report Completed (Percentage of Completed Projects)
Average Time to Complete Construction41
Median Time to Complete Construction
Average Time to Complete Final Report after Construction Complete42
Median Time to Complete Final Report after Construction Complete
403
210(52.1%)
109 (51.9%)
3. 51 years
3. 29 years
1.69 years
1.53 years
457
188 (41.1%)
48 (25.5%)
3. 22 years
2. 86 years
2. 14 years
1.69 years
Note: this table includes project duration information for all unique EPA projects where data was available and that had milestone information
in STARS: 403 of 462 DWIG-TSA projects (87 percent) and 457 of 554 CWISA projects (83 percent).  Project data was available as far back as
1993 for some CWISA projects and 1997 for DWIG-TSA projects.
Project duration is an important factor of project implementation and progress for both EPA and IMS.  Having
projects completed in a timely way ensures that homes receive needed water infrastructure as soon as possible. In
addition, any possible reductions in the length of time a project  takes to complete yields  cost savings in  both
construction costs and project staffing resources.  One of IHS's strategic measures is to have  projects constructed
in an average of four years or less and IMS has programmatic strategies to help achieve this target.43'44  Based on
the data available for this evaluation, EPA projects are, on average, within this IMS target.

ARRA Implementation

The American Recovery and  Reinvestment Act of 2009 (ARRA) provided  the CWISA and  DWIG-TSA Programs with
significant  additional  one-time  funding ($60  million for  CWISA and  $30  million for  DWIG-TSA).   Guidance
associated  with ARRA funding directed that  funds be managed in  a  manner  similar to  routine  program
implementation, with three major changes.
  >     Rather than enter into separate lAs with IMS for each project, EPA administered one national IA with IMS for
       CWISA and one IA for  DWIG-TSA.  In addition, the lAs were administered by EPA headquarters instead of the
       regions. (Tribes still had the choice of receiving a direct grant instead of funds through IMS, but EPA regions
       were to be the administrator  of the direct grants if tribes selected that option; no tribes  selected this
       option.)
  >     To the extent that there were eligible projects, 20 percent of each allocation was to be provided to projects
       that, "address green  infrastructure,  water or energy  efficiency  improvements or other environmentally
       innovative activities."
  >     Regions were asked  to work with tribes and IMS to identify projects that could be  under contract or
      construction within twelve months.
                                        45, 46
41 From when an MOA is signed with the tribe to when construction is complete.
42 From when construction is complete to when the final report is complete.
43 Indian Health Services. Strategic Plan 2006-2011. Available:
http://www.ihs.gov/PlanningEvaluation/documents/IHSStrategicPlan20062011.pdf.
44 Indian Health Services. FY2011 Online Performance Appendix. Available:
http:/
jix.pdf.
  Bergman, Ron, US EPA. Memorandum to Regional Drinking Water Branch Chiefs. "Drinking Water Infrastructure Grants-Tribal Set Aside
(DWIG-TSA) Program Guidance for projects funded using the American Recovery and Reinvestment Act of 2009."
46 Frace, Sheila, US EPA. Memorandum to Regional Water Division Directors. "Clean Water Indian Set-Aside (CWISA) Program Guidance for
projects funded using the American Recovery and Reinvestment Act of 2009." (July 9, 2009).
Evaluation Report-March 2011
        Page 27

-------
    EVALUATION OF THE DRINKING WATER AND CLEAN WATER INFRASTRUCTURE TRIBAL SET-ASIDE GRANT PROGRAMS
For ARRA projects oversight, in EPA headquarters and the regions have  held monthly calls and two in-person
meetings. As a part of this process, EPA and IMS also worked together to develop standard IA terms and conditions
that all regions must use going forward for routine annual implementation.

As of  December 2010, all ARRA funds had been assigned to projects, with  CWISA funding 96 projects and DWIG-
TSA funding 64.47

Table  8 provides a breakdown of number of ARRA projects, level of funding, and homes served by EPA regions.

                                     Table 8. ARRA Projects Summary
         10
        Total
                    $668,800
                    $698,000
                   $1,100,000
                   $1,692,700
                   $2,865,600
                    $765,700
                   $3,139,600
                   $8,448,500
$10,620,900
$30,000,000
                   13
                                     16
                                     64
                              2,051
                              1,311
                              2,163
                              387
                              1,480
                              858
                              2,916
                              5,604
                                                5,588
                                                22,358
 $346,670
 $1,343,330
 $1,700,000
 $1,590,010
 $4,430,760
  $7,160
 $6,417,660
$22,300,460
$21,863,950
$60,000,000
                                                            46
                                                                              23
                                                                              96
                                                                         50
                                                                        228
1,170
                                                                         59
                                                                        500
                                                                        125
1,583
6,178
2,159
12,052
      Note: The "homes served" value likely double counts some homes as sometribes received multiple grants from one or both programs.

Appendix E (Tables 31-37) provides funding and initial deficiency level information of ARRA projects and compares
the information to other projects within the evaluation's universe (2003-2009). On average, projects funded under
ARRA had significantly higher average funding levels than projects funded under routine program implementation.
In terms of the DLs of ARRA-funded projects, DWIG-TSA projects funded, on average, slightly higher DL needs with
ARRA projects than under routine implementation; for CWISA projects, the ARRA-funded DLs were not significantly
different than under routine implementation.

Interviewee Perspectives on ARRA Implementation

Interviewees identified the following ways  in which the differences in ARRA implementation have changed or
informed more routine implementation for both programs.

Coordination within EPA: Overall, interviewees were positive about the added communication and coordination
within EPA, and between EPA and IMS, as a result of ARRA.  They anticipated that this improved communication
will continue with  future routine  program implementation.  One regional coordinator said,  "I think the added
interest that headquarters has taken [as a result of ARRA] will hopefully lead to better policies  and processes with
the various EPA regions and IMS area offices." In December 2010, the EPA headquarters and regional coordinators
and  IMS staff met  to  discuss these programs, and feedback from participants in this meeting were generally
positive, with a number of comments asking for the  in-person meetings to continue after  ARRA projects are
  During the course of the evaluation, an additional DWIG-TSA project was funded with ARRA funding in Region 6. This project is included in
Table 8, but is not included in analysesforthis evaluation or in thetables in Appendix E.
Evaluation Report-March 2011
                                                                                Page 28

-------
    EVALUATION OF THE                AND            INFRASTRUCTURE TRIBAL SET-ASIDE

completed.  However, in phone interviews, when asked about communication under the ARRA processes, two
regional coordinators cautioned against having meetings for the sake of meetings and to encourage discussion
rather than simply updates.

National IA: Interviewees held a variety of perspectives on whether a  national IA, such as was done for each
program under ARRA, would be appropriate for routine program implementation. All who discussed this idea did
note that the different timing of project selections within  the  regions could be a problem for a national IA, as
funding decisions would need to be more coordinated and on the same timeline across all the regions.

Coordination with IMS: As discussed earlier in this report, the ARRA process has also spurred more coordination
between EPA's and IHS's headquarters staff, such as filling funding data gaps and developing the standard terms
and conditions for lAs.  Many interviewees, from both IMS and  EPA, indicated that the development of standard
terms and conditions for IMS and EPA  lAs was useful and that EPA regions will be using these in routine program
implementation. These standard terms and conditions may help alleviate some questions about IA expectations
noted earlier in the report.

Coordination with Tribes:  Under ARRA,  more travel money was available to  headquarters and the regions to
conduct site visits.  Many  interviewees from across the interview groups  indicated that travel by EPA to visit
project sites was useful to understand  needs and track project progress, and that having these funds under routine
program implementation would be useful.

Green  Projects: No  interviews noted the  "green" project requirement  under ARRA, so it is not clear to  the
evaluators whether this requirement caused any differences in project selection.
Evaluation Report - March 2011                                                                         Page 29

-------
             OF THE              AND
FINDINGS: PROGRAM GOALS  AND MEASURES


Core Evaluation Questions II-IV:

   To what extent are the DWIG-TSA and CWISA programs achieving their stated goals?

   What factors influence whether the programs achieve their stated goals?

   Are the current DWIG-TSA and CWISA program performance measures accurate
   indicators of EPA's progress?

The evaluators examined available information to understand how well the two programs are meeting their overall
program goals as articulated through the strategic measures, what factors could be influencing progress toward
the strategic measures, and whether the strategic measures are good indicators of progress for these programs.
The strategic measures for these programs  support the higher-level EPA strategic goal, "Protecting America's
Waters", in the FY2011-2015 EPA Strategic Plan.

The Compliance Strategic Measure: Progress and Factors Influencing Progress

EPA uses the compliance strategic measure (SP-3) to communicate the impact of the DWIG-TSA program,  as well
as the general SDWA program implementation in Indian  country.  The compliance strategic measure  is one
measure that gauges progress toward EPA's long-term drinking water compliance goal of 100 percent compliance
with SDWA as well as EPA's Strategic Goal 2,  "Protecting  America's Waters."  This section discusses reported
performance  on  the compliance  strategic measure; factors influencing progress on the compliance strategic
measure (and thus on the long-term drinking water compliance goal and EPA's strategic Goal 2); and whether this
performance measure is an accurate indicator of program progress.

Reported Performance on Compliance Measure

EPA's current target for its compliance strategic measure  is  to have 88 percent of the population served by tribal
community water systems  in compliance with all health-based rules by 2015.  (In the EPA FY2006-2011 Strategic
Plan, the strategic target was for 86 percent of the tribal population in compliance by 2011.)  Since FY2003,
reported progress on this strategic measure has been mixed, with the most recently reported  (FY2009)  data
showing the lowest percentage of population in compliance over the previous seven years (Table 9). Note that this
measure includes water systems for which EPA has SDWA direct implementation responsibility.  The measure,
however, excludes Alaska native systems, which are administered by the State of Alaska.  It also  excludes  non-
community water systems.
Evaluation Report - March 2011                                                                   Page 30

-------
    EVALUATION OF THE DRINKING WATER AND CLEAN WATER INFRASTRUCTURE TRIBAL SET-ASIDE GRANT PROGRAMS

                       Table 9. EPA Reported Performance on Tribal CWS Compliance
SP-3
, „ Total Systems out of , _ . . Population Served by _ ,
Fiscal Year ,. Total Population . ,. Reported
Systems Compliance Systems out of Compliance
2003
2004
2005
2006
2007
2008
2009
750
739
752
747
736
731
733
81
76
84
68
96
111
117
684,649
792,481
737,149
755,432
761,623
834,970
861,695
99,658
77,459
101,321
101,296
95,713
142,721
162,133
85.4%
90.2%
86.3%
86.6%
87.4%
82.9%
81.2%
The evaluation  looked specifically at water systems that had EPA-funded DWIG-TSA projects to see if there is a
difference in performance between the nationally reported data for all tribal CWSs and the systems that were the
recipients of EPA DWIG-TSA funding.48

Table 10 shows the SP-3  results for EPA-funded systems for calendar year (CY) 2008,  the last year for which this
evaluation had detailed violation data.  The evaluators compared the SP-3 results for EPA-funded systems against
the overall SP-3 results for FY2009, which, given  the reporting window for this strategic measure, is the most
accurate comparison for  the SDWA data the evaluators have.  The analysis showed that compliance at systems
with EPA funding  was lower than  the overall universe of tribal CWSs, which  is not surprising,  given  earlier
information that more systems funded by EPA had health-based violations than non-funded CWSs.  In addition, the
compliance rate for systems after EPA-funded  projects have been  completed is not markedly improved over the
average compliance rate for all CWSs, suggesting a marginal or minimal influence on compliance as a result of the
new infrastructure investments.  When Alaska systems are removed from the results, as it is for EPA reporting
purposes on SP-3, the compliance rate for systems with  completed projects  is slightly lower than the overall
universe of  tribal  CWSs.   This information is, however,  a  'snapshot' in time  that does  not show trends in
compliance,  and, importantly, does not explain the drivers of compliance  problems, which may not be related to
infrastructure or be solvable with funding from these programs.

                               Table 10. SP-3 Results for EPA-Funded Systems
EPA Funded EPA Funded SP-3 Result for EPA-
EPA Funded Systems Systems: Systems: Population funded systems
All Systems
All Systems, excluding Alaska
Systems with Completed
Projects49
Systems with Completed
Projects, excluding Alaska
214,686
196,899
44,566
40,002
164,093
152,297
34,701
30,683
76.4%
77.3%
77.9%
76.7%

SP-3 (2009) Result
for Tribal CWSs
81.2%
The nationally reported results for SP-3 for 2009 are calculated on four quarters of data, spanning July 2008 to June 2009. To more accurately
compare to this information with available detailed SDWIS data, this evaluation looked at the compliance information across the four quarters
of calendar year 2008.
  EPA funding has been provided primarily to CWSs, but funding has been provided to 12 non-profit non-CWSs (e.g., systems that serve schools
or other non-permanent populations).
49 Completed as of the end of CY2008.
Evaluation Report- March 2011
Page 31

-------
    EVALUATION OF THE                AND             INFRASTRUCTURE TRIBAL SET-ASIDE

This evaluation also examined whether there was any impact on compliance after a project was completed, rather
than just for systems that had received funding but had not necessarily had  projects completed.  From the SDWIS
information available for this evaluation, it appears that, on the whole, EPA-funded systems had fewer violations
after projects were completed when compared to their pre-project compliance.50
      34 of 92 completed drinking water projects (36.6 percent) were at systems that had health-based violations
      prior to when construction was completed. On average, these systems had 0.11 health-based violations per
      year and 1.08 monitoring and reporting violations per year prior to when construction was completed.
      15 of 92 completed drinking water projects (16.1 percent) were at systems that had health-based violations
      since construction has been  completed.  On average, these systems had 0.03 health-based violations per
      year and 0.47 monitoring and reporting violations per year after construction was completed.

By comparison, 16.1  percent of non-funded systems had  health-based violations in 2008. These systems had an
average of 0.39 health-based  violations for that year and 0.16 monitoring and reporting violations for that year.
(The evaluation used a single year to evaluate the number of health-based violations for the non-funded systems,
as there is not a similar "before" and "after" universe for these systems to use as a comparison group with funded
systems.)

However, given the data available and the likely influence of other factors on system compliance (see discussion
below),  this  evaluation  cannot conclude  definitively that  DWIG-TSA infrastructure projects  have  directly
contributed to improved compliance at funded  systems. Even  with the modest pre- and post-project compliance
rates described above, different scenarios could be at play.  For example, it is possible that compliance rates are
typically better just after a new project is completed  because new staff have just been trained and are  more
actively engaged in ensuring that compliance is supported.

Factors Influencing Progress on Achieving Compliance
Four factors suggested by interviewees and available data may  influence the extent to which EPA has been able to
improve in tribal system compliance with health-based standards through  the DWIG-TSA program: operations and
maintenance  (O&M) at tribal systems, compliance with new drinking water rules, the selection of DWIG-TSA
projects that  may address existing compliance issues at a system, and  the duration of projects to have an impact
on compliance.  Other factors  may also influence progress on the compliance strategic measure, but not enough
information was available to this evaluation to identify them.

O&M at Tribal Systems

Adequate O&M of tribal systems is essential to both ensure that the infrastructure funded by EPA is maintained
over its projected design life, and to maintain SDWA compliance.  Typically, O&M at tribal water  systems has been
poor, resulting in failures in infrastructure and public health protection. As part of an effort to address this issue,
the Infrastructure  Task Force Access Subgroup noted that, in areas  with  low population densities  or remote
geography, the technical,  managerial, and financial capacity necessary for adequate system O&M can  be difficult
to develop and  maintain.  However, the Subgroup did  also note that there is no data available to quantitatively
understand the relationship between the quality of management at  a tribal water system and the impact on
  As discussed previously, the SDWIS data that provides detailed violation data is only available for 2004 to 2008 and does not include ongoing
violations that began before 2004. In addition, the data does not tell us whether violations were infrastructure-related or O&M related.
Evaluation Report - March 2011                                                                           Page 32

-------
    EVALUATION OF THE                 AND             INFRASTRUCTURE TRIBAL SET-ASIDE

infrastructure resources.51  The Subgroup recommended a number of actions for the involved federal agencies,
such as better coordinating technical assistance programs to build workforce  capacity.   EPA and other federal
agencies have done some work to begin  addressing these recommendations through the formation of a Tribal
Technical Assistance Workgroup.

Interviews with most regional coordinators, as well with a tribal government representative, echoed that adequate
capacity is  essential for maintaining water system infrastructure,  and that there is a general lack  of tribal and
federal funding available to address capacity and O&M issues at tribal water systems. Statutorily, EPA is prohibited
from directly funding O&M through the DWIG-TSA program.  A few regional coordinators noted that many tribal
compliance  problems are less related  to  infrastructure than to  inadequate O&M at  tribal  systems and thus
infrastructure funding would not itself improve compliance.

The regional coordinator in Alaska  examined the significant non-complier (SNC) list (systems considered by EPA to
be significantly out of  compliance) in that state to determine the extent  to which compliance issues could  be
addressed with DWIG-TSA funding. As of May 2010, in Alaska, 60 percent of the systems (43 out of 72) at Alaska
Native Villages on this SNC list had violations solely related to operations.  Of the remaining systems on the SNC
list, seven had violations related to unfunded hardware needs,  five did not have adequate capacity to qualify for
DWIG-TSA funding, and only the remaining two systems would qualify for DWIG-TSA funding.  Therefore, only
three percent of the Alaska Native Village systems on the SNC list had infrastructure-related compliance problems
and  were eligible  for  DWIG-TSA  funding.53   This analysis covered only one  state in  one  EPA region,  and it
represented a 'snapshot' in time,  but its results are potentially significant, calling  into question the assumption
that DWIG-TSA funding could solve many SDWA compliance problems at tribal water systems.  Similar analyses in
other EPA regions could provide invaluable information about where DWIG-TSA funding could support compliance
and where compliance  of other systems could be better supported through other types of assistance, such  as
training or enforcement actions.

No source of sustained federal funding currently exists for the O&M of water infrastructure on tribal lands. Given
this constraint, interviews suggested the following strategies to help address tribal O&M issues for EPA to consider
or to continue to support.
      Appropriately-scaled infrastructure: Some interviewees discussed the importance of building simple water
      infrastructure for smaller tribal systems or tribes with a history of compliance problems to enable easier
      maintenance and sustained compliance.  A few expressed concern that  DWIG-TSA projects (and  also CWISA
      projects) have on occasion been  "over  built" (i.e., too technically complicated or large for the needs and
      capacity of the tribes that will have to maintain them) and that this alone could lead to future compliance
      violations.
      Technical assistance:   EPA  and other  federal agencies have  technical assistance  programs that include
      training for tribal system operators. For example, as of FY2010, EPA has funded  ten positions in IMS that
      work directly with tribal systems in six  EPA regions through the Public Water System Supervision (PWSS)
      program. Interviewees familiar with these positions thought  that this  program is a  very useful way  to
  Federal Infrastructure Task Force Access Subgroup. Meeting the Access Goal: Strategies for Increasing Access to Safe Drinking Water and
Wastewater Treatment to American Indian and Alaska Native Homes (March 2008).
52 Tribal Technical Assistance Workgroup, Federal Infrastructure Task Force on Tribal Access to Safe Drinking Water and Basic Sanitation.
Strategies for Improving Technical Assistance Delivery in American Indian and Alaska Native Communities (March 2010).
53 D. Wagner, US EPA Region 10. Personal Communication. December 23, 2010. Note: this analysis was performed to inform the region, but
was provided to this evaluation as additional information.
Evaluation Report - March 2011                                                                             Page 33

-------
    EVALUATION OF THE                AND             INFRASTRUCTURE TRIBAL SET-ASIDE

      support tribal system compliance, and also to keep EPA and IMS in close communication about compliance
      needs and issues. EPA also has a large number of on-line training programs, which interviewees noted were
      useful for helping enhance tribal understanding of compliance issues before violations occur.
      Usage fees:  Five interviewees identified usage fees as a barrier to proper O&M:  tribal systems were not
      charging usage  fees to  customers, were not charging enough, or were not enforcing their collection.  As a
      result, even  systems with adequate capacity were hampered from  addressing O&M needs to due to lack of
      funds.   One regional coordinator said, "I don't  necessarily advocate federal funding [for  tribal  O&M],
      because the problem is so big, but maybe the tribes need to increase their user fees so that they have the
      appropriate  amount of funds. Right now, tribes are charging fees that are 10 years old."
      Asset-Management Plans: One EPA  regional  coordinator recommended that  DWIG-TSA projects include
      funding for asset management plans  at tribal systems to ensure that there is a strategy in place to manage
      infrastructure adequately. Asset management can help reduce operating costs by enabling water systems
      to anticipate and plan for both the operational and capital costs of its infrastructure. Expanded EPA support
      for asset management planning could potentially be another form of technical assistance.

Compliance with New Drinking Water Rules

The evaluators sought to  understand whether the advent of additional or revised drinking water rules result in
influence system compliance, including those systems  that received infrastructure funded through the DWIG-TSA
program in the past.  During  the 2004-2008 timeframe for which we had data on tribal compliance with  SDWA,
only one drinking water rule came into effect, the 2006 Arsenic Rule.54 EPA adopted a new standard for arsenic in
drinking water  of ten parts per  billion (ppb), replacing the old standard of  50 ppb.  Systems  were required to
comply with this new rule by January 2006.  Two other  rules were promulgated during the timeframe, but have not
been monitored yet by EPA for compliance.

Interviews with EPA regional  coordinators,  an IMS area staff member, and a tribal government staff member all
suggest that the 2006 Arsenic Rule increased tribal  system non-compliance.  The tribal staff member said that,
under the new Arsenic Rule, "All  of the sudden, lots of systems were no  longer compliant... Now, the dollar
amount of needed  improvements is huge."

Data analysis supported the same conclusion. Fifteen of 113 (13.3  percent) systems that had EPA-funded projects
prior to the introduction of the Arsenic Rule in 2006 had a total of 68 arsenic maximum contaminant level (MCL)
violations between CY2006 and CY2008 (all of the systems with violations were CWSs).  Of these systems, none had
health-based violations associated  with arsenic in CY2004  or CY2005. Looking at the broader universe of tribal
CWSs, there were  24  arsenic MCL  violations in FY2007, 75 in FY2008, and 92 in FY2009—there were no arsenic
MCL violations from FY1998 to FY2006 for any tribal CWSs.  Though this  is a  small universe of violations for both
the EPA-funded systems and the overall universe of tribal CWSs, it does seem that the 2006 Arsenic Rule  had an
impact on tribal system compliance, thereby affecting reported performance on the compliance measure.
54 Detailed SDWIS data was only available for 2004-2008 for this evaluation, as compared to the evaluation timeframe of 2003-2009.
55 (1) The Stage 2 Disinfectants and Disinfection By-products Rule: This rule addresses risks from microbial pathogens and
disinfectants/disinfection by-products, specifically total trihalomethanes (TTHM) and five haloacetic acids (HAAS). Compliance monitoring for
this rule begins between 2012 and 2016, so this rule likely did not have any impact on compliance rates during the evaluation time scope. (2)
The Long-Term 2 Enhanced Surface Water Treatment Rule: This rule addresses Cryptosporidium and other disease-causing microorganisms.
Monitoring related to this rule began to be required between 2006 and 2008, but systems had three years to comply with implementing
treatment techniques. As such, this rule likely did not have much impact on compliance rates during the evaluation time scope.
Evaluation Report - March 2011                                                                            Page 34

-------
    EVALUATION OF THE                AND             INFRASTRUCTURE TRIBAL SET-ASIDE

Project Selection

As already discussed, the DWIG-TSA compliance strategic measure assumes that DWIG-TSA projects can and will
address and correct health-based compliance issues caused  by infrastructure deficiencies.  However, it appears
that some health-based compliance issues are related  to O&M capacity at tribal systems and the introduction of
new drinking water  rules, rather than  infrastructure.  In this section, we explore whether project selection and
project duration may also influence SDWA compliance.

Our review  of regional  project selection guidances and  processes  indicates  that the EPA regions use  tribal
compliance history or anticipated public health impacts of projects as one factor in determining whether to fund
projects.  However, other factors such as economic feasibility of the project and the availability of funding may in
some  cases play a more  prominent role in determining project selection. These other factors limit the influence
selection has on addressing system non-compliance.

Project Duration

Given the typical length of time needed for project completion, the effect on compliance at a system is likely to
take years to be realized. DWIG-TSA projects have taken an average of 3.51 years to be completed, and there have
been some projects that have taken eight or nine years. Therefore, the yearly compliance data as measured by SP-
3 is not likely to reflect recent infrastructure investments. As  discussed in the section on reported progress on the
compliance measure, systems with completed projects have fewer compliance violations after  a construction is
completed than before the project is complete.  This  result  supports the idea that we would not expect to see
changes in compliance at a system until after a project is completed if the violations are infrastructure-related.

Compliance Measure as Indicator of Progress

Given the findings described above, it appears as  if factors outside the scope or control of the DWIG-TSA program
funding and  implementation, such as system O&M, the introduction of new drinking water rules, project selection
and subsequent duration of the project, influence compliance at systems and thus the reported performance
indicated by the SP-3 compliance strategic measure.

Interviews and observations of the data  additionally indicate that the SP-3 strategic  measure is imperfect  as an
indicator of  DWIG-TSA progress as  the measured universe is different than the universe  of potential DWIG-TSA
influence. SP-3 measures the compliance of EPA-regulated tribal CWSs, rather than the universe served by EPA
projects, which has included non-CWSs and state-regulated systems.

Notably, systems in Alaska serving Alaska  Native Villages eligible under the DWIG-TSA program received 25 percent
of DWIG-TSA funding between 2003 and  2009. SP-3 does not capture any of the compliance improvements that
may have been made as a result of these projects because the State of Alaska implements the SDWA Alaska Native
Village water systems. (Note  that in this report we provide the data with and without  Alaska for comparison
purposes.)
In addition, between 2003 and 2009, DWIG-TSA projects have provided funding to tribal systems that account for
approximately one quarter of the population served by CWSs.56 How*
compliance at systems where DWIG-TSA funding has not been provided.
approximately one quarter of the population served by CWSs.56 However, the SP-3 strategic measure includes
  In 2008, 214,686 were served by EPA-funded PWSs out of an overall population of 834,970 (28.7 percent). Removing Alaska-funded PWSsfor
a more direct comparison to reported numbers, 196,899 were served (27.2 percent).
Evaluation Report - March 2011                                                                          Page 35

-------
    EVALUATION OF THE DRINKING WATER AND CLEAN WATER INFRASTRUCTURE TRIBAL SET-ASIDE GRANT PROGRAMS
The Access Strategic Measures: Progress and Factors Influencing Progress

Reported Performance on the Previous Access Strategic Targets
As discussed in the Background section, in  the EPA FY2006-2011 Strategic Plan, EPA listed two  "strategic targets"
to indicate progress on providing access to water infrastructure in Indian country:
  >   By 2015, in coordination with other  federal agencies, reduce by 50 percent the number of homes on tribal
      lands lacking access to safe drinking water.  (This target used a 2003 baseline of 38,637 homes lacking
      access to safe drinking water out of an estimated 319,070 tribal homes.
                                                                       ,57
  >   By 2015, in coordination with other federal agencies, reduce by 50 percent the number of homes on tribal
      lands lacking access to basic sanitation.  (This target used a 2003 baseline of 26,777 homes lacking access to
      basic sanitation out of an estimated 319,070 homes.)
                                                      58
According to reported numbers  (Chart 1), the overall progress toward these targets was stagnant or trending
downward.  In 2003, 12.1 percent of all identified tribal homes lacked access to safe drinking water, and in 2009,
12.1 percent of all homes still lacked access, indicating that no progress has been made on this measure.  In terms
of clean water, the reported trend was the  reverse of intended progress: In 2003, 7.9 percent of tribal homes
lacked access to basic sanitation and in 2009, 10.1 percent lacked access. These trends reflected the number and
percent of homes lacking access, rather than the number of homes provided access.

      Chart 1. Number of Tribal  Homes that Lack Access to Safe Drinking Water and Wastewater Services
             50,000
             45,000
             40,000
             35,000
             30,000
           o
             25,000
           ra
          .a
             20,000
             15,000
             10,000
              5,000
The orange and red lines show the intended
targets for these measures, while the blue
and green show the actual results of homes
lacking access.
                        •CWTargetSP-15
                        •CWActualSP-15
                        •DW Target SP-5
                        •DW Actual SP-5
Source: EPA Office of Water. "Tribal Drinking Water and Clean Water Access Measure Modification" (January 2010). Based on Community
Deficiency Profile (CDP) data from IMS STARS.
  The Annual Commitment System code for this target was SP-5.
 8 The Annual Commitment System code for this target was SP-15.
Evaluation Report-March 2011
                                      Page 36

-------
    EVALUATION OF Til	IE                AND             INFRASTRUCTURE TRIBAL SET-ASIDE
The current (FY2011-2015) and former (FY2006-2011) access strategic measures59 that the DWIG-TSA and CWISA
programs work toward result from EPA's commitment to reduce the number of tribal homes lacking access to safe
drinking water and basic sanitation  as part of the broader US access goal made at the 2002 World Summit on
Sustainable Development. Providing access, as defined by these measures, is moving a home from  a DL-4 or -5
designation to a  lower DL These measures also provide an indication of progress toward EPA's strategic Goal 2,
"Protecting America's waters." This  section discusses reported performance on these strategic measures;  factors
influencing progress  toward achieving these  measures  (and  thus the higher-level goals); and whether these
performance measures are accurate indicators  of DWIG-TSA and CWISA program progress.

Over the same time period, if instead the data  analyzed is the cumulative number of tribal homes provided  access,
the number of homes that that received access to drinking water and wastewater facilities exceeded the stated
target of 50 percent  of the 2003 baseline:  as of 2009, 80,941 tribal homes had been provided access to drinking
water,  as compared to  the target of 19,319 homes,  and 43,562 tribal homes had been provided access to basic
wastewater sanitation, as compared  to the target of  13,289 homes (Chart 2).60 According to a paper prepared by
EPA OW staff on these strategic measures, these trends affecting reported numbers could be due to:
      The addition of tribal homes to STARS (from 2003  to 2009, the number of homes in STARS increased from
      319,070 to 360,000, causing the relative  percentage to appear smaller due to an increasing denominator);
      "Backsliding" of DL designations (e.g., from a  DL-3 to a  DL-4) due to aging infrastructure or due to water
      systems falling out of compliance with drinking water rules; and
      Renewed interest by tribes in federal funding and thus tribes working with IMS to get project needs listed in
      STARS.61

As previously  discussed, the FY2006-2011 access  strategic  targets were replaced during the course  of this
evaluation by the strategic measures in the  FY2011-2015 EPA Strategic Plan. EPA explained that the reason for this
replacement was, "the changing [number of homes]  baseline distorted OW's reported performance in increasing
access."62   EPA  OW kept the previous access strategic targets as  "indicators" in the  EPA OW FY2011 NPM
Guidance,  meaning that EPA OW will  continue to report on the number of tribal  homes lacking access to safe
drinking water and basic sanitation in annual reports in future, but not against an annual or long-term target.

In addition to the issue of the increasing number of identified homes relative to the 2003 baseline, the past access
strategic measures relied on data reported in the  STARS Community Deficiency Profile (CDP), which, as previously
noted,  is not  automatically linked to project-specific data in the STARS system.  Interviews with some program
stakeholders indicate that the CDP  is not necessarily current with  all  known  needs and funded  projects in  a
community.
  In the FY2006-2011 EPA Strategic Plan, these commitments were referred to as strategic targets.
60 US EPA Office of Water. "Tribal Drinking Water and Clean Water Access Measure Modification" (January 2010).
61 Ibid.
62 US EPA. FY2011 National Water Program Guidance, Appendix D: Explanation of Key Changes Summary (April 2010). Available:
http://www.epa.gov/ocfo/npmguidance/2011/owater/nwp program guidance appendix d 508.pdf.
Evaluation Report - March 2011                                                                           Page 37

-------
    EVALUATION OF THE DRINKING WATER AND CLEAN WATER INFRASTRUCTURE TRIBAL SET-ASIDE GRANT PROGRAMS

     Chart 2. Cumulative Number of Tribal Homes Provided Access to Safe Drinking Water and Wastewater
                                         Infrastructure, 2003-2009
      10,000
                                                                                 DWTarget: 19,319

                                                                                 CW Target: 13,389
Source: EPA Office of Water. SP-5 and SP-15 Tracking Spreadsheet (Excel) (January 2010). Based on Project Data System (PDS) data from IMS
STARS.

To understand whether EPA projects would have influenced the community-1 eve I  DLs, and therefore changes in
access as it was  previously calculated at the community level, the evaluators conducted a correlation analysis of
the relationship between changes in DLs in homes provided a service through EPA  projects (at the PDS level) and
the changes in DLs at homes (at the CDP level).  The analysis results showed no significant correlation,  meaning
that a provision of access (in  PDS) for a home served by an EPA project would not be expected to result in a change
in access designation for a home at the community level (in CDP). This result could either be because the effect on
the community is small, or because the update to the community designation of homes is not immediate.63 (See
Correlation 6 in Appendix D for more details.)  In other words, EPA was reporting based on number of homes from
community/CDP  data, but EPA-funded projects were not influencing the changes in the CDP data.

Future upgrades to STARS will likely help  address this CDP-project level linkage issue, though there will still be
limits in the impacts of individual  projects, as homes may be associated with  multiple infrastructure needs.   In
2010, EPA decided to use project  level  (PDS) information as the basis for tracking performance on the access
strategic measure.  This improvement will alleviate  the  problem of lack of connectivity to the community-level
influences,  but it does introduce a double counting issue,  and also, as we describe in  more detail below, the
assumptions  behind the  term  "access" as   it  is used  by  EPA for performance measurement  purposes are
problematic.
  Forthis analysis, we looked only at EPA projects; the results of the correlation analysis may be different when looking at all projects and all
communities tracked by IMS in STARS.
Evaluation Report-March 2011
Page 38

-------
    EVALUATION OF THE                AND             INFRASTRUCTURE TRIBAL SET-ASIDE

Reported Performance on the Current Access Measures
In the FY2011-2015 EPA Strategic Plan, EPA replaced the former access strategic targets with two new strategic
measures.
      By 2015, in coordination with other federal  agencies, provide safe  drinking water access to  136,100
      American Indian and Alaska Native homes.64
      By 2015, in coordination with other  federal agencies, provide basic sanitation access to 67,900 American
      Indian and Alaska Native homes.65

The new strategic measures rely on a count of the number of homes provided  access, rather than the percent and
number of homes lacking access. These measures are cumulative (i.e., in 2009, 80,900 homes have been provided
access to safe drinking water; 55,200 more should be provided  access by 2015). These new strategic measures
address the issue of a "changing baseline," but some interviewees pointed out that the there is no longer context
(e.g., the proportion of the entire population) from which to gauge progress and know its relative impact.

According to the latest STARS data available for this evaluation (2009), the number of tribal homes lacking access
to safe drinking water was 43,437 homes, which was fewer than the remaining  number of homes (55,200) that EPA
has set for itself and other federal agencies to provide safe drinking water access to by 2015. In other words, EPA's
2015 measure identified more homes to provide access to then there were homes that needed access as of 2009.
EPA may, however, be anticipating that more homes with high drinking water  needs will be added to STARS (e.g.,
new SDWA rules causing systems to fall out of compliance, new tribal homes being added to STARS).

According to  STARS data, there were 28,052 homes lacking access  to basic  sanitation in 2009; this number is
slightly higher than the cumulative target (19,990  homes) that EPA has set in the new strategic measure.

EPA has not yet reported performance on these measures, but, from discussions with EPA, we expect that EPA will
use data from the Project Data System (PDS) from STARS to report on this measure alleviating the problem with
the previously reported progress based on data from CDP in STARS.

Factors Influencing Achievement of Progress Toward Access
Information collected through the interviews and data analysis suggest that  six factors influence the extent to
which EPA has been able to make progress on the access strategic measure:  O&M at tribal  systems, compliance
with new rules, available funding,  EPA's scope of influence, data availability  and quality, and  project selection.
Other factors  may also influence progress on the access strategic measures, but information with which to analyze
potential additional factors was not available.

O&M at Tribal Systems

As discussed in the section  on factors influencing  progress on the compliance measure, adequate O&M capacity is
vital for ensuring the  infrastructure funded by  EPA is functional  and maintained over its projected life.   The
Infrastructure Task Force Access Subgroup has noted that sufficient tribal capacity for O&M is crucial for ensuring
access to water and sanitation infrastructure, and that sub-optimal capacity is  a factor in infrastructure becoming
unusable before its design  life is over. The Subgroup also noted that, in areas with low population densities or
  The Annual Commitment System code for this measure is SDW-18.
 5 The Annual Commitment System code for this measure is WQ-24.
Evaluation Report - March 2011                                                                         Page 39

-------
    EVALUATION OF THE                AND              INFRASTRUCTURE TRIBAL SET-ASIDE
remote geography, sufficient capacity for O&M can be difficult to attract and sustain, and as discussed earlier in
this  report,
addressing.6
this report,  made recommendations related to O&M  capacity that the Infrastructure  Task Force has begun
          66
Interviews with most EPA regional coordinators, as well with  a tribal government representative,  echoed that
adequate tribal capacity is necessary  to maintain infrastructure, and that there is a general  lack of tribal and
federal funding and capacity available to address O&M issues at tribal systems. If existing infrastructure degrades
due to inadequate O&M, homes previously provided access to drinking water or basic sanitation could no longer
have access, given the way access is currently defined.

The lack of O&M capacity is a daunting issue, and no source of sustained federal funding currently exists for the
operation  and  maintenance of water infrastructure  on tribal lands.   The earlier  report section on factors
influencing the progress on  the  compliance  strategic measure  discusses a few possible strategies suggested  by
interviewees for addressing tribal  O&M issues, which are  applicable  to  both  drinking water  and  clean water
infrastructure maintenance.

Compliance with New Drinking Water and Clean Water Rules

Compliance with new drinking water rules influences performance on the drinking water access measure. Critical
health-based violations resulting from  new rules can cause IMS  staff to re-designate homes from a lower DL to a
DL-4,  assuming IMS is aware of the compliance problem. Compliance with new rules associated with the  Clean
Water Act  (CWA)  may  affect  whether tribal homes  lack access to basic sanitation; however, the evaluation
received no  data or information about this issue, and  we therefore cannot say  whether CWA compliance would
affect progress on access to basic sanitation.

Available Funding

The Infrastructure Task Force Access Subgroup report referenced previously stated that sufficient federal funds are
not available to meet the US access goal by 2015. This paper estimated  that a 40 to 50 percent increase over
current project funding is needed to meet the US access goal by 2015, or approximately $50 million to $63 million
more  federal dollars over current funding levels.  Interviewees with EPA headquarters staff and  some regional
coordinators emphasized that a more funding than the current level is needed to address all infrastructure needs
in Indian country.  As one EPA headquarters staff person said, "If I look at the data from IMS and the needs out
there... it looks like we would need a significant boost in funding to address the total need." In addition, this need
for funding for access to water infrastructure does not account for the additional funding needed to maintain that
infrastructure, to upgrade infrastructure if EPA promulgates new rules that would require that, or to expand to
account for new population growth.

Scope of Influence

EPA's access strategic measures are explicit that achieving these measures will be done in coordination with other
federal partners. The DWIG-TSA and CWISA programs  are limited in the amount of impact that they can have on
the reported progress of the programs' strategic measures.  As discussed in the program background section of
this report, the DWIG-TSA and CWISA programs have limitations on which projects are eligible for funding. Despite
the limits on what these EPA infrastructure programs can fund, ineligible projects and associated tribal homes are
  Federal Infrastructure Task Force Access Subgroup. Meeting the Access Goal: Strategies for Increasing Access to Safe Drinking Water and
Wastewater Treatment to American Indian and Alaska Native Homes (March 2008).
Evaluation Report - March 2011                                                                           Page 40

-------
    EVALUATION OF THE                AND             INFRASTRUCTURE TRIBAL SET-ASIDE

still included in the IMS STARS database and thus are included in reported numbers of homes provided or lacking
access. As EPA funds cannot be used for these projects, EPA cannot directly impact these homes.

In addition to not having the influence to fund all projects that would affect water infrastructure access at tribal
homes, EPA is not the direct implementer of these projects; tribes often work with IMS to at least administer the
funds for these grants and  IMS staff often serve as the engineers. Thus, EPA cannot directly influence the time
projects take to complete and thus the provision of access to homes.  As an IMS area staff person noted, "EPA is
put in a difficult situation being held accountable for work that IMS ends up doing."

Data Availability and Quality

For the current access strategic measures, EPA has addressed the changing baseline of homes issue, as well as the
issue of connecting project data to performance gauged at the community level.67 However, there are several
data-related issues associated with these strategic measures  that will impact EPA's ability to fully affect both
reported and actual on-the-ground performance.
      Double-counting of homes: Due to the way the PDS system is designed and how the access numbers are
      calculated, double counting of homes provided access will likely occur if EPA or IMS fund multiple projects in
      a particular community. Homes may also be double counted if funded through an infrastructure project and
      provided  "access" (by moving from a DL-4 or DL-5 to something lower), only later to return to a DL-4 or DL-5
      and be funded again. As an IMS staff person said, "Looking at access as changes in  DLs may be skewing
      things, because from year to year, those numbers can change a lot and we don't necessarily know if it's new
      homes or it's the same homes."  As discussed, IMS is working on improving the ability to uniquely identify
      individual homes, which will hopefully enable prevention of double counting in the future.
      Deficiency level "backsliding":  Related to the previous point, DLs can change from one year to the next.
      EPA relies on the provision of access (as defined  as a  change in DLs) in reported numbers on the measures,
      but these DL designations can shift following project completion.
      Timing of Project Impact: According to the STARS manual, when a project is funded, the homes receiving
      the service should immediately have their DL changed  to the final DL (FDL) associated with the project.68
      Given that most projects take years to complete, the  reliance on a metric that requires immediate  progress
      in reducing deficiencies is misleading, yet EPA also relies on the same immediate provision of "access" when
      reporting performance.
      Direct Grant Inclusion: Between both programs, approximately eight percent of all  EPA-funded projects
      (and  eight percent  of EPA  funding) between 2003  and 2009 were provided as direct grants to  tribes.
      Although this is a relatively small amount overall, some regions, including those with large populations, have
      a significant portion  of their funding directed to direct grants. For example, 60 percent of the DWIG-TSA
      projects between 2003 and 2009 in  Region 8 were direct grants to tribes.  Direct grant projects have not
      (and will  not) contribute to the reported number of homes provided access unless the homes associated
      with these projects are already being tracked by IMS due to  their  involvement in implementing the projects
      or active communication by EPA or the tribe.
 This statement assumes that the DLs of homes will be tracked using only PDS data, not CDP data. Our understanding from talking with EPA
staff is that only PDS data will be used in reporting on the new access measures.
68 Indian Health Services. STARS User Manual (September 2008).
Evaluation Report - March 2011                                                                           Page 41

-------
    EVALUATION OF Til	IE                 AND             INFRASTRUCTURE TRIBAL SET-ASIDE

Project Selection

As discussed in the section on implementation, many of the projects selected  by the CWISA and  DWIG-TSA
programs are not addressing DL-4 and DL-5 needs or DL-4 and DL-5 homes.  For the CWISA program, EPA regions
generally fund projects in order down the SDS list, where deficiency level is only one factor of eight of a project's
ranking score on the list.  Similarly, for DWIG-TSA projects, regional coordinators consider several factors—not only
providing access—when deciding which projects to fund. However, for both programs, we also note that DL-2 and
DL-3 needs are often necessary to address for tribes to have quality water system infrastructure and outputs.69

Access Strategic Measures as Indicator of Program Progress
In addition to the factors discussed above that may influence progress that are outside the control or scope of
these programs'  funding  and implementation,  the  current access strategic measures are not specific to  EPA's
activities. They rely on EPA's "coordination with other federal partners" and  thus the actions and influence of
other federal agencies.  This evaluation did not analyze the extent to which the other federal agencies involved
with the Infrastructure Task Force could affect progress on these measures, but a significant part of IHS's activities
are to work with tribes on  developing drinking water and  basic sanitation infrastructure.  In addition, there  are
HUD and USDA funds available to tribes that would  affect infrastructure projects.   To put it simply, the lack of an
EPA-specific strategic measure means that it is difficult to assess the impact of the EPA programs on tribal access to
safe drinking water and basic sanitation.

For comparison, IHS  has four strategic measures related to both serving and providing access to tribal homes, with
three of these related directly to IHS-specific activities, rather than more general indicators of  progress across
multiple federal agencies:70
      Number of new or like-new tribal homes and existing  homes provided with sanitation facilities;
      Percentage of existing homes served by the program at DL-4 or above;
      Average project duration; and
      Percentage of tribal homes with sanitation facilities.
  For example, IHS designates systems that do not meet national secondary drinking water standards as having DL-2 water needs, and lagoons
in need of repair as having DL-2 sewer needs. Source: Indian Health Services. Sanitation Deficiency System (SDS) Guide for Reporting Sanitation
Deficiencies for Indian Homes and Communities. May 2003. Available: http://www.ihs.gov/dsfc/documents/SDSWorkingDraft2003.pdf.
70 Indian Health Services. Strategic Plan 2006-2011.  Available:
http://www.ihs.gov/PlanningEvaluation/documents/IHSStrategicPlan20062011.pdf.
Evaluation Report - March 2011                                                                             Page 42

-------
              OF THE               AND
CONCLUSIONS AND  RECOMMENDATIONS
Conclusions

Based on the findings presented throughout  this report,  the  evaluators  have  drawn the following summary
conclusions about the implementation of the DWIG-TSA and CWISA programs, the progress on achieving goals as
tracked by the programs' strategic measures, and the use of these performance measures as indicators of program
progress.
>    The goals and priorities for these programs could be clearer and more focused. The goals and priorities for
    both programs lack precision and clarity: EPA staff have different ideas of what the term "goal" means for
    these programs (e.g., is it the US access goal? Or EPA Strategic Plan Goal 2? Is the drinking water program
    more focused on a compliance goal or an access goal, or both?) and how the drinking water program priorities
    of compliance and access relate (e.g., does access automatically support compliance and vice versa?). EPA has
    the  opportunity to more clearly articulate and focus the goals of both programs, and then to focus program
    design and implementation accordingly in a set of "cascading" decisions and actions that would follow clearer
    goals and priorities.
>    Funding allocation for both  programs could be improved to  be more in  line with  EPA's  priorities.  The
    method used to allocate funds for each program has an understandable basis and history; however,  neither
    allocation process is an  ideal match for the current program priorities and strategic measures.
           The DWIG-TSA program allocation of funds is not ideal for supporting the programs' current program
           strategic  measures and maximizing progress toward these measures.  For example, if compliance with
           SDWA is the highest  priority for the drinking water program, funding could more directly be allocated
           based on compliance needs. In addition, the use of the SDS list as a way to calculate overall drinking
           water needs in a region means that ineligible projects and projects addressing lower deficiency needs
           are contributing to the DWIG-TSA allocation.  In  addition, the two percent  base allocation under
           DWIG-TSA for each region is necessarily a strategic process for making progress toward the  program's
           strategic  measures.
           For the CWISA program allocation of funds, the SDS list is a valuable tool for EPA, but it does not align
           with the  CWISA program's ability to fund projects  or with  the strategic measure related to basic
           sanitation.
           We support the work that is already underway to allow the fund transfer between programs to more
           effectively address the most pressing water infrastructure problems in a given year.
>    For  both programs, project selection could be more clearly and consistently tied to  EPA's priorities, while
    still  maintaining regional discretion and flexibility.
           Project selection under the DWIG-TSA program differs  from region  to region, and these  processes
           have developed due  to differing regional circumstances. Though regional coordinators have the  best
           understanding of regional conditions, these varied project selection  processes may mean differential
           treatment of tribes across the EPA regions and that some regions are less consistent with  helping
           meet EPA headquarters program priorities than are others.
Evaluation Report - March 2011                                                                       Page 43

-------
    EVALUATION OF THE                AND             INFRASTRUCTURE TRIBAL SET-ASIDE

            For the CWISA program, the IMS SDS list is an  invaluable tool for the CWISA program  in terms of
            guiding project selection, but using the SDS list to fund in rank order is not ideal for CWISA's priorities
            and ability to fund projects.
    Communication and coordination between EPA headquarters and EPA regions had been inconsistent until
    the recent ARRA funding.  Based on interviews and our own observations, it appears that communication
    between EPA  regions  and  EPA  headquarters  is  generally  less  than optimal  for  facilitating  mutual
    understanding  of  national  and regional priorities and  progress,  program implementation, a unified EPA
    approach to these programs, and strategic learning  and  improvement.  The management processes used to
    implement ARRA funding led to improved  communication and coordination within EPA  and it appears that
    some of these processes will persist into routine program implementation.  From interviews, it seems that a
    single  IA for all DWIG-TSA  projects and  one for all CWISA projects, as was done under ARRA,  could  have
    advantages for coordination  However, trying to coordinate simultaneous funding and closure for all projects
    would likely be challenging if EPA pursued this funding mechanism for routine program implementation.
    Despite  improvement in recent years, communication between EPA and IMS is inconsistent and not optimal
    for strategic program management or strategic learning and improvement.  Though interviewees reported
    that communication  on the whole between  the  two agencies is  currently working well, it has been
    inconsistent in the past. The two  agencies could enhance communication  at the regional/area office and
    headquarters  levels to  better understand  each  others' priorities and roles and the  details of the new
    standardized IA.  For the DWIG-TSA program, increased communication around SDWA compliance problems
    would be worthwhile, and it appears from the December 2010 meeting of DWIG-TSA staff that  EPA is already
    committed to doing this.
    IMS is a vital partner for EPA to implement these programs.  EPA relies on IMS to implement these programs
    and help achieve improvements in  access at tribal homes.  EPA  could  not realistically take  over  IHS's
    responsibilities:  EPA does not have the in-house expertise, field presence, or relationships with  tribes that
    would be needed to do so.  In addition,  the IMS STARS database is an  important tool for the grant program
    implementation.  As noted by the  Infrastructure Task Force and interviewees for this evaluation, there are
    issues with using the STARS  database as a reporting mechanism  and project selection  tool for these EPA
    infrastructure grant programs.  However,  this database is also generally considered by program staff and other
    stakeholders to be the best source of information on tribal water infrastructure information and is thus an
    invaluable tool for these EPA programs.
    There are opportunities for improved communication with tribes about these programs.  Based  on the
    interviews with  tribal government  staff,  EPA, and IMS, the evaluators find that tribal communication about
    these programs is inconsistent and  "light" on the whole.  EPA could strengthen its direct communication with
    the tribes to enhance knowledge  of tribal  contexts and needs and more consistently  provide  tribes with
    information about funding options (lAs versus direct grants).  Increased communication  could help EPA
    understand program accomplishments and identify areas for program improvements.
    EPA has limited ability to make progress on the current strategic measures focused on  providing access to
    drinking water and clean water to tribal homes. EPA's limitations include:
            Lower program funding relative to the funding needed to provide for all identified access needs;
            A history  of funding  a substantial  proportion of projects that  do not provide  access (as  defined by
            moving from a higher deficiency  level (DL) to a lower one) due to project selection processes;
Evaluation Report - March 2011                                                                         Page 44

-------
              OF THE               AND

            Restrictions on EPA's authority to fund some projects that would provide access, such as by funding
            individual drinking water wells or providing sanitation infrastructure services to homes that are not
            on reservation land;
            Reliance on a metric which assumes changes in DLs (from a DL-4 or -5 to a lower DL) that are
            immediate and permanent, when in reality, DLs may not change in this manner as a result of funding
            (e.g., because a DL-4 project only partially funds the infrastructure needed to change the associated
            homes to a DL-3 or lower) and there is sometimes "backsliding" of DL designations due to compliance
            violations, lack of infrastructure maintenance, or other reasons;
            Substantial reliance on other federal agencies, particularly IMS, to contribute toward progress on the
            access measure; and
            Disconnects between reported progress on providing access (reported to be immediate upon project
            funding) and  actual  progress on  the ground, which typically occurs years later, after construction is
            completed and the new or upgraded infrastructure is operating.
i    EPA appears to be quite  limited  in its ability to make progress toward the current compliance strategic
    measure, though the extent of the limitations is not clear due to a lack of measurable data. According to the
    evaluation  interviews, an analysis of compliance data from  Region  10 (Alaska area), and the Infrastructure
    Task Force Access Subgroup report, the drivers of some if not most compliance problems may be related to a
    lack of sufficient O&M capacity at tribal systems or the introduction  of new drinking water rules, rather than
    to infrastructure problems that the DWIG-TSA program can address.  In addition, the allocation of DWIG-TSA
    funding has not been directly tied to compliance needs or opportunities, nor has project selection clearly and
    consistently been linked to compliance support on a national  scale.  EPA  has the opportunity to conduct
    additional data analysis that will further explore the relationship between SDWA violations and  the DWIG-TSA
    program and then to focus project selection on projects that that will be more likely to support compliance
    improvements over time.
i    EPA's meaningful contributions to meeting tribal infrastructure needs are not well reflected by the reported
    program performance.  Despite the identified challenges surrounding making  progress toward the current
    strategic measures, both  programs have made significant contributions over  the years to  Indian country
    through these programs by providing funding for vital water infrastructure needs. These contributions—and
    the public  and environmental health  benefits they have provided—have not been clearly recognized in a
    public or formal way.  As one EPA regional coordinator said, "I  don't think we hear enough about what the
    achievements are—we know what the goals are, but I don't know what the achievements have been."

Recommendations

Core Question ¥:  What implementation improvements or innovations could be made by
EPA to make the DWIG-TSA and CWISA programs more effective in meeting the  water and
wastewater  infrastructure needs of tribes?

Based on the evaluation's findings and conclusions, we offer eight recommendations and three additional items for
consideration.  We consider the first four recommendations to be policy  recommendations as they would require
approval from senior EPA management and, in the case of at least recommendations 1-3, approval from the Office
of Management and Budget (OMB).  Recommendations 5-8 are  more program-oriented and therefore could  be
considered and implemented by program staff. We present the  recommendations in this order because we feel
that they are iterative considerations for the programs.
Evaluation Report - March 2011                                                                        Page 45

-------
    EVALUATION OF THE                AND             INFRASTRUCTURE TRIBAL SET-ASIDE
1.  Clarify the goals and priorities of both programs.

To reduce confusion and  increase program focus and  management, we  recommend that EPA identify clear,
specific, and measurable goals and priorities for both the DWIG-TSA and CWISA programs. We encourage EPA to
clarify, for example, whether providing "access" (or perhaps simply "serving homes that currently lack access"
which might be preferable—see Recommendation 2) is the highest level priority for both programs,  and, if so, to
identify clear program-specific goals that correspond to this priority.  A set of cascading changes to both programs'
processes (e.g., allocation of funds, project selection) and performance measures could then logically follow as
described in  the following recommendations. For DWIG-TSA, if compliance with SDWA is the highest priority, then
making this clear would be helpful for program planning, implementation, and reporting. We are aware that EPA
has not consistently shared compliance information with IMS, so it  is likely that many compliance-related  issues
that would result in a lack of "access" as  defined by IMS are not reflected  in reported numbers. Ideally, clearer
program-specific priorities and goals would involve specific targets and performance measures that EPA can readily
track and report on.

2.  Consider changing the access strategic measures by which EPA tracks its performance to measures that
    more directly reflect EPA's  mandate, authority, and scope of influence; take into consideration data
    availability and quality; and reflect changes or clarifications to the programmatic goals made in response to
    recommendation 1.

We offer several options for EPA's consideration regarding the current access performance measures, which we do
not think serve EPA or reflect progress toward meeting tribal needs as well as other options would. Of the options
we list below, most are variations on  the current  access strategic measures, but they address some or all of the
challenges described in this report. (One option is to simply leave the current measures unchanged.)

Note that, in this recommendation, we use the term "measure" (to be consistent with the usage of this term in the
FY2011-2015 EPA Strategic Plan) for what would generally be considered a 5-year "goal" or "target" against which
performance would be "measured" from year to year.  Also, if EPA's highest priority for the DWIG-TSA program  is
to support SDWA compliance,  EPA may want to consider dropping the drinking water access strategic measure
altogether in support of a measure that is  more clearly related to compliance for the program  (see  also
recommendation 3, below).

Table 11 below provides a list  of options  for the  access-related strategic measures, including pros and cons for
each option.  This list and associated pros/cons are not necessarily exhaustive; they are intended to illustrate
tradeoffs between alternatives.

Options 1-5  assume that the number of homes will be counted using the PDS data of the IMS STARS database; the
STARS database, while imperfect for  EPA's uses, is currently the best available national data on the number of
tribal homes and water infrastructure  needs. Options 6-7 assume use of the EPA Safe Drinking Water Information
System (SDWIS) by connecting EPA project data on the water systems served with tribal population data available
inSDWIS.
Evaluation Report - March 2011                                                                         Page 46

-------
    EVALUATION OF THE DRINKING WATER AND CLEAN WATER INFRASTRUCTURE TRIBAL SET-ASIDE GRANT PROGRAMS
                          Table 11. Options Pertaining to Access Strategic Measures
Options -Both Programs Pros
1. By [YEAR] provide infrastructure services71 to [X]
tribal homes
2. By [YEAR] provide infrastructure services to [X%]
of tribal homes
3. By [YEAR] fund [X%] of projects [or could be a %
of funding need] with initial DL-4 or DL-573
4. By [YEAR] provide infrastructure services to [X%]
of homes with DL-4 or DL-5 designations
5. By 2015, in coordination with other federal
agencies, provide access75 to safe drinking water
for 136,100 AI/AN homes and basic sanitation to
67,900 AI/AN homes (Current measures)
Options - DWIG-TSA only
6. By [YEAR] provide infrastructure services to [X%]
of all tribal community water systems
7. By [YEAR] provide infrastructure services to [X%]
of the tribal population served by community
water systems
A, B, C, G
Note: One IMS measure is very
similar72
A, C, G
D, G
D, G
Note: One IMS measure is very
similar74
B, D, E, F
^^ifl^^^^^H
I,J, 0
I,J, 0
1, (M -depending on %
chosen), 0
I,K L, M, 0
1, J, K, L, M, 0
I^S
More concrete, comprehensive
measure of EPA's influence,
would provide a numerical
context (total tribal population
served by PWSs), would not rely
on external data source
Would require additional
data analysis by EPA,
would not reflect AI/AN
population not served by
public water systems
Pros
    A.  Relatively clean, clear measure of homes provided some type of infrastructure service each year with EPA
        funding; recognizes and "gives credit" to all work done by EPA
    B.  Does not require  a  percent or other  relational number that would be tied to a potentially-fluctuating
        number of homes  in the homes universe (STARS homes inventory), which, given its fluctuations, can make
        EPA's performance look poor even if  it is serving more than its original target number and percentage
        (relative to the original universe) of homes
    C.  Does not assume that EPA is or should  be exclusively funding DL-4/5 needs, or that all IDL-4/5 projects will
        result in "access" as defined by moving to  DL-3 or lower
    D.  Supports the US access goal
    E.  May be able to reduce  burden on EPA for performance and accountability given that the measure is "in
        conjunction with other federal agencies"
    F.  Has been approved by senior EPA managers and OMB
    G.  Is not dependent  on other agencies' funding, project  selection, or implementation  (in some cases other
        agencies may co-fund projects with EPA, but the measure could be tracked only in relation to EPA-funded
        work)
    H.  Provides context  through a percentage  or other relational  number intended to  represent the larger
        population
Cons
    I.   Homes will sometimes be double counted (extent unknown)
71 Infrastructure services (or a similar term) would coverall projects funded, rather than just projects serving homes with high deficiency levels,
and would include both significant upgrades to systems, as well as partial upgrades.
72 IMS measure (FY2006-2011 IMS strategic plan): Number of new or like-new AI/AN homes and existing homes provided with sanitation facilities
(Targets: FY2010: 21,811; FY2011: 21,500).
73 Some projects with IDLs of 4 or 5 also have FDLs of 4 or 5 due to other existing issues.
  IMS measure: Percentage of existing homes served by the program at DL-4 or above (Targets: FY2010: 37%; FY2011: 37%)
75 "Access" is defined as a shift in deficiency levels (in the IMS STARS data system) from a 4 or 5 to a 1-3.
Evaluation Report-March 2011
Page 47

-------
    EVALUATION OF THE                AND            INFRASTRUCTURE TRIBAL SET-ASIDE

    J.   Would not have a  numerical context (such as the total universe of homes)  from which to gauge
        relative/proportional performance
    K.   Assumes immediate "access" results even though projects take years to complete and therefore for the
        access or benefit to be realized on the ground immediately after funding
    L   Does not account for factors, such as compliance violations that result overnight from the introduction of
        new EPA rules or lack of adequate maintenance that could cause home to move from a DL-1-3 to  a DL-4-5
        and in so doing overlook the improvements EPA has invested in and achieved toward the measure
    M.  Does not reflect or give "credit" to the good work being done through projects that  are serving DL-1-3
        homes (a sizable proportion of projects have IDL designations of DL-2 or -3)
    N.   Puts pressure on EPA to fund a lot (or essentially all, depending on the  numbers chosen) of DL-4/5 homes
        regardless of whether EPA has the authority to (e.g., individual  home wells; sanitation projects not on
        reservation land), cost feasibility, funding levels, availability of access needs, etc.
    O.   Reporting on the measure from STARS does not necessarily include direct grants, and therefore does not
        reflect all EPA work or influence.

3.   Consider changing the DWIG-TSA compliance strategic measure to one that more directly reflects the
    drivers of compliance problems and EPA's scope of influence over these problems.

A revised compliance measure targeted specifically to infrastructure-solvable compliance problems that  EPA has
the ability  to  address  would more directly  and clearly reflect  the  influence  of the DWIG-TSA  program on
compliance. We offer three options for such a measure.  All three options assume that DWIG-TSA funds (or some
portion of these funds) would be targeted to those public water systems (community water systems and non-profit
non-community water  systems)  that, based on  additional data analysis  conducted by EPA regions, would
reasonably  be expected to  see  improved compliance as a  result of  receiving new  infrastructure  through  this
program.

Option 1:   By  [X YEAR], achieve [x] reductions in health-based compliance violations at systems funded by the
DWIG-TSA program.

This option  would likely require tracking over some years the compliance improvements that would be expected to
result after new infrastructure is completed and up and running.  Therefore, it could be difficult to track direct
compliance improvements from a particular  project in the year it was funded.

Option 2: By [X YEAR], target [X] percent of DWIG-TSA funds to those systems that have had compliance violations
that can be solved with new infrastructure funded through DWIG-TSA.

This relatively straight-forward option would aim to ensure that funding is focused (to an agreed  upon  extent) on
compliance improvements.   One advantage of this option  would be that it could be done  annually—without
waiting for years for the compliance benefit  to be realized. A disadvantage would be that it is more focused on the
activities of the DWIG-TSA program, rather than the expected or desired outcomes.

Option 3:  By [X YEAR], [X] percent of the population in Indian country served by community water systems that
have received DWIG-TSA funding will receive drinking water that meets all applicable health-based drinking water
standards.

Option 3 is  a variation on the current compliance strategic measure.  It would  limit EPA's commitment to DWIG-
TSA funded community water  systems  (rather than  all tribal  community water  systems) to more  directly
understand and track influence  of the program.  Like the other  options above, this would also ideally involve
Evaluation Report - March 2011                                                                         Page 48

-------
    EVALUATION OF THE                AND             INFRASTRUCTURE TRIBAL SET-ASIDE

increased  analysis of compliance drivers and funding targeted  specifically to systems that have problems that
DWIG-TSA funding could  help to solve or could help to prevent.  (Without these additional  data analysis and
targeted funding steps, EPA would continue to have a  measure for compliance that risks being outside of the
DWIG-TSA program's sphere of influence.)  Like option 1 above and the current compliance measure, compliance
improvements would take time (usually years) to be realized, creating a time lag between each year's DWIG-TSA
funded work and  compliance outcomes.   EPA would need  to consider  how and if  to include compliance
information for non-profit non-community water systems that receive funding, as well as state-delegated water
systems that receive funding (e.g., systems serving Alaska Native Village).

4.  Reassess the national annual budget allocation for both programs to be more clearly tied to EPA's goals and
    priorities.

Once  EPA has determined whether and how to update  each program's goals, priorities, and measures, deciding
whether and how to update the allocation formulas for one  or  both programs will be easier; however, we
recommend updating the  allocation formulas even if EPA  makes no changes in response to recommendations 1-3.
CWISA Allocation
For the CWISA program, EPA current allocates resources based on sewer infrastructure needs identified in each IMS
area's SDS priority list.  Per the CWISA national guidelines, EPA regions select projects in rank order down their
respective IMS area office SDS list(s), unless there are reasons to skip  projects as described earlier in this report.

Allocating funding based on the SDS list does not align directly with DL designation (i.e.,  "access") or with EPA's
authority to fund (e.g., projects listed that are not within tribal jurisdiction; projects that would require plumbing
or water connections from a home).  The CWISA allocation formula is thus allocating based on needs in the IMS
areas that do not necessarily support progress on the access strategic measure for basic sanitation.

A more direct way to support access (assuming access continues to  be a priority and strategic measure) would be
to use the DL designation alone and divide funding proportionately according to those EPA regions or IMS areas
that have  the highest proportion of DL-4/5 projects from the SDS list.   If, however, EPA were  to focus more on
homes served, rather than homes provided  access, the current allocation method may be the best method, given
that it prioritizes the highest nation-wide sanitation  needs. There may also be ways to factor in EPA priorities such
as decentralized infrastructure, though we are not clear from the work done for this evaluation  how EPA could do
this.
DWIG-TSA Allocation
For the DWIG-TSA program,  allocation is based on a  combination  of information from DWINS and the SDS list, after
each region is first allocated  two percent of the annual funding available.

The current two percent  base allocation  means that all  regions  with federally recognized tribes receive funding,
even if there are no projects that would have significant public health impacts, otherwise improve compliance with
SDWA, or provide access to  safe drinking water in a region. We recommend that EPA consider removing this base
allocation  in  favor of a formula that would be more needs- and results-focused based on (clarified) program goals
and priorities.

The DWINS  results, once completed in  the next few years, will  be a  helpful resource from which  to better
understand tribal infrastructure needs and consider allocation.
Evaluation Report - March 2011                                                                          Page 49

-------
    EVALUATION OF THE                 AND              INFRASTRUCTURE TRIBAL SET-ASIDE
In the meantime, if EPA considers provision  of access to be the highest program priority, a more direct way to
support access would  be to use the DL designations alone and divide proportionately according to  those EPA
regions that have the highest proportion of DL-4/5 projects from the SDS list.

The current allocation  method does  not consider SDWA compliance history or issues. Assuming compliance with
SDWA is a priority, if not the highest  priority, for the DWIG-TSA program, EPA could allocate funding to the regions
in proportion to health-based  compliance violations  in a given year.  If this method were used, it would be
extremely important for EPA to have a deeper understanding of the drivers of the compliance problems, isolating
infrastructure-related violations that would be appropriately solved through  DWIG-TSA funding versus through
other technical assistance or enforcement options. As EPA staff are also aware, focusing funding on systems with
infrastructure-related compliance problems may create a disincentive for systems to comply in the first place. EPA
could also choose to allocate based on compliance vulnerabilities  (versus violation histories);  however, this too
would require deeper  analysis  by EPA headquarters  and regions to understand  systems that are vulnerable to
infrastructure-related compliance issues.  We recommend that, to the extent that EPA considers SDWA compliance
as a priority, EPA shift allocation to more directly support compliance.

5.  For both programs, update and clarify expectations for project  selection to be more clearly in line with the
    program priorities, allow for regional flexibility and discretion,  and promote increased consistency and
    transparency.

The project selection processes for both  programs have  their bases and merits, but neither process is ideal: For
the DWIG-TSA program, consideration of both compliance needs and provision  of access varies from region to
region; regions often choose projects based on other  considerations, making it difficult for EPA to make progress
on its strategic measures.  For CWISA, exclusive use of the SDS list is limiting and does not recognize EPA's distinct
mission, authority to fund, or priorities, including provision of access.

Updating the project selection guidances to  focus on  a  simple set  of core program elements  while allowing for
regional flexibility and  discretion would provide more  coherency across the regions, particularly for the DWIG-TSA
program, and more direct linkage to program priorities for both programs.  In the guidances,  EPA could provide
guidelines on use of the SDS76 and other data  sources.

EPA headquarters  and regional staff could create  a simple form that lists each program's priorities for funding.
Regions could use this form to  show how the projects they have selected match with the identified priorities or
core program elements. (We are suggesting that the regions list only those projects they select for funding, not all
potential projects that  are not funded.) This form could include basic questions or  checklists such as the following:
  The IMS SDS list is an important resource with no clear substitute, but given its limitations for EPA's purposes, SDS is not the ideal information
source from which to pick projects without considering other information sources. For example, the SDS list does not directly represent EPA
priorities such as SDWA compliance, infrastructure security, and decentralized wastewater infrastructure, and it includes tribal homes and
projects that are not eligible for EPA funding. Further, choosing projects in rank order from the SDS list is not likely to support strong
performance toward the access measure.
Evaluation Report - March 2011                                                                            Page 50

-------
               OF THE                AND
      1.   Which of EPA's priorities are addressed through this project (check all that apply)?
              a.   Priority 1
              b.   Priority 2
              c.   Priority 3
              d.   Etc.
      (Questions 2-3 may not be needed if EPA headquarters staff can quickly access this information directly
      from STARS)
      2.   Was this project on the IMS SDS list? (Y/N) Optional comments:	
              a.   If yes, which IMS area?
              b.   If yes, what is the SDS ranking?
              c.   If yes, what are the project's estimated IDL and FDL?
      3.   In your assessment (or based on an IMS or other assessment), does the tribal community receiving
          funding have adequate capacity to manage the results of the project? (Y/N) Please describe.
      4.   Has the direct grant option been communicated to the funding recipient?
      5.   Will this  project be funded through a direct grant or an interagency agreement with IMS?
      6.   What are the estimated construction start and construction completion dates?
      For DWIG-TSA:
      7.   Was past compliance history (where relevant) evaluated?
      8.   Are past compliance violations able to be addressed through funding for infrastructure?  (Y/N)
          Optional comments:	
      9.   Have other compliance improvement options been considered and pursued?
Regions could submit the populated form to a central information source.  Headquarters staff would not be
selecting or rejecting projects for funding, but could ask questions to clarify why certain projects were selected.
This type of transparent,  priority-driven, yet flexible approach would likely make EPA less vulnerable to scrutiny
from national decision makers or auditors who could otherwise cast doubt on how selected projects support EPA's
priorities.

To further support the selection of DWIG-TSA projects that address compliance issues, EPA regional staff could also
continue working with regional compliance programs to support a more unified EPA approach, whether that be
through considering how to promote prevention of compliance problems, considering  enforcement options in lieu
of DWIG-TSA  funding, or other options.  During the meeting of EPA regional coordinators in December 2010, the
DWIG-TSA coordinators decided to more proactively share compliance information with IHS area offices to ensure
that priority compliance needs are reflected on the  SDS list.  This increased information sharing between the
agencies could help IHS to be better aware of the compliance issues and improve the reflection  of compliance
issues in the SDS  ranking.

6.  For both  programs, routinely collect and analyze data to enhance transparency and strategic coordination
    and improve EPA's ability to report on, advocate, and improve both programs through learning and
    adaptation.
Evaluation Report - March 2011                                                                          Page 51

-------
    EVALUATION OF THE                AND             INFRASTRUCTURE TRIBAL SET-ASIDE

Improved information availability will enable EPA to better understand its contributions through these programs
and support learning over time to improve program effectiveness.  Given there is currently very little information
collected by EPA nationally other than basic funding statistics, even a modest increase in data collection would
provide a markedly improved ability to understand, explain, and advocate for these infrastructure programs at the
regional and  national levels.  The type of information that could be beneficial  would not be complicated or
extensive. It would likely include the following.
    Summary information about projects funded  similar to what  is currently collected and supplemented with
    information from the project selection form described  in recommendation 5.
    The current status of the project on an annual  or semi-annual basis. This information could be taken from the
    STARS milestone information or from grant reports if projects are administered through a direct grant with no
    IMS involvement.
    For the DWIG-TSA program, an annual  summary of the current compliance trends at  tribal water  systems in
    the regions and a "diagnosis" about which of  these problems are potential candidates for DWIG-TSA-funded
    solutions.   This type of information  would  help the DWIG-TSA program  better target  funding toward
    compliance issues that can be addressed with infrastructure funding, as well  as understand when there are
    issues that cannot  be addressed with  this funding but could be supported by other  EPA programs, such as
    technical assistance.

An on-line tool that is professionally developed using current technology would allow for easy data entry, data
access, and reporting. With an on-line reporting tool, the reporting burden  on the regions would be modest and
infrequent,  and the benefits to the program would be worth the effort (including some decreased burden on the
regions from headquarters' staff making as many ad hoc inquiries).

EPA headquarters made a previous attempt to use  an on-line database to track  information for the  DWIG-TSA
program. We understand from  interviews  that users found the database to be difficult to use  and update and
therefore it was not  consistently populated and was  ultimately abandoned.   In addition, some regional
interviewees indicated that the EPA Integrated Grant Management System (IGMS) already has some of the kind of
data that we have suggested collecting, and that any additional reporting beyond what is already required would
be duplicative. It is our understanding that IGMS does have some basic data on EPA grants, but that the data on
lAs  is  very  limited, and on  the  whole,  that IGMS does  not  have most of the  information we would suggest
collecting. We would, however,  encourage  EPA headquarters staff to directly address and clarify this question of
dual reporting from IGMS, and wherever possible (whether from IGMS, STARS, or another source), gather data
from existing information first before asking the regions to  report  data.

A retrospective on the reporting done under ARRA would further help to identify when ARRA-like reporting may
helpful for routine program implementation.

7.   Update the national guidelines for both programs.

Following the consideration  of the above recommendations and  any resulting changes, we recommend that EPA
headquarters work with EPA regions to update the  guidelines for both programs to:
    Reflect any changes or clarifications in goals, priorities, and measures;
    Reflect any changes or clarifications to expected project selection processes;
Evaluation Report - March 2011                                                                         Page 52

-------
    EVALUATION OF THE DRINKING       AND            INFRASTRUCTURE TRIBAL SET-ASIDE GRANT

    Clarify  (or reiterate)  what projects  are  eligible for  program  funding,  based  on  experiences  with
    implementation over the course of the program;
    Clarify reporting expectations;
    Clarify the expected communication and roles regarding lAs (including interactions with IMS) and direct grants;
    Make the guidelines as concise and clear as possible; and
    For the DWIG-TSA program, also clarify expectations around:
            Identifying and communicating about compliance violations to IMS area staff;
            Tracking the cause of compliance violations (i.e., whether infrastructure related  or not) and whether
            there is a possible infrastructure fix to the problem; and
            Working with other regional compliance staff to pursue available options for improving compliance.

In addition, EPA headquarters should  work with  the EPA regions to update any communication  materials about
these programs. For example, the CWISA program has a Frequently Asked Questions document developed in 2007
that should be updated with new information about eligible projects.

8.  Continue to enhance and improve communication within EPA, between EPA and IMS, and between EPA and
    the tribes.

As discussed in the findings section, communication between  EPA headquarters and EPA regions, between EPA
and IMS, and between EPA and the tribes is important to effectively implement and measure the impact that these
programs have on tribal communities.  Although the evaluators heard from interviewees many positive reflections
on communication, we also heard that communication improvements in the following areas would be useful:
    Improving the "we are in this together as one agency" collaboration between EPA  headquarters and EPA
    regions;
    Ensuring that calls and meetings on these  programs have clear objectives and substantive exchanges, rather
    than holding them just for the sake of holding them;
    Increasing the frequency and quality of communication between the EPA regional staff and IMS area office
    staff where communication is currently sparse.  This communication is particularly important around sharing
    of  compliance history  for inclusion into the  IMS area SDS lists, but communication would also be valuable
    around each agency's (and region/area's) priorities, project selection considerations, and knowledge of tribal
    specifics;
    Encouraging EPA regional staff to  have  some direct contact with the tribes where they do not already.  At a
    minimum, this contact could be a  letter sent to tribes reminding them of  EPA's role in funding, including the
    process for project selection and the direct grant/IA options. There may also be the opportunity or need for
    more direct communication with tribes around compliance issues, though in some regions, this could likely be
    handled through other regional compliance program channels; and
    Considering ways to  highlight the accomplishments of the program  that may not be  reflected through
    numbers, both within EPA and to  external  stakeholders, such as through the issuance of an  annual program
    report.
Evaluation Report - March 2011                                                                         Page 53

-------
    EVALUATION OF Til	IE                 AND              INFRASTRUCTURE TRIBAL SET-ASIDE

Additional Considerations

Below we offer three additional considerations which reflect our support for efforts that are already underway or,
in one case, previous work done to reflect on many of the same issues that this evaluation has covered.

1.  Continue to focus on the critical issues surrounding infrastructure sustainability, including tribal O&M
    capacity issues.

It is clear that maintaining water  infrastructure for the duration of its expected lifespan is critical for ensuring the
best use of taxpayer dollars and  for best supporting tribal needs in the long run.  In  2010, EPA issued  its Clean
Water and  Drinking Water Infrastructure Sustainability  Policy,  which encourages all communities to develop
sustainable  systems that employ effective utility management  practices to build and  maintain  the level  of
technical, financial, and  managerial capacity  necessary to ensure  long-term sustainability.77 As  this evaluation
found, tribal systems have  particular challenges  associated with adequate O&M capacity to sustain and  maintain
infrastructure.

Given the number of underlying factors that contribute to infrastructure longevity, including ensuring appropriate
design and the role of ongoing maintenance and operation, the evaluators recognize that, to a large extent, the
long-term effectiveness  of these infrastructure programs will be determined by tackling  these bigger-picture
challenges.   At the same  time,  we  recognize  EPA's statutory  limitations  (e.g.,  these  programs  cannot  fund
operation and maintenance)  and resource constraints.  Therefore we  encourage EPA to work with other federal
agencies as well as with tribes to  identify and address these complex, but critically important, sustainability issues,
including identifying any needed changes to  federal statute(s).  Reports from the Infrastructure Task  Force on
meeting the access goal and on improving technical assistance to AI/AN communities have recommendations for
all the involved federal agencies to consider to best support infrastructure longevity, including addressing funding
for O&M and coordinating current technical assistance programs to maximize  their reach.

2.  Continue to communicate about IA requirements, expectations, and inter-agency roles.

We heard from IMS and one  of  the tribal interviewees that  clarifying IA requirements, expectations,  and roles
would be useful. Some, if not all of these individuals were already familiar with the new standardized IA terms and
conditions, yet still expressed concerns about  different interpretations of the requirements and expectations, and
at least two IMS staff noted continued concerns that lAs are perceived incorrectly as grants by EPA. Additional or
reiterated clarification of requirements, expectations, and roles may be helpful.

3.  Reflect  on other reports and  recommendations that pertain to these programs.

Other efforts have or are examining several of the same issues that this evaluation has covered, including statutory
limitations of EPA to provide funding for certain needs (e.g., operations and maintenance, individual wells), the gap
between the funding  needed  to  meet access needs and actual funding, and the challenges of tracking progress
toward  the  United States'  access goal using the available data.  The  Infrastructure Task  Force  Access  Subgroup
  U.S. EPA. Clean Water and Drinking water Infrastructure Sustainability Policy. Available:
http://water.epa.gov/aboutow/upload/Sustainabilitv-Policv.pdf.
78 Federal Infrastructure Task Force Access Subgroup. Meeting the Access Goal: Strategies for Increasing Access to Safe Drinking Water and
Wastewater Treatment to American Indian and Alaska Native Homes. (March 2008).
Tribal Technical Assistance Workgroup, Federal Infrastructure Task Force on Tribal Access to Safe Drinking Water and Basic Sanitation.
Strategies for Improving Technical Assistance Delivery in American Indian and Alaska Native Communities (March 2010).
Evaluation Report - March 2011                                                                             Page 54

-------
    EVALUATION OF THE                AND             INFRASTRUCTURE TRIBAL SET-ASIDE

developed a report in 2008, "Meeting the Access Goal: Strategies for Increasing Access to Safe Drinking Water and
Wastewater Treatment  to American  Indian  and  Alaska  Native  Homes."   This  paper made  a  number of
recommendations for the involved federal agencies to consider in the areas of infrastructure funding, operational
and maintenance cost and capacity, and programmatic coordination. The Infrastructure Access Task Force Tribal
Technical Assistance Workgroup also developed a  report on strategies for improving technical assistance delivery
in tribal communities. Several recommendations in  both of these reports are still valid and  consistent with the
findings and recommendations in this evaluation, and  we encourage  EPA to reflect upon this work and other
efforts, as well as the recommendations  in this report to determine best steps forward for these programs and
their interaction with other federal work underway.
Evaluation Report - March 2011                                                                          Page 55

-------