Local Government Auditing Quarterly

Local Government Auditing Quarterly The Journal of Local Government Auditing | Fall 2015 Auditing Housing & Homelessness Featured Articles • Auditi...
Author: Gloria Wright
0 downloads 1 Views 1MB Size
Local Government Auditing Quarterly The Journal of Local Government Auditing | Fall 2015

Auditing Housing & Homelessness

Featured Articles •

Auditing Housing Through a Performance Management Lens

Homelessness and Affordable Housing Challenges in Paradise, Audit Subjects Abound

Auditing Homelessness: Uncovering the Human Element

Frequent Utilizers of Health, Emergency, and Other Services, and the Impact of Lack of Housing

Plus Others…



About the Quarterly The Local Government Auditing Quarterly (LGAQ) is published four times a year – in September, December, March, and June – by the Association of Local Government Auditors (ALGA) Association of Local Government Auditors 499 Lewis Hargett Circle Suite 290 Lexington, KY 40503 (859) 279-0686 Opinions expressed in the Local Government Auditing Quarterly are those of individual authors, and they may differ from ALGA’s policies, official statements of ALGA committees, or those of an author’s employer.

LGAQ Editor Emily Jacobson City and County of Denver, CO

Social Media Follow ALGA at algaonline.org Follow ALGA on Twitter at twitter. com/ALGA_Gov

From the Editor When I joined the ALGA Publications Committee three years ago, I did not anticipate that I would eventually take on the role of Editor of the LGAQ. In fact, I was somewhat unsure about how much I could even offer the committee as a non-auditor. My role in the Denver Auditor’s Office as Communications Specialist is largely to edit our audit reports, so I knew I could bring my eye for detail and knowledge of grammar and style. Thankfully, I realized quickly that contributing to an ALGA committee has more to do with your willingness to pitch in and take on projects than your degree of Yellow Book knowledge. During my time as a Publications Committee member, I have been involved in a variety of initiatives, including developing a more proactive approach to author outreach, crafting the first ALGA annual report, creating a style guide for use by committee members when proofreading LGAQ articles, establishing guidelines and tools to carry out ALGA’s communications strategy, and moving the Quarterly to an all-digital format. Between working on those projects and serving as an Assistant Editor of the LGAQ, ALGA has become a big part of my professional life. In fact, I almost feel like an auditor. Coming on as Editor could not have come at a more exciting time. As Justin Anderson shared with you in the Summer 2015 issue, we have been working on a makeover of the LGAQ, which is being debuted in this issue. (Thanks again to Justin for serving as Interim Editor while I was on maternity leave from May through July!) Although both the online and PDF versions of the articles have an updated look and feel, the changes are most noticeable in the PDF. This updated look ultimately benefits the reader with the following features: • • • • • • •

A redesigned cover highlighting what’s inside A reformatted table of contents for easier navigation An ALGA training calendar More photos for greater visual interest Better use of white space for ease of reading Pull quotes to highlight salient points Brief author professional biographies

The entire Publications Committee is indebted to Kristine AdamsWannberg and A.D. Lewis who came up with the new design and to Glenda Johnson from Member Services who created a new template to realize their creative vision. The result is a clean, professional publication that feels more modern while preserving the ALGA brand. We hope you enjoy it and welcome your feedback on the new layout. LGAQ Fall 2015 | Page i



FEATURES (cont.)

2 Training Calendar

19 The Audit that Never Dies Niki Raggi

COLUMNS 3 Opportunities for Improvement Gary Blackmer 7 AudiTechie Nicole Dewees FEATURES

24 Developing a Methodology to Examine Non-Governmental Organizations (NGOs) Receiving Local Government Funds to Provide Homeless Support Services Julie A. Lebowitz, CIA, CFE 31 Auditing Homelessness: Uncovering the Human Element Carl Halvorson and Abby Musfeldt

11 Auditing Housing through a Performance Management Lens Kari E. Guy

35 Frequent Utilizers of Health, Emergency, and Other Services, and the Impact of Lack of Housing Sean DeBlieck

15 Homelessness and Affordable Housing Challenges in Paradise, Audit Subjects Abound Susan Hall

SUBMIT TO THE QUARTERLY 40 Submitting Abstracts, Articles, and Member News

From the Editor (cont.) With all the changes going on behind the scenes, I don’t want to understate the quality of the content you’ll find in this issue. The theme is Housing and Homelessness and, collectively, the corresponding feature articles provide a solid base of understanding about common issues identified through audits in this area. Affordable housing is reaching a point of crisis in many communities. You’ll read about relevant audits from Austin, Denver, Honolulu, Portland, and Washington, D.C., and from western

Washington. Additionally, we have excellent content from two of our regular columnists, Gary Blackmer and Nicole Dewees. I’ll close by inviting you to get in touch with me. What do you want to read about? What types of articles do you get the most out of? Which issue have you (virtually) dog eared? I’d love to know. Call me in Denver at 720-913-5073 or email me at Emily. [email protected]

LGAQ Fall 2015 | Page 1



October 6-7, 2015 Calgary, Alberta

The New IPPF: What to Expect October 13, 2015 2:00 - 3:00 p.m. Eastern time

Areas of focus will include: • Developing a Dynamic Risk Based Audit Plan • Incorporating Fraud Risk into a Dynamic Audit Plan • Best Practices for Overcoming Common Data Analytics Challenges • Data Analytics Panel Discussion

How to Audit Wireless Communication Devices: Can You Hear Me Now? November 10, 2015 2:00 - 3:00 p.m. Eastern time Topic: Benford’s Law December 9, 2015 2:00 - 3:00 p.m. Eastern time

Fees Both Days One Day

$350 $250

October 8-9, 2015 Portland, Oregon Areas of focus will include: • From Audit Practitioner to Elected Official • Writing Reader Friendly Audit Reports • Program Evaluation: Techniques and Applications for Performance Auditors • Lean Government: What It Takes for a Successful Program • Using Lean as an Auditor: What Is Our Role? • IT Frameworks: What Are They and When To Use Them • Fraud Detection and Documentation • Project Management Techniques for Efficient Audit Projects • Data Visualization and Online Reports Fees Members Both Days One Day

$300 $150

Non-Members Both Days $350 One Day $175

LGAQ Fall 2015 | Page 2

ANNUAL CONFERENCE 2016 ALGA Annual Conference May 23-24, 2016 Sheraton Austin at the Capitol Austin, Texas Pre-conference workshops will be held on May 22; post-conference workshops will be held on May 25.

EVENT REGISTRATION AND MEMBERSHIP MANAGEMENT PORTAL ALGA’s event registration site can be found at alga.membershipsoftware.org. At this site you can: • • • •

Register for any ALGA event Update your contact information and renew your membership Access ALGA’s online directory Access ALGA’s members-only online training resources

Questions about the event registration and membership management portal may be directed to ALGA Member Services at (859) 276-0608.


AUDIT RISK AND BENEFIT By Gary Blackmer Director, Oregon Audits Division

Good audits take risks. I would define a good audit as one that has a big positive impact on operations or decisions. I also recognize that a sound understanding of audit risk is important in deciding what to audit. If we think there is a high likelihood of encountering serious problems that we can’t mitigate, then the wise auditor should reconsider that audit topic. Our audit profession, and yes, the character of those it attracts, is often risk averse. We spend many hours finding errors, seeing dangers, predicting problems, and imagining failures. We shouldn’t be surprised if we turn that tendency on ourselves. Risk is a part of life, but I’ve always been intrigued by the contrast in how it is seen between business and government. Business managers often talk about ‘risk appetite,’ while government managers often use the phrase ‘risk tolerance.’ As more public scorn is directed at government, I think the gulf widens between those two worlds. The IT field is a great example of this difference in risk-taking and avoidance. Developers in the private sector will now build an application, run it, get user feedback, and fix the problems in a quick iterative process. To avoid the risk of criticism, government software developers are expected to produce a final, fully operational project that is bug-free on the day of release. There are many reasons why government systems fail, but I think risk-aversion plays a role during development, and is also the reason agencies decide not to upgrade old systems. If you think your audits take too long, or you are choosing small, lower risk issues, perhaps you have the same aversion. LGAQ Fall 2015 | Page 3


A sound understanding of audit risk is important in deciding what to audit.

As we started developing a procedure for addressing audit topic selection, I got to thinking about the other side of the equation, the benefits of audits. Every gambler knows that you need to judge the reward relative to the risks. Barb Hinton, from the Washington State Auditor’s Office, gave me a great idea with a matrix she developed on audit risks. I modified her table, hopefully for the better, and added a benefits table. Developing its counterpart on the benefits side made perfect sense. Below is the Audit Risks table that is included in our work papers. As we end our scoping phase, the team proposes a specific audit finding to plan, gather evidence, and report on. It makes them think about the difficulties they are likely to encounter in pursuit of that topic, and to begin thinking about a plan and methodology to mitigate those risks. We ask them to shade where their topic falls in each row, adding notes in a cell if they like, but ultimately we can talk it through and feel comfortable that we all understand the risks and steps to reduce them.

LGAQ Fall 2015 | Page 4


Discussing benefits before undertaking an audit can help everyone understand the reason for accepting the risks.

Below is the counterpart, the Audit Significance table. I started it, and our redesign team fleshed it out. Again, the audit team thinks through the value and impact of the potential audit topic and shades the appropriate cells, though every row doesn’t need to be completed. Each audit faces a wide variety of risks, but the benefits of the audit are narrower and more known. Discussing them before undertaking an audit can help everyone understand the reason for accepting the audit risks, and the audit work hours. You’ll see that the bottom row of each table is the same, addressing the context of risks and impacts. It is kind of a catch-all for those issues that don’t fall into any of the other rows or cells. These tables are written with descriptions or examples to help auditors see some of the differences between low, medium, and high categories. If the wording doesn’t work for your audit organization, feel free to reword it! Yellow Book standard 6.04 generally addresses audit significance, but it’s much more restricted to the activities within the scope of the audit, rather

LGAQ Fall 2015 | Page 5


than the impact of the audit itself. As I’ve said before, the Yellow Book is missing a chapter – one that would address the considerations and methods in choosing audit topics. If we are expected to follow the spirit of standard 2.10 then we must set standards to help us apply our efforts where we can produce the greatest value for the public. Most importantly, if the members of the audit team have a sense of the benefits that can result from the audit, it can strengthen their resolve to undertake the challenges of that topic.

About the Author Gary Blackmer has been conducting audits for 30 years, currently serving as Director of the Oregon Audits Division. The Division conducts performance, financial, and information technology audits, monitors financial audits of local governments, and responds to hotline allegations. Previously, Blackmer served 10 years as the elected Portland City Auditor, eight years as elected Multnomah County Auditor, a management auditor, and analyst for a variety of state and local agencies. Blackmer is a past-Chair of the Pacific Northwest Intergovernmental Audit Forum, and past-President of the Association of Local Government Auditors. He received the ALGA Lifetime Achievement Award in 2015.

LGAQ Fall 2015 | Page 6



Auditors love to analyze data, so one of our greatest frustrations is receiving data that needs to be processed before we can analyze it. I worked on a property tax audit last year and the data needed to be cleaned up before we could get to the fun part (the analysis). The property tax data was only available as a complicated comma delimited text file. The Assessor’s Office gave us a file with about half a million tax accounts and many of the properties qualified for one or more of the roughly 100 different tax exemptions. The file listed each of the property exemptions with a comma separating them. The corresponding value of the exemption was in a different column with a comma separating the values. FINDING THE TOTAL VALUE OF EACH EXEMPTION I have created a much simpler version of this to use as an example. In Exhibit A, each property has an account number and the property can receive a tax exemption for being a farm, forest, and/or recreation property. Some properties qualify for more than one exemption. I need to know the total value of each exemption type. In the following exhibits the header “Exemption Amount” refers to the value of each exemption type. Exhibit A

LGAQ Fall 2015 | Page 7



The Text to Columns Wizard in Excel is the key tool to separate

The first step would be to parse the comma delimited data. In this example, I know that there are up to three exemptions available, so I move the exemption amount over so that there are three columns available for the exemption types (columns B, C, and D). I click on column B and then I go to Data and then select “Text to Columns.” (See Exhibit B) Exhibit B

and process comma delimited data.

Then I select comma as the delimiter. I click “Finish.” (Exhibit C) Exhibit C

I repeat this step for the exemption amount and add titles to the columns. My spreadsheet now looks like Exhibit D.

LGAQ Fall 2015 | Page 8


Exhibit D

CALCULATE EXEMPTION AMOUNTS Now I add three new columns to calculate the exemption amount for each type. Next I want to add a formula that will look for each type and show the value that goes with it. In Exhibit E, you can see the formula I typed. This formula says, “If the exemption type is farm, then show me the corresponding amount. If it doesn’t say farm, then put zero.” The formula looks at each of the three columns where farm might exist. I made sure to include dollar signs in the formula to indicate an absolute reference. This means that if I drag the formula to adjacent columns and rows, it will still point to the same thing. For example, the dollar sign in $B2, means that I always want that part of the formula to reference column B, but I want the formula to reference different rows if I drag the formula down. Exhibit E

SUM THE COLUMNS In order to drag the formula down and to the right, I click on the cell H2 and move my mouse until my cursor becomes a black plus sign. Then I drag the formula across and then down. Then I sum the columns to get the total exemption amount for each type. The formulas should looks like Exhibit F.

LGAQ Fall 2015 | Page 9


Exhibit F

The end results should look like Exhibit G.

This formula can easily be adapted if you have more exemption types. Now that the parsing and processing is complete, the real fun can begin.

About the Author Nicole Dewees is a Senior Management Auditor in the Multnomah County Auditor’s Office. She has a Master’s degree in Business Administration and is a Certified Internal Auditor. Prior to joining Multnomah County, Nicole worked as a management analyst for the Washington State Employment Security Department and as a research analyst for the Washington Small Business Development Center.

LGAQ Fall 2015 | Page 10



In 2002, Portland’s Audit Services Division completed an audit called, “Managing for Results: A Proposal for the City of Portland.” The audit proposal to the city was to improve management by setting goals, allocating resources and managing to achieve those goals, and measuring and reporting results. Radical, I know, and the proposal has only intermittently caught on at the city. Yet, those performance management principles provide the high-level criteria for many of our performance audits. Our audits of the City of Portland’s housing programs illustrate this use of a performance management lens, and demonstrate the lack of connection between goals, funding and management, and measurement and reporting. The City of Portland encourages development of affordable housing with tax abatements and financial assistance. Since 2008, the City Auditor has reviewed these programs three times. Our 2008 audit focused on tax abatements, and our 2012 audit focused on risks of a newly reorganized Housing Bureau. Our 2014 audit focused on loans. The objectives in each audit reflected the performance management steps below:

LGAQ Fall 2015 | Page 11


Portland has

Some common themes related to these performance management steps were evident in all of our housing audits:



had challenges articulating the goals of its public housing programs.

Portland has consistently had challenges articulating the goals of its public housing programs. The need for affordable housing is well documented, and the city has many general policies and broad goals around ensuring a supply of housing for all residents. But setting goals for specific housing programs is more challenging. When it gets to the ”why” of providing new housing, such as to decrease homelessness, to provide workforce housing, to redevelop blighted neighborhoods, or to increase use of a transit hub, the discussion is even harder. Looking at the 2008, 2012, and 2014 audits, it is clear Portland has made progress clarifying goals and aligning broad goals with specific program objectives. In 2008, when we audited the city’s tax abatement programs, that was not the case. These programs were created to stimulate the construction of new housing development in certain areas in need of redevelopment. Criteria were added to the programs over time related to affordability and other public amenities. Layers of requirements, shifts in the housing market, and changing interests of policy makers had created a program that no longer aligned with the original redevelopment goals. We recommended and the city committed to clarifying the goals and objectives of the tax abatement programs. The housing risk assessment completed in 2012 identified a different situation. By that point, the Portland Housing Bureau had developed a strategic plan with specific objectives. The risk identified in this audit was that the Bureau was working towards numerous, diverse objectives with limited resources and staffing cuts. The audit recommended that the Bureau prioritize goals and objectives. The housing loan audit completed in 2014 determined that the Housing Bureau had established investment priorities for the loans and set specific goals related to affordability. We reviewed loans issued over the previous three years, and determined that the projects funded reflected the affordability goals. However, the types and costs of the housing funded varied widely. It was unclear whether investments were focused and cost effective. IT IS EASY TO LOSE SIGHT OF PROGRAM GOALS AND POLICIES WHEN MAKING FUNDING DECISIONS Providing funding to third parties to develop housing is inherently opportunistic. The city is limited to the proposals received, and there is always pressure to get money out the door and demonstrate units built.

LGAQ Fall 2015 | Page 12


New supportive housing for disabled adults.

Renovated senior housing.

We reviewed the project application and approval process as part of our audits for the 2008 tax abatement and the 2014 housing loans. In the tax abatement audit, in particular, projects were approved that had little relation to program goals. One project applied for and received the tax abatement after the project was built – in a program intended to stimulate development. Other projects provided only market-rate housing in areas that had some of the hottest housing markets in the city. In part, these diverse projects were due to shifting goals (i.e., decisions were made based on the dominant goals of that moment). In the housing loan audit, all approved projects met stated affordability criteria. The scoring and project evaluation process, which determined who was funded, was highly judgmental. Not all projects met program requirements in the loan guidelines. Over 80 percent of loans we reviewed were issued with either limited or no expectation of repayment. In some cases, staff noted that the projects could not support a loan, but approved them anyway because there was no grant program available. Some loan amounts exceeded the appraised value of the project. Furthermore, construction costs varied widely, with some units costing almost double the costs of similar projects. Approving loans that are not cost effective or will not be repaid decreases the money available for future projects and threatens the ability to meet long-term housing goals. FUNDING NEW PROJECTS OFTEN TAKES PRIORITY OVER MONITORING EXISTING HOUSING With limits on staff and resources, and pressure to keep developing new housing, ongoing monitoring may be short-changed. In the tax abatement and housing loan audits, we reviewed a sample of project monitoring files for compliance with program and contract requirements. In both audits, monitoring was extremely limited. We identified projects that did not make loan payments, projects that did not build the amenities defined in the contracts, and one project that did not provide the affordable units promised. Ongoing challenges include ensuring the sound physical structure of housing and providing units and services to eligible tenants. Another risk identified was proper management of the built-housing loan portfolio. Without adequate policies, tools, and resources, the loan portfolio may not be effectively managed, and the impact of funding decisions may not be maximized. In the most recent housing loan audit, we identified improvements in the monitoring program, and a clear commitment by the city’s Housing Bureau to address system limitations. However, most changes were too recent to be evident in the sample we reviewed. Sounds like a good follow-up audit topic!

LGAQ Fall 2015 | Page 13


If the jurisdiction has not been clear about what it is trying to accomplish with its housing programs, it will be very difficult to measure whether it is succeeding.

DETERMINING HOUSING OUTCOME MEASURES IS EVEN MORE CHALLENGING AND DESERVES DISCUSSION Housing production programs naturally focus on housing units produced as a performance output measure. Portland’s Housing Bureau routinely reports on the number of units produced in all programs. The next step -- developing outcome measures in a complicated social services environment -- is much more difficult. Should the city be measuring the length of time a family remains in city-funded housing, to demonstrate the ability to keep families housed and off the streets? Or is the goal to move families out of rent-restricted housing, in which case long stays in rentrestricted housing would be a negative? If the goal is to produce housing to increase transit use, how will this be measured? Again this ties back to setting clear, specific goals. If the jurisdiction has not been clear about what it is trying to accomplish with its housing programs, it will be very difficult to measure whether it is succeeding. CONCLUSION Performance management steps were not the only criteria used in our housing audits. We also reviewed housing program policies and rules, compared them to best practices, and evaluated internal controls. But evaluating the programs through a performance management lens helped frame the audits and keep the focus on program effectiveness and results – the bottom line for the public.

About the Author Kari E. Guy, Senior Management Auditor, has worked with the City of Portland Audit Services Division since 2007. Before joining the city, she spent ten years with Washington State Senate Committee Services, both as a fiscal analyst to the Ways and Means Committee, and as a policy analyst with various natural resources committees. Kari has a Master of Public Administration degree and is a Certified Government Auditing Professional.

LGAQ Fall 2015 | Page 14



HOMELESSNESS IS A GROWING CRISIS IN HONOLULU Hawai`i’s state capital, Honolulu, is located on the island of O`ahu and has a population of almost one million. Like many other cities, Honolulu has a serious problem with homelessness, which is worsened by a backlog in transitional shelter, public housing, and affordable housing units. Over the past five years, the Honolulu City Council has appropriated nearly $212 million for the homeless and affordable housing initiatives. Many residents report the homelessness crisis is worsening and affecting the quality of life and economy of Honolulu. In a recent survey, 90 percent of Honolulu residents rated homelessness as a major or moderate problem; 80 percent reported the problem of homelessness as essential or very important for the city to address. Recent statistics for 2015 indicate Honolulu’s homeless population has increased four percent to 4,093 individuals, veterans, families, and children. Over 40 percent are not in homeless shelters. Many are chronically homeless. Others suffer from mental health and/or substance abuse. One-fourth reported living in Hawai`i for less than five years. Honolulu’s homeless population comes from throughout the United States and the Pacific basin. Many arrive without any safety net to provide housing, employment, food, or shelter. The City and County of Honolulu initiated several programs and received federal grants to address homelessness problems. For example, U.S. Department of Housing and Urban Development grants provided funding for affordable housing and support services. Mayoral initiatives and strategies for easing the crisis include constructing temporary modular housing, implementing rapid re-housing LGAQ Fall 2015 | Page 15


More than 40 percent of Honolulu’s homeless are not in shelters and many are chronically homeless.

for the chronically homeless, building hygiene centers in city core areas such as Chinatown, and implementing coordinated programs for the homeless. These programs augment the services and housing provided by homeless shelters and other non-profit organizations throughout the city. The city council passed several resolutions and ordinances in response to citizen complaints, business concerns, and the impact on the Honolulu’s major economic sector of tourism. Laws now prohibit people from sitting, lying, or sleeping overnight on sidewalks, parks, beaches, and other areas. Other resolutions addressed transit-oriented development districts, strategies for developing affordable housing, and coordinating permitting efforts to expedite renovations of a backlog of 175 unusable public housing units that could house 600 people. Despite these efforts, the problem of homelessness continues, and some report the crisis is growing. HONOLULU’S AFFORDABLE HOUSING SHORTAGE IS A LONGSTANDING CONCERN High housing costs in the 1990s, prompted the city to adopt rules for inclusionary zoning, which required developers seeking residential zoning changes to adopt an affordable housing plan as a condition for approval. Conditional zoning, called “unilateral agreements,” set aside low- and moderate-income units in market-rate housing developments. Our report, Audit of the City’s Management of Unilateral Agreements in Affordable Housing, Report No. 07-05, examined issues that directly affected the number of affordable housing units actually built. This audit found a poorly administered program that did not proactively monitor developers’ compliance with their affordable housing commitments. Reliable counts of affordable housing credits by developer, and excess affordable housing “credits” banked and redeemed were unavailable, not tracked or verified by the department. In-lieu fees paid by developers (i.e., developers can also pay fees in-lieu of building affordable units) had not been expended for affordable housing since 1998. The department updated and formalized program rules in 2010. However, an audit recommendation to include verified affordable housing data in the department’s annual report continues not to be implemented due to insufficient staffing. THE MARKETPLACE IS NOT BUILDING ENOUGH AFFORDABLE HOUSING UNITS TO KEEP UP WITH DEMAND Honolulu needs to add an average of 5,000 housing units annually to catch up with demand and address the formation of new households. Under the affordable housing rules, the marketplace produced 68 units in FY 2014. To build more units, the mayor proposes extending new affordable housing requirements to all construction above a proposed threshold.

LGAQ Fall 2015 | Page 16


Honolulu has improved housing options for lowincome individuals through the development of micro units,

The housing program is important for several reasons. Many residents live in multi-family and/or multi-generational homes, and spend more than 45 percent of household income on housing and transportation. Honolulu’s cost of housing is high, with the median housing value currently at $700,000. Fair market rent for a one-bedroom apartment is over $1,500, and a two bedroom apartment is over $1,600. To afford the rent, individuals would have to earn about $31.61 per hour in a city where the minimum wage of $7.25 per hour is prevalent. The city developed other initiatives for increasing the workforce housing inventory and providing diverse housing options for low-income persons, the homeless, and the elderly. The initiatives included: •

Developing “micro units” such as those underway in California cities. “Micro units” are smaller housing units that are designed to provide the basics for a high-quality living environment within about 300 square feet. Designed to be occupied by no more than two persons, the units would provide housing for the elderly, homeless, and individuals newly entering the work force.

Introducing Transit-Oriented Development (TOD) that provides affordable housing in close proximity to transit systems. These developments would reduce transportation expenses and commute times, offer more housing choices close to employment centers, and provide mixed-use housing and commercial developments near rail stations.

Allowing Accessory Dwelling Units (ADU) in existing neighborhoods. This program would increase the supply of housing by updating zoning codes to allow additional dwelling units such as cottages, additions, or converted garages on existing single-family lots.

introducing transit-oriented development and allowing Accesory Dwelling Units in existing neighborhoods.

AUDIT OPPORTUNITIES ABOUND In Honolulu, homelessness affects the quality of life for everyone – residents and visitors. It is a complex and challenging problem that has grown over the decades and poses many problems for the city council and mayor despite many attempts to resolve the crisis. Honolulu is trying to provide relief for the homeless and those at risk of becoming homeless. To succeed, it must carefully coordinate the efforts of city, state, and federal agencies (e.g., the U.S. Department of Veterans Affairs, U.S. Department of Housing and Urban Development, the State of Hawai`i Public Housing Authority, the Community Development Authority, and the Office of Planning). It must also consider community groups, non-profits, faith-based organizations, businesses, concerned citizens, and many others. For cities confronted with similar problems related to homelessness and affordable housing, there are many opportunities for audits of the following:

LGAQ Fall 2015 | Page 17


Effectiveness and efficiency of homelessness initiatives and programs.

Assessment of grant recipients’ expenditure of funds for homelessness initiatives and services delivered.

Homeless shelters and services to the homeless to determine if they are well implemented and coordinated.

The viability of affordable housing programs and strategies.

Transit-oriented development efforts.

Efficacy of city department planning and permitting processes needed to support and implement city and mayoral plans and strategies for resolving homelessness and affordable housing issues.

City’s affordable housing assistance projects and programs that serve low- and moderate-income households.

TRANSPARENCY AND ACCOUNTABILITY ARE EFFECTIVE TOOLS FOR ADDRESSING HOMELESSNESS Successfully implementing homelessness and affordable housing initiatives requires effective programs, economical use of resources (such as city funds and federal grants), and performance measures for evaluating the efficacy of strategies and programs. Timely reporting of reliable and accurate data to policymakers and taxpayers is essential. The issues of homelessness and affordable housing pose many challenges to the City and County of Honolulu and many other cities throughout the nation. These issues also provide many opportunities for value-added audits that impact local, state, and federal entities and communities everywhere.

About the Author Susan Hall CFE CRMA, is a Deputy City Auditor with the City and County of Honolulu’s Office of the City Auditor (OCA). She is a member of ALGA’s Education Committee. She is also project manager for OCA’s annual Service Efforts and Accomplishments, National Citizen Survey, and Citizen Centric reports. Over the past four years, Honolulu’s OCA has documented the high priority that residents place on the need for affordable housing and assisting the homeless and homelessness through their annual National Citizen Survey of Honolulu residents.

LGAQ Fall 2015 | Page 18



MAINTENANCE EVER AFTER Local governments have the fundamental responsibility for delivering public services that citizens see and use every day. Some of these services are obvious, such as providing public safety services or managing parks. Some other services … not so much. For example, only through an audit I came to learn that my city, among its many responsibilities, is tasked with operating cemeteries. This is quite a complex business, which requires a wide variety of skills. For example, there are real estate-related activities such as selling burial lots and recording the transaction with the county; financial transactions for recording revenues from the sales; interment activities; and then there’s everything that goes into maintaining historic monuments. In the City of Austin, the Parks and Recreation Department has been responsible for managing/overseeing activities related to the operations of its five municipal cemeteries since 1987. From 1990 through 2012, cemetery maintenance and management was outsourced to a vendor who provided grounds maintenance, tree maintenance, and burial operations on behalf of the city, including gravesite opening and closing, funeral set-up, and the installation of grave liners. In 2013, the city began directly managing the sales, operations, and maintenance of cemeteries, while continuing to outsource interment services. WAIT, YOU ARE AUDITING WHAT? Needless to say, cemetery operations are not a popular audit topic; though it is surely a topic that raises a lots of sentiments. Indeed, it was initially brought to our attention by several citizens who voiced their concerns regarding poor

LGAQ Fall 2015 | Page 19


conditions of the city cemeteries to the City Council. This prompted our office to do a first audit in 2010 which found significant issues with maintenance. We then followed-up in 2012 and found that the overall conditions of the cemeteries had not significantly improved. By then, the parks department had initiated the development of a Historic Cemeteries Master Plan that, as stated in its introduction “will provide our city with an innovative approach to restoration and programming for these historic and sacred landscapes.” The pictures below, which were shown in our audit report, should give you a good idea of the reasons for such attention to this topic…and yet another cemetery audit. OPENING A CAN OF WORMS In 2014, while the city was developing the Master Plan, we performed yet another audit of cemetery operations in light of the risks observed in earlier audits and investigations, as well as concerns raised by citizens. Since the city had just brought the management of cemetery operations in-house, we began our audit by identifying all the critical relevant processes, which, as shown below, are more than we would have imagined.

LGAQ Fall 2015 | Page 20


Department management was very optimistic about us reassuring them and the public about the new oerations being much better, but we were not too optimistic based on what we had observed to date.

We conducted a high level review of the three processes outlined in the chart below. Based on the risks identified and some mitigating factors (such as the master planning process and another project self-initiated by the parks department related to capital projects) we decided to focus our audit on internal operations for sales, administration, and management. Department management was very optimistic about us reassuring them and the public about the new operations being much better, but we were not too optimistic, based on what we had observed to date. The Good •

Revenue collected were trending up ($2.3 Million).

Customer complaints (which used to take a significant amount of the department’s time) were going way down.

The Bad •

The recently created Cemetery Operations Group, organizationally placed in the Grounds Operations division, was operating in a separate location from all other depart-ment offices, with limited supervision.

LGAQ Fall 2015 | Page 21


The accountant was the only one with access to the financial system.

The Ugly •

Financial transactions were recorded in an off-the-shelf accounting software, which did not interface with the city financial systems; sales of burial plots were marked in a lot book, which had missing pages.

Any staff could conduct a sale transaction, including the accountant who could perform the sales transaction, receive the funds, and record it in the system. Oh, and the accountant was the only one with access to the financial system; his supervisor did not have the system loaded on his computer, nor did he know how to run reports. BY NOW YOU SHOULD BE DYING TO HEAR WHAT WE FOUND To get a sense of what was really going on, we designed fairly straightforward audit procedure, such as tracing sales order numbers through deposit and deed creation, tracing deeds back through deposit to original sales order, testing transactions to determine if a deed was been cre-ated and if there were duplicates, etc. This testing led us to identifying errors and inaccuracies in all steps of the process (see highlighted areas in the chart and examples below). In each instance, a lack of adequate supervision resulted in mistakes and irregularities going undetected or un-remediated for long periods.

LGAQ Fall 2015 | Page 22


We identified

We identified instances where the Cemetery Operations Group sold the same burial space to two different people. And in at least one instance, a duplicate sale resulted in the city having to disinter a body and relocate it to another burial space at a different cemetery.

We identified instances in which the Cemetery Operations Group had not recorded a deed. Without a deed, customers who purchased a burial space are not registered as the legal owners of that space.

We also could not account for all financial records. We identified a significant violation of cash handling policy and issues with the review and approval of vendor charges.

instances where the Cemetary Operations Group sold the same burial space to two different people.

So, here is my take-away: sometimes you don’t need to be fancy with your audit procedures. Just follow the audit mantra of “trust but verify” and, needless to say, do not leave any stone unturned. And a word of caution: audits like this one can distort your sense of humor.

About the Author Niki Raggi is an Assistant City Auditor with the City of Austin, responsible for managing all types of audit and special request projects. Niki has been with the city since 2004 and previously worked for both the Texas State Auditor’s Office and KPMG’s Corporate Governance group in Milan, Italy. Niki holds a degree in political science from the University of Genoa and a master’s degree in Public Affairs from the University of Texas. She is a certified government auditing professional (CGAP) and a certified internal control auditor (CICA) and holds a Certification in Risk Management Assurance (CRMA).

LGAQ Fall 2015 | Page 23



INTRODUCTION Despite the national trend of decreasing rates of homelessness in the years following the Great Recession, state and local government policymakers still prioritize addressing the human, societal, and financial costs of homelessness. Recent estimates have counted a total of 578,424 individuals in the United States experiencing homelessness,2 and in recognition of the magnitude of the challenge, the federal government provided $4.5 billion in fiscal year (FY) 2015 for various programs to address homelessness.3 In contrast to declining rates of overall homelessness in 34 states from 2013 to 2014, the District of Columbia (District)4 experienced a 12.9% increase in the homeless population during the same time period. This unfortunate trend, in spite of increasing District financial support provided to non-governmental organizations (NGOs) for homeless support services, led to a request from a member of the D.C. Council for an independent review of locally-funded programs operated by NGOs and intended to serve the homeless population. This article discusses the methodology developed by the Office of the District of Columbia Auditor (ODCA) to review homeless support programs provided by NGOs and funded by local District appropriations in FY 2014. The examination strategies discussed should help overcome challenges faced by most state and local governments in reviewing the propriety and efficacy of their respective homeless support services.

LGAQ Fall 2015 | Page 24


The audit methodology allowed us to identify and focus on the riskiest financial and contract management practices.

METHODOLOGY The inherent nature of the homeless population – geographically transient, difficult to quantify, and in need of a wide range of social services provided by diverse agencies and NGOs – presented our team with unique challenges. The following methodology allowed us to identify and focus on the riskiest financial and contract management practices. STEP 1: IDENTIFY THE LOCAL GOVERNMENT AGENCIES, NGOS, AND CONTRACTS UTILIZED TO PROVIDE HOMELESS SUPPORT SERVICES While seemingly simple, the task of identifying the locally-funded agencies, NGOs, and relevant contracts that provided homeless support services was a difficult one. The District had no comprehensive homeless services budget and no individually responsible department; many programs commingled local and federal funds; some programs fell outside the scope of our review because homelessness was incidental to the service provided,5 and some programs were omitted because of their relatively nominal cost.6 After applying these narrowing factors, our review focused on the District’s Department of Human Services (DHS) and Child and Family Services Agency (CFSA) as the only agencies that provided targeted homeless services solely through local funds provided to NGOs. By far, the single largest area of expenditure was through a contract between DHS and a local 501(c)(3) homeless services organization (Prime NGO) that had been designated as manager of the District’s Continuum of Care (CoC).7 The Prime NGO held two separate contracts related to the CoC – the Management Contract and the Sole Source Contract. In total, the Prime NGO was responsible for 76% percent of the local dollars spent on homeless services during FY 2014. Management Contract (MC) The MC established a competitive system for the Prime NGO to award subcontracts to other NGOs (Sub NGOs) for carrying out CoC responsibilities, such as street outreach, shelter security, and low-barrier emergency shelters.8 Among other duties, the Prime NGO was responsible for providing oversight of Sub NGOs’ scope of services. The Prime NGO received a fee for administrative oversight not to exceed 8% of the total cost of the contract. Rather than subcontracting out all of its responsibilities, the Prime NGO chose to provide some contract components itself, thus serving as a subcontractor under the MC. Of the 68 subcontracts to the MC, 58 were subcontracted to Sub NGOs and 10 were provided by the Prime NGO itself – which then supervised its own performance. LGAQ Fall 2015 | Page 25


Sole Source Contract (SSC) The SSC was distinct because services were required to be provided at specific sites, such as church-based meal programs or private shelter locations, and therefore subcontracts could not be competitively bid. Unlike the MC, the SSC provided no administrative oversight fee to the Prime NGO. Yet similarly to the MC, the Prime NGO sometimes managed its own performance under the SSC. Of the 63 subcontracts, 53 were subcontracted to Sub NGOs and 10 were self-provided by the Prime NGO, which then supervised its own performance. Other Programs and NGOs DHS and CFSA utilized a small number of vendors outside of its relationship with the Prime NGO. Motels were sometimes compensated directly for providing emergency housing when shelters were at capacity. While the Prime NGO managed a permanent housing program, DHS self-managed and directly contracted with nine NGOs to provide case management services for a separate Permanent Supportive Housing Program. Similarly, while the Prime NGO managed a rapid rehousing program, CFSA directly contracted with four NGOs to provide a separate rapid rehousing program for CFSA clients. The table below provides an overview of the agencies, programs and the local dollars expended to NGOs in FY 2014

LGAQ Fall 2015 | Page 26


A single staff person was responsible for the District’s management of 151 contracts and subcontracts.

STEP 2: CONDUCT A RISK ASSESSMENT Having identified the complex universe of agencies, NGOs, and contracts utilized to provide locally-funded homeless services, we requested FY 2014 spending reports for the Prime and Sub NGOs under the MC, SSC, and all other aforementioned contractual relationships. Given the vast amount of data and limited resources available for review, we honed in on the riskiest areas of operations. It is inherently risky for governments to contract out complex program management functions to private vendors. We were particularly concerned with the convoluted contract structure in place with the Prime NGO, which was not only responsible for managing the performance of 111 NGOs, but also 20 performance areas that the Prime NGO itself provided. In the former category, the District had no contracts in place with the Sub NGOs, relying entirely on the diligence of the Prime NGO to ensure program compliance. In the latter category – where the Prime NGO acted as a subcontractor to itself – no contract whatsoever was in place to monitor the Prime NGO’s performance. Regardless of the structure of contractual relationships to utilize private vendors for the provision of public services, governments are always responsible for actively managing vendors to ensure taxpayer funds are spent wisely. At this point in our review we had significant concerns regarding the ability of District agencies to monitor program performance and fiscal management, particularly on the part of Sub NGOs and the Prime NGO when it provided services itself without an independent contract manager in place. STEP 3: REVIEW LOCAL GOVERNMENT AGENCY MANAGEMENT OF VENDOR CONTRACTS We next reviewed all locally funded contracts with NGOs to provide homeless services, with a particular focus on performance and financial data required to be submitted to the contract administrators and the deadlines for such submissions. In scrutinizing the process for submission and approval of invoices through interviews with key DHS personnel, a primary weakness in oversight became evident: a single staff person was responsible for the District’s management of 151 contracts (including subcontracts) comprising of approximately $107 million in expenditures and 98% of all District funds spent to provide homeless services. Our initial impression was that this task could not be responsibly carried out by a single staff person.

LGAQ Fall 2015 | Page 27


The monthly invoices from the Prime NGO for both the Management Contract and Sole Source Contract were automatically approved by the District.

We then learned that the monthly invoices from the Prime NGO for both the MC and SSC were automatically approved by the District, meaning no supporting documentation was submitted along with the invoices to justify expenditures. Both the single contract administrator and the agency’s chief financial officer signed off on these invoices based on only summary information; no verification of subcontractor expenditures or invoice review took place. Based on this red flag, we reviewed the management contracts, subcontracts, budgets, actual spending reports and performance data for all agencies and NGOs. To ensure the reliability of the spending reports we requested general ledger detail from the Prime NGO’s accounting systems. We compared the amounts the Prime NGO paid to Sub NGOs to the billing and payment amounts provided by District, keeping a close eye on the administrative fees charged by the Prime NGO. In interviews with the Prime NGO’s staff, we learned of its practice to bill the District monthly with the first and last month’s invoices of the FY submitted as estimates – ostensibly for cash flow purposes – with no yearend reconciliation to ensure the return of excess payments to the District’s General Fund. INITIAL FINDINGS We identified three major areas of concern during our examination. 1. The Prime NGO improperly billed the District for estimated rather than actual costs. The Prime NGO’s billing method based on estimates without year-end reconciliation led to District overpayments of more than $5.3 million. This amount represented approximately $4.2 million or 6% of the total for the MC and approximately $1.2 million or 7% percent of the total for the SSC in FY 2014. In addition, the method of estimate-based billing resulted in the Prime NGO over-charging for its administrative fee. This finding was confirmed after comparing actual annual spending reports for all NGOs to the total the Prime NGO had billed (and received payment from) the District. The Prime NGO claimed over-payments were to be used in subsequent FYs, but neither it nor DHS could provide evidence of reconciliation. This finding creates serious legal concerns because District and federal law prohibit the obligation or expenditure of funds in advance or in excess of an appropriation. It is well established that when resources remain unspent at the end of a fiscal year, they may not be carried over without an appropriation for the following fiscal year.

LGAQ Fall 2015 | Page 28


By allowing the Prime NGO to provide the homeless service under the SSC, the District paid the vendor an additional 18% above direct costs.

Finally, District policy prohibits the payment for good or services in advance of performance, subject to very limited exceptions. By leaving the funds with a vendor, the District is paying for goods and/or services for the next fiscal year, in advance of receiving them. 2. The Prime NGO may have improperly billed the District for thirdparty contract management when it provided services itself, independent of Sub NGOs. The Prime NGO billed the District for overhead, in addition to a contract management fee for managing a homeless shelter in the District. In addition to costs, it charged a 10% percent indirect rate of their direct costs – the same cost to the District as if the Prime NGO had engaged a Sub NGO and managed that contract. In addition, the Prime NGO also charged the District an 8% percent management fee for essentially managing itself. By allowing the Prime NGO to provide the homeless service under the SSC, the District paid the vendor an additional 18% above direct costs. If a management vendor wants to provide services rather than subcontract the services, the contract must be clear on whether that is acceptable and, if it is, the vendor should be treated like all other subcontractors under the contract. The vendor should be able to charge a 10% indirect rate but they should not be able to charge an additional 8% management fee. 3. Program efficacy concerns were raised by significant variances across NGOs in the ratio of total budget to direct benefit to homeless individuals. We analyzed all the spending reports and prepared spreadsheets that documented projected and actual outcomes, how long the subcontractors provided the services, the difference between the budget and actual costs, and the total costs and percentage of program costs versus non-program costs for each subcontractor. Our analysis of the Sub NGOs’ spending reports identified a wide variance of ratios for non-program costs to total actual costs within the same category of service. This variance raises questions regarding whether the maximum possible amount of taxpayer funds are expended to directly benefit homeless individuals. This concern indicates a need for greater monitoring of what is an allowable ratio of direct to indirect costs or a maximum percentage that is allowable to be spent on non-program related expenses. GAGAS AUDIT TO FOLLOW The initial findings of this examination compelled us to engage in a Generally Accepted Government Auditing Standards (GAGAS) audit of the

LGAQ Fall 2015 | Page 29


The initial findings compelled auditors to engage in a

management contract between the District and the Prime NGO. We are currently reviewing all relevant contract terms, the specific appropriations authorized for FY 2014, the services actually rendered as compared to estimate-based invoices, and the appropriateness of fiscal practices in light of relevant District and federal laws. This comprehensive audit is currently underway and will likely be released in the fall of 2015.

comprehensive audit, which will

About the Author

likely be released in the fall of 2015.

Julie Lebowitz is a Supervisory Auditor with the Office of the District of Columbia Auditor (ODCA). She has been with ODCA since 2013. She has a Bachelor of Science in Accounting from St. John Fisher College. Her previous work history includes six years of government auditing, three years as the Internal Audit and Control Manager for Monroe County in Rochester, New York, and three years as a Senior Examiner for the New York State Comptroller’s Office. Prior to her experience in government, she worked for two years as a Forensic Auditor at a regional CPA firm in Rochester, New York. She is a Certified Internal Auditor and a Certified Fraud Examiner. Notes Supervisory Auditor, Office of the District of Columbia Auditor (ODCA). The author is grateful for the support of Kathleen Patterson, District of Columbia Auditor; Ingrid Drake and Tia Clark, members of the team that carried out this examination; and Anovia Daniels for her assistance in developing this article. The full report is available at: http://www.dcauditor. org/reports/examination-non-governmentalorganizations-ngos-receiving-local-district-fundsprovide. Questions may be directed to Julie. [email protected] 1

The U.S. Department of Housing and Urban Development (HUD) defines “experiencing homelessness” as when a person sleeps outside or in an emergency shelter or transitional housing program. HUD employs a methodology known as a point-in-time count (PIT) to identify the size of homeless populations, which takes a snapshot of the population for a ten day period. In January 2014, a PIT identified 578,424 individuals experiencing homelessness throughout the United States.


National Alliance to End Homelessness, The State of Homelessness in America 2015, accessed at: http://www.endhomelessness. org/library/entry/the-state-of-homelessness-inamerica-2015.


The District is unique in that it has organizational and operational characteristics in common with city, county and state governments. The current structure of the District’s municipal government – with an independently elected Mayor and District Council


LGAQ Fall 2015 | Page 30

– was established by the U.S. Congress with the District of Columbia Home Rule Act of 1973. For the purposes of this article, the reader may consider the District government as functionally similar to other major U.S. cities, except that District appropriations and local laws are subject to review by the U.S. Congress. Programs for veterans and victims of domestic violence oftentimes address issues of homelessness, but the connection was considered too attenuated for the purposes of our review.


In FY 2014, the District’s Department of Behavioral Health spent approximately $2,000 with two mental health providers to support outreach during the declared cold emergencies.


HUD defines a CoC as a program that “is designed to promote communitywide commitment to the goal of ending homelessness; provide funding for efforts by nonprofit providers, and state and local governments to quickly rehouse homeless individuals and families while minimizing the trauma and dislocation caused to homeless individuals, families, and communities by homelessness; promote access to and effect utilization of mainstream programs by homeless individuals and families; and optimize self-sufficiency among individuals and families experiencing homelessness.”


Low-barrier shelters do not impose identification, time limit or other program requirements.




Like many other cities across the United States, homelessness remains one of the country’s most complex and significant social issues. In April 2015, the Audit Services Division of the City and County of Denver released an audit of the city’s homeless program, Denver’s Road Home (DRH). Our objective was to evaluate DRH’s effectiveness toward meeting goals and outcomes in its effort to reduce homelessness in Denver. DRH was born out of Denver’s 10 Year Plan to End Homelessness (Plan), which outlined strategies not only to end homelessness by 2015 but also to prevent homelessness by improving access to housing and enhancing services for special-needs populations. The Plan was developed in 2005 as part of the U.S. Interagency Council on Homelessness’ challenge to mayors throughout the nation to develop a plan to end homelessness in their cities within ten years; it became the guiding document for DRH. METHODOLOGIES WITH A HUMAN TWIST We released our audit during the final year of the Plan, and we incorporated various methodologies to assess DRH’s progress towards meeting its goal of ending homelessness. These methodologies, however, did more than help us meet our audit objectives. They helped us illustrate the human element of homelessness that might not have come through in a typical performance audit. These methodologies also allowed us to observe and interact with the people providing services on the ground to understand how challenging it is to serve the homeless population and how difficult it is for their clients to get back on their feet. Every facet of the issue of homelessness comes down to people. By using specific methodologies to gain an understanding of the human element, we feel we were able to produce more than just another audit of a city program.

LGAQ Fall 2015 | Page 31


Every facet of the issue of homelessness comes down to people. By using specific methodologies to understand the human element, we were able to produce more than just another audit of a city program.

Rather, we were able to spotlight the complexities and challenges of such a sensitive social issue, which many people feel strongly about on both ends of the spectrum, by complementing our audit with stories about the day-to-day struggles of being homeless in Denver. The methodologies we used to gain a unique perspective included interviewing homeless individuals, conducting a focus group with area homeless service providers, and observing a citywide homeless outreach event. Illustrating the Human Element of Homelessness—We conducted several interviews with homeless individuals to gain an understanding of how homelessness has affected their lives and to dispel some of the negative stereotypes associated with being homeless. We were surprised to learn how easily a seemingly everyday problem could leave a person homeless and wanted to inform our readers of this humbling reality. The team decided that speaking directly with individuals experiencing homelessness would provide additional context regarding DRH and the population it serves. Recognizing the sensitivities and possible risks associated with identifying and approaching someone on the street who appears to be homeless, we instead worked with local service providers who helped connect us with people who they believed would be willing to share their stories. The service providers not only identified individuals who would be willing to speak to us, but they also coordinated logistics and meeting space to help alleviate hesitation individuals may have had. Without this facilitation, the interviews may not have been possible and we would not have gained this valuable perspective. Auditors sitting down with people who are experiencing homelessness in Denver was unique and emotional for both parties. Interviewees were able to share their stories about becoming homeless and the impact that service providers have had on their wellbeing. Although their stories varied, the individuals we talked to were all grateful to have the opportunity to share their stories. Even without access to everyday conveniences, such as cell phones, computers, and transportation—even without personal identification—they were all optimistic about the future. Each conversation left us more humbled and grateful for the little things we have taken for granted. DRH is one of the many avenues from which service providers receive funding; DRH does not directly provide any services to Denver’s homeless population. Therefore, we did not consider our interviews with homeless individuals a direct reflection of DRH’s performance. However, we decided to include their stories as a way of connecting readers to the City’s ultimate goal of ending homelessness. We presented these stories, or vignettes, in side bars throughout our audit report. The identities of the individuals were protected by changing or omitting any identifying information, and included

LGAQ Fall 2015 | Page 32


pictures of homelessness in Denver. The photographs were taken by audit staff and did not depict the people who were interviewed. Figure 1 is an example of one of the vignettes we included in the audit report. Listening to Local Service Providers—In addition to speaking with homeless individuals, we also facilitated a focus group with local service providers to better understand their relationships with the City’s homeless program and homeless individuals. These organizations offer a wide variety of services, many of which are targeted at specific groups of people, ranging from emergency shelter space for single men to temporary housing for families or unaccompanied women. Initially, we conducted interviews with three service providers who received funding from DRH. After gaining more insight into their challenges than we expected, we wanted to learn more. Facilitating a focus group allowed us to have personal contact with many providers at once in a group setting. Conversations focused on perceived strengths and weaknesses of DRH, the role of DRH, and barriers to measuring performance. We then used results from the focus group to corroborate the information we learned from previous audit steps. Hosting a focus group gave us more insight into the service provider’s perspective, but the method had its limitations. We decided to host the focus group participants at one of their offices rather than at DHS or another city office in order to foster a safe environment. Despite this effort, some participants still did not feel comfortable being completely candid among their peers. This could have been compounded by an unforeseen issue with the timing of the focus group, as many of the service providers were in the process of submitting funding applications to DRH. In fact, one provider did not feel comfortable sharing her opinion and instead opted to email our team at a later date. Observing Interactions—We also attended a homeless services event co-hosted by DRH and Mile High United Way called Project Homeless Connect. More than 120 vendors and 1,000 volunteers supported this annual event where homeless individuals and families can obtain legal and medical services, employment assistance, food, clothing, haircuts, and more, free of charge in a single location. Project Homeless Connect furthered our understanding of how service providers, businesses, and other groups work together

LGAQ Fall 2015 | Page 33


Auditors sitting down with people who are experiencing homelessness was unique and emotional for both parties. Each conversation left us more humbled and grateful for the little things we have taken for granted.

to serve Denver’s homeless population. By walking through the event and seeing attendees visit different vendors, we were able to observe firsthand how the City financially supports a broad array of services. We were also able to speak with homeless individuals about the event. Although many individuals were grateful for the event and the many service providers, some expressed frustration. For example, one woman said that she was tired of attending these types of events year after year, only to be put on another waiting list. She stated that she just needed a place to live. This helped illuminate the finite nature of services available to assist Denver’s homeless population, which further complicates such an intractable issue. Informing the Citizen—At the end of our audit report, we highlighted a tangential issue that was outside the scope of our audit but central to the broader issue of homelessness in Denver. The city has an ordinance prohibiting unauthorized camping, which makes it illegal for individuals to camp on any private property without consent or on any public property except in locations where camping is allowed. Similar to our approach of highlighting stories of homeless individuals throughout the report, explaining the camping ban was yet another way to inform the reader and add context to the broader policy landscape. While this controversial issue was not part of our audit objectives, we felt it was important to cover because of its daily impact on homeless individuals’ lives. We included this section to give the reader a general understanding of the unauthorized camping ordinance and how its existence could place an additional burden on homeless individuals, law enforcement, and taxpayers alike. Although incorporating these methodologies into our audit work required additional time and effort, we believe they helped us produce a report that not only added value to DRH but also helped educate citizens and raise public awareness on the issue. The additional context we obtained from interviews, the focus group, and attending Project Homeless Connect helped us understand the complexities of what it means to be homeless in Denver and what the City can do to further its ultimate goal of ending homelessness.

About the Authors Carl Halvorson is a Senior Auditor at the City and County of Denver Auditor’s Office. He has auditing experience in the public and private sector and holds a B.A.s in Economics and International Studies from Miami University (OH). Abby Musfeldt is a senior auditor at the City and County of Denver Auditor’s Office. Her background is in program evaluation and data analysis at the state and local levels, and she holds master’s degree in public administration from the University of Colorado Denver. LGAQ Fall 2015 | Page 34



Emergency service workers know that there is usually a small number of people in the community who use a disproportionate amount of services. However, until recently, very little data had been developed to better understand these usage patterns, as well as their proximate causes. As it turned out, the anecdotal evidence was supported, and further, the data suggested that lack of housing was a significant driving factor in the findings. In 2013, I was involved in a research project seeking to address some basic questions about users of community services, including health, emergency, and correctional services. Specifically, we wanted to learn how often frequent users of one type of service were also using other services, and what this told us about the costs of these treatments. The overarching goal was to begin examining the resources being consumed by frequent users, and what could be inferred about their needs. To do this work, my team looked at three groups of people and their use of services over a 10-month study period in western Washington. 1. A jail group of 23 people who had 9 or more bookings in county jail during the 10-month study period; 2. A crisis center group of 30 people that had 4 or more visits to a mental health emergency center during the study period; and 3. An emergency medical group of 29 people who received 11 or more mobile emergency visits from firefighters or EMTs during the study period. Using queries and matching functions, we were able to piece together serviceusage histories of the people in the 3 groups, and helped to show that there was indeed truth behind what service providers were seeing. We also found that LGAQ Fall 2015 | Page 35


A rough estimate

people frequently using one type of service (jail, crisis center, or emergency medical) were also frequently using other services in the county.

for the 23-person

The scale was often surprising. For example:

jail group would

The jail group had 200 bookings in the county jail and also registered 399 healthcare or emergency service encounters.

The crisis group had 158 visits to the emergency room, 120 other emergency medical service encounters, and multiple bookings in the county jail.

The emergency medical group had 420 emergency room transports, 155 other emergency room visits, and more than a dozen jail bookings.

approach $2,000 per person per month in services.

Another metric we looked at was time—days in jail or service days in a health care setting. Again, the numbers were pretty staggering. •

The 23 people in the jail group had a total of 2,723 days in jail.

The crisis group had nearly 1,200 service days in the crisis center, 343 days in jail, and 70 nights in homeless shelters.

The emergency medical group had 1,049 days of health treatment and 444 days in jail.

Although our research not assign costs to these services, it is clear that the amount of money being paid to help these people can add up quite quickly. For illustrative purposes, a rough cost estimate for the 23-person jail group ($100 for a night in jail, $500 for other visits) would approach $500,000, or about $2,000 per person per month in services. Perhaps these dollars would be worth it if the treatments were effective. However, our research and others

LGAQ Fall 2015 | Page 36


have found that: 1) frequent utilizers typically have complex, co-occurring disorders; and 2) jail clinics and emergency settings generally do not meet those needs. What we can reasonably expect from the status quo is ongoing, high-level use of costly-yet-ineffective interventions. WHAT ARE PEOPLE DOING ABOUT IT? Some jurisdictions are being proactive in addressing the needs of people using these services, and using research like this to bring together various community actors to address individual and systemic issues. For example, in the county I worked for, the Human Services Department now has staff working directly with inmates, helping them sign up for benefits and putting them in touch with community services before they leave the jail. Another effort is to use data from multiple systems to identify frequent utilizers and design interventions to address their needs. Addressing frequent utilization makes sense—in a nutshell, replace the expensive, ineffective, and inappropriate care with something that works. Unfortunately, addressing frequent utilization is not simple or cheap. These programs require significant financial investment, local resource capacity, partnerships, and long-term commitment. WHAT CAN INTERVENTIONS DO? Perhaps one of the best documented efforts to address frequent utilization was the Frequent Users of Health Services Initiative (FUHSI). This five-year, $10 million project was piloted in six California counties and aimed to direct people with frequent emergency room visits to lower-cost community care. An evaluation found that these types of interventions were successful; there were significant reductions in emergency departments and hospital visits, and some reductions in cost.

LGAQ Fall 2015 | Page 37


Coordinating across systems takes time and effort.

WHAT DOES THIS HAVE TO DO WITH HOMELESSNESS? The results of the FUHSI project indicate that homelessness is key to understanding frequent utilization and designing interventions that work. About half of the participants in the FUHSI project were homeless, and the evaluation found a strong correlation between housing and the use of services. Homeless clients that found housing had fewer days in the hospital, while those that remained homeless showed increased use of services. This adds weight to the growing body of evidence supporting the housing first movement. The basic philosophy of housing first is that a homeless person’s primary need is to obtain stable housing, and that other issues are addressed more effectively once housing is obtained. WHAT CAN AUDITORS DO? One topic that auditors might explore is the cost and benefits of frequent utilizer projects. Some might be tempted to oversell the financial savings of care coordination interventions. However, targeted, coordinated care is still costly (albeit cheaper than frequent emergency care).The costs form an investment that the research cited above indicates will lead to a better result for the client. Coordinating across systems takes time and effort, with years of ongoing, coordinated care for the client and between organizations. Another area to examine is how costs are spread out under the status quo, and how costs could be shared under a more enlightened (but still expensive) approach. Local governments in our state typically do not have deep reserves to push into new or expanded human services, but perhaps others have been successful in getting stakeholders such as hospitals, health plans, and fire districts to help.

LGAQ Fall 2015 | Page 38


Finally, it would be interesting to see the types of projects that communities are using to address frequent users, and what types of results are coming out of this work. WHERE CAN I LEARN MORE? The best place to start is the US Interagency Council on Homelessness, which has a database of effective and promising interventions. The section on frequent utilizers is here: http://usich.gov/usich_resources/solutions/ explore/frequent_users_of_health_services The results of the project I worked on in western Washington: http:// snohomishcountywa.gov/1053/HSD-Reports-Publications

About the Author Sean DeBlieck is a member the King County Auditor’s Office (KCAO). His path to performance auditing began in Albania, where he evaluated international police reform efforts, and led him to positions with the Government Accountability Office and the Minnesota legislative auditor. Prior to joining KCAO, Sean provided management and policy analysis for several departments in Snohomish County, including its human services division.

LGAQ Fall 2015 | Page 39

SUBMITTING ABSTRACTS, ARTICLES & MEMBER NEWS ABSTRACTS 1. Log on to algaonline.org. You must log on to view and access the form. 2. Navigate to the online submission form: Resources > Abstracts > Submit an Abstract. 3. Enter (type or copy/paste) and submit your abstract details as prompted in the form. Abstract Summaries and Descriptions will not be published in the Quarterly but will be displayed and searchable online. For the Summary, please provide a single paragraph describing what you did and what you found. The description may include scope, background, objectives, significant findings, unique methodologies, recommendations, or other information you wish to share with ALGA members.

ARTICLES Articles may be submitted for any topic, but those relating to advertised themes will receive first priority in any given Quarterly. Upcoming themes and submission titles are as follows: • • • •

Winter 2015: Training and Travel (due November 15) Spring 2016: Utilities (due February 15) Summer 2016: Education (due May 15) Fall 2016: Fleets (due August 15)

How to Submit: 1. Log on to algaonline.org. 2. Navigate to the online submission form: Publications/Media > Local Government Auditing Quarterly > About the Quarterly. 3. Click on the “Submit Your Article” llink. Fill out the “Submit Your Quarterly Article” form. Attach your article and provide details as prompted in the form. Format Guidelines: • • • • • • • •

One page is roughly equivalent to 450-500 words. Articles must be at least 1,000 words and no more than 2,000 words. For each article, please include a suggested headline and a short (3-4 sentences) author biography. Article text should be submitted in an unformatted Word file. To maximize the quality of graphics for both print and web presentations, graphics must be submitted separately from the article text in .JPG form. Indicate the desired location of any graphics within the article by enclosing the instruction in brackets. For example, [INSERT GRAPH 1 here]. Please include the proper attribution of any copied elements within your article to their respective sources. Remember to thoroughly proofread your article.

MEMBER NEWS ITEMS News from ALGA member organizations may be submitted for promotions, certifications, awards, retirements, or other relevant ALGA member news. Email your updates to [email protected]

LGAQ Fall 2015 | Page 40




Past issues of the Local Government Auditing Quarterly are available to members on the ALGA website at www. algaonline.com

Justin Anderson Chair King County, WA

© 2015 ALGA

Kristine Adams-Wannberg Portland, OR

A.D. Lewis City of Chicago, Office of the Inspector General, IL

Lisa Callas Edmonton, AB

Ruth Riddle Port of Seattle, WA

Angela Darragh Clark County, NV

Paula Ward Washoe County School District, NV

All works published in this journal are originals of the individual authors and have been provided for the express purpose of publishing the work in ALGA media. Individual authors hold clear title to their work, and the agreement to publish does not grant ALGA ownership of the work. The individual authors have granted permission to publish their work, and this permission is not limited and shall not expire. The individual authors have agreed that ALGA may assert, without notice to the individual author or any another party, ALGA’s title and copyright as to any use, reprinting, or reference to the work on any ALGA media.

Emily Jacobson LGAQ Editor Denver, CO

Daniel Genz Dallas, TX Chris Horton Fairfax County Public Schools, VA


Kymber Waltmunson, President King County, WA

Kip Memmott, At-Large Member Denver, CO

David Givans, President-Elect Deschutes County, OR

Alexandra Fercak, At-Large Member Portland, OR

Tina Adams, Secretary Charlotte, NC Kristine Adams-Wannberg, Treasurer Portland, OR

Pamela Weipert, At-Large Member Oakland County, MI Van Lee, At-Large Member Honolulu, HI

Corrie Stokes, Past President Austin, TX

LGAQ Fall 2015 | Page 41

ALGA Member Services 449 Lewis Hargett Circle, Suite 290 Lexington, KY 40503-3590 www.algaonline.org