Loading...
The URL can be used to link to this page
Your browser does not support the video tag.
Home
My WebLink
About
10/18/2017 - STAFF REPORTS - 00 (2)
• • �u�p • \'� AWAY FORWARD • FISCAL REFORMS • • • • IN • • • • MEETING PALM SPRINGS' • • • FISCAL CHALLENGES • • IN FY 17/18 • • • • AND • • • • • BEYOND • • • • Submitted by Daryl Terrell on October 17, 2017 at the City of Palm Springs City Council • meeting • • 1 • BACKGROUND • Despite an improving economy, despite General Fund discretionary revenues increases, the city continues to face a fiscal challenge. The city's fiscal challenges are brought on by our dysfunctional state in Sacramento and federal government in Washington, DC through unfunded mandates, using the city, the residents, the small businesses, and the college students as their personal ATM, by higher costs in pensions and by the city operating costs outpacing discretionary • revenue. The residents can no longer afford to pay more- and get less from- the city. As the • permanent fiscal storm batters us, it is time for the city to saddle up and ride. It is time to radically change the way the city operates- to shift from top- down bureaucracy to a city that empowers the residents and communities to change the city from the bottom up. • Governor Franklin D. Roosevelt said it best on May 22, 1932 in a commencement address at Oglethorpe University, "Our country needs and demands bold persistent experimentation. It is common sense to take a method and try it. If it fails, admit it • frankly and try another. But above all try something." i Crisis brings opportunities to new ideas and new approaches. We must try something • new other than the same old solutions. • Let's use this crisis to break the mold. Let's start rethinking how we operate the city, so we can deliver the results the residents want at a price they're willing to pay. • How the city weather this storm may define the city into the future. But success will require the mayor and city council to think outside of the box. It will require city • employees, who are encouraged and supported to use their valuable knowledge to • continuously improve their operations. It will require city manager's executive management and the city council and the mayor • willing to invest our first dollars in change, even when they are cutting General Fund Budgets. And it will require city employee to continue to partnering with city manager to shape • innovative changes that improve services, save money, and improve working conditions for employees. EXCUTIVE SUMMARY The city faces increasing mounting unfunded pension liability threaten the city's ability to provide basic city services and the long-term sustainability of the retirement funds. Also, the city faces increases in demand and cost in public safety and unfunded • mandates and take away in revenues from the state. 2 • • Without fundamental General Fund Budget reform, the continuance growth of city expenditures outpacing city revenue and increasing mounting unfunded pension liability will destroy the city's ability to deliver basic services to the community. • Without fundamental changes, the residents are facing the prospect of even more devastating cuts in the years ahead. • The city will have to end the broken budget cycle by instituting common sense reforms that would reinvent the city government, provide in the short-term, a balance General Fund Budget, replenish the General Fund reserves, maintain the city's bond rating, fund ther residents' priorities and maintaining city services. • PROBLEM Over the next several decades, the city owes $110 million to pensioners and $107 • million in retiree healthcare costs. To get ahead of those costs, officials agreed in June • to find an extra $6.5 million every fiscal year over a six-year period through new revenue or cuts. • The city's pension and retiree healthcare costs have grown substantially in recent years — the result of having more employees than similarly sized cities and the loss of significant California Public Employees' Retirement System assets during the last recession. The city's annual pension and retiree healthcare obligations were expected to rise to $25.6 million by fiscal year 2022-23. In addition to the extra $6.5 million, the city is • planning to spend $17.2 million in the next fiscal year. In addition, according to city documents, demands for public safety as increased, which has lead the General Fund expenditure to outgrow General Fund revenue. For example, • there was 46% increase in 911 calls from 2000-2016. Also, 73% of all 911 calls to the fire department require emergency medical attention. In addition, In 2016, the police department dispatch received 81,624 calls. Also, the state has continued its practice of balancing their budget on the backs of cities in the form of unfunded mandates or revenue take-aways. • For example, the state is sending programs and services to local governments for them to provide those services or program to the residents, but the catch is that the state doesn't provide funding for them. So, the city, like Palm Springs, are left to fund those services or programs with their own funding source, which in some cases are scarce. To illustrate my point, according to city documents, the state has already taken $30 million dollars from the city over the last five years. Furthermore, city will be required to . fund another$30 to $40 million dollars in unfunded mandates by the state. • 3 As the city council and the mayor understands and community understand, the city • faces continuing fiscal challenges in FY 17/18 and beyond that must be corrected. . Success in address these challenges will take several years, however, and will require purposeful actions on the part of the mayor and city council and the city manager's • executive team, with the support of the community. When the city recognized that General Fund revenues were not keeping pace with • General Fund expenditures, the mayor and city council began taking steps to cut costs. Yet, the problem and challenges continues to haunt us fiscal year after fiscal year. And it is quite possible that the city's fiscal challenges will become even larger. Like as many other cities throughout Riverside County, the city has entered an era of seriously constrained resources. This means that the city have a real and urgent need to make difficult choices. These choices will require re-evaluating the role of city hall as . a provider of services and identifying the specific services that are the city's core, or • basic, services. The city must determine the service levels the residents are willing to fund and the price • they are willing to pay through taxes and fees. The city must keep the cost of doing city's business down through cost, controls, improved service delivery methods, and strategic investments in technology and employee training and development. Tonight, I will introduce a comprehensive plan to build the city's road long-term fiscal solvency. The road to a balance budget and long-term fiscal solvency is through new ideas, empowering city departments and holding department heads for results through • performance measurements in every city department. • This plan lays out a roadmap for the city's future, but the road will not be easy, and reaching your city's destination will take discipline and perseverance. It describes the • fundamental problems the city needs to solve and suggests the steps the mayor and city council and the city manager executive team should take to ensure fiscal and organization stability for the future. The plan is a policy guide and an operational blueprint for actions that, if taken, can achieve fiscal stability and organizational sustainability. There are no simple solutions. Rather, a combination of actions will enable the city to meet its fiscal challenges now , and beyond. Since there are no other parties that can or will solve this problem, it is up to us collectively to understand and think outside box using best practices. • CHALLENGES TO MEETING THE CITY'S FISCAL CHALLENGES The process which has led to the development of this road map to the city's long-term • fiscal solvency. The root cause of the city's fiscal challenges is the fact that operating • 4 • • General Fund expenditures is growing faster than the revenue base, which supports them. All city services provide an important service to someone. Cutting existing programs and services are always unpopular in a community, especially' when there are so many needs to be met and when other units of government such as the State and • Federal are cutting funding for services or passing down new responsibilities without funding to operate them. • Ironically, individual and community needs are most acute when the overall economy is • weak and city revenues are also down. Dealing with these facts is complicated by various environmental and budget process factors. In order to address these challenges, this Plan embodies new approaches aimed at constructively dealing with the challenges, which exist. Actions • Adopt and Implement "The Road Map To Palm Springs's Long-Term Fiscal • Solvency Plan". • GOAL 1. To balance General Fund Budget, replenish the General Fund reserves, maintain • the city's bond rating 2. To put the residents first when it comes to the residents' budgeting priorities. 3. To provide transparency, hold city department managers and city employees accountable for results and improve efficiency in the management and delivery of city services to the residents. 4. To give city departments greater freedom and flexibility to manage their own budgets, personnel, and purchasing, but in return, to demand greater • accountability for results: 5. To let the city department managers manage: untie their hands, give them clear missions and real authority, hold them accountable for results, and free them • from the constraints of hidebound rules. ACTIONS 1. Implement Priority Based Budgeting, an approach that puts the • residents first, That Focuses Delivering Outcome Not Creating Or Maintaining Programs. • 5 Priority Based Budgeting is an approach to budgeting that puts residents first, • asking the residents their priorities for tax dollars as a first step to assembling a General Fund Budget. It takes budget writing out of the city and put it right into the neighborhood - into the resident's hands. A Priority Based Budgeting approach includes specific outcomes measurements to ensure progress is being made toward the goals set by the residents and requires the city to report back on whether taxpayer actually got their money's worth. When we put the residents first and focus on results, the • outcome is always the same: a common-sense budget based on measureable goals. We will know, month by month week-by-week, with specific data whether or not the city is being effective. • Instead of starting with previous year's General Fund Budget and justifying increases or cuts, let's build the General Fund Budget that delivers the results the residents want at a price they are willing to pay by adopting a new • budgeting approach called Priority Based Budgeting that calls for the mayor 9 9 PP Y 9 9 Y and city council to set the price of the city government: Decide up front how much residents are willing to spend, find out what results the residents most value using the residents input, allocate resources among high priority results • and lower priority would not be funded, conduct analysis to determine what strategies, programs, and activities will best achieve desired results, budget available dollars to the most significant programs and activities, set measures of annual progress, monitor, and close the feedback loop And check what . actually happened using performance measures and communicate performance results. BENEFITS OF PRIORITY BASED BUDGETING . While most obvious benefit of Priority Based Budgeting is to help maintain fiscal responsibility almost just as important is how Priority Based Budgeting pushes the city to rethink their priorities and managers to rethink the way they deliver service. In addition, Priority Based Budgeting would help the city: • • Eliminate Obsolete Or Low Value Activities. • Find The Revenue For Important New Investments. If an investment is important, it rises to the top of the list, whether it is old or new. Other spending, which offers less value, falls off the list. • • Ensure Accountability For Performance. Programs must deliver results or they don't get funded. 6 • • • Talk About The Budget in Common Sense terms. Most traditional budgets are so complicated they seem to be in another language, but Priority Based Budgeting is designed for anyone to understand. • 2. Implement A Chartering Department Initiative. Under this new initiative, departments and agencies would be offered a trade-off; if they volunteered for this program, their annual appropriation would be cut, but in return, they • would get more flexibility in how they operate. Participating departments would be exempt from mandated across-the board budget cuts, but they would have to commit to producing measurable benefits • either by saving money or earning revenues. To achieve those goals, they • would be given incentives to act entrepreneurially. . The key to more effective management is to let the managers manage. Untie • their hands. Give them clear missions and real authority, hold them accountable for results, and free them from the constraints of hidebound rules. Chartering city departments would get them thinking about what they • can do, not what they can't do and look at results rather than rules. 3. Implement The GovStat Initiative, a management accountability tracking initiative designed to monitors and determine the effectiveness and • efficiency of city service delivery in all city departments on a weekly basis, set performance goals, hold city employees, managers accountable for their performance on a weekly bases and measures result not yearly, quarterly or monthly, but week to week, allow follows- up to ensure problems were actually fixed in a timely matter and deploy resources quickly to fix a problem. • The goal of the GovStat Initiative is to save the city revenue and improve service delivery in every department. GovStat is an innovative computer- based management-tracking program that is based on accountability, efficiency and transparency. By assuring accountability and determining the best deployment and use of city resources, GovStat approach will provide the necessary platform for • tacking accurately how city services are being delivered and help the city • government create greater efficiencies in every city department and hold city employees and department manager accountable every week. • GovStat approach is an accountability tool that ensures that failures in city services delivery are acknowledged amended in timely fashion. The GovStat Initiative will lead to greater efficiency in the delivery of city services, greater accountability for all city employees and streamlining city government. The GovStat Initiative will serve as the cornerstone and centerpiece of the city commitment to improve and responsive services throughout government. The city will establish a system where residents can ask a question of one of • the city departments that would be answered during a GovStat meeting. • Accountability will be a primary goal as member of the city council or the mayor, and publicizing and broadcasting GovStat meetings and streamlining the meetings on the city website, I want the residents to be part of the • process. SUMMARY • With a GovStat approach, department heads and representatives meet regularly with executive officials to discuss performance data and develop . strategies to improve or sustain the performance of each participating department. Meetings enable department heads, staff, and executive officials to review and analyze measurements of department performance and efficiency. • The GovStat approach is based on the idea of holding department manager responsible for results through a regular review of performance data. If any . programs or services are not performing as expected, strategies are then developed to quickly correct any problems. Because the GovStat approach system relies on real-time access to operational level data, and resources to develop strategies to quickly correct problems, it requires an investment in • technology and commitment from executives. The GovStat approach at times are viewed as confrontational as department • managers are "put on the spot" for poor performance. This however, is not S their intent. The purpose of a stats program is to focus decision-makers attention on operational level data to identify and correct unwanted trends before they become big problems. The GovStat approach is most effective • when operational reviews are connected with achievement of defined goals and objectives. HOW DOES IT WORK This is how the GovStat Initiative should work: Bring top city officials together • frequently, arming them with timely data that allows them to track everything • from service-delivery to work force trends. Encourage them to challenge program managers. And perhaps most important, remind the managers that they'll be held accountable. PRINCIPLES PROCESS OF SUCCESS 1. Provide timely, accurately and relevant data 2. Analyze data and develop effective solutions that responds to 8 �' • emerging issues 3. Deploy resources quickly to address issues 4. Relentless follow-up and assessment OBJECTIVES • Ensure accountability, reliability and productivity of city workforce. • Develop and execute effective strategies and tactics for all delivery systems. 0 Accurately gauge the impact and effectiveness of city service delivery. • Identify and implement any new technologies needed to improve service • delivery. • Maintain consistent follow-up of all city functions. • Guarantee swift deployment of resources where needed. • 4. Develop And Implement Performance Based Contracting PURPOSE The purpose of performance based contracting is to align incentives in order to improve results. It is a method to focus contractors on supporting the strategy and goals of the contracting organization. Performance contracting with suppliers would focus on changing the relationship from one based on paying for inputs or meeting specifications to one based on producing outcomes important to the state. This relationship should be accompanied by simplification of reporting and less red tape. SUMMARY w Across the globe, governments are opening up traditional public services to •, the competitive marketplace. The government of the future will be more a • focuser of resources than an owner, and more a purchaser and manager of services than a supervisor of personnel and direct service provider. This dramatic change in the nature of government requires fundamental organizational changes. For governments to avoid failures and mishaps, they must concentrate on becoming smarter shoppers. This means creating contracting systems that are outcome based; writing • contracts that contain clear performance standards; incorporating financial . 9 incentives and penalties into the contract; and developing advanced • measurement techniques. Such state-of-the-art contracting is often referred to as "performance-based-contracting." When properly structured, performance- based contracting holds great promise to reduce contracting costs while increasing service quality. • Performance-Based Contracting has several key components: • 1. Outcome Based Contracts Performance contracts clearly spell out the desired end result expected of the , contractor, but the manner in which the work is to be performed is left to the • contractor's discretion. Contractors are given as much freedom as possible in figuring out how to best meet government's performance objective. 2. Performance Standards The process of drawing up the Request for Proposal is a great way to focus a manager's mind on exactly what it is the agency wants accomplished from the . delivery of a service, operation of an enterprise, or running of a program. 3. Financial Incentives and Penalties Performance-Based Contracting gives city officials the freedom to creatively design contractor payments to correspond with certain performance pegs. Incentives to increase productivity, cut costs and raise service quality can be built into the contract. Incentive-based contracts shift much of the risk onto • the contractor, who is rewarded for productivity improvement and penalized for poor performance or rising costs. 4. Advanced Monitoring and Measurement Techniques • The monitoring plan defines precisely what a city government must do to guarantee that the contractor's performance is in accordance with contract • performance standards. • Different services require different types and levels of monitoring. For highly visible services that directly affect citizens such as snow removal and • garbage pickup, poor service will be exposed through citizen complaints. For • complex or technical services, it may make sense to hire a third party to monitor the contractor. The contractor should be considered a strategic partner and given incentives to innovate, improve, and deliver better customer • service 10 • i 5. Create Savings incentives Program. Challenge divisions and departments • to save and reward their efforts by making some or all of the savings available for investments in improved performance. City Departments would be able to roll over 50% of what they do not spend on internal operations during a fiscal year. Department and divisions should be permanently allowed to rollover i 50% of unobligated year-end balances in all appropriations for operations 6. Provide Flexibility To Reallocate Money Across Budgets In Response To Changing Circumstances. The best solution to the problems of budget • rigidity is to give City department managers Lump-sum operating budgets without line items and let them allocate resources as they see fit and holding them accountable. 7. Create An Innovation Fund. Innovation funds are pots of money organizations can use to invest in efforts to improve service, reduce costs or both. These funds are most effective when every investment is connected to • a specified return whether financial, so the fund can be repaid, or in terms of • service quality. 8. Create An E-Commerce Infrastructure. Use electronic bidding, e-Bay-Like • purchasing, electronic catalogues and ordering, and electronic funds transfer. In effect, many governments have put their catalogues of approved commodities and services online and have asked vendors to compete with • one another by adjusting their prices every day. 9. Partner With Vendors And Get Them To Share The Risk And Rewards. Create long-term partnerships with suppliers to harness their best thinking • about productivity improvements. Especially when a project is complex or uncertain, pay them a share of savings or revenues than a guaranteed fee. • 10. Enhance Competition. For commodities, managers should qualify products • from multiple vendors and force continuous competition on price. For specialized procurements, ensure that multiple vendors are included, as procurement offices have traditionally done, and move toward results based • specifications. This will encourage more vendors to think of themselves as • possible bidders, while encouraging traditional bidders to think more creatively about how to deliver the desired results. • 11.Simplify And Automate The Purchase Of Commodities, While Expanding The List Of Items That Qualify. Purchasing an off-the shelf commodity should be much simpler than purchasing other things. There is no need for an elaborate process of writing specifications, reviewing them, qualifying vendors, and so on. And there are usually enough vendors of commodities that market competition provides equality goods at reasonable prices. Use master contracts of approved commodities, create electronic catalogues, put • 11 commodity purchases online, and insist that vendors update prices often. • Continuously review what's included, since yesterday's specialty items • become tomorrow's commodities. 12.Establish Biennial Budgeting. A two-year (or biennial) budget strengthens planning b forcing them to account for longer-range the cit s financial Yp g Y 9 issues. For example, short-term grants that may fund parts of your operation (e.g., the "COPS Fast" program) may reduce or disappear, forcing the city to find other means of support for the program. Adding a new position mid-way through a year may make the cost appear less in the first year, but the city would have to reckon with the full cost the second year. But the primary reason for going to a two-year budget is the time it saves. • Most local government line items don't change much from year to year. State-imposed limits on revenue often make revenue projections simple (in a sick way). Union contracts can be negotiated to line up with the budget • period, and personnel costs are relatively easy to project. There is no point • pretending that local budgets are "zero-based" when they are, in reality, very incremental. The City may suffer from some large but erratic revenue sources (e.g., sales tax that excludes groceries in the base), or be hit with some completely unexpected expenses. There are two options here: 1. Stabilize the city's revenue side by basing the next biennium's revenue estimate on what you actually collected in the past biennium. Then if revenues actually come in higher than the estimate, the city can—in the next • budget cycle—allocate this "surplus" to increased fund balance (reserves), • capital improvements, or investments in increased efficiency. Similarly, you can budget a General Fund Contingency Reserves to cover most unexpected expenses. • 2. Amend the budget at the one-year point. Even here, the city will only need to deal with a few problem programs or funds. PROVEN RESULTS i Here are some specific examples of where these reforms made their mark on communities that used them. • In 2003, with three years in a row of declining revenues, the State of Iowa implemented "the charter agencies" initiative. The governor negotiated with six agencies a flexible performance agreement, including what measurable results they will produce, lower net expense targets, and flexibilities they will receive. • 12 • • this initiative saved the taxpayers $92 million dollars over the first three years, while improving services dramatically. • In 2004, Snohomish County, faced with a $13.4 million-dollar General • Fund budget deficit implemented Priority Based Budgeting. This reform not only produced a structurally balanced budget, but a growing reserve, an upgraded bond rating, and improved services. In 2000, the City of Baltimore, faced with huge budget deficits, an unresponsive city hall bureaucracy and a high crime rate, implemented GovStat. Since its inception, this reform has produced dramatic improvements in city services and efficiency, with savings of$350 million dollars. In the first year, it saved the city $13.2 million—$6 million in overtime pay alone. Nonpublic safety overtime fell by 40% within the first three years, and absenteeism reduced as much as 50% in some agencies. Let's start rethinking how you can reinvent city hall, so you can deliver the results your residents want at a price they're willing to pay. � 13 I � BEST PRACTICE Budgeting for Results and Outcomes (2007) (BUDGET) • Background.The National Advisory Council on State and Local Budgeting(NACSLB)has identified four essential principles of effective budgeting. The specific principles include: (1) set broad goals to guide decisions, • (2)develop strategies and financial policies, (3)design a budget supportive of strategies and goals and(4) focus on the necessity of continually evaluating a government's success at achieving the goals that it has set for itself (i.e.,performance). The Government Finance Officers Association(GFOA)has officially adopted the recommendations of the NACSLB. GFOA also has issued separate recommended practices on strategic planning • and performance measurement. All of these documents underscore GFOA's longstanding support of strategic planning and performance measurement as part of the budget process. • Consistent with the NACSLB principles,a growing number of governments use the budgeting for results and outcomes approach. Rather than starting with the prior period's budgeted programs and activities,they begin with • available revenues,continue with a consideration of desired results and strategies,and then conclude by deciding • what activities and programs can best achieve desired results. This approach is a marked departure from the incrementalism often characteristic of budgeting. Budgeting for • results and outcomes links strategic planning,long-range financial planning,performance measures,budgeting, and evaluation. It also links resources to objectives at the beginning of the budgetary process,so that the primary focus is on outcomes rather than organizational structure. Recommendation. The Government Finance Officers Association(GFOA)recommends that governments consider budgeting for results and outcomes as a practical way to achieve the NACSLB objective of integrating • performance into the budgetary process. GFOA believes that the following steps should help a government in • making this successful transition: (1) Determine how much money is available. The budget should be built on expected revenues. This would • include base revenues, any new revenue sources,and the potential use of fund balance. (2) Prioritize results. The results or outcomes that matter most to citizens should be defined. Elected leaders • should determine what programs are most important to their constituents. (3) Allocate resources among high priority results. The allocations should be made in a fair and objective manner. (4) Conduct analysis to determine what strategies,programs, and activities will best achieve desired results. (S) Budget available dollars to the most significant programs and activities. The objective is to maximize • the benefit of the available resources. (6) Set measures of annual progress, monitor, and close the feedback loop. These measures should spell out • the expected results and outcomes and how they will be measured. • • (7) Check what actually happened. This involves using performance measures to compare actual versus budgeted results. • (8) Communicate performance results. Internal and external stakeholders should be informed of the results in an understandable format. • Budget professionals may play a number of roles in budgeting for results and outcomes, including the following: • Facilitating government-wide results and analytic support. • • Providing"reality checks"on budget allocations and expected revenues in the light of adopted financial policies. • • Advising on allocations for administrative support functions,which provide necessary organizational • infrastructure for achieving community goals,but do not typically emerge as high priorities on their own. • • Analyzing work product to evaluate the process of budgeting for results and outcome. • Serving as an advocate for outcomes and the process in general rather than for any particular • department. Budgeting for results and outcomes is not just a one-year exercise,but also a multi-year effort that should improve the budget process. • References • GFOA Best Practice, "Performance Management: Using Performance Measurement for Decision Making • (2002)-Updated Performance Measures," 1994. • GFOA Best Practice, "Recommended Budget Practices of the National Advisory Council on State and Local Budgeting(NACSLB)," 1998. i • GFOA Best Practice, "Adoption of Financial Policies,"2001. • • GFOA Best Practice, `Establishment of Strategic Plans,"2005. • • GFOA Best Practice, "Managed Competition as a Service Delivery Option,"2006. • Approved by the GFOA's Executive Board,March 2, 2007. Am -'Er, 1 p '(IPIP MMi fi 0 fi , i'i i I 4 - i iii � 11� i� yi I i Iry 1 l Gil - I IIT o ti i I i� III � �( fii: I iI 1 III i 11 r �' li a i P ii it Tii j. 41 " ) i Ir 1 p l o W - p— 1 _ = r riJ�,` i� ip o k,i IIIII'= .uv', ��6 �'l iil fill -1 1�.i li 11 i (�i� :r'ii r III I�� '.'�:I= p� - �"���� ::t P i _�rlyi y�.'lli,s ,I III i? lip .,v �1�� 1 ail rr) :.d Ir °-,� (' Iligi "a,i - �' ( — 1 I� H. ii �•• o _ - tl y" i'. Ili (iii 6 ", J ( �. tI i. P r ill " Ill — i, -ii (, =11-2 1 "` (i r I1, rim ( i�� 1i ( Ili II�(ii IV ia1 - 11Ti l i ry l 'I u o - r 6'i 11il itl Ili IIIII "I4ii iirl = _ °ihlo"yiy 1 WC g It it - ` li( _ II 9i r,r< Iir ii II it rili� r _ ,i it 1 ii �21 �. I Ili 1 III` it fir �i•_ 1 'Ili r r r.mjp _�� 1) L:T11 Ili 'ilk _ i illy li ." i _ y 31 B o 0 6 I _.yi d 3 'vi' i � 1 �� ,ph �F • • • • • i r � i 1 i 1 r � � I' i �i I �i � j��t� �! p i �� � i III ,:. �.t_ III fi€I p9IE It � '�l�'i al ", �.c,r l� 1 �.. • • • Credits . This paper was written by Shayne C. Kavanagh, Jon Johnson, and Chris Fabian. Kavanagh is Senior Manager of Research for the GFORs Research and Consulting Center in Chicago, Illinois; he can be reached at • skavanagh@gfoa.org. Johnson is a Senior Manager, Research and Advisory Services, at the Center for Priority Based Budgeting; he can be reached at jjohnson@pbbcenter.org. Fabian is a Senior Manager, Research and Advisory • Services, at the Center for Priority Based Budgeting; he can be reached at cfabian@pbbcenter.org. • • The following individuals provided valuable contributions to this paper: • Marcia Arnhold Finance Director,Mesa County, Colorado • Mike Bailey Finance Director, City of Redmond, Washington • Kindle Bowden Office of Management and Budget Manager, City of Lakeland, Florida • Steven G Chapman 11 Director of Finance, City of North Lauderdale, Florida • Ed Hacker • Strategic Planning and Continuous Improvement Manager, City of Lakeland, Florida Stanley Hawthorne • Assistant City Manager, City of Lakeland, Florida Anne Kinney • Director, Research and Consulting Center, GFOA Fran McAskill • Director, Finance and Strategic Planning, Polk County, Florida Christopher Morrill • City Manager, City of Roanoke, Virginia Roger Neumaier, CPA • Finance Director, Snohomish County, Washington • Jay Panzica Chief Financial Officer, City of Ventura, California • Walter C. Rossmann Assistant Budget Director, City of San Jose, California • Lorie Tinfow Assistant City Manager, City of Walnut Creek, California • Doug Thomas City Manager, City of Lakeland, Florida • Kim Walesh Economic Development and Chief Strategist, City of San Jose, California • Wanda Williams Research and Budget Director, City of Savannah, Georgia • 14N, i �a . � !F } • �D'u���y� ���I�� t �� is - 'i� € Q �� - • • • • � I +� � ii III��ifj�7i I� (� • III ..I • • • Contents • Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Leading the Way to Priority-Driven Budgeting . . . . . . . . . . . . . . . . . . . . . . . . 2 • Steps in Priority-Driven Budgeting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 • 1. Identify Available Resources . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 • 2. Identify Your Priorities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 • 3. Define Your Priority Results More Precisely . . . . . . . . . . . . . . . . . . 8 • 4. Prepare Decision Units for Evaluation . . . . . . . . . . . . . . . . . . . . 10 • 5. Score Decision Units Against Priority Results . . . . . . . . . . . . . . . . 11 • 6. Compare Scores Between Offers or Programs . . . . . . . . . . . . . . . 13 • 7. Allocate Resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 • 8. Create Accountability for Results, Efficiency, and Innovation . . . . . 17 • Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 • Appendix 1: Building a Program Inventory . . . . . . . . . . . . . . . . . . . . . . . . . . 20 • Copyright 2011 by the • Government Finance Officers Association 203 North LaSalle Street, Suite 2700 • Chicago, Illinois 60601 www.gfoa.org • • • • • • • • • • Anatomy of a Priority-Driven goals and objectives that are of greatest value to Budget Process the community. In a priority-driven approach.,a • government identifies its most important strate- Introduction gic priorities,and then,through a collaborative, • The traditional approach to governmental budg- evidence-based process,ranks programs or serv- ices according to how well they align with the • eting is incremental:The current year's budget priorities.The government then allocates funding • becomes the basis for the next year's spending in accordance with the ranking. plan,and the majority of the organization's ana • lytical and political attention focuses on how to The purpose of this paper is to describe factors • modify this year's spending plan based on rev- that have led governments to adopt priority . enues anticipated in the next year.'An incremen- budgeting and to identify the essential concepts tal approach is workable,if suboptimal,in peri- and steps in such a process,including the adap- • ods of reasonably stable expenditure and revenue tations individual governments have made to • growth because the current level of expenditures customize priority-driven budgeting to local con- can be funded with relatively little controversy. ditions.The paper is based on the experiences of • However,the incremental approach to budgeting the governments below,which were selected for • is not up to the financial challenges posed by the variety in organization size,type of government, new normal of relatively flat or declining rev- and approach to budgeting.'This paper builds on • enues,upward cost pressures from health care, prior publications about priority-driven budget- pensions,and service demands,and persistent ing by taking a step back from specific approach- structural imbalances.' es to budgeting and describing the major steps in • Priority-driven budgeting'is a common sense, the process and then outlining options for put- ting those steps into operation.It is GFOA's • strategic alternative to incremental budgeting. hope that this paper will give those who are new • Priority budgeting is both a philosophy of how to to priority-driven budgeting a solid base from • budget scarce resources and a structured, which to get started,and to provide veterans of although flexible,step-by-step process for doing priority-driven budgeting with ideas for further • so.The philosophy of priority-driven budgeting. is that resources should be allocated according to adapting and sustaining priority-driven budget- how effectively a program or service achieves the ing in their organizations. • • • • • n • • • • • . GFQA Research and Consulting I www.gfoa.org gfoa.org ! • • Leading the Way to Priority-Driven time,they must not be overly committed to any • Budgeting particular budgeting technique or process.They • Priority budgeting represents a fundamental. must remain adaptable and able to respond to the . change in the way resources are allocated.The circumstances while remaining true to the philos- • governing body and the chief executive must ophy.If the organization doesn't have this type of understand and support the process and commu- leadership,it might be better to delay priority- • nicate that support throughout the organization. driven budgeting or look to another budgeting • In addition,these officials must be willing to reform that has greater support.The"Philosophy carry out their decision-making responsibilities in of Priority-Driven Budgeting"sidebar describes . a way that is consistent with a priority-driven the philosophy of priority-driven budgeting and • process.The change an organization desires to its central principles.Use these principles to test bring about by virtue of implementing priority- the support among critical stakeholders and to • driven budgeting won't happen overnight,so build a common understanding of the tenets the • those leading the move to priority budgeting budget process will be designed around. • must make it clear that this type of budgeting is not aone-time event-it is the"new normal."To Of course,not everyone in the organization can be • leaders expected to immediately accept priority-driven see the change through for the long-term, . budgeting with the same enthusiasm.The leader- must have a passion for the philosophy underly- • ship must articulate why apriority-driven budget ing priority-driven budgeting,but at the same • Eq M i� � n i �� �h�i � i �� did""�i �i i ( ��i iI 0 i i�i�(` ��ii • y' Ii p i k E Ii TOA P, IPA 40Will es s � � ii • • GFOA Research and Consulting l www.gfoa.org 2 • s s • s • • is something worth actively supporting and voting l.y say one of the main reasons they endorse it is • for,rather than just a"least-worst"outcome in a because it allows them to achieve what inspired . time of revenue scarcity.'The leadership must also them to run for office in the first place-identify- create a sense of urgency behind priority-driven ing the results and implementing the policies that budgeting by showing the financial forecasts, are most important to their community. analysis,and other information that supports the need for a new approach to budgeting.Ensuring Senior staff must support the process as well that a priority-driven budgeting process is suc- because priority-driven budgeting requires a sig- • cessfully adopted requires organization-wide nificant time commitment from staff.If the board • acceptance and a shared understanding of the and CEO are behind priority-driven budgeting,it entity's financial condition.For example,the City will go a long way toward getting senior staff of Savannah,Georgia,shared trends in major rev- engaged. Staff members who have experienced • enue sources,reserves,and long-term forecasts to priority-driven budgeting say they support it show that the city's revenues were entering a peri- because it gives them a greater degree of influ- od of protracted decline.Of course,the case need ence over their own destinies. Staff no longer not hinge on financial decline.A case can also be passively awaits judgment from the budget made based improving the value the public office;instead,they create their own solutions • receives from the tax dollars government spends. because priority-driven budgeting invites them to articulate their relevance to the community. • Two groups in particular that must be recruited . to support priority-driven budgeting-elected To raise awareness about the move to priority- officials and senior staff.Elected officials need to driven budgeting and to build support for it show consensus and support for priority-driven among all stakeholders,the governments that • budgeting to make it through the challenges in shared their experiences for this paper emphasize the budget process that will inevitably occur. the importance of a communications and risk mit- Ideally,at least one or two elected officials will be igation strategy.The strategy identifies major • attracted to the philosophy so they can champion stakeholders,their potential concerns,and mes- the idea with other officials.Elected officials may sages and actions that can assuage those concerns. • be particularly drawn to the fact that priority- For example,employees might want to know if . driven budgeting allows them to set the organza- their job tenure will be affected,and citizens . tion's key priorities and see how services align or might want to know the implications for service don't align with their priorities.This puts elected offerings.The need for transparency in the process officials in an influential policy-making role-per- cannot be emphasized enough-many organiza- • haps more powerful than under a traditional. tions create a specific Web page to provide budgeting system.Elected officials who have employees and citizens with regular and timely • experienced priority-driven budgeting consistent- updates on the process as it unfolds.Involving key _ stakeholders-such as the Chamber of Commerce, m,Iua79i1 yf �� �� labor union leaders,editorial staff from the media, o and leaders of community groups and neighbor- hood groups-at appropriate stages in the process often provides the best form of"informal"commu- � �V ncation to the rest of the public.In communities such as Boulder,Colorado,and Fairfield, �, California,a town hall format was used as a com- munication device.The first group was asked to invite others to subsequent meetings,and not only �� �� did they invite friends and family,but they '� h brought them to the event. • • • i. GFOA Research and Consulting/www.gfoa.org 3 s • Perhaps the primary risk to successful priority- demonstrate that this kind of budgeting is consis- • driven budgeting that officials and other stake- tent with best practices,but,most of all,devote holders might reject of the process because they time to intensely study priority-driven budgeting. see it as insufficiently legitimate-the process is Some of the research participants for this article thought to be flawed in some way that makes it a studied it for two years before moving forward. poor basis for allocating resources.Mitigate this While two years of study will not be necessary for • risk by conferring"democratic"and substantive every government,becoming fluent in priority - legitimacy onto priority-driven budgeting.' budgeting allows the leadership to • speak Democratic legitimacy means that the process is convincingly on the topic and lead an honest dis- consistent with the will of the public.Engage the cussion about the feasibility of priority-driven • elected officials,the public,and employees in the budgeting for the organization.If the organization process to achieve democratic legitimacy.When decides to move forward,the leadership's expert- a budget process is seen to have democratic legit- ise will allow it to design a credible process,define • imacy,it gives elected officials permission to the roles of staff in priority-driven budgeting,lead resist narrow bands of self-interest that seek to others through it,and adapt to the pitfalls and overturn resource allocation decisions that are curveballs that will be encountered. based on the greater good. . The next section describes the major steps in a Substantive legitimacy means that priority-driven priority-driven budgeting process and provides • budgeting is perceived to be based on sound tech- options for answering the six questions-listed nical principles.Use Government Finance Officers below-for customizing priority-driven budget- Association(GFOA)training and publications to ing to your organization. r1 i � �, �i�ii.��i7�ii �� ��-.pi q 2� ;III �(� �ii �- �iir tlq iii � ��.�pb � :'� �i� • 6 Ut p �r s r� � 11 r i YO MH • GF©A Research and Con'sulting I www.gfoa.org q • • i • • • Steps in Priority-Driven Budgeting tal shift in its approach to budgeting.This shift, There are eight major steps in a priority-driven while subtle,requires that instead of first having wbudget process.Exhibit 1 provides a map for how the organization identify the amount of resources S the eight steps fit together,and the steps are "needed"for the next fiscal year,it should first more fully described in the following pages.'As clearly identify the amount of resources that are • the exhibit shows,the eight steps are not com- "available-to fund operations as well as one-time • pletely linear.Steps 1 and.2 can begin at the same initiatives and capital expenditures. time,and Step 8 comes into play at many differ- ent points of the process. As their first step in budget development,many • organizations expend a great deal of effort in 1. Identify Available Resources completing the analysis of estimated expendi- • Before embarking on priority-driven resource allo- tures to identify how much each organizational • cation,the organization must undergo a fundamen- unit will need to spend for operations and capital E i irp i ( r i I I ) O I I • = h1( �� 1 III (� �� t x ��i P �� h rVu'i � r 1 li I' epI° ,r pig( ��� Id��i� - III3 IMF) d (� = i�0� 4 � _ �i P4 �i 4It r ,z IN llP ` • �� �� �l p (i �4 Tlh il, ii dI a ii Wi r a hip ill, III (i q �Iii i i5i i3 1i i i E n II r ip �G 5 • �'� �' _ uPl' 111 i � �� h �i I i Ii i a - i� I �i 9i7 Yll N-MIN i i • • GFOA Research and Consultini!www.ofoa.org 5 • in the upcoming fiscal year.Once that"need"is dispels the belief that there are"secret funds" • determined,then the organization looks to the that will fix the problem and establishes the level • finance department or budget office to figure out of trust necessary to be successful. how these needs are to be funded.An integral • part of the priority-driven budgeting philosophy In the first year,an organization might choose to is to spend within your means,so the first step in focus attention on only those areas that do not have • developing a budget should be focusing on.gain- true structural balance.For most organizations,this . ing a clear understanding of the factors that drive will often include the general fund,but the jurisdic- revenues and doing the requisite analysis to tion might decide to include other funds in the • develop a reasonably accurate and reliable rev- process.Both Polk County,Florida,and the City of. . enue forecast in order to understand how much Savannah took steps to limit the scope of imple- is available to spend for the upcoming fiscal year. mentation.For example,Polk County concentrated on the general fund,and Savannah excluded capital • projects from the process. Step 1 Intended Result:Adopt a"spend within your means approach-meaning there is a com- mon understanding of the amount of resources available and that there is a clearly established limit on how much can be budgeted for the upcoming fiscal year. st- 04, 2. Identify Your Priorities Priority-driven budgeting is built around a set of organizational strategic priorities.These priori- Resources must also be clearly differentiated in ties are similar to a well-designed mission state- terms of ongoing revenues versus one-time ment in that they capture the fundamental pur- sources.The organization must be able to identi- poses for which the organization exists and are fy any mismatch between ongoing revenues and broad enough to have staying power from year to • ongoing expenditures(operations)as well.as year.A critical departure from a mission state- between one-time sources and one-time uses ment is that the priorities should be expressed in (one-time initiatives,capital needs,fund balance terms of the results or outcomes that are of value reserves).This analysis will ensure that the enti- to the public.These results should be specific ty can pinpoint the source of its structural imbal- enough to be meaningful and measurable,but not ance and address it in developing its budget.This so specific as to say how the result or outcome will also ensure that a government does not will be achieved or become outmoded after a • unknowingly use fund balance(a one-time short time.Below are the five priority results • source)to support ongoing expenditures. determined by Mesa County,Colorado. Notice how these results are expressed in the`voice of . Once the amount of available resources is identi- the citizen." fied,the forecasts should be used to educate and inform all stakeholders about what is truly avail- A strategic plan,vision,and/or mission statement • able to spend for the next fiscal year.The organ- can serve as the ideal starting point for identifying • zation must understand and believe that this is the priority results.If you have an existing strate- truly all there is as it begins developing the budg- gic plan,it might be helpful to ground the priority et. Sharing the assumptions behind the revenue results in these previous efforts to respect the projections creates a level of transparency that investment stakeholders may have in them and to GFOA Research and Consulting I www.gfoaaorg 6 . i • I Want plans and 1 ture that mairdain • quality of I". Create Well Planned and de-vetoped communities, ,€ r • l want a healthy • Mesa Gouty Promote Public Heatth give the priorities eater legitimacy.If you don't so that the governing board must full support g p � �' g� � Y� Y g g Y pp have an existing plan,developing one as a prelude them.The role of an elected official is to set the to priority-driven budgeting can provide a results the organization is expected to achieve. stronger grounding for the priorities.It might also Developing the priorities might also be a good help increase the enthusiasm of elected officials place to involve citizens.Some communities have . and senior staff for priority-driven budgeting,as used traditional means of doing this,such as citi- . they seek a way to connect the new plan to deci- zen surveys,focus groups,and town hall meet- sions about annual resource allocations. ings to engage citizens in helping establish the expected results for their community.Others are The governing board also needs to be closely being innovative.The City of Chesapeake, involved in setting the priorities.The priorities Virginia,recently asked citizens viewing a result- are the foundation of priority-driven budgeting, setting exercise on their public access channel to 111 4 l I 1 1 I I I � 1 i. a oNAPLly e• „ eat w Ii — rr II IPI(ilil r 9 illiq I�s — '�"ri — r ��dl W, 6 pp i I i 4Y 0 d ! I I � y,tW - . GFOA Research and Consuttltlg/wym gfoa.org 7 participate online and share their thoughts on contribute to the safety of the community in • "what does the city exist to provide"Cities such Walnut Creek. Hence,the specific definitions of as Walnut Creek,California,and Blue Ash,Ohio, the community's results is where the identity of • set up kiosks in city facilities and asked citizens your community and the objective meaning of . to participate in a brief survey that helped vali- what is relevant is revealed. date the city council's established results and to "weight"the relative importance of those results to the community. � iaii Ali Step 2 Intended Result:A set of priorities expressed in terms of measurable results that are of value to citizens and widely agreed to be tegiti mate by elected officials, staff, and the public. ' l 3. Define Your Priority Results More Precisely The foundation of any prioritization effort is the M`�` �� Il5��� a' v �� results that define why an organization exists. Organizations must ask,"What is it that makes A powerful method for defining results was estab- us relevant to the citizens?"Being relevant-pro- lished in Strategy Maps by Kaplan and Norton." viding those programs that achieve relevant Strategy mapping is a simple way to take a com- • results-is the key purpose and most profound plex and potentially ambiguous objective-like outcome of a priority-driven budgeting process. achieving a safe community-and creating a pic- ture,or map,of how that objective can be achieved. The challenge with results is that the terms can Sometimes referred to as cause-and-effect diagrams be broad,and precisely what they mean for each or result maps,strategy maps provide an effective individual community can be unclear.For way for an organization to achieve clarity about i instance,take a result like"Providing a Safe what it aims to accomplish with its results. Community,"which is shared by most local gov- Strategy maps should be developed using cross- ernments.Organizations talk about public safety functional teams.Teams consist primarily of staff or providing a safe community as if it is an obvi- (both with subject matter expertise relating to the ous and specific concept.But is it? priority result and without),but they can also include elected officials and citizens. In the City of Walnut Creek,citizens and city leadership identified building standards for sur- Exhibit 2(on the following page)provides an viving earthquakes as an important influence on example of a strategy map from the City of • providing a safe community.In the City of Savannah for"high-performing government" • Lakeland,Florida,however,not a single citizen (Savannah's equivalent of the"good governance" or public official discussed earthquakes to define result described in the earlier sidebar).Savannah's . the very same result.In the City of Grand Island, map includes performance indicators to help Nebraska,the city highlighted community gauge if the priority result is being achieved. acceptance and cohesiveness as intrinsic to achieving a safe community(acknowledging Exhibit 3(on the following page)is a picture of a their initiatives to help integrate a growing and slightly different style of strategy map from the important population of their community- City of San Jose,California,for its"Green, • immigrant farm workers).However community Sustainable City'priority result.The center of integration was not a relevant factor that would the map is the result,and the concepts around 0 GFOA Research and Consulting,t www.gfoa,.orsg g . s 1911MMMEME,OEM= 101i�l -I�I�� Ei nor+' 31 - i i W4� in�iyi� 6 t a l — UI 1 iii6 �i ii3j1� e pi i�4, Gn G °i as,, � ���r� • - ; 07 l� t, oV6 6 ii �K � ooi �� 3 41 7 API«�GiC ( �i itl���� R I[ WOO B z � Ii 7 14 1 I i r E _ m �� i�'li 6,� it setlrih4 � . y dl fi In M — ( W 510 I i i I IT tl �������MR �G.. - _ - 0 tb . • the result are the definitions-they help the city G i �I,�' h4 olj u�C'���'",clearl articulate its riorities:"When the Cit of Y p Y01ORSan Jose (fill in the blank with any of � the result definitions),then we achieve a Green, ' ' "? ° ' ��CH Sustainable City." �JosE • Consider San Jose's result map relative to your CMI rG OF Kl NALLEY arrs and deaigne the Promoees amtaappw a own community.Would your community define City'sgmwdrto .�a,.�e�atswadda minimize emisa;wu, the relevance of your organization by its ability to energynaage,and ofb—inron nenta7 achieve a green,sustainable community?Would imp-& _ your community define a result like a green,sus- Gin, - — Sustainable tainable community in a similar or different way? city g Prorr Ang buV.."S dM One of the challenges local governments face is trying to address what can seem like a growing (and seemingly limitless)expectation for pro- grams and services.One of the benefits of devel- �m r _ '�. oping strategy maps is that local governments . can give citizens a more precise description of i• GFOA Research and Consulting/www.gfoaa.org 9 • • the results that make local government relevant. clear to participants that this ranking process is This will establish a shared foundation,a com- not a budget allocation exercise(whereby the mon context for evaluating and prioritizing the budget of a certain result is determined by the programs and services the jurisdiction.offers.A votes given to a result).Through such a ranking, service's relative priority can be evaluated only participants can express that certain results(and through a common belief about the results local therefore the programs that eventually influence government is striving to achieve. these results)may have greater relevance to the • II community than others. The City of Walnut Creek knew that citizens and community stakeholders needed to be involved in defining the priority results.The Step 3 Intended Result: Reveal the identity of rationale was that the city's priority results your community and the objective meaning • what is relevant to it through the process of defin- . would be legitimate only if community members ing priority results. were responsible for establishing the results and their definitions.The city reached out to the community on the radio,in the newspaper,and 4. Prepare decision Units for Evaluation through the city's newsletters and Web site to The crux of priority-driven budgeting is evaluat- . invite any citizen to participate in one of several ing the services against the government's priority town hall meetings.At the meeting,citizens results.Thus,the decision unit to be evaluated were asked to submit answers to the question: must be broad enough to capture the tasks that "When the City of Walnut Creek go into producing a valued.result for citizens,but then they achieve [the result the citizen was not so large as to encompass too much or be too • focused on]."The response from citizens was vague.Conversely,if the decision unit is too tremendous and generated a host of answers. small,it may only capture certain tasks in the City government staff members(who participat- chain that lead to a result and might overwhelm • ed in the meetings)were then responsible for the budget process with details.Our research . summarizing the citizen's responses by develop- subjects took one of two approaches to this ing strategy maps. issue:"offers"or"programs." � �� � � ''��i� P .' � � Offers.Offers are customized service packages OF prepared by departments(or perhaps designed by cross-functional staff teams or even private firms or non-profits)to achieve one or more pri- ority results.Offers are submitted to evaluation teams(typically comprising a cross functional group of staff,but possibly citizens as well)for consideration against the organization's priority results.Often,the evaluation team will first issue • Lastly,when defining the priority results,consid- a formal"request for results"that is based on the er whether some results might be more impor- strategy map and defines for departments,or tant than others.This could have an impact on others who are preparing offers,precisely what • how programs are valued and prioritized. Elected the evaluation team is looking for in an offer, officials,staff,and/or citizens can participate in . ranking exercises,where each artici ant is t p p r given a quantity of"votes" (or dollars,or points, ti t etc.)and can allocate their votes among all the '' priority results to indicate the relative value of one result versus another. It is important to make Frol GFOA Research and Consulting/www.gfoa.org 10 • • • Offers are purposely intended to be different conceptual leap,or perhaps not a leap at all.This from existing organizational subunits(like means less work and process risk.However,even . departments,divisions,programs)to make a when the concept of programs is familiar,be sure direct connection between the decision-unit the"programs" (or offers)are sized in a way that being evaluated and the priority results,to allows for meaningful decision making. Programs encourage outside-the-box thinking about what that are too big are often too vague in their pur- goes into an offer,and to make it easier for out- pose to be accountable for results,and it can be side organizations to participate in the process. difficult to fairly judge the impact of a program For example,multiple departments can cooper- that is too small.Generally speaking,if a pro- ate to propose a new and innovative offer to gram equates to 10 percent or more of total achieve a result instead of relying on past ways of expenditures of the funds in which it is account- doing things.A private firm could submit an ed for,then the program should probably be bro- offer to compete with an offer made by govern- ken down into smaller pieces.If a program ment staff. equates to either 1 percent or less of total expen- ditures or$100,000 or less,it is probably too The drawback of offers is that they are a more small and should be combined with others. radical departure from past practice and may be • too great a conceptual leap for some.This could Also,be aware that using programs might pro- increase the risk to the process,but if the leader- vide less opportunity for outside organizations to ship's vision is for a big break from past practice, participate in the budgeting process because the • then the risk might be worth taking.For exam- starting point is,by definition,the existing port- ple,Mesa Coun.ty's board is very interested in folio of services.For that same reason,radical having private and non-profit organizations par- innovation in service design or delivery method is ticipate fully in its budget process at some point less likely. in the future,so the offer approach makes sense • for Mesa County. S Step 4 Intended Result: Prepare discrete decision- units that produce a clear result.Think about eval- Programs.A program is a set of related activities uating these decision units against each other and intended to produce a desired result. not necessarily about evaluating departments Organizations that use the"program"method gainst each other. inventory the programs they offer and then com- pare those to the priority results.Programs are an established part of the public budgeting lexi- 5. Score Decision Emits Against Priority con and some governments already use programs Results in their approach to financial management,so Once the organization has identified its priority . thinking in terms of programs is not much of a results and more precisely defined what those . results mean,it must develop a process to objec- tively evaluate how the program or offer achieves or influences the priority results.Scoring can be approached in several.ways. The first variation to consider is if a program or , a kltNri �r �o offer will be scored against all the organization's priority results or just the one it is most closely associated with.The cities of Lakeland,Walnut Creek,and San Jose scored against all of the prior- i, � t ity results.The belief was that a program that influenced multiple results must be a higher prior- !* GFOA Research and Consulting'/www.gfoa.org 11 i • • • • • • ity—every tax dollar spent on a program that if there was an anticipated increase in demand achieved multiple results was giving the taxpayer for the program or that program received fees or the"best bang for the buck."Alternatively,organ- grant dollars to significantly cover the costs to zations like Mesa County,the City of Savannah, provide it. Finally,if the citizen.had to rely solely Polk County,and Snohomish County matched on the government to provide the program or each program or offer with only one of the priority service and there was no other outside option • results and evaluated it against its degree of influ- available,then a program was believed to be of a • once on that result.Under this scenario,guidelines slightly higher priority. should be established to help determine how to • assign a program or offer to a priority area as well The next variation is how to actually assign i as provide some sort of accommodation for those scores to programs or offers.One approach is to programs or offers that demonstrate important have owners of the programs or offers(e.g., • effects across priority result areas.Both of these department staff)assign scores based on a self- approaches have been used successfully,so the assessment process.This approach engages the right choice depends on which approach resonates owners in the process and taps into their unique more with stakeholders. understanding of how the programs influence the priority results.Critical to this approach is a • In addition to scoring the offers or programs quality control process that allows the owner's against the priority results,some organizations peers in the organization(other departments) have included additional factors in the scoring and/or external stakeholders(citizens,elected process.Examples include mandates to provide officials,labor unions,business leaders,etc.)to the service,change in demand for the service, review the scoring.The peer review group chal- • level of cost recovery for the service,and reliance lenges the owner to provide evidence to support on the local government to provide the service the scores assigned.A second approach to scor- (as opposed to community groups or the private ing establishes evaluation teams that are respon- sector).The governments believed that a pro- sible for scoring the programs or offers against gram should be evaluated more highly if there their ability to influence the priority results. was a mandate from another level of government, Owners submit their programs or offers for the . • ♦ �MOJ i W o �,In H. ;R • • • GFOA Research and Consulting!www:gf©a arg 12 • • • s � • teams to review,and the teams score the pro- their belief in the relative influence that program grams against the results.The priority-driven or offer has on the priority results it has been budgeting process becomes more like a fonnal evaluated against.In other organizations,the • purchasing process,where the departments are process can be implemented as a staff-only tool analogous to vendors and the evaluation teams that is used to develop a recommendation to the are like buyers.Evaluation teams could be made governing body. Snohomish County uses this up entirely of staff,with representation both approach,as its culture and board-staff relation from staff members who have specific expertise supports it. related to the result being evaluated and others who are outside of that particular discipline.An Regardless of which variations are selected,there alternative team composition would include both are three important points to establish.The first staff and citizens,to gain the unique perspectives is that to maintain the objectivity and trans- of both external and internal stakeholders.This parency of the process,programs or offers must second approach brings more perspectives into be evaluated against the priority results,as they the initial scoring and encourages cross-function- were defined collectively by stakeholders(see al teamwork via the evaluation teams. step 3). Secondly,scores must be based on the demonstrated and measurable influence the pro- grams or offers have on the results.Finally,the �» results of the scoring process will be provided as recommendations to the elected officials,who �- �' hold the final authority to make resource alloca- tion decisions. Step 5 Intended Result: Each decision unit offer 4 4 �''�i or program)should have a score that indicates its �I relevance to the stated priorities. ------------- Another consideration is the particular scoring 6. Compare Scores Between Offers or • method to be used.For example,will evaluators Programs have to use a forced-ranking system where pro- It is a"moment of truth"in priority-driven budg- grams/offers are fit into a top-to-bottom ranking eting,when the scoring for the offers or programs . or will each program be scored on its own merits, is compiled,revealing the top-to-bottom compar- . with prioritization as a natural byproduct?Each ison of prioritized offers or programs.Knowing system has its advantages,but the important this,an organization must be sure that it has athing is to make sure the scoring rules are clear done everything possible up to this moment to . to everyone and applied consistently. ensure that the final scores aren't a surprise and that the final comparison of the offers or pro- The role of the elected governing board in this grams in priority order is logical and intuitive. step is another point of potential variation in the scoring.In some organizations,the board is heav- The City of San Jose engineered a peer review ily integrated into the process and participates in process through which the scores the depart- the scoring and evaluation step.They have the ments gave to their programs were evaluated, opportunity to question the scores that have discussed,questioned,and sometimes recom- li been assigned by the owner or the evaluation mended for change.The city established a review team,ask for the evidence that supports that team for each.of its priority results.The team score,and ultimately request that a score be first reviewed the strategy map to ensure that changed.based on the evidence presented and each member of the team was grounded in the s • GFOA Research and Cofisultinss/www.gfoa,or 13 city's specific definition of the result.Next,the ask them to decide which programs should be • review teams were given a report that detailed cut or which ones should be preserved.They every program scored for the particular result framed the discussion very simply:Evaluate how under review.The teams met to discuss: our programs help us achieve our results,and to what degree.The outcome of prioritization was • whether they understood the programs they therefore expected and self-evident,based on the were reviewing; common understanding of the programs and hove • whether they agreed with the score given by the programs influence results. the department(the departments scored their oven programs); Stakeholders could be concerned that their • whether they required further testimony or favored programs might lose support in the evidence from the department to help them course of priority-driven budgeting. Even when a better understand the score given;and program director or a citizen who benefits from a • whether the score should stand,or if the particular program understands why that pro- team would recommend an increase or gram ranked low,they are not going to be decrease. pleased about it.Invite stakeholders from all sides,from within the organization and even the All programs were evaluated in this manner until community,to understand the process. Include • a final recommendation was made on program stakeholders at various points in the process so scores. they might influence the outcome.Constantly communicate progress,throughout the process. • The city invited the local business community, Program directors,stakeholders of a particular citizens representing their local neighborhood program,organizational leadership,and staff commissions,and labor leaders to review the might not enjoy seeing their program prioritized a i �a MIT- scores.Walter Rossman,from San Jose's City below other programs,but if they understand it, • Manager's Office,described their effort this way: if they've had a chance to influence the process, "The participants found the effort informative as and,most importantly,if they are aware of • to what the city does;they found it engaging actions they might take to improve the priority . with respect to hearing staff in the organization ranking of their program,the process will have a discuss how their programs influence the city's great chance for success. results;and,most interesting,they found it fun." Lastly,consider if the scoring of the programs or San Jose's story is important because it demon- offers will be used only to decide where to make strated how stakeholders from various perspec- budget reductions.Organizations such as the tives and political persuasions can all productive- cities of Lakeland and Walnut Creek have used ly participate in the priority-driven budgeting prioritization not only to balance their budgets, process. San Jose didn't ask these stakeholders to but also to understand how services that might come together and rank programs.They didn't appear less relevant to the city government might • GFGA Research and Consulting I WWW.gf©a.org f 4 be relevant to other community groups.These instance,the designer should think about ways groups might take responsibility for supporting priority result areas can share information during • or preserving a service.There could be great the evaluation of programs or offers,and/or ways potential in engaging other community institu- to jointly fund programs or offers. tions-businesses,schools,churches,non-profits -about partnership opportunities. t } 0111! Peter Block has focused much attention on this �,��0i ��` s issue in his book,Community:The Structure of ���e a�,,,o �� _ "' Re Belonging."Citing the way we sometimes unduly �i He rely on government to meet the community's �, m r� � WIN needs,he highlights citizens'experiences of tak- NO ing accountability for the results they hope to see achieved.This occurs when cohesion is built 1-1,1 between local government,businesses,schools, ��ii �,_- „"T d � di . social.service organizations,and churches.A Vl,,,1 @ �, � �i� complete and successful priority-driven budget- , ing process doesn't conclude when the budgets �� "��M for low-priority services are reduced-rather,it - brings together otherwise fragmented institu- tions in society to find ways of providing services that may still be relevant to the community,even Then,the offers or programs can be ordered if they are less important to the priority results a according to their prioritization within a given local government seeks to achieve. priority result area and the budget staff draw a • line where the cost of the most highly prioritized .Step 6 Intended Result:The prioritized ranking of offers or programs is equal to the amount of rev- programs is a logical and well-understood product enue available(see Exhibit 4).The offers or pro- of a transparent process-no surprises. grams above the line are funded,and the ones that fall below the line are not.The board and staff will have discussions about the programs on 7. Allocate Resources either side of the line and about moving those Once the scoring is in place,resources can be allo- offers or programs up or down,redesigning them cated to the offers or programs.This can be done to make more space above the line(e.g.,lowering in a number of ways.One method is to first allo- service levels),or even shifting resources among cate revenues to each priority result area based on priority results.Variations on the approach are historical patterns or by using the priority's rela- possible-for example,there could be multiple tive weights,if weights were assigned..Allocating lines representing multiple levels of funding cer- resources to a priority result area can be contro- tainty.In the City of Redmond,Washington, versial because,as we will see,this allocation programs above a top line were categorized as determines the number of offers or programs that "definitely fund,"while programs in between the will be funded under that priority area(e.g.,how top line and a bottom line were open to addition- many public safety programs will be funded). al scrutiny. There are no easy answers to this issue.As such, the designer of the process should look for ways Another method is to organize the offers or pro- to mitigate controversies associated with how grams into tiers of priority(e.g.,quartiles)and • much funding is allocated to one result versus then allocate reductions by tier.For example,pro- another and to prevent these allocations from grams in the first tier might not be reduced,while becoming new types of organizational silos.For programs in the lowest tier would.see the largest GFOA Research and Consulting l www.gfoa.org 15 reductions.The programs could be forced to reduction applied to any particular support serv- i make assigned reductions,or each department ice was based on its priority relative to other • could be given an aggregate total reduction target, support services.A couple of our participants based on the programs under its purview(with envisioned moving to a system wherein the cost . the implication being that the department will of support services would be fully distributed to • weight its reductions toward the lower-priority operating programs so support services would be programs,although it would have more flexibility affected according to the prioritization of the to decide the precise reduction approach than if operating services they support. the cuts were not done within the department). This tier approach generates discussion among Another question is how to handle restricted • board and staff about how much money is spent monies(e.g.,an enterprise fund).One option is on higher versus lower tier services in aggregate, to handle special purpose funds(where there are as well as on resource allocation strategies for restrictions on how the money can be used)sep- individual departments and programs.Exhibit 5 arately.For example,enterprise funds or court • presents an example of the value this analysis can funds inight be evaluated on a different track or provide.It shows the total amount of money one budgeted in a different way altogether.Another city had historically spent on its highest priority option is to rank programs or offers without programs(e.g.,the top tier)versus the others. respect to funding source,but then allocate This city was spending significantly less on the resources with respect to funding source. top tier than it was spending on the second tier, Knowing the relative priority of all the offers or and less than it was spending on the third tier,as programs might generate valuable discussion, • well.This raises interesting questions about even if there is no immediate impact on funding. spending patterns in the organization and builds For example,if a low-ranking offer or program is a compelling case for change. grant funded,is it still worth providing,especial- ly if that grant expires in the foreseeable future? Organizations also need to consider the funding Ideally,participants will become less fixated on • of support services. Many of our research partici- funding sources,realizing that the government pants elected to fund support services based on has more flexibility than it might think. For historical costs,making some reduction that was example,if a low-priority service is funded by a r consistent with the reduction the rest of the special earmarked tax,is there a way to reduce or organization was making.The magnitude of the eliminate that service and its tax,and increase a i Ali � G, ,m in o 2 777 j $(millions) $10 $20 $30 $40 $50 $60 PIT s r+ N 01, 11W t • • GFOA Research and Consulting!www.gfoa.org 16 • 0 • HEEMENNEENOM general tax by an analogous amount?As the gov- grams or offers,and innovation in the design of ernment becomes more proficient at expressing programs or offers.Although priority-driven the value it is creating for the community,it budgeting will identify which programs or offers should be better able to articulate these potential are best for achieving priority results,it does not trade-offs to the community. speak directly to the efficiency with which those • programs or offers are delivered or to innovative • Of course,no matter what method is selected to approaches to program delivery(although it allocate resources,remember that priority-driven might indirectly encourage these things). budgeting,like any budgeting process,is still a political process.As such,it will not and should , not lead to"scientific"or"apolitical"allocation of 0 resources—rather,it should change the tone of budget discussions,from a focus on how money was spent last year to a focus on how the most Basic Needs �4 value can be created for the public using the money that is available this year.10 • Step 7 Intended Result:Align resource allocation '� „ � " � � �` �" ��" consistent with the results of priority-driven scor- 8. Create Accountability for Results, �_ ! � � I �i�l �'� - Fi �p�i`,�� ' �N Efficiency, and Innovation The owners of the programs or offers being evalu- 11 „"' �� �� 0 ated might over-promise or over-represent what • they can do to accomplish the priority result.To As such,the designers of the process might need • address this potential moral.hazard,create meth- to consider specific techniques for ensuring pro- ods for making sure programs or offers deliver the gram efficiency.A proven model for improving • results they were evaluated on.Many of our efficiency helps avoid cost-cutting techniques research participants anticipate using perform- that also cut productivity and degrade the results ance measures for this purpose.For example,a a program produces.For instance,a systematic 0 program or offer might have to propose a standard method for reviewing and improving business 0 of evidence or a metric to be evaluated against,so processes could be implemented along with pri- the organization can see if the desired result is ority-driven budgeting.One such method that being provided.Exhibit 6 is Polk County's con- GFOA research has shown to be effective for local • ceptual approach for connecting its priority result governments is"Lean"process review—a system areas to key performance indicators.However, for identifying and removing or reducing the non- none of the research participants have reached value added.work that can be found in virtually • what they would consider a completely satisfacto- any business process.You can learn more about ry state in this area.For those just starting out,the Lean at www.gfoaconsulting.org/lean. 0 lesson is to understand where evidence is needed • in your process design,but also to be patient with Business process improvement can also be incor- • respect to when this part of priority-driven budg- porated into a more comprehensive approach to etng will be fully realized. reviewing program efficiency.Exhibit 7(on the • following page)provides a sample program Other issues to consider as part of the priority- review decision tree that is inspired by work driven budgeting design are the efficiency of pro- from the City of Toronto,Ontario.As the exhibit 0 0 15FOA Research and consulting I www.gfoa.ore 17 shows,a program is subjected to a series of tests process should consider hove to encourage new to see if it is being provided efficiently.For exam- ways of structuring programs or offers to best ple,can the service be shared with other govern- achieve the government's priority results. Some ments?Can greater cost recovery be achieved research argues that innovation is a"discipline, • through fees or fund raising?Can the private sec- just like strategy,planning,or budgeting."" for provide the service more efficiently?Can Lean Public managers who want to encourage innova- process improvement techniques be applied? tion will need to develop and institutionalize • Exhibit 7 also shows how the review might be dedicated processes to generate ideas,select the linked to priority-driven budgeting—discre- best ones,implement them,and spread the bene- tionary sen ices are subject to a relevance test fits throughout the organization.Along the way, that asks the above questions about each priority public managers will need to make use of a vari- program,while non-priority programs go ety of implementation strategies,including those through a divestment test. that rely on the organization s own resources and • those that seek to harness resources from out- Finally,innovation tends to be the exception side.Public managers will also have to create an rather than the rule in the public sector,so the organizational culture that is not just conducive designers of the priority-driven budgeting to innovation,but actively encourages and even • Et —11 iv `fi-• 'l��,. „m � M=Ip��)IrI h Y�ill. ilia iti'- �I i��s �i i�hrill Ii,IIGi��l'jI) ��i, k ,�i h D) $eniae l�eif ` a; li do. HYe ie�afssram marwdpt Me? No % Discretionary Services Olvesiment Tlest �9FDA Sample (service exceeds mandate) , , Program s " de ar (E_ Review Tree No i'. Yes Yes Mandatory Services j fept.tple (service within mandate) govenrimertt twe Identify organization& +alloup+#rapmeptYlsm 25eeSireofNovidet7 terms of transfer pa�ae0dsxeli Futtel?Reguiat2 of �7 ! 'I Assess impact& ��i��l there iC# abandon program ibA ttr Sid k3w� N Cflrfb� tYlFNe S o dira,i rl t Tk, Does the the OXW Teti tt r fn6veo,r- oc uro Uskca hesp #ttedin 00vemm is • s ,r riu u k Odied ai nw-10 rs to 9wkf Ne zsrusfr �� ward Gaff betY4eIm Ns dtiz�Yoluri Concern d t av zrteer s? Proaram Improvement Plan •Analysis of current situation . Analysis of options Recommendation • • • GFOA Research and Consulting!www.gfoa.org 18 • • • • • • demands it.The Public Innovator's Playbooh driven budgeting,outlined in the eight steps pre- describes one approach to encouraging innova- sented in this paper.A priority-driven budgeting tion in this kind of systematic way.`Z process can be approached in several ways,so keep in mind the major levers and decision points to create a process that works best for Step 8 Intended Result:Make sure that those who your culture and environment,and that embraces received allocations are held accountable for pro- ducing the resutts that were promised. Find ways the concepts of democratic and substantive legit- to directly encourage efficiency and innovation. imacy.The governments that participated in this research show that there are opportunities to introduce flexibility in the process-but keep in Conclusion mind that with that flexibility comes risk,if Priority-driven budgeting represents a major changes are made that don't embrace the basic shift from traditional budgeting methods.A clear principles of priority-driven budgeting. understanding of the priority-driven.budgeting philosophy should be in place before proceeding Research what other organizations have done down this path,along with a strong level of sup- and ask them about their long-term success in port-especially from the CEO(whose role is shifting to the"new normal"in local government • normally to propose the budget)and,ideally,the budgeting. Understand that priority-driven governing board(whose role is to adopt the budgeting is a process that will evolve and budget).Priority-driven budgeting is not a improve over time-don't expect perfection in • process that is brought in to fix a structural the first year.Engage outside help where needed • deficit;instead,it becomes the way an organza- to design.the process,develop successful commu- tion approaches the resource allocation process. nication plans,incorporate citizen involvement, • It brings with it an important cultural shift- and institute a process.Enjoy new conversations moving from a focus on spending to a focus on that were not possible before,and embrace the achieving results through the budget process. transparency in decision-making that accompa- Priority-driven budgeting should be perceived by nies the priority-driven budgeting process.As • all stakeholders as a process that improves deci- your organization adapts to the new normal,the Sion-making and changes the conversations process will guide decision-makers in making • around what the organization does(programs resource allocations that fund the programs that and services),how effective it is in accomplishing are most highly valued by the organization and, its priority results,and how focused it is on allo- more importantly,by the citizens who depend on • cating resources to achieve its results. those programs and services for their well being, • The success of your process design rests on a comfort,and expected quality of life. clear understanding of the principles of priority- GFOA Research and Consulting l www.gfoa.org 19 • Appendix 1' understand and can be digested by the Building a Program Inventory average citizen(i.e.,not too detailed). • Drawing distinctions between the results Introduction (that matter to citizens)provided by dif- Financial constraints have forced many govern- ferent programs.To achieve this,programs ments to take a hard look at the services they cannot to be too large or vague. offer.A fundamental step is to inventory all the • Beginning to show the true cost of doing service programs a government offers.A program business by describing what government inventory clarifies the breadth of services provid- does on a meaningful level,and then iden- ed and,ideally,highlights key characteristics of ti.fying costs for those programs. each program(e.g.,the full cost of prodding the • Laying the groundwork for priority-driven program and the level of revenues that program budgeting,where programs receive budget i directly generates to support its operations).The allocations based on.their contributions to the government's priority objectives. inventory provides the basis for discussion about the services that should be provided. • Laying the groundwork for program review,where programs are subjected to Steps to Take efficiency tests to determine if the service 1. Define your objectives and goals for the pro- delivery method employed is optimal. gram inventory.Identifying a program is as 2. Decide what information the program inven- much art as it is science-an inescapable tory should contain, in addition to the basic amount of subjectivity is involved.Therefore, to make judgments as effectively as possible, description the program.Options to con- make sure you are clear on why you are devel- sider include: make • Full cost.The full cost of the program is its a o ing program inventory. Some of the p b p g D' direct cost plus its indirect cost overhead p ( potential purposes are: charges).Full-cost accounting makes the • Understanding the complete scope of true cost of offering a service transparent, services government provides. which allows better planning and decision • Communicating the scope of services to making. It also helps show that the organ- the public in a format that is easy to ization is achieving the expected level of • t � i � cf` i P b f��c f__ 1 _00 phi i.o- 1JIlii i�.wxs,.w. �� 4�,' �' = i� ihiirF � REN . r • GFOA Research and Consulting/www.1#64.org 20 • cost recovery for a given service. Full cost- with verbs-programs are action-oriented. ing is especially important if the govern- For example,programs in a sheriff's office • ment envisions eventually going to a prior- might include crime investigations,deten- ity-driven budget process. tions,and court security.However,programs Alignment with strategic goals.Knowing should not be described in terms of overly • how programs contribute to priority goals detailed tasks.For instance,"supplying a • enables organizations to develop more bailiff for court rooms"is a task within the strategic cutback strategies. court security program,not a program itself. • • Service level. Describe the level of services • provided to the public.If service is being 5. Find the right level of detail.A program is a provided at a premium level,perhaps serv- set of related activities intended to produce a • ice levels can be lowered to reduce costs. desired result.When constructing a program • • Mandate review.List and clearly define inventory,it can sometimes be challenging to any mandates a program is subject to. find the right level of detail. If a program is • Then review the current service level too big or encompasses too much,it will not • against the mandate requirements. provide sufficient information-that is,it«rill Perhaps the service level being provided is be very difficult to describe the precise value • higher than what the mandate requires. the program creates for the public or to use • • Demand changes.Is demand for a service program cost information in decision making. going up or down?If demand is going However,if program definitions are too small, • down,perhaps the program can be cut decision makers can become overwhelmed back and resources shifted elsewhere.If with detail and be unable to see the big pic- demand is going up,steps can be taken to ture.In addition,tracking program costs for • manage demand. For example,perhaps very small programs is generally not cost- means testing can be applied to a social effective. services program. • • Support from program revenues.Describe Generally speaking,if a program equates to 10 . the extent to which the program is sup- percent or more of the total expenditures of ported by its own user fees,grants,or the fund in which it is accounted for,then the • intergovernmental revenues.Is there an program should probably be broken down • opportunity to achieve greater coverage of into smaller pieces.And if a program equates the full costs of the program? to 1 percent or less of total expenditures,or to • $100,000 or less,it is probably too small and • 3. Develop forms and templates.Create tools should be combined with others.This is just a departments can use to describe their pro- guideline-there could be valid reasons for • grams in a manner that is consistent and that going outside of these parameters.For exam- captures the information needed to fulfill the ple,a small program could be much more purpose of the inventory.Consider testing the important than its cost suggests.Here are • forms and templates with one or two depart- some other points that have proven helpful in • ments and then distributing them to a wider identifying programs: group.Also consider providing training and • an official point of contact for questions. e A program is a group of people working together to deliver a discrete service to 4. Differentiate programs from functions. identifiable users. • Departments or divisions(i.e.,public health, • A program groups all tasks that a cus- • courts,public works,sheriff)are often tomer of that program would receive and described as functions or nouns.These are does not break one program or service into not programs,which are more often described multiple items based on tasks. • GFOA Research and Consulting/www.gfeaa.©rg Z 1 • As far as possible,a program is individual are combined with other existing programs. • -a program with its own name,cus- . A program uses an existing name that is tomers,and staff team.Each program familiar to customers and staff,and/or it • stands alone and is distinct from like pro- uses a name that could stand on its own • grams in a similar service area. and would be understandable to the aver- • Programs that are handled by less than 1 ETE age reader. • 10 !a �"� (IP���' (i"'il�l�iliil� t i � � _ � '� i �� 'IP�� • I Ili 17 s q®y■��,, i (i � �yi i ���) (f I�i�i�iti ( it W li li�i�li�� .i i I s P i li i I i � d 9 _ � " T"'� ��i L (� i i 3(I ��i—=•£ r (�ii Ii � ����OIi ��i���l � Ci • ii n i I I i � I u IFS ( y Off. FAY M01 ISZ �! (i i 'Se yy a p�i ri =',r iii P i6 wi i. rfu3 i ii;i tyi I I li it �i di ftli its � i i iir i "� 14� i�4 r� A" Wt i i�a��ii�) I� ( i:) ti ��� � yik i hCi ay a IT INiN 0 Ali " pp.,�ygyp �p F �i Ali Niels PIS, r �' min R#trnav;Wd� i F(ii �i( i � ,I a I G �ry ���°tilu� _ �nsilk l � • • GFQA Research and Consulting!www.gfoa.org 22 • Notes University Press,1997). 1 The concept of incremental budgeting was 7 Diagram inspired by Eva Elmer and • developed by Aaron Wildavlsky. See,for Christopher Morrill,"Budgeting for example:Aaron Wildaysky,The Politics of the Outcomes in.Savannah,"Government Finance • Budgetary Process(Boston:Little,Brown,1964). Review,April 2010. . 2 Robert Behn discusses the shortcomings of 8 Budgeting for outcomes was the subject of • incremental budgeting in a cutback environ- The Price Of Government:Getting the Results We ment in the following article:Robert D.Behn, Need in an Age of Permanent Fiscal Crisis by David • "Cutback Budgeting,"Journal of Policy Analysis Osborne and Peter Hutchinson(New York: • and Management,Vol.4,No. 2(Winter,1985). Basic Books,2004). 3 Priority-driven budgeting is also known as 9 Robert S.Kaplan and David P.Norton, • "budgeting for results"and"budgeting for Strategy Mays:Converting Intangible Assets into outcomes,"although the latter is used to Tangible Outcomes(Boston:Harvard Business • describe a specific method of priority-driven Press,2004). • budgeting. 10 Peter Block,Community:The Structure of • 4 Personal interviews were conducted with the Belonging(San Francisco:Berrett-Koehler managers who led priority-driven budgeting Publishers,2008). • at these entities. 11 William D.Eggers and Shalabh Kumar Singh, 5 Behn. The Public Innovator's Playbooh:Nurturing Bold • 6 Mark Moore emphasizes that these two Ideas in Government(New York:Deloitte, • sources of legitimacy are essential to making 2009). • any big public policy change. Mark Moore, 12 Eggers and Singh. Creating Public Value(Boston:Harvard �I • GFOA Research and Cansult9ng f www.gf©a.org 23 • co µ t�Tot (__,incinnati - r Roxanne Qualls COUncilmernber • • • February 7, 2012 • MOTION • We move that City Council and the City Administration follow a Priority-Driven Budgeting • process when developing the 2013/2014 Biennial Budget. A Priority-Driven Budgeting process • is a broadly inclusive process that involves City Council, citizens, community leaders, and business leaders who work through an extensive and intense process to determine strategic • priorities for the City. Once Council adopts the strategic priorities,the administration uses • those priorities to determine the allocation of resources. • In the 1980's, Cincinnati's Policy Budget Process was an early version of Priority-Driven • Budgeting. The process included numerous public hearings on service areas like Parks, • Recreation, and Public Safety. At each meeting, the department would provide details about • their operations and budget. There was extensive citizen involvement to determine priorities • for each service area. There were voluminous requests for follow-up information which resulted in follow-up sessions for each service area. Based on the results of the public hearings • and information, City Council established the policy direction for the upcoming budget. • Priority-Driven Budgeting a different a •g g'is approach than anything recently used by the City, and • therefore • We move that the administration retain a consultant through an RFP process who has experience in facilitating a Priority-Driven Budgeting process. The consultant chosen must be • able to reach a broad base of people beyond the usual participants who always take part in civic • engagement processes in Cincinnati. The consultant must be able to work well with groups • such as, but not limited to community councils, neighborhood business associations,the • NAACP, the Urban League, the League of Women Voters, the Cincinnatus Association, WIN, and • local churches, etc. In addition,the consultant must have a demonstrated capacity to directly reach citizens who are not participants in civic, community, and religious organizations. • • Roxanne Qualls, Vice Mayor Wendell Yo` ng, Vice • Chair, Budget and Finance Committee Budget and Fin ce Com • • • • City Hall m 801 Plum street « Suite 34 A - Cincinnati, Ohio 45202 • 511352,3604 - ro an;ie.qualls cincinnati-oh,gov v, �" • " Equal Opportunity Employer 1,7 'I The Philosophy of Priority-Driven Budgetingi The underlying philosophy of priority-driven budgeting is about how a government entity • should invest resources to meet its stated objectives. It helps us to better articulate why the • services we offer exist, what price we pay for them, and, consequently, what value they offer • citizens. The principles associated with this philosophy of budgeting are: • Prioritize Services. Priority-driven budgeting evaluates the relative importance of individual programs and services rather than entire departments. It is distinguished by • prioritizing the services a government provides, one versus another. • • Do the Important Things Well. Cut Back on the Rest. In a time of revenue decline, a traditional budget process often attempts to continue funding all the same programs it funded last year, albeit at a reduced level (e.g. across-the-board budget cuts). Priority- driven budgeting identifies the services that offer the highest value and continues to provide funding for them, while reducing service levels, divesting, or potentially • eliminating lower value services. • • Question Past Patterns of Spending. An incremental budget process doesn't seriously question the spending decisions made in years past. Priority-driven budgeting puts all the money on the table to encourage more creative conversations about services. • Spend Within the Organization's Means. Priority-driven budgeting starts with the revenue available to the government, rather than last year's expenditures, as the basis • for decision making. • • Know the True Cost of Doing Business. Focusing on the full cost of programs ensures • that funding decisions are based on the true cost of providing a service. • Provide Transparency of Community Priorities. When budget decisions are based on a well-defined set of community priorities,the government aims are not left open to interpretation. • 1 Shayne C.Kavanaugh,Jon Johnson,and Chris Fabian,Anatomy of a Priority-Driven Budget Process. Research and Consulting Center of the Government Finance Officers Association • • • • • Provide Transparency of Service Impact. In traditional budgets, it is not often entirely • clear how funded services make a real difference in the lives of citizens. Under priority- • driven budgeting,the focus is on the results the service produces for achieving community priorities. • • Demand Accountability for Results. Traditional budgets focus on accountability for • staying within spending limits. Beyond this, priority driven budgeting demands • accountability for results that were the basis for a service's budget allocation. • • Examples of Jurisdictions that use Priority-Driven Budgeting • • City of Savannah, Georgia (pop. 131,000) • City of Walnut Creek, California (pop. 64,000) • Mesa County, Colorado (pop. 146,093) • City of San Jose, California (pop. 1,023,000) • Polk County, Florida (pop. 580,000) City of Lakeland, Florida (pop. 94,000) • Snohomish County, Washington (pop. 683,655) • Blue Ash, Ohio (pop. 12,114) • • i • • • • • • • • • • • • • • • • • • • • • • , Fa IFS -1� 1� B �C : r e _ IN f u 77 � 1 ZA wa _ P � i � 1 e yy m�A i s • • • • • • • • Credits • This paper was written by Shayne C. Kavanagh,Jon Johnson,and Chris Fabian. Kavanagh is Senior Manager of Research for the GFOXs Research and Consulting Center in Chicago, Illinois;he can be reached at skavanagh@gfoa.org. Johnson is a Senior Manager, Research and Advisory Services,at the Center for Priority Based • Budgeting; he can be reached at jjohnson@pbbcenter.org.Fabian is a Senior Manager,Research and Advisory Services,at the Center for Priority Based Budgeting;he can be reached at cfabian@pbbcenter.org. • The following individuals provided valuable contributions to this paper: Marcia Arnhold • Finance Director,Mesa County,Colorado Mike Bailey • Finance Director,City of Redmond,Washington Kindle Bowden . Office of Management and Budget Manager,City of Lakeland,Florida Steven G Chapman II • Director of Finance,City of North Lauderdale, Florida Ed Hacker Strategic Planning and Continuous Improvement Manager, City of Lakeland, Florida Stanley Hawthorne Assistant City Manager, City of Lakeland, Florida Anne Kinney Director, Research and Consulting Center,GFOA Fran McAskill . Director, Finance and Strategic Planning. Polk County, Florida Christopher Morrill City Manager, City of Roanoke,Virginia Roger Neumaier,CPA Finance Director,Snohomish County,Washington Jay Panzica Chief Financial Officer, City of Ventura,California Walter C. Rossmann • Assistant Budget Director,City of San Jose,California Lorie Tinfow Assistant City Manager,City of Walnut Creek, California Doug Thomas • City Manager.City of Lakeland, Florida Kim Walesh • Economic Development and Chief Strategist,City of San Jose,California Wanda Williams • Research and Budget Director,City of Savannah, Georgia • GFOA's Research and Consulting Center The Research and Consulting Center(RCC)is the management analysis and consulting arm of the Government • Finance Officers Association.Since beginning operations in 1977,the RCC has provided management and technol- ogy advisory services to hundreds of local,county,and state governments; public utilities;elementary and sec: ondary education systems;and transit authorities.The RCC is nationally recognized for its comprehensive analyti- cal and advisory services, as welt as for specialized research on state and local government finance. You can team more about us and contact us at www.gfoaconsulting.org or 312.977.9700. • i • �I • r • 4 • • • Contents • Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 i Leading the Way to Priority-Driven Budgeting . . . . . . . . . . . . . . . . . . . . . . . . 2 • Steps in Priority-Driven Budgeting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 • 1. Identify Available Resources . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 • 2. Identify Your Priorities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 3. Define Your Priority Results More Precisely. . . . . . . . . . . . . . . . . . 8 4. Prepare Decision Units for Evaluation . . . . . . . . . . . . . . . . . . . . 10 5. Score Decision Units Against Priority Results . . . . . . . . . . . . . . . . 11 • 6. Compare Scores Between Offers or Programs . . . . . . . . . . . . . . . 13 7.Allocate Resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 8. Create Accountability for Results, Efficiency, and Innovation . . . . . 17 • Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Appendix 1: Building a Program Inventory . . . . . . . . . . 20 • • i Anatomy of a Priority-Driven goals and objectives that are of greatest value to Budget Process the community.. In apriority-driven approach,a government identifies its most important strate- iaic priorities,And then,through a collaborative, ntroduetion cvidenrr based process,ranks programs or scrv- ✓� The traditional approach to governmental budg� ices according to how well they align with the sting is incremental:The current yeas budget priorities.The government then allocates funding becomes the basis for the nest year's spending in accordance with the ranking. plan,and the majority of the organization's Ana lytical and political attention focuses on how to The purpose of this paper is to describe factors modify this year`s spending plan based on rev• that have led governments to adept priority . enues anticipated in the next year.'An incremen- budgeting and to identify the essential concepts • t.il approach is workable,if suboptimal,in peri• And steps in Such a process,including the adap- ods of reasonably stable expenditure and revenue tations indi+ridual governments have made to gra+vth because the current level of expenditures customize priority-driven budgeting to local con- can be funded with relatively little controversy. ditions.The paper is based on the experiences of However,the incremental approach to budgeting the governments below,which++sere selected for • is not up to the financial challenges posed by the variety in organization size,type of government, new normal of relatively flat or declining rev- And approach to budgeting.'Tlu�paper builds on enucs,upward cost pressures frrnn health care, prior publications about priority.driven budget- . pensions,and an service demands.and persistent ing by taking a step back from specific approach- es to budgeting and describing the major steps in the process and then outlining options for put- Priority-driven budgeting'is a commonsense. ring those steps into operation.It is GFO�'s strategic alternative to incremental budgeting. hope that this paper will give those who are new • Priority budgeting is both a philosophy of how to hope priority-driven budgeting a solid base from budget scarce resources and a structured, which to get started,and to provide veterans of .►lhhough flexible,step by step process for doing priority.driven hudgeting with ideas for further • so.The philosophy of priority driven budgeting adapting and sustaining priority driven budget is that resources should be allocated according to ing in their organizations. how effectively a program or service achieves the • Our Research Participants City of Savannah,Georgia (pop. 131,000) City of Walnut Creek,California (pop.64,000) Mesa County,Colorado (pop. 146,093) City of San Jose,California (pop.1,023,000) Polk County, Florida (pop.580,000) City of Lakeland, Florida (pop.94,000) • Snohomish County,Washington (pop. 683,655) • GFOA Research and Consulting/www.gfoa.org 1 • Leading the Way to Priority-Driven time,they must not he overly conunimd to any Budgeting particular budgeting technique or process.They Priority budgeting represents it fundamental must remain adaptable and able to respond to the change in the way resources are allocated.The circumstances while rcmaii-Ling true to the greater form t philos- ophy.if the organization doesift have this type of governing body and the chief executive must leadership,it might be better to delay priori I ry understand and support the process and comma- - driven budgeting or look to another budgeti'Rg • nicate that support throughout the organization. that has g In addition,these officials must Ix willing to re , support.The-Philosophy carry out their decision-making responsibilities in of Priority-Driven Budgeting'sidebar describes the philosophy of priority-driven budgeting and • a way that is consistent with a priority driven proLcss.The change an organization LICSircs to its central principles.Lse these principles to test • bringabout by virtue of implementing priority- the support among critical stakeholders and to n driven budgeting won't happen overnight,so build a common understanding of the tenets the •those leading the move to priority budgeting budget process%-All be designed around. • must make it clear that this type of budgeting is not a one-time event—it is the"new normal."To Of course,not everyone in the organL-ation can be • ,,cc the change through for the long=term,leaders expected to immediately accept priority-,driven d must have a passion for the philosophy underly- budgeting g with the same enthusiasm.The leader ,ship must articulate achy a priority-driven budget ing priority driven budgeting,but at thesame The Philosophy of Priority-Driven Budgeting The underlying philosophy of priority-drtven budgeting is about how a government entity should invest resources to meet its stated objectives. It helps us to better articulate why the services we offer exist,what price we pay for them,and,consequently,what value they offer citizens.The principles associated with this philosophy of budgeting are- • Prioritize Services.Priority-driven budgeting evaluates the relative importance of individual programs and services rather than entire departments.It Is distinguished by prioritizing the services a govem= • ment provides,one versus another. • Do the Important Things Welt.Cut Back on the Rest. In a time of revenue decline,a traditional budg- et process often attempts to continue funding all the same programs it funded last year,albeit at a reduced level(e.g.across-the-board budget cuts).Priority-driven budgeting identifies the services that • offer the highest value and continues to provide funding for them,white reducing service levels, divesting,or potentially eliminating lower value services. • Question Past Patterns of Spending.An incremental budget process doesn't seriously question the spending decisions made In years past. Priority-drivert budgeting puts all the money on the table to encourage more creative conversations about services. • • Spend Within the Organization's Means. Priority-driven budgeting starts with the revenue available to the government, rather than last year's expenditures,as the basis for decision making. • Know the True Cost of Doing Business.Focusing on the full costs of programs ensures that funding decisions are based on the true cost of providing a service. • • Provide Transparency of Community Priorities.When budget decisions are based on a welt-defined set of community priorities,the government's aims are not left open to interpretation. • Provide Transparency of Service Impact. In traditional budgets,it is often not entirely clear how funded services make a real difference in the lives of citizens.Under priority-driven budgeting,the focus Is on the results the service produces for achieving community priorities. • Demand Accountability for Results.Traditional budgets focus on accountability for staying within spending limits. Beyond this, priority-driven budgeting demands accountability for results that were the basis for a service's budget allocation. • GFOA Research and Consulting I www.gfoa.org 2 • • • 0 • • is something worth actively suppotting and voting ly sav one of the main reasons they endorse it is for,rather than just a"least-worst"outcome in a because it allows them to achieve what inspired . time of revenue scarcity.'The leadership must also them to run for office in the first place—idenrify- ercatc a sense of urgency behind priority-driven ing the results and implementing the policies that . bucketing by showing the financial forecasts, are most important to their community, analysis,and other iiformation that supports the need fora new approac h to budgeting.Ensuring Senior staff must support the process as ,N cfl • that a priority-driven budgeting process is suc- because priority-driven budgeting requires a sig- cessfully adopted requires organLation-%Hide nificant time commitment from staff.if the beard acceptance and a shared understanding of the and C:.EU are behind priority-driven budgeting,it entity's financial condition,For example,the City will go a long way toward getting senior staff of Savannah,Georgia,shared trends in major rev.' ev engaged.Staff members who have experienced enue sources,reserves,and long-term forecasts to priority driven budgeting say they support it show that the ciry:s revenues were entering a peri- because it gives them a greater degree of influ od of protracted decline.Of course,the case need ence over their own destinies.Staff no longer • not hinge on financial decline.A case can also be passively awaits judgment from the budget made based improving the value the public office;instead,they create their own solutions receives from the tax dollars government spends. because priority-driven budgeting invites them to articulate their relevance to the communal}'. Two groups in particular that must be recruited to support priority driven budgeting—elected To raise awareness about the move to priority • officials and senior staff,Elected officials need to driven budgeting and to build support for it show consensus and support for priority-driven among all stakeholders,the governments that budgeting to make it through the challenges in shared their experiences for this paper emphasa_e the budget process that will inevitably occur. the importance of a communications and risk mit- ideally,at least one or two elected officials will be igation strategy.The strategy identifies major attracted to the philosophy so they ran champion stakeholders,their potential concerns,and men- the idea with other officials.Elected officials may sages and actions that can assuage those concerns. • he particularly drawn to the fact that priority, For example,employees might vv ant to know if driven budgeting allows them to set the organi_a- their job tenure will be affected,and citizens tion's key priorities and see how services align or might want to know the implications for service don't align vv,ith their priorities.This puts elected offerings.The need for transparency in the process officials in an influential policy making role—per- cannot be e►nphasizcd enough—many organiza- haps more powerful than under a traditional tions create a specific Web page to provide system.Elected officials who have employees and citizens with regular and timely experienced priority-driven budgeting consistent updates on the process as it unfolds.Involving key stakeholders—,such its the Chamber of Commerce, • Do You Have a Strategic Plan? labor union leaders,editorial staff from the media, ind leaders of community groups and neighbor- If you already have a strategic plan that identifies hood groups—at appropriate stages in the-process community priorities,you may be able to use it as launching pad for priority-driven budgeting.Elected often provides the best form of"informal"conunu officials wilt likely be interested in a budget system nicarion to the rest of the public, In communities that promises to decisively connect resource use to such as fioulder,Colorado,and Fairfield, • their priorities.in fact,some officials might be frus- California,a town hall fonnat was used as a com trated with an incremental budget system that munication device.The first group was asked to doesn't effectively align resources with evolving P strategic priorities.This dissatisfaction with the sta invite others to subsequent meetings,and not only • tus quo provides a natural segue to priority-driven did they invite friends and family,but they budgeting. • bnuight them to the event. • • GFOA Research and Consulting/www.gfoa.org 3 • • • i w i 5 Perhaps the primary risk to successful priority- demonstrate that this hind of budgeting is consis • driven budgeting that officials and other stake. tent with best practices,but,most of all,devote holders might reject of the process bccausc they time to intensely study priority driven budgeting. see it as insufficiently legitimate-the process is Some of the research participants for this article thought to be flawed in some way that makes it a studied it for two years before moving forward. poor basis for allocating resources.Mitigate this While nvo years of study%vill not be necessary for risk by mrnferring"democratic"and substantive every government,becoming fluent in priority legitimacy onto priority,driven budgeting, driven budgeting allowx the leadership to speak democratic legitimacy means that the process is convincingly on the topic and lead an honest dis- consistent with the will of the public.Engage the cussion about the feasibility of priority,driven elected officials,the public.and employees in the budgeting for the organi_ation.If the organisation process to achieve democratic legitimacy.When decides to move forward,the leadership's expert- a budget process is seen to have democratic legit= ise wrill allow it to design a credible process,define impay,it gives elected officials permission to the roles of staff in priority driven budgeting,lead resist narrow hands of self-interest that seek to others through it,and adapt to the pitfalls and . overturn resource allocation decisions that are curveballs that will be encountered based on the greater good. The next section describes the major steps in a Substantive legitimacy means that priority driven priority-driven budgeting process and Provides • budgeting is perceived to he based on sound rech- options for answering the six questions-listed nical principles.Use Government Finance Officers below-for customizing priority driven budges- Association(GFUA)training and publications to ing to your organization. Be Adaptable Snohomish County,Washington, met with some resistance from the County Court.To move the process forward, . the county designed a separate but parallel version of prioritydriven budgeting for the courts.With time and the delivery of a consistent,transparent message,it effectively became the"new normal"in making resource allocation decisions. Designing a process that is fair,accessible,transparent,and adaptable is a challenge. However,it is also an opportunity to customize a priority-driven budgeting process that fits your organization best.This research has Identified six key customization questions you should answer as you design a process: • i. What is the scope of priority-driven budgeting?What are the fundamental objectives of your process?What funds and revenues are included?What is the desired role of non-profit and private- sector organizations in providing public services? 2. How and where will elected officials,the public,and staff be engaged in the process? Engagement is essential for democratic legitimacy.Giving stakeholders a clear understanding of • their role In the process gives them greater confidence in the process and eases the transition. 3. What is the dedslon-unit to be evaluated for alignment with the organization's strategic priorities? Functional units,work groups,programs?Something else? 4. How will support services be handled?The research participants agreed that budgeting for support services like payroll and accounting was one of the foremost challenges of designing a process. Support services need to be perceived as full participants in priority-driven budgeting,but at the • same time, accommodations must be made for the fact that they potentially exist to achieve differ- ent results than those services that have a direct impact on the public. 5. How will decision-units be scored;and who will score them?The scoring mechanism and process Is key implementing priority-driven budgeting successfully. 6. What is the role of priority-driven budgeting in the final budget decision?What method will be used to allocate resources to services?Will the methodology lead to"formula-driven"allocations or . allow for flexibility and discretion in formulated recommendations? GFOA Research and Consulting 1 www.gfoa.org 4 Steps in Priority-Driven Budgeting tal shift in its approach to budgeting.This shift, There are eight major steps in a priority-driven while subtle,requires that instead of first having budget process.Exhibit I provides a map for how the organtation identify the amount of resources the eight steps fit together,and the steps are 1'necdcd"for the next fiscal year,it should first more fully&SLribcd in the following pages.'As clearly identify the amount of resources that arc the exhibit sho-%vs,the eight steps arc not com- "aNmlable"to fund operations as well as one-time pletely linear.Steps I and 2 can begin at the same initiatives and capital expenditures. time,and Step 8 LOMCS into play at many differ, As their first step in budget development,many ent points of the process. organizations expend a great deal of effort in 1, Identify Available Resources completing the analysis of estimated expendi- G)Cfore embarking on priority-driven resourccallo- tures to identify how' much each organj=ational cation,the organization must undergo a fundamen- unit will need to spend for operations and capital Exhibit 1: Process Map for Priority-Driven Budgeting 1. Identify Available 2. Identify Your Resources Priorities,, 4 3. Define Your • Priority Resutts' More Precisely,, 4. Prepare Decision 8.Create Service Units for Efficiencies and • i Evaluation Innovation • S. Score Decision Units Against, Priority Results 6.Compare Scores Between Offers and Programs 7. Allocate Resources 4 B.create C� Accountability for Results GFOA Research and Consulting www.gfoa.org i in the upcoming;fiscal year.Once that"need"is dispels the belief that there are"secret funds" • determined,then the organization looks to the that will fix the problem and establishes the level finance department or budget office to figure out of trust necessary to be successful. • how these needs are to be funded.An integral part of the priority-driven budgeting philosophy In the first year,an organization might choose to f� is to spend within your means,so the first step in focus attention on only those areas that do not have developing a budget should be focusing on gain- true structural balance.For most organisations.this ing a clear understanding of the factors that drive n-ill often include the general find,but the jurMic,. revenues and doing the requisite analysis to don might decide to include other funds in the develop a reasonably accurate and reliable rev- process.Both Polk County,Florida,and the City of enue forecast in order to understand how much Savannah tool:steps to limit the scope of imple- is available to spend for the upcoming;fiscal year. mentation.For example,Polk County,concentrated on the general find,and Savannah excluded capital The Price of Government projects from the process. . The"price of government"is a concept originated by i David Osborne and Peter Hutchinson.'Government Step 1 Intended Result:Adopt a"spend within takes economic resources from the community to your means"approach-meaning there is a com- provide services and,hence,the total revenue that mon understanding of the amount of resources government receives is really the"price of govern available and that there is a clearly established ment,"from the perspective of the citizen.This can limit on how much can be budgeted for the be a useful concept in the first step of prkx ity-driven upcoming fiscal year. budgeting because it asks decision-makers to think about the total tax and fee burden they are willing to place on the community to fund services-thus, put-. • ting revenues before expenditures. 2. Identify Your Priorities Priority-driven budgeting is built around a set of organizational strategic priorities.These priori- Resources must also he clearly differentiated in tics are similar to a well-designed mission state- terms of ongoing revenues versus one-time ment in that they capture the fundamental pur_ • sources. Fhe organization must be able to identi- poses for which the organization gists and are fy.tnv mismatch between ongoing revenues and broad enough to have staying power from year to ongoing expenditures(operations)as well etas year.A critical departure from a mission state- between one time sources and one-time uses ment is that the priorities should be expressed in . (one-time initiatives,capital needs,fund balance terns of the results or outcomes that are of value reserves).This analysis will ensure that the enti to the public.These results should be specific ty can pinpoint the source of its structural imbal- enough to be meaningful and measurable,but not .tncc and address it in developing its bud;;et.This so specific as to say how the result or outcome will also ensure that a government does not will be achieved or become outmoded,titer a unknowingly use fund balance(a onetime short time. Below are the five priority results source)to support ongoing expenditures. determined by Mesa County,Colorado.Notice . how these results are expressed in the`V0icc of Once the amount of available resources is identi the citizen." Pied,the forecasts should he used to educate and inform all stakeholders about what is truly avail- \strategic plan,vision,and/or mission statement able to spend for the next fiscal year.The organ- can serve as the ideal starring point for identiRring =ation must understand and believe that this is the priority results.if you have an existingtrate- . truly all there is as it begins developing the budg- gic plan,it might he helpful to ground the priority et.Sharing the assumptions behind the revenue results in these previous efforts to respect the projections creates a level of transparency that investment stakeholders may have in them and to GFOA Research and Consulting/www.gfoa.org 6 • • • N o� 1 want plans and ' '1 y Infrastructure that maintain +Cd � e quality of tile. S t b 2,1, TJTJ� ZZ IVA Create well planned and developed cc munlries " yV r_r, !vnittYfieets> a ar►yt6lfie, 1 want a healthy • + � i Cry,. Mesa County J k' Pronvild Public$"a€ely. PJontore Public health m give the priorities greater legitimacy. if you don't so that the governing board must fully support have an existing plan,developing one as a prelude them.The role of an elected official is to set the • to priority driven budgeting can prm ide a results the organization is expected to achieve. stronger grounding for the priorities.It might also Developing the priorities might also be a good help increase the enthusiasm of elected officials place to involve citizens.Some communities have and senior staff for priority,driven budgeting,as used traditional means of doing this,such as citi- they seek a way to connect the new plan to deci- zen surveys,focus groups,and town hall meet- sionvs.about annual resource allocations. ings to engage citi_ens in helping establish the expected results for their community.ethers are The governing board:also needs to be closely being innovative.The City of Chesapeake, • involved in setting the priorities.The priorities Virginia,recently Asked citizens vievving a result are the foundation of priority-driven budgeting, setting exercise on their public access channel to a Are Support Services a Priority? Our research subjects offered two alternatives for prioritizing support services.Most commonly,entities created a"good governance"priority that addressees high-quality support services.This gives support services a clear place in priority-driven budgeting and allows the relevance of these services to be tested against the organiza tion's priorities. Here is how the City of Walnut Creek,California,defined its governance goals. • Enhance and facilitate accountability and innovation in all city business.. Provide superior customer service that is responsive and demystifies city processes. Provide analysis and long-range thinking that supports responsible decision making. • Proactively protect and maintain city resources," • Ensure regulatory and policy compliance. Alternatively,other participants envisioned moving to a system that would fully distribute the cost of support . services to operating programs so support services would be affected according to how the operating services they support are prioritized. • GFOA Research and Consulting/www gfoa.org 7 participate:online and share their thoughts on contribute to the safety of the Community in .what does the city exist to provide."Cities such Walnut Creek.Hence,the specific definitions of as Walnut Creek,California,and Blue:ash,Ohio, the community's results is where the identity of 5ct up kiosks in city facilities and asked citizens your community and the objective meaning of to participate in a brief survey that helped v ali, what is relevant is revealed. date the city council's established results and to "weight"the relative importance of those results , to the community. Staff.Teams in Priority-Driven Budgeting Step 2 intended Result:A set of priorities Creating strategy maps is the first significant role for • expressed in terms of measurable results that are cross-functional staff teams in the process.Such teams of value to citizens and widely agreed to be legiti- have repeated and important uses,so their members mate by elected officials,staff,and the public. need to be highly skilled and sufficiently supported.A number of our research subjects engaged consultants to train and/or directly assist the teams.Many organiza- tions use that as an opportunity to involve the"up and 3. Define Your Priority Results More coming"leaders in the process to ensure its tong-term • Precisely sustainabitity.- The foundation of any prioriti=ation effort is the results that define why an organization exists. Organi_ations must ask,"What is it that makes A powerful method for defining results was estab- us relevant to the citi=cns?"Being relevant-pro- fished in Strutgy.laps by Kaplan and Morton." vid ng those programs that achieve relevant Stratdigy mapping is a simple way to take a coin- results-is the key purpose and most profound plex and potentially ambiguous objective-like outcome of a priority-driven budgeting process, achieving a safe community-and creating a Pic i turd,or map,of how that objective can be achieved. The challenge with results is that the terns can Sometimes referred to as cause-and-effect diagrams he broad,and precisely what they mean for each or result maps,strategy maps pmvide an effective • individual community can he unclear.For way for an organization to achieve clarity about • instance,take a result like"Providing a Safe what it aims to accomplish with its results. Community,"which is shared by most local gov Strategy maps should be developed using cross- ernments.Organizations calk about public safety functional teams.Teams consist primarily of staff or providing a safe community as it it is an ohvi (both with Subject matter e%-pertise relating to the ous and Specific concept. But is it? priority result and w,ithout),but they can also include elected officials and citi=ens. In the City of 4,Valnut Creek,citizents and city leadership identified building standards fur sur- Exhibit,(on the following page)provides an giving earthquakes as an important influence on example of a strategy map from the City of proN iding a safe community.In the City of Savannah for"high,perforning government" Lakeland,Florida,however,not a single citi_en (Savannah's equivalent of the"good governance" or public official discussed earthquakes to define result dcscnbcd in the earlier sidebar).Savannah's the ven same result.In the City of Grand Island, map includes performance indicators to help i Nebraska.the city highlighted community gauge if the priority result is being achieved. acceptance and cohesiveness as intrinsic to achieving a safe community(acknowledging F-,hibit 3(on the following page)is a picture of a their initiatives to help integrate a growing and slightly different style of strategy map from the important population of their communist,- City of San Jose,California,for its"Green, immigrant farm workers). However community Sustainable City"Priority result.The center of • integration was not a relevant factor that would the map is the result,and the concepts around GFOA Research and Consulting/www.gfoa.org g Exhibit 2: High Performing Government Strategy Map from the City of Savannah Accoontabgity and Integrity • Balanced Budget • Long Range Strategic Planning • Lang Range FiacaiPlanning • Transparency • ProfanlonalBestPraa:ife- • hi fagMvernm�y Adwwq �� in3 a...• CommuMcathn 'e e•'e • Aoceabilty • Roponslveresa Prdesdonstism= Asset Planning and Management • compattttve IsasitrnarN Capital Sbalogy and In estaMnt Waddoree Developmod hdlabn: Sustatnstdtty • CreathAty and huwvation. CndtRB*V - CoNaborstion • Succesdon Manning and Ch m Sianfa anwilh CityMdeSw*e DN�y . Ettgapemerd PerPerao O*af Gaaemmsnt Enp"R"-*n Rms • CtiiaenSfarlon5urvey the result are the definitions-they help the city clearly articulate its Priorities:"When the City of Exhibit 3: Green City Strategy Map • San Joffe (fill in the bl°ank writh any of from the City of San Jose • the result definitions),then we achieve a Careen, Sustainable City.- E Consider San Jose's result map relative to your SANJOS ^ .a• �.N z am and de,19m eA• NON"Oft ad a t)\t'n community.Would your community define city o owd'ro ,.a•,w•ro,,,e,,,ya„ mdwl. the rcic:iance of your organization by its ability to uri•.•mr;mr••"terms" • .nm hanahw ichievc a.*been,sustainable community'Woulcl ; your community define a result like a green,sus. Green. Sua tauiable community in a similar cat different way' city one of the challenges local governments face is trying to address what can seem like a growing (and seemingly limitless)ea 3ectation forpro- Mak. grams and services.One of the benefits of devel- oping strategy maps is that local governments can Rive citizens a more precise description of GFOA Research and Consulting/www.gfoa.org 9 the results that make local government relevant. clear to participants that this ranking process is This gill establish a shared foundation,a com- not a budget allocation exercise(whereby the mon context for evaluating and prioritizing the budget of a certain result is determined by the programs and services the jurisdiction offers.A rotes given to a result).Through suL h a ranking, service's relative priority can be evaluated only participants can express that certain results(and , through a common belief about the results local therefore the programs that eventually influence �Tovcrnment is striving to achieve. these results)may have greater relevance to the community than others. The City of Walnut Creek knew that citizens and community stakeholders needed to be e involved in defining the priority results.The Step 3 intended Result:Reveal the identity of your community and the objective meaning of rationale was that the city's priority results what is relevant to it through the process of defin- vvould he legitimate only if community members ing priority results. i were responsible for establishing the results and their definitions.The city reached out to the community on the radio,in the newspaper,and 4. Prepare Decision Units for Evaluation through the city's newslerters and Web site to The crux of'priority driven budgeting is evaluat- invite any citizen to participate in one of several ing the services against the government's priority town hall meetings.At the meeting,citizens results.Thus,the decision unit to be evaluated were asked to submit answers to the question: must be broad enough to capture the tasks that -When the City of Walnut Creek so into producing a valued result for citizens,but then they achieve[the result the citizen was not so large as to encompass too much or be too focused on]."The response from citizens was vaguo.Conversely,if the decision unit is too tremendous and generated a host of answers. small,it may only capture certain tasks in the City government staff members(who parricipat chain that lead to a result and might overwhelm ed in the meetings)were then responsible for the budget process with details.Our research summarizing the citizen's responses by develop- subjects took one of two approaches to this ing strategy maps. issue:"offers"or"programs." Offers.Offers are customized service packages Define Your Priorities: A Quick Win prepared by departments(or perhaps designed • if the organization has not already dearly defined its by cress functional staff teams or even private priorities,just getting through this step could be a firms or non profits)to achieve one or more pri- major accomplishment. Knowing the priorities can help ority results.Offers are submitted to evaluation an organization make better resource allocation deci- sions,even in the absence of atrue priority-driven teams(typically comprising a cross functional budgeting system, group o(staff,but possibly citizens as,,vell)for . consideration against the on anization's priority . results.Often,the evaluation team will first issue Lastly,%A hen defining the priority results,consid a formal'tequcst for results"that is based on the cr whether some results might be more impor- strategy map and defines for departments,or Cant than others.This could have an impact on others who are preparing offers,precisely what how programs are valued and prioritized. Elected the evaluation team is looldrig for in an offer. • officials,staff,and/or citizens can participate in ranking exercises,where each participant is 11"5 a quantity of'ti ores"(or dollars,or point,, Now Many Offers Are There? ctc.)and can allocate their votes among all the Our research participants who used the"offer" priority results,to indicate the relative value of approach averaged one offer for every$1.5 million in one result versus another.It is important to make revenue that was available to fund offers. GFOA Research and Consulting/www.gfoa.org Ili • • • • • • a • • Offers are purposely intended to be different conceptual leap,or perhaps not a leap at all.This • from existing organizational subunits(like means less work and process risk.Ho-vvever.even departments,divisions,programs)to make a when the concept of programs is familiar,be sure • direct connection between the decision-unit the'programs"(or offers)are sized in a way that • being evaluated and the priority results,to allows for meaningful decision making. Programs encourage outside--thc box thinking about what that are too big are often too vague in their pur- • ;goes into an offer,and to make it easier for out- pose to be accountable for results,and it can be Side organizations to participate in the proces,�s. difficult to fairly judge the impact of a program • for example,multiple departments can coolner that is too small.generally speaking,if a pro • ate to propose a new and innovative offer to ,ram equates to 10 percent or more of total achieve a result instead of relying on past ways of expenditures of the funds in which it is account- • doing things.A private firm could submit an ed for,then the program.should probably he bro- • offer to compete with an offer made by govern, ken down into smaller pieces.If a program ment staff. equates to either I percent or less of total expen • ditures or$100,000 or less,it is probably too The drawback of offers is that they are a more small and should be combined with others. radical departure from past practice and may be • too great a conceptual leap for some.This could Also,be aware that using;programs might pro increase the risk to the process,but if the leader vide less opportunity for outside organizations to ship's vision is for a big break from past practice, participate in the budgeting process because the then the risk might be worth taking. for exam- Starting point is,by definition,the existing port- ple,Niesa County's board is very interested in folio of services.For that same reason,radical having private and non-profit organizations par innovation in service design or delivery method is ticipate fully in its budget process at some point less likely. in the future,so the offer approach makes sense • for Mesa County. Step 4 Intended Result: Prepare discrete decision- units that produce a clear result.Think about eval- Programs.A program is a set of related activities uating these decision units against each other and intended to produce a desired result. not necessarily about evaluating departments . Organizations that use the"program"method gainst each other. inventory the programs they otter and then con- . pare those to the priority results.Programs are an established part of the public budgeting lexi.• 5. Scare Decision Chits Against Priority con and some governments already use programs Results • in their approach to financial management,so Once the organization has identified its priority thinking in terms of programs is not much of a results and more precisely defined what those results mean,it must develop it process to objee • tivcly evaluate how the program or offer achieves or influences the priority results.Scoring can be • Program Inventory: A Quick Win approached in several ways. • if the organization does not have a sense of the pro- grams it provides, then simply developing a fully costed The first variation to consider is if a program or >� (direct plus indirect costs)program inventory should offer tivill he scored against all the organization's provide immediate benefits.A program inventory can be used to help decision-makers understand the full priority results or just the one it is most closely breadth of services provided and their costs,and might associated with.The cities of Lakeland,Walnut help the organization recognize immediate opportuni• Creels,and San Jose scored against all of the prior- • ties for efficiency.Appendix t provides additional irnfor- ity results.The belief N as that a program that mation on how to build a program inventory. influenced multiple results must he a higher prior • • • • GFQA Research and Consulting/www.gfoa.org i 4. • • • • • • • • ity—every tea dollar spent on a program that if there was an anticipatccl increase in demand • achieved multiple results wits giving the taxpayer for the program or that program received fees or the"best bang for the buck."Alternatively,organi grant dollars to significantly cover the costs to =ations like Mesa County,the City of Savannah, provide it.Finally,if the citizen had to rely solely • Poll:County,and Snohomish County matched on the government to provide the program or • each pro„Tram or offer with only one of the priority service and there was no ether outside option results and evaluated it against its degree of influ- available,then a program vas believed to l+e of a once on that result.Under this scenario,guidelines slightly higher priority. should be established to help determine how to assign a program or offer to a priority area as well The nett variation is how to actually assign • as provide some sett of aecommoclation for those scores to programs or offers,one approach is to programs or offers that demonstrate important have owners of the programs or offers(e.g., effects across priority result areas.Both of these department staff)assign scores based on a self. • approaches have been used successfully,so the assessment process.This approach engages the . right choice depend-,on which approach resonates owners in the process and taps into their unique more with stakeholders, understanding of how the programs influence the priority results.Critical to this approach is a In addition to scoring the offers or programs quality control process that allows the owner's • against the priority results,some organizations peers in the organi=ation(other departments) have included additional factors in the scoring and/or external stakeholders(citizens,elected process.Examples include mandates to provide officials,labor unions,business leaders,ctc.)to the service,change in demand for the service, review the scoring.The peer reticw group chal level of cost recovery for the service,and reliance lenges the owner to provide evidence to support on the local government to provide the service the scores assigned.A second approach to scor • (as opposed to community groups or the private ing establishes evaluation teams that are respon- sector).The governments believed that a pro Bible for scoring the programs or offers against gram should he evaluated more highly if there their ability to influence the priority results. • was a mandate from another level of government, owners submit their programs or offers for the • What about Capital Projects? • For most organizations,outlays for capital projects and one-time initiatives are a significant part of their budget- ing process.A priority-driven budgeting process can be used to prioritize these major one-time expenditures in the same way it is used to evaluate ongoing programs and services.The starting point is a capital improvement plan(CIP)that includes ail the potential capital projects from across the organization. Ideally,it should include not only major capital construction,capital improvement,or capital equipment purchases,but also significant one-time expenditures items such as major studies,comprehensive plan updates,and software upgrades that are . planned for the next five years. In addition to the strategic results,other evaluation factors for capital projects might include: • Is the project mandated by some other governmental agency? • Is it a continuation of an existing project that has already been approved? • Is it an integral component of the organizations Comprehensive Plan for future community growth? • Is it being fully or partially funded by another agency or private interest? • is the project responding to an emergency situation or critical need of the organization? When evaluated in this way, projects that are of a higher priority have assurance of funding in the next five-to- ten year period over those that are of a lower priority, especially when there are limited one-time resources available to fund them.This method also avoids funding a current-year project that is of a low priority instead of • setting aside funds to ensure the successful completion of the higher-priority capital need in a future year. • • • GFOA Research and consulting/www.gfoa.org i2 • • • • s teams to review,and the teams score the pro- their belief in the relatix a influence that program • _grams against the results.The priority driven or offer has on the priority results it hits been budgeting process becomes more like a formal evaluated against. In other organizations,the • purchasing process,where the departments are process can be implemented as a staff-only tool ro ll analog ous to vendors and the evaluation teams that is used to develop a recommendation to the �I arc like buyers.Evaluation teams could be made governing body.Snohomish County uses this up entirely of staff,with representation both approach,as its culture and board-staff relation from staff members who have specific expertise supports it. related to the result being evaluated and others who are outside of that particular discipline.An Regardless of which variations are selected,there alternative team composition would include both are three important points to establish.The first staff and citizens,to gain the unique perspectives is that to maintain the ohjectivity and trans• • of both external and internal stakeholders.This parency of the process,programs or offers must second approach brings more perspectives into he evaluated against the priority results,as they the initial scoring and encourages cross function were defined collectively by stakeholders(see A teamwork via the evaluation teams. step 3).Secondly,scores must be based on the demonstrated and measurable influence the pro- grams or offers have on the results.Finally,the Scoria Support Services � PP results of the scoring process will be provided as As mentioned earlier,a number of our research sub- recommendations to the elected officials,who • jects established a priority result for"good gover- hold the final authority to make resource alloca nance."Those programs that provided internal services were scored against these governance results in a par- Met evaluation process.These governments believed that internal services were important, but were expect- ed to achieve different results than those programs or step 5 Intended Result: Each decision unit(offer offers intended for citizens. or program)should have a score that indicates its relevance to the stated priorities. P knother consideration is the particular scoring b, Compare Scares Between Offers or • method to be used.For example,voll evaluators Programs have to use a forced-ranking system where pro- It is a"moment of truth"in priority-driven budg- gramsioffcrs are fit into a top-to-bottom ranking etinr,when the scoring for the offers or programs • or will each program he scored on its own merits, is compiled,revealing the top-to-bottom cornpar- a ith prioritization as a natural byproduct?Each ison of prioritized offers or programs.Knowing system has its advantages,but the important this,an organization must be sure that it has • thing is to make sure the scoring rules are clear clone everything possible up to this moment to to ever gone and applied consistently, ensure that the final scores aren't a surprise and Mthat the final comparison of the offers or pro The role of the elected governing board in this grams in priority order is logical and intuitive,. step is another point of potential variation in the aroring. In some organizations,the board is heav, The City-of San Jose engineered a peer review ily integrated into the process and participates in process through which the scores the depart the scoring and evaluation step.They have the meats gave to their programs were evaluated, . opportunity to question the.scores that have discussed,questioned,and sometimes recom been assigned by the owner or the evaluation mended for change.The city established a reN iew team,ask for the etidence that supports that team for each of its priority results.The team score,and ultimately request that a score be first reviewed the strategy map to ensure that changed based on the evidence presented and each member of the team was grounded in the GFBA Research and Consulting/www.gfoa.org 13 I � I, • 0 m i city's specific definition of the result. Next,the ask them to decide\N°hich programs should be review teams were given a report that detailed cut or which ones should be preserved.They every pre?gram scored for the particular result framed the discussion eery simply:Evaluate how under review.The teams met to discuss: our programs help us achieve our results,and to what degree.The outcome of prioritization was • whether they understood the programs they therefore expected and self evident,based on the i were reviewing; common understanding of the programs a nd how • whether they agreed with the,core given by the programs influence results. the department(the departments scored their o«m programs); Stakeholders could be concerned that their • whether they required further testimony or favored programs might lose support in the evidence from the department to help them course of priority-driven budgeting. Even when a better understand the score given,and program director or a citizen who benefits from a • • whether the score should stand,or if the particular program understands why that pro- team would recommend an increase or gram ranked low,they are not going to be decrease. pleased about it.invite stakeholders from all aides,from within the organization and even the All programs were evaluated in this manner until community,to understand the process.Include a final recommendation was made on program stakeholders at various points in the process su scores. they might influence the outcome.Constantly communicate progress,throughout the process. The city invited the local business community. Program directors,stakeholders of a particular cin=ens representing their lo-cal nci�hborhood program,organizational leadership,and staff commissions.and labor leaders to review the might not enjoy seeing their program prioritized San Jose framed the discussion very simple: Evaluate how our programs help us achieve our results, and to what degree. scores.Walter Rossman,from San Jose's City bcloxv other programs,but if they understand it, Managers Office,described their effort this 4vay: if they've had a chance to influence the process, • "The participants found the effort informative as ,end,most importantly,if they are aware of to what the city does;they found it engaging actions they night take to improve the priority N%ith respect tc hearing staff in the organization ranking of their program,the process\\-ill have a • discuss how their programs influence the city s great chance for success. results;and,most interesting they found it fun" Lastly,consider if the scoring of the programs or • San jose's stop.is important because it demon offers dill he used only to decide where to make strated how stakeholders from various perspec- budget reductions.Organizations such as the • tikes.and political persuasions can all productive cities of Lileeland and Walnut Crecl:have used ly}participate in the priority driven budgeting hrioriti=anon not only to balance their budgets, process.San Jose didn't ask these stakeholders to but also to understand how services that might dame together and rink programs.They didn't Appear less relevtnt to the city government might GFOA Research and Consulting/www.gfoa.org 14 . • I • m be relevant to other community,;roups,These instance,the designer should think about ways roues might take responsibility for supporting priority result areas can share information during or preserving a service_There could be great the evaluation of programs or offers,andlor ways potential in engaging other community institu- to jointly fund programs or offers. tions—businesses,schools,churches,non-profits —about partnership opportunities. Exhibit 4: Drawing.the Line Peter Block has focused much attention on this issue in his book,Community:The Stnicture of Gdnr>Ging.=Citing the way we sometimes unduly g rely on government to meet the community's needs,he highlights citi=ens'experiences of tak- ing accountability for the results they here to see achieved.This occurs when cohesion is built between local government,businesses,schools, social service organ=ations,and churches.A complete and successful priority-driven budget ing process doesn't conclude when the budgets for low-priority services are reduced—rather,it brings together otherwise fragmented institu- tions in society to find ways of providing services • that may still be relevant to the community,even Then,the offers or programs can be ordered if they are less important to the priority results a according to their prioritization within a given local government seeks to achieve priority result area and the budget staff draw a line where the cost of the most highly prioritized offers,or prop ams is equal to the amount of rev, • Step 6 Intended Result:The prioritized ranking of programs is a logical and welt-understood product enue available(see Exhibit 4).The offers or pro.- of a transparent process-no surprises. grams above the line are funded,and the ones that fall below the line are not.The hoard and • staff will have discussions about the programs on 7. Allocate Resources either side of the line and about moving those Once the scoring is in place,resources can be alto. offers or programs up or down,redesigning them rated to the offers or programs.This can be done to make more space above the line(eq.;lowering in a number of ways.one method is to first allo- service levels),or even shifting resources among • rate revenues to each priority result area based on priority results.Variations on the approach are historical patterns or by using the priority's rela- possible—for example,there could be multiple rive weights,if weights were assigned.Allocating lines representing multiple levels of funding set f, resources to a priority result area can be contro tainty. In the City of Redmond.\Vashington, versial because,as ,vc will see,this allocation programs above a top line were categori=ed as deterinines the number of offers or programs that "definitely fund;'while programs in between the N,,ill be funded under that priority area(e.g.,how top line and a bottom line were open to addition many public safety programs\\ill be funded). al scrutiny. There are no easy answers to this issue=As such, ithe designer of the process should look for ways ;Another method is to organize the offers or pro- to mitigate controversies associated with now ;rams into tiers of priority(e.g.,cluartiles)and much funding is allocated to one result versus then allocate reductions by tier.For example,pro. ' another and to prevent these allocations from grams in the first tier might not be reduced,while • becoming new types of organizational silos.For program in the lowest tier would see the largest CFOA Research and Consulting/www.gfoa.org 13 • • reductions.The programs could be forced to reduction applied to any particular support sen,- mace assigned reductions,or each department ice was based on its priority relative to other could he given an aggregate total reduction tanget, support services.A couple of our participants based on the programs under its pur0evv(with envisioned moving to a system wherein the cast the implication being that the department will of support services would be fully distributed to weight its reductions toward the lower-priority operating programs so support services would be programs,although it would have more flexibility affected according to the prioritization of the to decide the precise reduction approach than if operating services they support, the cuts were not done within the department). This tier approach generates discussion among Another question is how to handle restricted board and staff about hover much money is spent monies(e.g.,an enterprise tend).One option is on higher versus lower tier services in aggregate, to handle special purpose funds(where there are as well as on resource allocation strategies for restrictions on hove the money can be used),sep- individual departments and program,.Exhibit 5 arately. For example,enterprise funds or court presents an example of the value this analysis can funds might be evaluated on a different track or provide.It shoti.s the total amount of money one budgeted in a different way altogether.Another40 city had historically spent on its highest priority option is to rank programs or offers without programs(e.g.,the top tier)versus the others. respect to funding source,but then allocate This city was Spending significantly less on the resources with respect to funding source. top tier than it was spending on the second tier, Knowing the relative priority of all the offers or and less than it was spending on the third tier,as programs might generate valuable discussion, well.This ruses interesting questions about even if there is no immediate impact on funding. Spending patterns in the organization and huilds For example,if a love ranking offeror program is a compelling case for change, grant funded,is it still worth providing,especial- ly if that grant expires in the foreseeable future' Organizations also need to consider the funding Ideally,participants will become less fixated on of support services,Many of our research partici- funding sources,realizing that the government pants elected to fund support senices based on has more flexibility service it might think. For historical costs,making some reduction that was example,if a love,-priority se �ce is funded by a consistent with the reduction the rest of the special earmarked tax,is there a way to reduce or • organization vvas making.The magnitude of the eliminate that service and its tax,and increase a • Exhibit 5: Spending by Priority Tier z 3 '36 • 4 u i S(rtxlstons)_._ .___ St4 __._..._ .__A. $24.d, __.�_S30_ ___. ��._ _�_ $50 ._.._$60 • • GFOA Research and Consulting/www.gfoa.org tb • • ,cneral tax by an analogous amount'As the gov- grams or offers,and innovation in the design of • crnment becomes more proficient at expressing programs or offers.Although priority,driven the value it is creating for the community,it budgeting will identify which programs or offers bhould be better able to articulate these potential are best for achieving priority results,it does not • tradeoffs to the community. speak directly to the efficiency with which those programs or offers are delivered or to innovative • Of course,no matter what method is selected to approaches to program delivery(although it • allocate resources,remember that priority-driven miglu indirectly encourage these things). budgeting,like any budgeting process,is still a political process.:�s Such,it will not and should • not lead to scientific"or"apolitical"allocation of Exhibit 6: Polk County Concept for resources—rather,it should change the tone of Key Performance:Indicators • budget discussions,from a focus on how money was spent last year to a focus on how the most value can be created for the public using the priority; • money that is available this year. People in Polk County who are at risk because of • their health or economic status will get their bask needs met,and are as self-sufficient as possible. • Step 7 Intended Result:Align resource allocation Indicators: consistent with the results of priority-driven scor- ftwoving • mg . •,. ` Maintaining • 8. Create Accountability for Results, improving • Efficiency, and innovation Droving The owners of the programs or offers being evalu • ated might overpromise or overrepresent what • they can do to accomplish the priority result.To :1s such,the designers of the process might need • address this potential moral hazard,create meth- to consider specific techniques for ensuring pro- rids for nuking sure programs or offers deliver the ;ram efficiency.A proven model for improving iresults they were evaluated on.luny of our etficiency helps avoid cost-cutting techniques • research participants anticipate using perform- that also cut productivity and degrade the results ance measures for this purpose.For example;a a program produces.For instance,a systematic • program or offer might have to propose a standard method for reviewing and improving business of evidence or a metric to be evaluated against,so processes could be implemented along with pri the organization can see if the desired result is ority-driven budgeting.One such method that • being provided.Exhibit 6 is Polk County's con- GFOA research has shown to be effective for local ccptual approach for connecting its priority result governments is"Lean"process review—a system • areas to key performance indicators.However, for identifying and removing or reducing the non • none of the research participants have reached value adcled Nvork that can be found in virtuaUv what they would consider a completely satisfacto- any business process.YOUcan learn more about • L1,state in this area. For those just starting out;the Lcan at .v%Nrw.gfoaconsulting.org/lean, . lesson is to understand where evidence is needed in your process design,but also to he patient with Business process improvement can also he incor. • respect to when this part of priority,driven budg ivrated into a more comprehensive approach to eting mil I be fully realized. reviewing program etficiency.Exhibit i(on the following page)provides a sample program • Other issues to consider as part of the priority- review decision tree that is inspired by work • driven budgeting design are the etficiency of pro from the City of Toronto,Ontario. .As the exhibit • CFOA Research and Consulting l www.gfoa.org 17 • • • bhows,a program is subjected to a.series of tests process should consider how to encourage new • ro see if it is being provided efficiently. For exam- ways of structuring programs or offers to best • ple,can the service be shared with other govern achieve the government's priority results.Some • ments?Can greater cost rerovcry be achieved research argues that innovation is a"discipline. through fees or fund raising?Can the private sec- just like strategy,planning,or budgeting."" • for provide the service more efficiently?Can Lean Public managers who want to encourage innova • process improvement techniques be applied? tion will need to develop and institutionali= Exhibit 7 also shows how the review might be dedicated processes to generate ideas,select the • linked tit priority,driven budgeting—discre- best ones,implement diem,and spread the bone tionary services are subject to a relevance test fits throughout the organization.Along the way, that asks the above questions about each priority public managers\mill need to make use of a vari. • program,while non,priority programs gocry of implementation strategies,including those through a divestment test. that rely on the organi_arion's own resources and those that seek to harness resources from out • Finally,innovation tends to be the exception side.Public managers will also have to create an • rather than the rule in the public sector,so the organi`ational culture that is not just conducive designers of the priority driven budgeting to innovation,but actively encourages and even • Exhibit 7: Sample Program Review Decision Tree What$oral of aenrks is • required?wnat do we Y Is the program provide? maridaW No • GFOA Sample Dgam—y SamaraP. Should tA program • be Program to Ure die program a delivered eriotlier Review Tree °O0"'""""� omani at on? Yes No MaMuary Samaa Yea Does pOrartYTaMIt hsv0POMW a Ol Identify organization d to be a dared provider? terns of transfer Do Po�wbskk ab dne acceptable levelsI Fes►Regulator? t .. Are there&tango In demand? Assess impact It Is there wAibWWU to consider$awe abandon program • service tevats? Can ft program be more&M. wAro rat?User fees? • Sporisorships? ? QMQUISIWILIM Community Co4liroduction Test Can die process be Does dis program meat Ow criteria far Do • redesipried fe remove or outsourcing:Taste tern be owe with other r go for altering reduce rio n-value-added advance?a s in service with other ems.or u9 s, - sappointlrq aorwaaors paMerenp vnth NG�,or usirnp work? can be replaced?Government Is ciftm voW"M? ooncemed with erida over raarie? - Procrem Imorovement Plen - •Matysis d curer&situation •Anatysie o}options - •Recommendalicn - • - CFOA Research and Consulting/www.gfoa.org is • - - p • demands it.The Public Innovator's Plu400k driven buct,eting,outlined in the eight steps pre. • describes one approach to encouraging innovaw gented in this paper.A prionty-driven budgeting tion in this kind of systematic way:" process can be approached in several ways,so keep in mind the major levers and decision • Step 8 Intended Result:Make sure that those who points to create a process that works best for received allocations are held accountable for pro- your culture and environment,and that embraces • ducing the results that were promised. Find ways the concepts of democratic and substantive legit- to directly encourage efficiency and innovation. imacy.The governments that participated in this research show that there are opportunities to • introduce flexibility in the process-but keep in Conclusion mind that with that flexibility comes risk,if • Priority-.driven budgeting represents a major changes are made that don't embrace the basic • shift from traditional budgeting methods.A clear principles of priority-driven budgeting. understanding of the priority driven budgeting • philosophy should be in place before proceeding Research what other organi_ations have done • down this path,along with a strong level of sup and ask them about their long-term success in • port-especially from the CEO(whose role is shifting to the"new normal"in local government normally to propose the budget)and,ideally,the budgeting.Understand that priority driven • ;owerning board(whose role is to adopt the budgeting is a process that will evolve and hudget).Priority-driven budgeting is not a improve over time-don't expect perfection in process that is brought in to fix a structural the first year.Engage outside help where needed • deficit,instead,it becomes the way an organza to design the process,develop successful commu rion approaches the resource allocation process. nication plans,incorporate citizen involvement, • It brings with it an important cultural shift- and institute a process.Enjoy new conversations • moving from a focus on spending to a focus on that were not possible before,and embrace the achieving results through the budget process. transparency in decision making that accompa.. i• Priority-driven budgeting should be perceived by nics the priority-driven budgeting process.As . all stakeholders as a process that improves deci- your organisation adapts to the new,normal,the Sion-making and changes the conversations process will guide decision-makers in making • around what the organization does(programs resource allocations that fund the programs that • and sen1ces),how effective it is in accomplishing; .are most highly valued by the organization and, its priority results,and how focused it is on allo- more importantly,by the citi_ens who depend on • eating resources to achieve its results. those programs and services for their Nwcll being, The success of your process design rests on a comfort,and expected quality of life. clear understanding of the principles of priority- CFOA Research and Consulting!www.gfoa.org 19 • -7 Appendix 1: understand and can be digested by the Building a Program Inventory average citizen(i.e.,not tot.)detailed). • * Drawing distinctions between the results ,'ritroduction (that matter to citizens)provided by dif Financial constraints have forced many govern- ferent programs.To achieve this,programs • ments to take a hard look at the services they cannot to be too large or vague. offer.A fundamental step is to inventory all the 9 lkginning to show the true Lost of doing service programs a government Offers.A program business does on a by describing what government • r inventory clarifies the breadth of services provid- meaningful level,and then iden- cd and,ideally,highlights key characteristics of tify�ing costs for those programs. • a Laying the groundwork for priority-driven g each program the full cost of providing the program and the level of revenues that program budgeting,where programs receive budget directly generates to support its operations).The allocations based on their contributions to • inventory provides the basis for discussion about the government's priority objectives. • the services that should be provided. * Laying the groundwork for program review,where programs are subjected to • Steps to Take efficiency tests to determine if the service 1. Define your objectives and goats for the pro- delivery method employed is optimal. gram inventory. Identifying a program is as • much art is it is Science—an inescapable 2. Decide what information the program inven- amount of subjectivity is involved.Therefore, tory should contain, in addition to the basic • to make judgments as effectively as possible, description of the program.Options to con • - make sure you are clear on why you are dcvcl- finder include: ' , a Full cost.The full cost of the program is its • oping a program inventory.Some of the direct cost plus its indirect cost(overhead potential purposes are: charges). Full-cost accounting makes the • Understanding the complete scope Of true cost of offering ascrvice transparent, • services government provides. which allows better planning and decision • Communicating the scope Of services to making.It also helps show that the organ- the public in a format that is easy to i=ation is achieving the expecrLd lc%rl of • Program Costing Tips Precise costs for each program might not be achievable without a great deal of work(or a new financial man • - agement system). For purposes of priority-driven budgeting,accessible and widely used cost allocation • methodologies allow for relatively accurate costing of each program Is possible. If you have a format cost allo- cation plan,this would be the best place to start assigning program costs. Otherwise,start with direct costs. Remove any one-time costs(e.g.,capital)to make sure you are capturing only ongoing expenditures related to a given program. However,you can assign the operating and maintenance costs of the assets employed by a • program to the direct costs,if doing so is logical and consistent with the way these costs are being handled for other programs. • Cost allocation plans may be the most cost effective way to produce a reliable overhead allocation figure. In the inventory document,displaying the overhead costs separately from the direct costs can provide flexibility to those who use the information. In making the transition from department or division budgets to program costs,use an allocation method that is intuitive and therefore would enjoy legitimacy among the users of the costing system(e.g.,the number of FrEs or percentage of employee time devoted to a program).Whatever the allocation methodology,the • finance or budget staff needs to be able to prepare a reconciliation. • • • GFOA Research and Consulting/www.gfoa.org 20 • • • • • • • cost recovery for a given service.Full cost- with verbs—programs are action oriented. • ing is especially important if the govern- For example,programs in a sheriff's office merit envisions eventually going to a prior- might include crime investigations,cleten • ity-driven budget process. tions,and court security.However,programs • • Alignment with strategic goals.Knowing should not be described in terms of overly how programs contribute to priority goals detailed tasks.For instance,"supplying a • enables organizations to develop more bailiff for court rooms"is a task within the • strategic cutback strategies. court security program,not a program itself. • Service level. Describe the level of services • provided to the public.if service is being 5. Find the right level of detail.a program is a • provided at a premium level,perhaps serv, set of related activities intended to produce a ice levels can he lowered to reduce costs. desired result,When constructing a program • • X landate review.List and clearly define inventory,it can sometimes he challenging to any mandates a program is subject to. find the right level of detail.If a program is • Then review the current service level too big or encompasses too much,it will not • ainst the mandate requirements. provide sufficient information—that is,it will Perhaps the service level being provided is he very difficult to describe the precise value • higher than what the.mandate requires. the program creates for the public or to use • • Demand changes.Is demand for a service program cost information in decision malting. going up or down? If demand is going However,if program definitions are too small, • clown,perhaps the program can be cut decision makers can become overwhelmed back and resources shifted elsewhere.if with detail and be unable to see the big pic. demand is going up,steps can be taken to ture.In addition,tracking program costs for • manage demand.For example,perhaps very small programs is generally not cost- • means testing can be applied to a social effective. services program. • • Support from program revenues. Describe generally speaking,if a program equates to l0 the extent to which the program is sup. percent or more of the total expenditures of • ported by its own user fees,grants,or the fund in which it is accounted for,then the • intergovernmental revenues.is there an program should probably be broken down opportunity to achieve greater coverage of into smaller pieces.And if a program equates • the full costs of the program? nn 1 percent or less of total expenditures,or to • Si00,000 or less,it is probably too Small and 3. Develop forms and templates.Create tools should be combined with others.This is just a • departments can use to describe their pro- guideline—there could be valid reasons for • grams in a manner that is consistent and that going outside of these parameters.For exam- captures the information needed to fulfill the ple,a small program could he much more purpose of the inventory.Consider testing the important than its cost suggests.Here arc • forms and templates with one or two depart- some other points that have proven helpful in meats and then distributing them to a%eider identifying programs: • group.Also consider providing training and . .In official point of contact for questions. • A program is a group of people working together to deliver a discrtte service to • 4. Differentiate programs from functions. identifiable users. • Departments or divisions(i.e.,public health, • A program groups a[]tasks that a cis- courts,public works,sheriff)are often tomer of that program would receive and • described as functions or nouns.These are does not break one program or service into . not programs,which are more otters described multiple items based on tasks. • • • GFOA Research and Consulting J www.gfoa.org 24 • • • As far as possible.a program is indi-%idual are combined with other e.xisting programs. -a prograin with its own name,cusr . A program uses an e.,isting name that is tamers,and staff team.Each pre,ram familiar to customers and staff,and;or it • stands alone and is distinct from like pro- uses a name that could stand on its own ;;rams in a similar serVicc area. and would be understandable to the aver- • e Programs that are handled by less than I FIE age reader. • Examples of Program Inventories • Sample Health and Environment Programs Sample Sherrtff Programs Environmental Planning Traffic Air Quality Control Patrol Precincts Water Quality Emergency Management Ambulance Licensing Transportation EIP FoodNet Court Security Compliance&Community Safety Work Release . Vital Statistics Inmate Food/Medical Service Immunization Grant Civil/Fugitive/Warrants • Emergency Preparedness Response Records Non-grant Immunization, Dispatch(Communicatkm Center) S Sexually Transmitted Disease(STD) Academy Food Protection Executive • Cities Readiness Initiative Directed Operations(DOU) Zoonosis Critical Incident Response • Canter Control Initiative Radio Maintenance Communicable Disease Grams Coordinator • Early Periodic Screening, Diagnosis and Treatment West Metro Drug Task Force Radon Crimes Against Children • Health Care Program for Children with Special Needs Crimes Against Persons • Women,Infants,and Children Victim Services Special Needs Nutrition Services Training and Recruiting • Family Planning Patrol Administration Recreation Criminalistics • Maternal&Child Health Block Grant Detentions Administration Prenatal Plus Crimes Against Property • Housing&Institutions Special Investigations Adult Substance Abuse Counseling Support • Fetal Alcohol Syndrome Laundry/Custodial Youth Substance Abuse Counseling Inmate Worker Program HIV Counseling&Testing School Resource Officers(SROs) Nurse Home Visitor Operations/Booking • Specialized Women's Services Animal Control Tobacco Cessation Inmate Welfare . Nutrition Services Evidence Adult Health Accreditation • Home Visit/Maternity Crime Analysis International Travel Clinic Investigations Administration • Heart Wise Grant Professional Standards Health Education Internal Affairs • Healthy Wheat Ridge Staff Inspection Public Health Communications Volunteer Programs . Home Visit/Children Community Relations • • • GFOA Research and Consulting/www.gfoa.org 22, • • • • • • • • • Notes University Press,1997). • I The concept of incremental budgeting was 7 Diagram inspired by Eva Elmer and • developed by Aaron Wildavlsky.See,for Christopher Morrill."Budgeting for example:Aaron Wildaysky,The Politics o(rhe Outcorncs in Savannah"Governrnerrt Finance Budgcrary Proccss(Boston: Little,Brown,1964). Rc%*—o,,April 2010. • 2 Robert Behn discusses the shortcomings of S Budgeting for outcomes was the subject of incremental budgeting in a cutback cnviron. The Price Of Government:Getting the Rcsults IVe • ment in the following article:Robert D.Bchn, 14ced in an Agc of Permanent Fiscal Crisis by DaAd "Cutback Budgeting,"Journal of PolicvAnal-sis Osborne and Peter Hutchinson(Newyork: • and,'4lan(gemenr,Vol.4.No.2(Winter,1985). Basic Books,2004). • 3 Priority-driven budgeting is also known as 9 Robert S.Kaplan and David P. Norton, "budgeting for results"and"budgeting for Srratcq Maps:Converrinq Intangiblc Asscrs into outcomes"although the latter is used to Tar{gible Outcomes(Boston:Harvard Business describe a specific method of priority-driven Press,2004). • budgeting. to Peter Block,Comrrrunity:The Structure of • 4 Personal interviews were conducted with the Belonging(San Francisco:Berrett-Koehler managers who led priority driven budgeting Publishers,2008). at these entities. tl William D.Eggers and Shalabh Kumar SmI h, • 5 Behn. The Public Innovator's Playbook:Nw rurinq Bold • 6 Mark Moore emphasi_es that these two Ideas in Government(New York:Deloitte, sources of legitimacy are essential to making 2009). • any big public policy change.lklark Iv3oorc, 12 Eggers and Singh. • Crcatiq Public Value(Boston:Harvard • • • • • • • • • I, • I! • • • GFOA Research and Consulting/www.gfoa.org 73 • • • • What All Mayors Would like • r ® to Know About Baltimore's • CitiStat Performance Strate • gy • t � S �II C� r E - b . r L bfJ • cC� . s • i • • • 2007 MANAGING FOR PERFORMANCE AND RESULTS SERIES • • • • • • • What All Mayors Would like • to Know About Baltimore s • CitiStat Performance Strate • 9 • • i Robert D. Behn Lecturer • John F. Kennedy School of Government Harvard University • • • • • • • • • • • • • • • • IBM Center for The Business _ of Government • • � TABLE OF CONTENTS ~~ � The Commitment ..........................................................................9 How Does a City Get Started?...........................................................ll The Beginning -------------------------11 � |nhiu| Progress -------------------------12 What Measures and Data Does aCity Need?....................................14 The Measures -------------------------'l4 The Data ----------------------------l5 Who Has W8xa 0t�e�------------------'l8 ' The Mayor -------' l8 ' -------------------' The [idStatStaff ------------------------20 The Agency Director and Managers ............................................2] What Kind of Infrastructure Does aCity Need? ................................23 The Room ---------------------------'2S The Technology ------------------------..26 The Budget ---------------------------27 What bthe Purpose, Operation, and Impact mf the Meetings?.........29 The Participants in the Meeting ...................................................29 The Conduct of the Meetings ......................................................30 The on for the Meetings ................................................32 �~ How Does CitGtat Affect Key Relationships?..................................... 4 City Council Members ................................................................ 4 �mp Agency Directors and Managers .................................................}4 Aaencies ----------------------------]S Unions -----------------------------]5 City Employees -------------------------]5 Citizens ------------------ ]6 ----------' journalists ...................................................................................]6 OtherStakeholders .....................................................................37 Cover photo:Seated at the center o/the table opposite the podium,former Mayor Martin O'Malley listens tva presentation ata Baltimore CitiStatmecting- 2 Photo courtesy of the City of Baltimore. | ��r What Did Baltimore Accomplish?......................................................38 Results ........................................................................................38 • Cause and Effect .........................................................................43 What Is the Future of CitiStat? ..........................................................45 Uniqueness ................................................................................45 Evolution ....................................................................................45 . Sustainability ..............................................................................46 Endnotes............................................................................................48 About the Author...............................................................................52 Key Contact Information....................................................................53 3 WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • • FOREWORD • • On behalf of the IBM Center for The Business of Government, we are pleased to present this report, "What All Mayors Would Like to Know About Baltimore's CitiStat Performance Strategy," by Robert D. Behn. • This report continues the Center's long interest in Baltimore's CitiStat program, an exemplar of managing for performance in government. In • 2003, the IBM Center for The Business of Government published a case . study of the program by Lenneal Henderson, "The Baltimore CitiStat Program: Performance and Accountability." In 2004, Baltimore's CitiStat i was selected as an Innovation in American Government Award winner. • I= As the CitiStat program received increased attention, Baltimore became Albert Morales a frequent destination for mayors from across the United States and the • world to visit so they could learn how the program worked. Based on these visits and ongoing discussions with CitiStat staff, Dr. Behn prepared '_ this report to summarize and present the questions most frequentlyI posed to CitiStat staff and to Mayor Martin O'Malley. In January 2007, Mayor O'Malley was sworn in as governor of Maryland and quickly ' began implementing a StateStat program.The new mayor of Baltimore, • Sheila Dixon, has continued the CitiStat program. o` While most visiting mayors were impressed with CitiStat and aspired to replicate the program, many were not sure how to proceed in bringing CitiStat to their city.This report explains how CitiStat should be viewed • as a leadership strategy rather than a management system.When viewed Michael C. Powell . as a leadership strategy, Dr. Behn argues, the program can be repli- • cated and customized to each mayor's individual needs and priorities. 4 I IBM Center for The Business of Government • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • • • • A key insight from this report is that there is no single, right approach • as to how to develop a successful management performance and accountability structure. Success depends heavily on clear goals, committed leadership, and persistent follow-up. As Dr. Behn says, • " ... those who would design a CitiStat for [their] city need to start • with their purpose." We believe this report offers a clear road map for other mayors and government leaders who may be interested in pursuing the CitiStat strategy. The lessons in this report are also clearly relevant to other government organizations at the federal, state, and local level.A recent • IBM Center for The Business of Government report, "The Philadelphia SchoolStat Model," shows how the strategy has been adapted to an • urban school district.The CitiStat strategy has been adapted for use in state government as well. Albert Morales Michael C. Powell Managing Partner Senior Consultant, State and Local Strategy . IBM Center for The Business of Government IBM Global Business Services • albert.morales@us.ibm.com mcpowell@us.ibm.com www.businessofgovernment.org 5 WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • • • • LETTER TO CITY MAYORS � Dear Mayor: So you've heard about CitiStat. Perhaps you saw Baltimore's former mayor, Martin O'Malley, extol its virtues.' Perhaps you heard Baltimore's current mayor, Sheila Dixon, explain how she is adapting it to her administration's priorities. Perhaps you learned about it at a conference, or by reading about it in a maga- zinez or report3 or online.'And you've heard that CitiStat can improve the performance of city agencies— • that it can motivate a city's managers and employees to produce the kind of results that citizens value. • Now you want your own CitiStat. But what exactly does this mean?What does it entail?What do you need to do?What do you need to do first?What do you need to worry about?What are the potential pitfalls?What can you expect to accom- plish? Should you really bother? I'm sure that you have lots of questions. . This report is designed to answer all of the questions that you have—plus some that may never have occurred to you. These answers will help you get started, recruit an effective staff, create the necessary (though minimal) infrastructure, achieve some successes, learn from these successes (and the inevitable failures too), and create a durable (if evolving) performance strategy that can motivate a city's managers and • employees to produce the kind of results that you promised during your election campaign. As with anything else you try to do in your city, the details matter—they matter a lot. And these details • inevitably depend on other particulars—everything from what your citizens expect your administration to accomplish, to the current capacity of your different city agencies, to your own governing style. Consequently, the answers to these questions cannot be definitive.You can't just copy the Baltimore "model."You will • need to recognize the core idea contained in each answer and then figure out how to make it work in your own city with its own unique problems and opportunities. Still, the CitiStat performance strategy is not something that is and must be unique to Baltimore. It can prove • effective in a variety of governmental jurisdictions—including yours. And if you have any more questions, don't call me. Call Governor Martin O'Malley or Mayor Sheila Dixon. i 6 I IBM Center for The Business of Government • sWHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • What Exactly Is CitiStat? s • CitiStat sounds like a wonderful idea. But what is it, really?A mayor may have seen the CitiStat room.A m" mayor may have watched a CitiStat meeting.A mayor may have talked with the CitiStat staff. A mayor may . have examined some CitiStat data.A mayor may have f I t admired a few CitiStat maps.Yet, this mayor may still not be sure exactly what CitiStat is. E T The Concept Q: What exactly is CitiStat? A: A leadership strategy! s • - L CitiStat is a leadership strategy that a mayor can View from the north gallery of the CitiStat room. Former employ to mobilize city agencies to produce spe- Mayor Martin O'Malley reviews CitiStat surveys. cific results.' aThis characterization is not a very demanding one. The obvious and operational components of Technically, all that it requires is "an ongoing series CitiStat are its meetings and questions, its targets of regular, periodic meetings" plus the actual "use" and data. But these visible features are only the of some "data"—all designed to improve the "per- vehicles by which the mayor focuses the personal attention the management energy, the operational formance" of city agencies. Given, however, that these meetings are "ongoing," "regular," and "peri- tactics, and the creative talents of the people in odic," one subtle feature of CitiStat is often missed individual city agencies on the task of producing by casual observers:This ongoing discussion of clearly specified results. performance involves much persistent follow-up • Consequently, one way for a mayor to think about on past performance deficits and previous commit- CitiStat might be: ments to fix specific problems, as well as follow-up . on decisions, commitments, and established expec- . A city is employing a CitiStat performance tations for future performance improvements. strategy if it holds an ongoing series of regular, periodic meetings during which the In practice, however, CitiStat is much more com- mayor and/or the principal members of the Alex. The key aspect of this way of thinking about mayor's leadership team plus the individual public management is the clear, express, detailed director (and the top managers) of different focus on performance.This focus is revealed in the city agencies use data to analyze the agen- effort to learn what the data reveal about the achieve- city and deficiencies of past performance; in the cys past performance, to establish its next performance objectives, and to examine its establishment of specific targets for future perfor- • mance; and in the development, testing, evaluation overall performance strategies. www.businessofgovernment.org 7 WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY and adjustments of operational tactics that can build This is bad news; it discredits the dream of simply • on past achievements, remedy past deficiencies, copying the Baltimore "system." At the same time, and bring about future improvements. Thus, CitiStat it is good news; it licenses a mayor to adapt the . is more than meetings and data. It requires: core concept of CitiStat to his or her specific pur- poses and the city's own, unique circumstances. • Targets (which provide benchmarks for judging successes and failure) Indeed, any mayor must adapt the core concepts of • • Tactics (which focus organizational efforts on CitiStat to reflect his or her own leadership strategy, • achieving the targets) with specific components that respond to the cur- rent needs of the city, the operational capacities of a • Data (which track the performance of agencies its agencies, and his or her own political and mana- • and subunits) gerial agenda. • Analysis (which, using the data, identifies the causes of both success and failure) The Purpose • Questions (which reveal what agencies are ? • Q g Q: What is CitiStat designed to accomplish. doing and not doing to achieve their targets) A: More and better results! • • Learning (which comes from these analyses, questions, and answers) CitiStat is designed to improve the performance of every city agency. Each city agency is charged with • • Collaboration (which helps the mayor's staff producing results. Otherwise the agency has no rea- and the agency's director and managers6 to son to exist.Thus, the purpose of CitiStat is to help, . determine what to do next) motivate, cajole, and, if necessary, pressure agency • • Experimentation (which creates new ways of managers to achieve more and better results. achieving success) • Meetings (which regularly review agency Q: Whose results? progress, targets, analyses, and strategies) A• The mayor's results! • Thinking (which can suggest how the entire As the city's elected chief executive, the mayor is approach can be improved) the official responsible for the overall management of the city—for establishing its strategic direction • If a mayor and his leadership team are doing these and producing results. And, obviously, the citizens things consciously, persistently, imaginatively, and elected this mayor because they expected that he or skillfully, they are undoubtedly doing something she would accomplish something—perhaps some that—even if it does not have the outward appear- very specific somethings that the mayor promised ance of Baltimore's CitiStat—accomplishes the during the campaign, perhaps just some general CitiStat's purpose: to improve the performance of somethings that now need to be translated into city government. specific operational improvements." Q• What is CitiStat not? At the same time, the mayor is not apt to have a long, detailed list of very pointed results to be pro- A: CitiStat is not a system. duced by every subunit within every agency in the • There is no correct, prescribed, fixed "model" for city. For some items at the top of his or her agenda, CitiStat. No one has created the "mold" from which the mayor may have very explicit ideas about the all other CitiStats must be cast.There exists no orga- specific results that a specific department, agency, i nizational "genome" from which to create a DNA test or unit needs to produce. In other circumstances, to determine whether a CitiStat is a true descendent the mayor will rely on the judgments of others—key of the original. No one has designed the template stakeholders, people on his or her own leadership that a city must methodically follow if it is to offi- team, as well as the political appointees and civil • cially qualify as practicing CitiStat.' servants in the departments. 8 I IBM Center for The Business of Government • . WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY In many circumstances, the mayor will preside over local office), the top manager cannot expect to some subtle negotiations—perhaps merely called produce results in absentia.To produce real results, • "discussions"—with the various parties. In an attempt the manager must be personally engaged in every- to decide what agencies need to produce what thing from establishing the targets to be achieved, results, the people engaged in these discussions will to monitoring progress, to analyzing failures, to attempt to answer a number of questions:What do rewarding success. the citizens expect?What does the mayor expect? What can we afford?What can the agency's person- The top manager can delegate some of these core nel deliver?What mental, strategic, tactical, and/or tasks to others.This requires, however, that the top operational changes would we need to make if we manager clearly establish that the deputies to whom wanted to produce more?What kinds of changes can these tasks have been delegated speak for the man- be expected to produce what level of improvements? ager. If subordinates learn that they can go directly • to the top manager—over the heads of these depu- • For some departments in Baltimore, Mayor O'Malley ties—and win, the deputies will be neutralized. had some very specific ideas about the performance They will be able to accomplish nothing. targets that he wanted them to achieve. For other . departments, he established the overall framework In Baltimore, Mayor O'Malley made it clear to all but was open to suggestions about what exactly the that First Deputy Mayor Michael Enright and CitiStat targets should be. For still other departments, he Director Matthew Gallagher spoke for him. Indeed, • permitted the selection of targets to be worked out after a few years, O'Malley rarely attended CitiStat by his CitiStat staff and the department's managers. meetings. Nevertheless, the CitiStat strategy remained effective because everyone knew that Enright and • Q: What kind of results? Gallagher—and, over time, the younger members of the CitiStat staff—were always operating with the A: Service-delivery targets. mayor's explicit, personal authority and approval. • In Baltimore, Mayor O'Malley and his staff estab- lished a set of key performance targets for every city The Commitment agency. Each target reflected a specific type of ser- vice that the city provided to its citizens that was to Q� What kind of commitment does CitiStat be completed within a specific time period. require? A: Areal, serious commitment. For example, O'Malley established the "48-hour pothole guarantee." If a citizen called in a request No mayor should initiate the creation of CitiStat to fill a pothole, the Department of Public Works without fully recognizing the implications of the • undertaking. After all, most city employees and would fill that pothole within 48 hours. many managers of city agencies will quickly inter- pret it as yet another management fad.They've seen Q: How does CitiStat produce these results? it all: management by objectives and total quality A• Through leadership! management, zero-based budgeting and perfor- mance-based budgeting, the balanced scorecard Again: CitiStat is not a mechanical system. It is not and the organizational dashboards.They aren't going an evaluation scheme. It is not a computer program. to get too excited about the mayor's latest little brain- CitiStat is a leadership strategy that permits the storm (or brief mental shower). From experience, mayor and his management team to track, analyze, they have determined how best to cope with the appraise, diagnose, and improve the results produced latest mayor's random neuron firings.Why bother, by every city agency. they have learned, when this will soon disappear, to be replaced by another mayoral impulse? This requires leadership—active leadership by the mayor, by the mayor's key deputies, and by agency Thus, a mayor who wishes to establish CitiStat not . managers. In any organization (public, private, or only needs to make a real commitment; he or she • nonprofit) of any size (a large corporation or a small also needs to dramatize this commitment. www.businessofgovernment.org 9 WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY Q: How can a mayor demonstrate his or her expectations for the specific improvements, and can commitment to CitiStat? tell from the data whether the agency is making A: By spending time. progress.The mayor has to be prepared to engage in • an intelligent discussion about what can be learned The most obvious way for a mayor to demonstrate from the data, about what can be learned from the • personal commitment to CitiStat (or to anything) is to agency's efforts and experience, and about the kinds spend time on it—serious, consistent, repeated time. of changes the agency and the mayor's office need to S make to ratchet up to the next level of performance. Indeed, as with anything else in government—or in life—people will not accomplish very much if Later, a mayor can delegate the task of conducting they do not commit their most valuable resource: the meeting to a deputy—signaling in some unam- time. Others will measure a mayor's commitment to biguous way that this aide speaks for the mayor. CitiStat(or to anything else) by the level of resources Initially, however, if a mayor wants agency directors • that the mayor puts into it. And again, because and managers, their staff, and their frontline workers everyone recognizes that a mayor's time is his or her to seriously work toward achieving specific perfor- scarcest resource,9 they will look to see how much mance targets, the mayor needs to dramatize that he time the mayor has invested in CitiStat. or she is also working on them—and seriously too. Q: To what should a mayor commit his or • I her time? • A: To participating in CitiStat meetings and to knowing key data. The most obvious way that a mayor can demonstrate commitment to CitiStat is to participate in the CitiStat meetings.A mayor may not be able to attend • every single one.The inevitable crises require may- ors to leave city hall to be on the scene. And any mayor has a host of other obligations and pressures. Nevertheless, if a mayor is to convince the city's S agency directors and managers to take the CitiStat process seriously, he or she has to attend—and actively participate—in many of the meetings. Eventually, a mayor can commission a key deputy • to run the meetings. Indeed, if that authorization is unambiguous, the mayor need not even attend many meetings. Still, at the beginning, the mayor • needs to be a participant. Moreover, the mayor needs to know the data—or, at least, some of the data. For whatever the performance targets the mayor cares about the most, he or she needs to • know the data. Otherwise, everyone will quickly comprehend that the mayor doesn't really under- i stand these targets, let alone care about them. • A mayor need not become a statistician. A mayor need not personally massage the data. Nevertheless, the mayor does need to demonstrate—repeatedly at . every meeting with every agency—that he or she • does understand what is going on, has some clear 10 I IBM Center for The Business of Government • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • • • How Does a City Get Started? • • Even if a city's mayor and its leadership team under- And they collected personnel data so that they • stand both Baltimore's overarching concept and its could keep track of their staff and comply with operational details, they will discover that it isn't various reporting requirements. obvious how to go from zero to a fully functional • and effective CitiStat strategy. After all, for any new From these data, the CitiStat staff quickly figured out • enterprise, getting started is a big challenge. that Baltimore had a big overtime problem.Thus, it began focusing its initial analyses and meetings on The Beginning the challenge of getting control of overtime.10 Each city will begin in its own way. It will begin with Q: How much equipment is needed to get Y g• Y• g� started? the data that are available and with the performance deficits that it considers most important or most ame- nable to some swift and significant improvements. Sure, it would be nice to have a fully equipped CitiStat room with the computers and the projectors Like much else with the CitiStat performance strategy, and all of that fancy technology. It would certainly where to begin is a judgment call. For example, a city be nice to have hired a competent complement of may face a glaring performance deficit—one that is analytical staff. And, obviously, it would be nice to recognized by the mayor, by the responsible city have all of the data templates filled in with abso- agency(or agencies), by journalists, and by citizens. • lutely perfect data. If a mayor waits that long, how- The mayor may conclude that no one will take • ever, he or she may no longer be mayor. CitiStat seriously if the mayor fails to attack this obvi- ous problem. In the absence of such a conspicuous SSo rather than wait until all 10,000 ducks are lined performance deficit, however, the mayor may have • up in a perfect row, just get started. the luxury of choosing from a variety of significant if less visible problems. In this situation, the mayor • Q• What should a city do first? might well choose to focus on a few smaller but eas- ily corrected (if not eliminated) problems. Faced with • A: Start with what it has. skepticism inside and outside of city government, a In Baltimore, the CitiStat staff began by asking each mayor could elect to demonstrate some quick wins agency to bring what data it already had. Not sur- that can silence the critics and convince others to see prisingly, most agencies brought two kinds of data: the benefits of the CitiStat strategy. financial data and personnel data. Most agencies were not really collecting performance data of any Q: How much resistance will a city get—and kind. For other administrative reasons, however, they from whom? were all collecting lots of data—particularly data A• Some active,mostly passive. • about money and people.They collected financial • data so that they could keep track of their budget Any new initiative in any large organization creates and comply with various reporting requirements. resistance. Some of this resistance comes from the • www.businessofgovernment.org 11 • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • passive pessimists who conclude "Why bother?" answering the "Why?" •p p h question will the city Y g y q y be able Some of this resistance comes from active malcon- to learn from its successes and its mistakes. tents who have become comfortable with their , existing procedures and routines and see no reason Initial Progress • to change; indeed, these individuals may be signifi- cantly inconvenienced by the new initiative. Q• How ambitious should a city be? The active malcontents will forcefully and publicly A: Enough to create some obvious successes. criticize the new initiative: "It won't work." "We One way to convert the passive pessimists into tried it before." "This new guy doesn't understand active adherents is to create some quick wins. how our organization works." "We don't have the The size of the wins is less important than that they • money or the people or the equipment to do that." be quick and obvious. Karl Weick of the University • They may even seek to sabotage the initiative with of Michigan has called this "the strategy of small a strategy of malicious compliance: Do precisely wins."72 Don't undertake to solve all of your prob- what is requested, but be completely undiscerning lems at once.You can't. So don't try. • in doing so. Instead, attempt to build upon a series of small suc- The passive pessimists will go through the motions. cesses, each one of which is an accomplishment by They will do what is requested, without making itself. At the same time, each small win also serves • any effort to undermine the enterprise. But they to convince others that the strategy is working and believe—actually "know"—that it won't last. It will continue to work in the future.Thus, the series won't do any good and will be quickly abandoned. of small wins gives confidence to committed sup- Or something else will come along—another brain- porters, converts some passive pessimists, and neu- storm, another fad—and this one will vanish. "Why tralizes the active malcontents. And each additional • look stupid?Why commit to an enterprise that is win—no matter how small—sends an unambiguous M bound to prove ephemeral? If you knew it would signal that this isn't going away. • last—sure—it would be worth helping to make it a success; but long before it has a chance to be suc- cessful, it will disappear. So why bother?" Q: How can a city get results quickly? • A: By focusing on eliminating obvious, simple . Q: What kinds of mistakes can a city make? obstacles to improve performance. • A• All kinds of mistakes. The quick wins will come by strategically selecting • opportunities to eliminate annoyingly small yet No one pulls off a new initiative without making clearly consequential barriers that are preventing • any mistakes.A mayor has to accept that a new city agencies from producing more or better results. • CitiStat will not be any different. If the city tries to Such a barrier could be obsolete rules that no one copy Baltimore's approach precisely, it will make has felt authorized to change. It could be the lack mistakes. No other city is precisely like Baltimore, of a key resource such as a piece of equipment or • and thus the details of what Baltimore did will be an individual with particular training. It could be the perfectly suited for no other city. If the city tries to inability of two agencies to collaborate effectively to copy Baltimore's approach, it will make mistakes— produce a common product. for it will miss one or more important differences between it and Baltimore, or it will misinterpret the Or it could simply be a lack of attention to the par- nature or magnitude of a key difference. ticular result that the mayor has now identified as being a core commitment of city government. After • Consequently, at the beginning the mayor's leader- all, every city agency has multiple tasks—and for ship team will have to manage the CitiStat process every existing task there exist some influential stake- by groping along." Some things will work. Some holders who believe that their task is the most won't.Thus, the mayor and the CitiStat staff will important in all of city government. Consequently, • have to be consciously analyzing not only what agency personnel are usually taught that every task, worked and what didn't, but also why. For only by 12 IBM Center for The Business of Government • • • • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • • every activity, every result is a top priority. City government has no low-priority jobs. • The consequence is that there are no priorities. No • one can assert that one activity is more important or less important than another. • Still, a CitiStat performance strategy requires focus. • It requires choices. Some things will be more impor- tant than others. Some results will be discussed at a CitiStat meeting; others won't. Some data will be distributed, analyzed, and debated; others won't. Some aspects of performance will warrant attention • and resources; others won't. • Consequently, the mayor and the CitiStat staff need • to focus on some specific aspects of performance that they want to improve.They need to choose. • And, at the beginning at least, they ought to choose opportunistically. They ought to choose aspects of performance for which they can easily obtain data and for which they can easily produce some signifi- cant improvements. • • • • • • • • • • • • • • • • • • www.businessofgovernment.org 13 • • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • • • • What Measures and Data Does • • a City Need? • • • Any mayor who chooses to employ a CitiStat strat- Still, given that the city will already be collecting a egy needs measures and data that focus attention on variety of miscellaneous data, the selection of measures the results produced by city agencies that the mayor should not be random.The collection of particular • seeks to improve. Consequently, much of the work measures should be disciplined by the results that of CitiStat staff and agency managers involves the these measures will help the mayor, CitiStat staff, and selection, analysis, interpretation, and revision of agency directors, managers, and employees to improve. these indicators of performance. Consequently, even if the initial search for measures The Measures imitates Baltimore's—"bring us what you've got"— • the choice from among these available measures is • Q• How does a city know what to measure? not arbitrary. It should still be disciplined by the results that the mayor seeks to achieve. Once the Y • A: It depends on what the mayor is trying to CitiStat staff start hunting for measures that agencies • accomplish. are already employing, they may unearth a greater As with every other choice about how to conceive number and variety than they originally expected to and create, then implement and adjust a CitiStat find. After all, agencies may well have found it to be • leadership strategy, this decision also depends in their own internal, operating interest to measure • upon CitiStat's purpose. Architects operate by Louis certain aspects of their work; at the same time, these same agencies may Sullivan's Important principle.. Form ever follows a g Y have found it not to be in their function."13 Architects can't make important deci- own external, political interest to publicize the exis- !i sions about the design of a building until they tence of these measures. • know what purpose the building will serve. Similarly, those who would design a CitiStat for a city need Thus, as with almost all of the other questions to start with their purpose. Only once the mayor about CitiStat, the answer depends upon what the • has established in his or her own mind what mayor seeks to accomplish. Like architects, public executives should always remember that form ever CitiStat should strive to accomplish—and is able • follows function. to clearly articulate this purpose to CitiStat staff . and agency heads—can they begin to decide what they will measure. Q: Who decides what to measure? • Unfortunately, the measures that the mayor needs to A: All city employees and stakeholders cancontribute. • accomplish the purposes underlying CitiStat may not • be available. Maybe no one is measuring what the The mayor has a monopoly on deciding what mayor needs measured. Maybe no one will ever be CitiStat should seek to accomplish.At the same able to measure what the mayor needs measured. time, when formulating this objective, the mayor , Maybe the mayor's CitiStat staff needs to start with needs to listen to a variety of people. Like everyone • whatever is currently being measured. else on the face of the planet, the mayor does not possess a monopoly on wisdom." • • 14 IBM Center for The Business of Government • • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • Q: Why did Baltimore keep measuring the days—might be unacceptable. As for most analytic • same things? tasks, the CitiStat staff can learn the most from the A: Because they continued to be important. data if they get it in its original, most unfiltered, disaggregated form. Producing results in city government is not a one- • time project. It will not be completed by the end of Q. What kind of data does Baltimore • the first—or second—fiscal year. It is an endless actually use? process of continuous, incremental improvement. A: All kinds of data including internal,admin- • Indeed, even if a city agency has somehow miracu- istrative data,plus data on how city • lously managed to achieve optimum performance— agencies responded to citizen requests in the eyes of the mayor, of all of the city councilors, for specific services. • of the voters, and of every journalist in the metro- CitiStat utilizes a variety of standard administrative • politan region—the agency cannot simply hold a data (usually for two-week periods). For the Depart- city-hall celebration to commemorate its champion- ment of Transportation, such administrative data ship. A city agency can't just have a career year, and include parking citations issued, vehicles towed, • cash in on its triumph. It has to repeat its successes and signs installed and repaired. For the Department again next year, and the year after that, and the year of Recreation and Parks, these data include number • after that. of trees pruned, stumps removed, programs for school • Citizens don't just care that the potholes are filled groups, and volunteer hours. For the Department in FY 2008.They also want them filled in FY 2009 of Housing and Community Development, these • data include the housing code enforcement inspec- and FY 2019. The same applies to fixing sewer tions (including the daily average by district and overflows or providing recreational services. Once area). Such data and more can be found on the • a city jumps on the performance treadmill, it can't CitiStat website.16 • jump off.15 CitiStat also utilizes a variety of personnel and SThe Data financial data, such as overtime hours, unscheduled • leave, and disciplinary actions; changes in fleet and • Q: What kind of data does a city really need? equipment inventory; and, of course, expenditures • A: Data that helps to reveal how well the city in various categories (and against budget). is doing in achieving the mayor's objectives. • Finally, CitiStat utilizes a variety of output data For example, if a mayor is focused on filling pot- driven by "service requests," or "SRs," from citi- holes, the mayor (and the CitiStat staff) needs data zens. For example, how did the Department of • on potholes. How many potholes were reported? Transportation do in responding to citizen requests • How many potholes were filled? How quickly were to fill potholes? How many such SRs were they filled? received, and how quickly where they "closed," that is, completed? • Moreover, the analytical staffers who work for • CitiStat need to be able to disaggregate the data Other city agencies get other SRs.The Bureau of in various ways: by districts of the city, by crews in Water and Wastewater gets SRs for low water • these districts, by time of day, by day of the week, pressure. The Forestry Division of the Department • by day and week of the year.They need to be able of Recreation and Parks gets SRs for pruning the • to use these data to calculate various summary over 300,000 city-owned trees. The Department statistics, such as the average time it takes to fill a of Housing and Community Development gets . pothole.They also need to be able to examine the SRs for investigating (and issuing violation notices • distribution of the data.The average time to fill a and/or fines) for high grass or weeds on private • pothole—say, less than two days—might be quite property.The Health Department gets SRs for its acceptable.Yet the upper tail of the distribution— Rat Rubout program. • how many potholes were not filled within, say, five • www.businessofgovernment.org 15 WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY Is What kind of targets does Baltimore set? •Q� g Q: How does a city ensure the integrity of • A: Specific completion targets for every type the data? of service request. A: Audit it. For each specific service request, the city has estab- Each agency closes its own SRs by entering this • lished a target time (measured in days) for how long information into the CitiTrack system. Consequently, it should take to close the SR. For example, Mayor Baltimore has to check to be sure that the citizen's • O'Malley established the 48-hour pothole guaran- request for service has indeed been satisfied. It does • tee: If a citizen calls 311 to ask that a pothole be this by randomly calling each week 100 citizens to filled, he or she will be told that this service request see if they are satisfied with the city's work. will be completed in two days. • Examples of a few of the other service-request Q: What kinds of data are nice to have but are • targets for other city agencies include: not essential? • A: Data that are available because someone • • The Bureau of Water and Wastewater: sewer is collecting them but which are not obvi- overflow, one day ously connected to the mayor's purposes. • Bureau of Solid Waste: graffiti removal, seven If the city has few potholes, if citizens are not com- • days plaining about how long it takes to fill a pothole, • • Health Department: dead animal pickup, three and, thus, if the mayor does not really care about • days potholes, data on how long it takes to fill a pothole are nice, but not particularly helpful. They may • Forestry Division: tree pruning, 300 days become helpful some day, when potholes become a • meaningful issue (either because the city has a par- • Q• Where does Baltimore get its service- ticularly bad winter or because a few vocal citizens . delivery data? have some neglected potholes near their residences). A: From its 311 and CitiTrack systems. At the same time, even if potholes are not at the • In March of 2002, Baltimore's 311 phone number— top of the mayor's, the city council's, or the citizens' • for all non-emergency calls for city services—and agenda, the mayor and the CitiStat staff may still its CitiTrack data system were linked. Consequently, wish to focus on the pothole data.Why? Because it • any citizen wishing to request a city service now may be that (despite the lack of political interest) dials just three digits: 311. A city operator takes the the pothole-filling process can be easily and quickly call, determines the type of service request, identi- improved significantly.This provides an opportunity • fies the appropriate SR template, fills in the neces- to demonstrate some quick if small wins" and to • sary information, and gives the citizen his or her teach the managers and employees of other agencies service-request number along with the completion that progress is, indeed, possible. target. For this purpose, Baltimore's 311 Call Center has 12 workstations, which are staffed round the Q: Does CitiStat employ any qualitative data? • clock, receiving approximately 3,000 calls per day.t' A• Yes. • The data from these service requests are entered into Neither the mayor, the first deputy mayor, the the CitiTrack database, which each city agency uses director of CitiStat, nor the CitiStat staff are hun- • to follow up on its SRs. When it closes an SR, the kered down in City Hall staring at their computer • agency also enters that data into the CitiTrack system. screens.They also live in the city.They observe city Consequently, this database provides the raw mate- operations themselves.They hear from constituents. rial that both agency and CitiStat staff can use to They read the newspapers and listen to the news. • analyze what is happening with performance on Consequently, when an agency fails to fulfill expec- • particular service requests. tations—either a citizen's, a journalist's, or their own—they quickly seek to fix the mistake. • 16 I IBM Center for The Business of Government • • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • • Moreover, they use the problem as an opportunity The fire department could have arrived at every • to learn whether there exists a more systemic, fire quickly, put out every fire efficiently, and res- • underlying difficulty that requires more analysis cued all endangered individuals safely. Still, the • and follow-up. In addition, the CitiStat staff includes city's families and businesses could have suffered an investigator with responsibility for photographing too much property damage from fires.The fire • (and re-photographing) egregious past failures, plus department may have failed to create a fire- new and troubling concerns. prevention campaign that reduced the need to Put out so many fires or rescue so many people. Indeed, often a CitiStat session will begin not with • a discussion of the data for a particular SR, but with For most city agencies, most of the available data • some questions about a recent newspaper story, or will be output data—measures of what the agency about a problem that someone on the mayor's lead- does, measures of its outputs. Consequently, this ership team has simply observed, or about a series may be the only place to start—with whatever out- of photographs that the investigator has taken. put data are available. Moreover, what citizens may care most about is that city agencies do produce Q• Can a city use outcome data or does it their required outputs.They may well care more that . have to rely on output data? the potholes are filled quickly, efficiently, and perma- nently than whether the city's pothole-filling strategy • A: Rarely will a city have outcome data that is has significantly reduced congestion.They may care available sufficiently quickly to be used man- more that broken tree branches are trimmed quickly agerially to make operational improvements. (and the mess removed) than how high the city ranks • The standard measurement mantra is: "Don't measure on some organization's beautiful-city index. A mayor • inputs. Don't measure processes. Don't measure cannot devote too much time and energy to the out- activities. Don't measure outputs. Only measure comes until the city agencies are doing a good job • outcomes. Unfortunately, in city government producing their assigned outputs. • (indeed, in any government) this is often difficult. Sometimes it is impossible. Consequently, a CitiStat Still, outputs are not the definitive measure of a strategy may have to rely more on output data than city agency's performance. How does a city know on outcome data. whether its streets are clean (the ultimate outcome • about which citizens care) if it only measures tons A government's output is what occurs at the agen- collected (the output that the sanitation department cy's border. It is the streets cleaned, the children produces)? It doesn't. Consequently, in addition to . vaccinated, the criminals arrested, the restaurants measuring how efficiently, effectively, and speedily • inspected, the trees trimmed, the fires extinguished. the department picks up the city's trash, the mayor All these are important activities of a city govern- might seek outcome measures for the cleanliness of ment.They are important, however, primarily because city streets. • we believe that, in producing these outputs, the city is accomplishing important public purposes. We For a mayor seeking to create a CitiStat performance • strategy, it makes sense to start with the available believe that, by producing these outputs, the city is creating public value." Still, the connection is data.The city will, of course, have a lot of input • rarely perfect.20 data. It may have some output data, though the mayor's CitiStat staff may need to do some signifi- • • The health department could have inspected cant work to create useful and revealing output every restaurant twice in the past year. Still, data. Only once the city has managed to create an • the department might have failed to protect assortment of productive output data should it move the city's citizens and visitors.The health to the more challenging task of creating outcome department could have checked for the wrong data that are both meaningful and useful. • diseases or the wrong unsanitary practices or it • could have checked for the right problems but • done this checking badly. • www.businessofgovernment.org I 17 • • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • • • • • • • Who Has What Responsibilities? • • • Given the multiplicity of real, nitty-gritty operational latest whim—"What do they call it? CityStat?"—will • responsibilities of any city government, given the be any different? challenge of getting anything done in government, • and given the complexity of the CitiStat strategy, Thus, any mayor who seeks to employ a CitiStat who needs to do what to ensure that performance strategy to improve the performance of city agencies • does improve? needs to convince the people who work for the city • that this mayoral initiative is, indeed, for real. In • The Mayor fact, unless the mayor makes a conscious, commit- ted, and consistent effort to do so, people will pay • Q: What is the mayor's primary responsibility? no more attention to CitiStat than they do any politi- • cian's BOMFOG oratory."They accept it as nice A: To convince people that CitiStat is for real. rhetoric but understand that it lacks any real opera- • In any government jurisdiction, the elected chief tional significance. • executive sets the tone. Whether it is the mayor, the county executive, the governor, or the president, Q: What can a mayor do to set the necessary • people are always looking to this individual for clues. tone? Does the mayor really care about this? Or is this • something that the mayor is doing merely to appease A: Invest personal prestige in CitiStat. • some important constituency? Does the mayor really What does a mayor take seriously?The things— believe in this? Or will this soon disappear (just like activities, policies, endeavors, projects, initiatives, • all of those other mayoral initiatives) to be replaced programs—in which the mayor has invested his or • by the next big thing? Does this mayor follow her reputation. If employees and citizens conclude through? Or does this mayor jump capriciously that the mayor has staked his or her political, profes- • from fad to vogue to trend to craze? Should I pay sional, and personal reputation in an undertaking, • any attention to what the mayor is currently espous- they will also take it seriously. ing? Or is it not worth my time? • This creates a dilemma. Without investing personal For most people in government, the default assump- prestige in a CitiStat strategy, a mayor cannot • tion is obvious: This too will pass. After all, expe- mobi- lize people to take it seriously. At the same time, by • rienced public employees have accumulated investing personal prestige in CitiStat, the mayor is significant empirical evidence to support this infer- gambling on its success. By not investing his or her • ence.Time after time, year after year, most of the reputation in CitiStat, a mayor is lowering the cost of • top priority Initiatives have possessed no more failure.At the same time, however, by not investing • permanence than a Fourth of July fireworks display. his or her reputation in CitiStat, a mayor is also low- They were dazzling wonders to watch.Yet they were ering its chances of success.As with any mayoral ini- • soon gone, mostly remembered for how brilliantly, tiative, by increasing his or her personal investment • and briefly, they lit up the sky. Why should city in CitiStat, a mayor simultaneously improves both the employees or citizens think that this latest mayor's probability of success and the costs of failure. • • 18 I IBM Center for The Business of Government • • s WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY Q: How can a mayor invest personal prestige get the message. If the mayor delegates the task of in CitiStat? running the CitiStat meetings to a junior aide, every- A: By spending money, staff, and time. one will get the message. Mayors spend staff just like they spend money, and people watch to see on Elected chief executives have numerous ways to what priorities a mayor is spending the best staff. • demonstrate that they are serious about an initiative. • The two most obvious are to spend money and to Mayors also spend time—their time, their most valu- • spend time. Both money and time are very scarce able commodity. And again, people are watching: On commodities.Thus, people will watch to see how what does the mayor spend time? Is the mayor spend- the mayor invests them. ing time on CitiStat? If so, the mayor must be serious • about it—deluded, maybe, but, nevertheless, undeni- On what does a mayor spend money?Any new ably serious.Thus, even if city employees believe that mayor faces a budget deficit. It is a law of urban the mayor is naive about the prospects for producing • government; no mayor leaves a successor with a results in city government, even if they believe that, budget surplus.23 Still, CitiStat does not have to cost CitiStat or no CitiStat, city government will never that much. Sure, there is the initial capital cost of improve performance, they will still take note of the the room and the technology, plus the ongoing mayor's willingness to spend time on CitiStat • operating cost for the staff. Compared with the . public-works budget, however, CitiStat is cheap. Q: What does the mayor not need to do? • Nevertheless, people will be checking on what finan- A: Attend every meeting. . cial resources the mayor invests in CitiStat. Does the If a mayor faithfully attends every CitiStat meeting, mayor convert an existing city-hall room into the people will quickly get the message. If the mayor is official CitiStat room? Or are CitiStat meetings con- always in a room, a lot of other people will want to • ducted in some undisclosed location away from city be in that room. Not everyone, of course; all those • hall?24 Does the mayor invest in the technology to city employees who prefer to remain anonymous will conduct the meetings—and the technology that the want not to be in that room.And, if the mayor is ask- CitiStat analysts need to do their work? Or do the ing pointed questions about agency performance (or • people with responsibility for collecting, analyzing, is merely present while others ask such questions), • and displaying CitiStat data have to scavenge for the number of people who would prefer not to be in computer hand-me-downs? Does the mayor invest in the room will increase. Nevertheless, if the mayor is • a talented staff of CitiStat analysts? Or does the mayor frequently attending the meetings, people will get the • give these CitiStat tasks to people who already have message:The mayor is serious about CitiStat. multiple, much-higher-priority assignments?21 • Unfortunately, mayors are human. Like the rest of Indeed, on what does the mayor spend talented us, they too have only 168 hours in their week. • staff? Mayors can attract ambitious, talented people Moreover, all mayors have a multitude of responsi- who believe that they can make adifference—and bilities—the numerous things that they want to do, their reputation—in city government. At the same plus the even more numerous obligations that other • time, the supply of such people—people with the people (always very important people) expect them • capacity and willingness to make a difference in city to fulfill. Mayors must attend meetings, deliver government—is not unlimited.A mayor can seek to recruit more of these people. At the same time, a remarks at portentous events, lobby state legislators, • listen attentively to impassioned lectures, massage • mayor has to decide what assignments to give the city-council egos, smile for photographs, hold city's top recruits. Again, people will be watching. neighborhood outreach sessions, shake hands with If the mayor recruits people with real analytical tal- visiting dignitaries, visit schools and toss students' • ent and assigns them to the CitiStat office, everyone hair, and cut pretty ribbons. Oh yes: And most • will notice. If the mayor doesn't even bother to mayors also have a family. attempt to attract analytical talent, or if the mayor assigns all the people with analytical skills to, say, All mayoral schedules require compromises. the budget shop (and not CitiStat), everyone will Consequently, no mayor can attend every meeting. • www.businessofgovernment.org 19 • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY No mayor should expect to attend every CitiStat It might be the director of the department of man- • session. Even if a mayor initially believes that he or agement. It might be the director of CitiStat. The she will be able to attend every CitiStat session, this formal title this person holds is less important than • promise will quickly be compromised. his or her informal but authoritative status. • Fortunately, a mayor need not attend every meet- The mayor may well enter office accompanied by ! ing—or even half the meetings. After all, a key this individual. He or she may have been the cam- • virtue of CitiStat—a necessary attribute that is often paign manager, law partner, college roommate— • missed by those who visit Baltimore for a morn- someone whom the press has already publicly ing26—is the continuity of the issues discussed at the identified as the mayor's alter ego. When Martin ! meetings.The most important aspects of a city agen- O'Malley became mayor of Baltimore, his first • cy's performance are examined repeatedly—at meet- appointment was Michael Enright as first deputy ing, after meeting, after meeting ... after meeting. mayor. Everyone knew (or quickly figured out) that Enright spoke for the mayor. • A mayor who cannot attend every meeting cannot personally provide this continuity. In other circumstances the ma p Yp Y � or will need to Y make this delegation of authority visible and explicit. ! Q• What does the mayor have to do? The mayor can do this by attending many of the • initial CitiStat meetings and frequently and explic- . A: Confer authority. itly ratifying with appropriate words and strategic A mayor cannot chair every CitiStat meeting. But silence the probing, suggestions, and directives of ! someone must.The responsibility for conducting the the individual who will also run the meetings in ! meeting cannot be randomly rotated among miscella- the mayor's absence. • neous members of the mayor's staff. It cannot be left to a third-level subordinate. If the CitiStat meetings— Finally, of course, the mayor will need to quickly and clearly strike down the first threat (subtle or and thus the entire CitiStat strategy—are to have any . impact on the behavior of city employees, these direct) to this individual's authority.When the first agency head seeks to appeal an instruction issued meetings must be chaired by a single individual upon ! whom the mayor has conferred significant status. at a CitiStat session, the mayor needs to squelch it quickly. Otherwise, everything that happens at • The purpose of a CitiStat meeting—or, more appro- future CitiStat sessions will be either ignored or priately, the purpose of a series of CitiStat meet- open to appeal. ings—is to focus the energies of city employees on ! fixing their key performance deficits.27 This, how- To ensure that CitiStat is effective, the one thing that • ever, will not happen after one meeting. It will take a mayor has to do is to endow the individual(s) who a series of meetings, months of meetings, perhaps will run the meetings and the staff who will do the years of meetings. Moreover, this will not happen analyses with the authority they need to do their jobs. • if each meeting examines a new performance prob- lem. If CitiStat meetings are to have an impact, they The CitiStat Staff . must continuously and consistently examine a core set of performance challenges: tracking the data, Q: What kind of staff does it take to make ! observing both advances and setbacks, debating CitiStat work? • alternative approaches, motivating people to experi- A• A smart, dedicated,hardworking staff. • ment with new tactics, learning from both failures and successes, and keeping people focused on the The skills needed by CitiStat staffers are the same as • results to be produced and improved. those required by any city employee with significant responsibilities. CitiStat staffers must be dedicated, ! One individual has to provide this consistent conti- willing to work long hours for significantly less pay • nuity. And the mayor needs to make it clear exactly than they could make in the private sector. CitiStat • who this individual is. It might be the mayor's chief staffers must be smart—smart along a variety of of staff. It might be the deputy mayor for operations. dimensions. They must be analytically smart, ! 20 IBM Center for The Business of Government • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY comfortable with numbers, and able to tease out crew in district one do compared with the pothole interesting trends, problems, and opportunities from crew from district six?Were potholes reported on • the available data. At the same time, they must be Wednesday morning filled more quickly—or less politically and organizationally smart.They must be quickly—than those reported on Saturday morning? able to present and explain their data in a manner Were potholes reported on Monday morning less that does not gratuitously offend the personal or pro- likely to be filled quickly—and, if so, was that • fessional competence of the agency's director and because of the backlog of potholes reported over key managers. If CitiStat staffers can explain their the weekend (when citizens had more time to report work clearly and cooperatively but can offer only them and fewer crews were working) or because • shallow analyses, they contribute little to the perfor- Monday is traditionally a low-productivity day in • mance of city agencies. If these staffers are brilliant the street maintenance division? analysts of the data but present their conclusions condescendingly, they will undermine the mayor's This simple example suggests the uncertainty in ana- • efforts to convince city agencies to focus on results. lyzing disaggregated data. First, in what ways should the analysts disaggregate the data? By geography? Most public-sector jobs require multiple intelli- By organizational subunits? By time? In the search gences.21 The job of the CitiStat staffer is no different. for something revealing, the analysts don't necessar- ily know along what dimension to disaggregate the • Q• What exactly does the CitiStat staff do? data. Mostly analysts guess, using either past experi- ence with the data or an educated hunch. But until . A: Identify performance deficits and suggest the analysts actually do the disaggregation for the • strategies for improvement. first time, they won't learn whether their work will The first task of the CitiStat staff is to look at the uncover anything unusual. • data. From whatever data are available, the staff needs to figure out what is working and what isn't. Moreover, even if the analysts do discover some- needs unusual in the data—something that suggests The task here is to identify the city's performance they might learn something from their disaggregated deficits—results being produced by city agencies • that are, in some way, inadequate. data—they may not immediately understand what • the data are revealing. Does the difference between This requires a comparison. It may be the compari- the crews in district one and district six reflect dif- ferences in their workplace dedication • son of the city's data with similar data from other ' in their similar municipalities. It may be the comparison of operational competence, in their crew supervisors, . the data for one city unit with the data for another or in their organizational intelligence? Or is the similar unit. It may be the comparison of the data difference due to the conditions under which they • for one city agency with some ideal—an ideal work? Or is the difference purely random, likely to expressed in a general mayoral aspiration or in a disappear in next month's data?29 • specific performance target. Whenever the CitiStat Finally, once the analysts have learned something staff reaches any conclusion about the performance significant from the data, they need to figure out of a city agency—positive or negative—they do so what to do about it. What might they recommend? based on some kind of comparison. • Should they suggest that the district-six crew reor- Sometimes these conclusions will be drawn from ganize its work to reflect the superior tactics of the aggregate data: How many potholes did citizens district-one crew? Or should the crew supervisor in • report last month? How many did the city fill district six be replaced? Or can the lower productiv- within its target time of 48 hours?And what was ity of the district-six crew be fixed with some simple • the average time it took for the city to fill a pothole? on-the-job mentoring for its supervisor? Such data provide a picture of an agency's city- • wide performance. The data do not answer these questions. Conse- quently, these kinds of questions—both what are the Often, however, useful conclusions can be drawn conclusions to be drawn from the differences in the only from disaggregated data. How did the pothole data and what kind of remedial action (if any) • www.businessofgovernment.org 21 • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • sop should be taken—need to be discussed with the An applicant need not have high-level econometric agency director at the next CitiStat meeting. training; CitiStat staffers run few regressions. But CitiStat staff do have to be comfortable with num- • Q• How many CitiStat staff does a mayor need? bers—able to look at data, uncover trends, and tease out revealing pieces of information. And A: Not many. they do have to be able to present their conclu- The size of the CitiStat staff depends upon size of sions in a respectful and convincing manner. • the city and the ambition of the mayor. If the mayor • of a large city (say, with a population of half-a-million Q: How does a mayor attract CitiStat staff? • people, and hundreds of propagating potholes) A• By promising them the opportunity to have wants to use CitiStat to really improve performance, • an impact. he or she will need several analysts in the CitiStat office.Yet Baltimore (with a population of 640,000) A mayor can make CitiStat hot. By committing . has only a director, half-a-dozen analytic staff, and city government to producing real results, and by an investigator. making CitiStat a key element of this performance strategy, a mayor can make joining the CitiStat Like city government itself, Baltimore's CitiStat staff staff both an exciting adventure and an intelligent is organized around city agencies. Each of its ana- career move. Young analysts just out of graduate lysts is assigned to cover several of these agencies, school will be attracted by the opportunity to have • and they become very familiar with their agencies' an impact on government's performance that work- • data, operations, and key managers. Given the two- ing on CitiStat offers. week cycle of the CitiStat meetings, each analyst usually covers just two agencies, preparing each This is, of course, the same strategy that the mayor • week for one CitiStat meeting for one agency.30 In will employ to recruit the director of any city agency. addition, the investigator roams the city looking for But the pool of talent from which agency directors trouble spots, often photographing them for display and CitiStat staffers are recruited may be quite differ- at a future CitiStat session. ent. Agency directors need managerial experience. • CitiStat staffers need analytical expertise. Both will • Still, the minimal number of CitiStat staff is probably be attracted to work for a mayor who is using two.Why? Because if the CitiStat office contains only CitiStat to make a difference. S one person, he or she will have no one with whom • to talk. No one with whom to check and debate Q: Where is the CitiStat staff located, organi- • ideas. No one with whom to commiserate. No one zationally, within city government? to help get out of a demoralizing rut or with whom • to celebrate a small yet meaningful success. Everyone A: /n the mayor's office. • needs colleagues. CitiStat staff are no different. For the CitiStat performance strategy to prove . effective in motivating the agencies to focus their Q: How does the mayor select the CitiStat staff? energy and intelligence on improving performance, A• By trial and error. the staff need to be seen by everyone working in . city government as a direct extension of the mayor. • Given the multiple talents required of any CitiStat Consequently, the director of CitiStat needs to • staffer, it is not always easy to determine whether an report directly to the mayor. Or, if the CitiStat applicant will be successful in the job.Thus, as with director is not the individual on whom the mayor • many public-sector jobs, the only way to discover if has conferred the authority to conduct the CitiStat • an individual can do it is to let him or her try. Some meetings, the CitiStat director should, at least, applicants will prove successful. Some won't. report to this individual. • Nevertheless, it makes sense to select people with Baltimore also emphasizes where CitiStat should not • a set of basic analytical skills combined with some be organizationally located. CitiStat should not be part people skills and then help them grow into the job. of the budget bureau. If it is, the people in Baltimore • 22 IBM Center for The Business of Government . WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • believe, the definition of performance will soon morph • This aspect of performance, as captured by these into: How much money did we save?Cost is not irrele- service-request data, is slipping or has not • vant in Baltimore, and it is not irrelevant for Baltimore's improved as fast as we would like. What has • CitiStat. Nevertheless, in Baltimore—unlike so many your agency done about this? other governmental jurisdictions—expenditures are Snot the sole measure of performance. The other week (month, quarter, year) we agreed on the need to employ this particular approach. What has your agency done to The Agency Director and Managers implement it? Q: What is the agency director's responsibility? • We've noticed this really big problem. What is A: To produce real results. your organization doing about it? Given that the purpose of CitiStat is to ensure that That is, the questions focus on some improvement • city government—and thus city agencies—produce that the agency needs to make. • results that citizens value, the primary responsibility of the manager of each city agency is to produce Agency directors and managers need to respond to those results. such questions in four ways: • But what results? Whose results?The manager of 1. They need to provide straightforward, factual any public agency receives multiple and conflicting information. What happened?When? How? instructions about what results to produce.The agen- Why? (Note: Denying the existence of the cy's authorizing legislation is simultaneously vague problem is not a politic rebuttal.) and contradictory. The pressures from members of the city council, organized constituency groups, 2. They need to explain what the organization has journalists, and individual citizens come with the done so far to mitigate the problem. Who did same ambiguity and tensions. what when?What happened? (Note: Reporting that no effort has been made to deal with the Given this reality, some agency managers may view problem is not a prudent reply.) • CitiStat as a blessing, because CitiStat eliminates 3. They need to explain what they have so far that ambiguity.The CitiStat process provides an learned from these efforts.What worked?What agency director with a clear understanding of the didn't?Why? (Note: Disclosing that nothing has specific results on which the agency should concen- been learned from any effort to fix the problem • trate. CitiStat may not eliminate all of the tensions. is not an astute response.) • Not every stakeholder will share the mayor's priori- ties. Nevertheless, by providing an agency with a 4. If they are not convinced that they have satis- clear definition of the results that are of the highest factorily mitigated the problem, the agency priority, CitiStat implicitly designates the other director needs to outline a new approach to remaining results as lower priority. CitiStat provides fixing it. What might work? Why?When will the director, managers, and employees of each city it be implemented? How much will it cost? Sagency with a focus for their work. When can improvements be anticipated? (Note: Revealing that no thought has been given to a •• Q: What is the job of an agency's director and new approach is not an advisable reaction.) • managers at a CitiStat meeting? At a CitiStat meeting in Baltimore (as at a CompStat • A: To answer questions, to explain existing meeting in New York City31), an agency director or approaches, and to offer new ideas. manager is responsible not so much for a specific . outcome but for having an intimate working knowl- A CitiStat meeting centers around a series of ques- tions asked by the mayor's staff to which the agency well-thought-out strategies for fixing them. director and managers need to respond.These ques- tions come in several basic forms: www.businessofgovernment.org 23 • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • Q: How do agency directors prepare for a CitiStat meeting? A: Byconducting their own internal AgencyStat session. When Baltimore launched CitiStat, agency directors came to their biweekly meetings unprepared.They • quickly learned, however, that they needed to get themselves ready to answer the inevitable questions. So the day before, they started getting together with their own managers, a meeting that quickly morphed . into an internal "AgencyStat." Initially, these meetings • were like debate preparation: "What are our data say- ing?What questions will I be asked?What is my best • answer?" But eventually, as agency directors began to • appreciate how the CitiStat strategy could improve • performance, they began to use their AgencyStat ses- sions to manage their own organization. In Baltimore, • AgencyStat sessions are not just designed to get ready • for tomorrow's meeting. Instead, agency directors use them to improve performance so that they won't have to sweat so much about a future meeting, three or six • months in the future. • • i • 24 I IBM Center for The Business of Government • • • • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • • • • • What Kind of Infrastructure Does • a City Need? • • • CitiStat may be a leadership strategy. Nevertheless, ' . the implementation of this leadership strategy takes place within a specific operational framework. It a • takes place in a room, depends upon some specific forms of technology, and (as always) needs a budget. r v E • m . The Room ° U d Q: What are the key characteristics of the • CitiStat room? c 0 A: Chairs, tables, and a podium.32 — g • d • The most obvious feature of Baltimore's CitiStat room Mayor Sheila Dixon conferring with Philadelphia Demo- is the podium. Behind this podium (see Figure 1 on cratic Party mayoral primary winner, Michael Nutter,on page 26) stands the director of the agency whose June 22,2007,during a CitiStat meeting. performance is being discussed. Occasionally, the first deputy mayor or the agency director will call CitiStat meetings. Michael Enright, the first deputy other agency managers to the podium. mayor, was, however, a permanent fixture at most • CitiStat meetings and usually led the discussion. This podium is not essential. A CitiStat session could Similarly, Matthew Gallagher, the director of CitiStat, be conducted with everyone sitting around a large participated in almost every meeting (and took the • conference table or in a variety of other settings. lead when Enright could not).The mayor's chief of staff rarely attended and rarely participated when he did. Nevertheless, the podium does possess symbolic significance.The individual at the podium is the Sitting on the two wings of this table were the direc- only one in the room who is standing. Everyone tors of the city's key support units: else in the room is focused (both visually and men- tally) on this person. (Or persons; sometimes two or • The Department of Finance was usually repre- maybe three people are standing behind the podium.) sented by its deputy director, Helene Grady. • Thus, the podium dramatizes who is responsible— The Mayor's Office of Information and who is responsible both for answering the specific Technology was represented by the city's chief question now being discussed and for the general information officer, Elliot H. Schlanger, who overall operation of the broader issue of perfor- almost always attended the meetings. mance being examined. • • The Department of Human Resources was Sitting at a table facing the podium are four chairs. represented by a variety of different people. One is for Mayor O'Malley, who during the second . The Office of the Labor Commissioner was usu- half of his seven-year tenure rarely participated in ally represented by Commissioner Sean Malone. • • www.businessofgovernment.org 25 • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY Figure 1:A Rough Schematic of Baltimore's CitiStat Room During the O'Malley Administration • Screen high on wall Agency Director Screen high on wall Seats for top agency managers Seats for top agency managers • Podium i • Legal Counsel Chief Labor Commissioner Table Information •Human Finance Officer Resources Matthew Mayor Michael Chief Director . Director Gallagher O'Malley Enright of Staff • Control'room with projector • The mayor's Law Department was usually repre- Today, Baltimore's CitiStat relies on four types sented by the legal counsel, Ralph S.Tyler. of technology: (1) the 311 phone system, (2) the . CitiTrack data system, (3) spreadsheet templates • Under each of the two screens on the wall were and analytical frameworks for analyzing data, and • half-a-dozen seats for the top deputies of the agency. (4) computers and projectors that display the maps, Both on the left side and the right side of the room charts, and data during a meeting. were another two dozen chairs for more officials • from the agency as well as for the usual collection Given that the objective of CitiStat is to improve the of out-of-town visitors. In a bench in front of the performance of city agencies, it needs data about control room sat the CitiStat analyst for the agency. this performance: What specific results is a specific • And in the control room were another two CitiStat agency supposed to produce? Until this question is staffers who projected the various maps, charts, and answered, it is impossible to decide what data to data on the screens. collect and analyze, let alone what technology is needed to facilitate these tasks. • The Technology • In Baltimore, Mayor O'Malley chose to focus on Q: What kind of technology does a city need? the delivery of city services. In particular, he chose S A: Enough so that the city can collect,analyze, to focus on the time it took for a city agency to sat- . isfactorily respond to a citizen's request fora spe- cific service. How long did it take to fill a pothole? Over time, Baltimore's technology has become To trim a tree?To get water out of a residential • more sophisticated. In the beginning, however, it basement? If this is the kind of results that the mayor was not particularly polished. As its initial search wants to produce, the city needs to be able to for data was pragmatic and opportunistic, so was measure the relevant agency's performance. Conse- . its choice of technology. It began with what was quently, Baltimore specified 250 different kinds of • available. "service requests," or SRs, that citizens could make • from city government. 26 IBM Center for The Business of Government • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • • Q: What is the 311 City Call system? chose to define performance as improvement in city A• A single centralized method for citizens to services, the CitiTrack database collects information . request services from the city. on how well different city agencies are doing on this mayoral dimension of performance. In addition to CitiStat, Baltimore created its 311 City Call system (an innovation pioneered by Chicago). Q. What are the data templates that Baltimore A citizen who wants a city agency to do something employs? no longer has to figure out what agency that is and what the phone number for that agency is. Nor does A: Excel spreadsheets. • the citizen need to resort to calling 911 with the pre- When Baltimore created CitiStat in 2000, it also cre- text that the request is an emergency. Instead, the citi- ated data templates using Excel spreadsheets. Since zen just dials a single number-311—for all such then, the CitiStat staff has occasionally updated the requests. Like a lot of other cities with a 311 number, spreadsheets to provide additional data or analysis. • Baltimore refers to this as "one call to city hall." Many of these data templates can be seen on the Baltimore's 311 call center contains 12 workstations CitiStat website (in PDF format).35 When Baltimore • and has a staff of 30 trained customer service agents. receives a request from another jurisdiction, it has • These agents employ 250 different templates on been willing to send a sample to others. which they record the details of the 250 different types of service request.There is the SR template for • a pothole, a different one for a tree trim, and still Q: Where does Baltimore obtain the special another for water in a basement. Citizens can also equipment it uses? make a 311 service request online.33 A• Off the shelf. Baltimore has obtained the various components of Q: What is CitiTrack? the information technology that it uses for CitiStat A• A customer relationship management strictly from standard sources. None of Baltimore's . system for keeping track of citizen equipment is proprietary.The computers in • requests for service and their fulfillment. Baltimore's CitiStat offices are no different from The analysis conducted by the CitiStat staff requires those in any other city hall. data. Consequently, Baltimore needs a mechanism— a database—that can be used to collect and com- The Budget pare data. CitiTrack is that database. Q: How much did it cost to create CitiStat? • Baltimore stores the details on each of these SRs A: $20,000. using yet another technology—the CitiTrack system. • Modeled after customer relationship management The initial setup cost for CitiStat—for the room and (CRM) systems in the private sector, CitiTrack the information technology—was just $20,000. Most records the details of all service requests made to of this went for the information technology.The facili- • the 311 system (either over the phone or online) ties—the room in which the CitiStat meetings are including when these tasks were completed. held and the offices in which the CitiStat staff work— Originally, this CRM software was developed by were created from underutilized parts of City Hall. Motorola in cooperation with Baltimore; it can Consequently, the start-up funds that were not spent now be purchased off-the-shelf from Motorola.34 on technology went for sheetrock, tables, and chairs. Any analysis of performance requires an agreement Q: What is the annual operating cost of CitiStat? about exactly what data is required.This, in turn, • requires an agreement about the exact nature of A: Half-a-million dollars per year. • performance that needs to be improved. Until those For FY 2007, the operating budget for Baltimore's issues are addressed, it makes little sense to create CitiStat was $509,000. All but $6,000 of this was a new database. In Baltimore, because the mayor for salaries and benefits. 0 • www.businessofgovernment.org 27 0 WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • Q: To what component of the city's budget is of revenue; to ensure that all funds are spent exactly • CitiStat assigned? as appropriated; to ensure that the city does not overspend its budget. For a city budget office, A: The mayor's budget. spending less is always good.36 The budget office CitiStat is a unit within the mayor's office. Conse- would, inevitably, want to get the same bang for • quently, the CitiStat operating budget is part of the fewer bucks.The budget office might be even hap- mayor's operating budget. pier with a smaller bang for fewer bucks. Q: How much does the 311 Call System and In contrast, the CitiStat office has a different set of CitiTrack cost? purposes. Primarily, the CitiStat office wants to improve the results produced by city agencies. A: $2.5 million in initial capital costs and$4.6 Like every other unit within city government, it is million per year in operating costs. constrained in what it can accomplish by the The capital costs included $2 million for Motorola's restrictions contained in the budget. Working within • customer relationship management system, and these constraints, the CitiStat office seeks ways to • another $500,000 for the city's Call Center. produce more or better results. The staff of CitiStat . or an agency may, however, obtain some inter- budget-item flexibility if they can make the argu- • Q: How does CitiStat influence the budget ment to the first deputy mayor that such flexibility • process? will permit an agency to improve performance. A• Indirectly. Then, using that flexibility, the agency will need to Once a month, an agency's CitiStat meeting will develop a new strategy exploiting this flexibility and to report back the resulting improvements. For what- • begin with a budget update and an examination of ever bucks are available, Baltimore's CitiStat staff is specific funds: Why is this running over budget? always looking for a bigger and better bang. • Why is it running under?Then, once a year, CitiStat • conducts a detailed review of each agency's budget. Of course, CitiStat analyses and discussion have a • lot of subtle influences on other aspects of the city's operations, everything from budget to procurement to human resources. CitiStat identifies difficulties • that need to be resolved, and with people from the agency and mayor's support offices in the room, these can be worked out on the spot. Moreover, the reputation for competence and cooperation that an . agency, its director, and its managers establish over • many CitiStat sessions affects their working relation- ships with other agencies, not just the budget office. Q: Should the CitiStat office be part of the budget department? A: No. t Baltimore emphasizes that if CitiStat is run out of a city's budget office, the sole measure of concern will quickly become dollars saved. The budget office has one set of purposes: to create • the mayor's annual budget proposal; to ensure that the city's expenditures are consistent with its sources 28 IBM Center for The Business of Government • • • • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • • • • • What Is the Purpose, Operation, • and Impact of the Meetings? • • The most visible feature of CitiStat is the meeting. Q: Who attends from the mayor's office? Mayors and others who visit Baltimore to learn about • CitiStat spend most of their time observing a meeting. A: About a dozen key mayoral appointees. • And although CitiStat is more than a series of meet- During the O'Malley administration, the attendees • ings, these meetings are, in fact, central to the strategy. from the mayor's office typically included: • The Participants in the Meeting • Michael Enright, the first deputy mayor, who • conducted the meeting • Q: Who attends the meeting? . Matthew Gallagher, the director of CitiStat • A: Several dozen people from the agency and (who conducted the meeting if Enright was • the mayor's office. unable to attend) • A CitiStat meeting is a deliberation that involves the • The CitiStat analyst responsible for the agency at • management team from a city agency and key per- the session sonnel from the mayor's office. . One or two additional data analysts from the • CitiStat staff, who were responsible for project- . • Inevitably, the agency is represented by the director, ing the proper graphics onto the wall who stands at the podium; 10 to 12 of the director's • key deputies, who sit under the two screens; and • A representative from the Department of Finance, • another dozen or so middle managers, who sit on usually the deputy director, Helene Grady • the wings. Every one of these individuals can be 0 Elliot H. Schlanger, the city's chief information called upon—either by the mayor's staff or by the officer and the head of the Mayor's Office of • agency director—to answer questions about some Information and Technology • aspects of his or her specific responsibilities. Some of the deputies may be repeatedly called upon ses- sion after session; others may need to respond only Human Resources • two or three times a year. • Sean Malone, the commissioner of labor • Yet, even if agency middle managers rarely need 0Ralph S.Tyler, the mayor's legal counsel and • to answer a question, they profit from the session. head of the city's Law Department • Kimberly Flowers, who served on the CitiStat staff • before she became director of the city's Department Q: Who does not attend the meeting? of Parks and Recreation, noted that she brought A: Journalists, stakeholders, and citizens. • her agency's top two-dozen officials to her CitiStat • sessions because she wanted each of them to under- CitiStat is an internal management meeting—the • stand the mayor's agenda and to appreciate the mechanism that the mayor uses to run city gov- specific results-focused expectations that the mayor ernment. Consequently, participation is limited to • and his staff had for the agency. those with direct operational responsibilities for the • • www.businessofgovernment.org 29 • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • specific agency under discussion.37 (Occasionally, working and what isn't; how they track the effective- journalists have attended a CitiStat meeting, but ness of the ideas, adjustments, and solutions that they they do this to write a feature story, not to provide have implemented; how they establish who in each S ongoing coverage.) agency is competent and who isn't; how they reward significant accomplishments; and how they drive Q• Do union officials attend the meeting? home their unhappiness with inadequate results. A: No. Q: How many CitiStat sessions does Baltimore • Mayor O'Malley's position was that CitiStat is an conduct? • internal management meeting. If a union would let someone from the mayor's office attend its internal A: Over half-a-dozen a week; several dozen in meetings, asserted O'Malley, then he would permit a month. • the union to attend CitiStat. Some sessions—specifically PoliceStat and School- • Stat—are held every week. CitiStat sessions for most In reality, of course, no CitiStat meeting is completely city departments are held on a two-week cycle; these • confidential.After all, in the room there would usu- include: Fire, General Services, Health, Housing, ally be a union member and probably someone who Solid Waste, Recreation and Parks,Transportation, was close to the union's leadership. Consequently, and Water and Wastewater. In addition, Baltimore after the meeting ended, the union would quickly has held (biweekly or monthly) CitiStat sessions for • learn what was discussed and decided. Capital Program Management, Minority Business • Enterprise (MBE) and Women's Business Enterprise The Conduct of the Meetings (WBE), Finance, Homeless Services, Information and Technology. Q: How frequently does an agency attend a CitiStat meeting? Q: Who sets the agenda for each meeting? A: Every two weeks. A: The first deputy mayor, the director of The CitiStat cycle is religiously regular. If an agency CitiStat, and the agency's CitiStat analyst— • is scheduled to appear this week at 1:00 on Thursday and the agency, too. afternoon, it will also appear at 1:00 on Thursday For every agency in Baltimore, the agenda for its afternoon two weeks later and every two weeks upcoming CitiStat meeting is hardly secret. It is • thereafter. If an agency is scheduled to appear next set by the performance targets specified in the week at 8:30 on Friday morning, it will also appear turn-around time for the agency's service requests, • at 8:30 on Friday morning two weeks later and or SRs—particularly the agency's priority service every two weeks thereafter. With the exception of requests. And it is set by the discussion at the pre- official holidays, nothing interrupts this timetable. vious CitiStat session as well as the follow-up memo • that the assigned CitiStat analyst sent to the agency Of course, this means that each week the mayor's staff attends multiple CitiStat meetings—at a minimum after that session. And it is set by new developments that emerged during the previous 14 days, such as a • four, sometimes as many as eight. Obviously, this is a newspaper article. significant commitment of one of the city's most valu- • able resources: the time of the mayor's key staff. Moreover, the agency has access to the same data • that the CitiStat staff is using. So agency managers • At the same time, CitiStat is how the mayor and his can easily predict the issues on which the CitiStat staff run the city.They devote so much time to con- staff will focus. Indeed, because the agency has ducting and participating in these meetings, and so appeared at a CitiStat meeting only two weeks • much time analyzing the data and preparing for before, and because the CitiStat staff has followed these meetings, precisely because this is how they and because the mayor's with a memo immediately after that meeting,focus is not spasmodic • learn what is going on within the various city agen- cies; how they track performance improvements; but doggedly consistent, an agency's director and • how they identify problems; how they learn what is 30 IBM Center for The Business of Government • • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY managers should have little difficulty predicting the session cover all of the issues raised in the analyst's menu of questions that they will be asked. Conse- previous day's memo. • quently, the agency can prepare itself to respond to • questions about its performance on various SRs, Q. How can the agency influence the agenda? particularly its priority SRs. A: By coming to the meeting prepared. • Of course, the agency might have been subjected to The agency directors cannot control the flow of the some media attention during the previous two weeks meeting.They can neither influence the questions that could raise new questions. But the agency will raised nor the order in which they are raised. Still, • know about any such issues and prepare for them, too. if they are prepared—if they know the questions Agency managers should be surprised by only one they are likely to be asked—they can have not only answers but also questions of their own.They can kind of problem. CitiStat staff—and particularly its take advantage of the topics introduced to raise their investigator—are always traveling around the city. own concerns and priorities. Some agency directors So are the mayor, first deputy mayor, and other mem- have established a reputation for entering their bers of the mayor's staff.And they are vigilant, always CitiStat sessions with their own agenda. Then, as on the alert for problems. And although agency the meeting progressed, they took advantage of the • managers may well be doing the same thing, they questions, problems, and issues to introduce their • might not observe (or recognize) the same problems. own ideas and specific requests. Thus, in reality, the agenda for any CitiStat meeting Of course, if the agency is undergoing some dogged is set by the performance of the agency. questioning over an obvious or repeated deficiency, Still, the mayor's staff cannot examine all of the its managers will need to find another, more auspi- • possible issues during an hour-and-a-half CitiStat cious time to introduce these ideas or requests. But given that as CitiStat evolved, and it was rarely as session.They have to make some choices. brutal as depicted in the popular press, most agency • The first choice is made by the CitiStat analyst, who managers could find an opportunity—perhaps intwo weeks, perhaps in four or six—to raise their • the day before the meeting prepares a memo (usually concerns. And given that as CitiStat evolved, and the about a dozen pages) to the first deputy mayor and defiant and ineffective agency directors left the city, CitiStat director (but not the agency) outlining the those who remained established enough credibility key issues and suggesting how much time should be and credit with the mayor's office to get a fair hear- spent on each.Then, in a brief discussion in the ing—if not always a yes. office of the CitiStat director(just down the hall from the CitiStat room), the three of them go over the issues, identifying the ones that they need to cover. Q: Who conducts the meeting? Of course, the actual flow of the session never A: Usually the first deputy mayor. • quite follows this plan. Some topics take less time During the administration of Mayor O'Malley, than originally allotted. More often, the discussion the majority of the meetings were conducted by becomes stuck on a significant issue.This may be Michael Enright, the first deputy mayor. because the problem is complex and thus requires time for the mayor's office to articulate its definition Q. What should be the mayor's role at the of the problem and for the agency to explain what meeting? it has done, is doing, and is planning to do.This may be because the agency does not have adequate A: Both a little and a lot. answers for the questions or even an adequate A little, in the sense that the mayor needs to neither explanation of what it plans to do to deal with the conduct nor attend the meeting.A lot, in the sense problem. This may be because it takes time for that the mayor needs to clearly confer authority on everyone in the room to work out and agree on a the person who does run the meeting. new approach to the problem. Rarely does a CitiStat • www.businessofgovernment.org 31 WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • Q: What is the role of the mayor's staff and This does not mean that the mayor's staff has an • the CitiStat staff at the meeting? incentive to ignore serious problems or to fail to press agencies to resolve them. Rather, it simply • A: To ask more questions, offer suggestions,and provide support. means that the mayor's office has an incentive, if it •wants to motivate the people in the room to do their Although the first deputy mayor (or the director of best to resolve these problems, to treat every indi- CitiStat) runs the meeting and controls the move- vidual with professional respect. Only those who . ment of the agenda from topic to topic, other mem- repeatedly demonstrate that they are unworthy of bers of the mayor's staff contribute their own questions, such respect—primarily because they deny that a comments, and suggestions. problem exists or because they refuse to attack the • problem intelligently—lose the presumption that • In addition, as the various participants in the they deserve respect. meeting analyze, discuss, and debate what should be done about a problem, one or more of the When an agency (or a subunit within an agency) central support agencies may be called upon for consistently fails to solve problems that have been ideas or resources. How can the human resources identified and repeatedly discussed, the mayor's staff department help solve an important personnel will be tough.They care about the results, and if S difficulty? Is there some way to resolve a ticklish one individual or one group repeatedly demon- • legal dilemma? strates an inability to produce them, they will not be • happy—and they will show it. No one in the room Indeed, during a CitiStat meeting, it is essential to will fail to get the message. Still, by the time that have in the room representatives ep esentatives from these key one individual has performed so badly as to deserve support agencies: budget, human resources IT public reproach, there will be little sympathy left in legal, and labor.As an issue is being discussed, it the room. Indeed, most of the failing manager's col- may become obvious that to fix a key problem, leagues will be embarrassed, because they believe • one of these agencies may need to do something, that their incompetent colleague is making their too.And if the agency is not part of the discussion, agency—and themselves—look bad. the issue cannot be resolved until later. Consequently, • CitiStat works best when each central support Q. What doesn't happen at a CitiStat meeting? agency is represented by someone who is autho- • rized and willing to make decisions and provide A: The agency director does not deliver a cooperation and assistance. formal presentation. In Baltimore, a CitiStat session is not show-and-tell. Q: What is the tone of a CitiStat meeting? The meeting does not begin with the agency direc- A• Professional even personal, skeptical but for delivering a PowerPoint presentation.The agency • . friendly, interrogatory yet occasionally director does not control the agenda. Rather, the congratulatory, but rarely abusive. first deputy mayor asks the lead question for each topic and determines when it is time to move to a • Some popular and academic depictions of CitiStat new one. have described the meeting as a "star chamber" affair.The clear implication is that the mayor's staff The Preparation for the Meetings treats agency managers and city employees with • abusive contempt and imposes on them arbitrary Q• How does the CitiStat staff prepare for the punishment. meeting? In fact, most of the discussion is professional, even A: By doing a lot of analysis, summarized in friendly. Everyone in the room knows everyone else; a day-before memo. they are all on a first-name basis.They know that The analyst assigned to the agency begins work for they will see each other again in two weeks, if not the next meeting after the previous meeting; by 5:00 • before. No one has a personal incentive to be arbi- that day, the analyst prepares a memo outlining the trary or abusive.38 decisions made and actions requested during that 32 IBM Center for The Business of Government • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • meeting—a memo that might be described as a Department of Recreation and Parks has created . preliminary draft of the agenda for the next meeting. ParkStat. And Baltimore Housing (which includes • Then, the analyst keeps track of what progress is the Department of Housing and Community • made on these various items, perhaps making a field Development, and the city's Housing Authority) visit or two to observe whether anything is really even created its own room in which to conduct its • happening.Then, when the next set of two-week SR HousingStat meetings.39 • data becomes available, the analyst examines these • data. Finally, the day before the next meeting, the analyst prepares a memo for the first deputy mayor • and CitiStat director outlining the key issues for the • next day's agenda. Q: How do the first deputy mayor and CitiStat director prepare for the meeting? A: They read and think. • They read the memo prepared by the CitiStat ana- lyst, contemplating which of the many issues dis- cussed in the memo deserve the most attention. • Then, 10 minutes before the meeting, they gather in the office of the CitiStat director to decide on the • issues on which they will focus. • Q: How do agency directors (and agency staff) • prepare for the meeting? • A: By conducting their own "AgencyStat" meetings. • After the launch of CitiStat, agency directors quickly learned that it was unprofessional, unacceptable, and • unpleasant to show up at their meetings unprepared. • Consequently, they started holding their own meeting the day before to review the issues raised at the last • meeting, to go over the latest service-request data, • and to discuss the questions they might be asked. In . essence, these initial meetings were like debate prep- aration: "What topics will be on the agenda?What • problems will be raised?What questions will I be • asked? How should I respond to each?" Slowly, however, the agency directors began to fig- ure out that just as the biweekly CitiStat meetings • were the mechanism that the mayor used to run the city, so their biweekly meetings could be the mech- anism that they used to run their agency. Thus, over • time, these meetings morphed from debate prepara- tion into AgencyStat. Some city agencies even have given these meetings their own "-Stat" name:The Health Department has HealthStat, plus DrugStat • and LeadStat(for lead poisoning in children).The • www.businessofgovernment.org 33 • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • • • • • How Does CitiStat Affect Key • • Relationships? • • • CitiStat appears to be a top-down management strat- who could not endure this pressure to improve per- egy. So what is the role of others in city government? formance resigned. Today, no one would accept a • How do the people who work in various parts of top management position in the city if he or she • the city react to the demands and implications of could not work within the CitiStat performance- CitiStat?And what about other stakeholders—how management structure. • do they respond? • Q: How do agency directors and managers • City Council Members react to getting questions from their peers? • Q: What is the role of the members of the city A: This depends upon the tone and validity of • council? the questions. • A• Not much. If the questions are asked in a civil way, and if the • .The City of Baltimore40 has a very "strong mayor" questions focus on crucial concerns about important results produced by the agency, agency managers • form of government.There are 15 members of the perceive them as legitimate. Conversely, if the city council (14 elected from single-member districts, will with the city council president elected at-large). questions are phrased in a nasty or haughty tone, or • if the questions are about trivial issues, the agency The members of the council can reduce items in managers will resent the grilling. If CitiStat becomes • the mayor's budget but not add to it.They need a a gotcha game, agency directors and managers will two-thirds vote to override a mayoral veto.To con- play it as agame—seeking primarily to escape being • duct CitiStat, the mayor needs neither approval nor cooperation from the city council. caught rather than trying to achieve their targets. • Agency Directors and Managers Q: How does a mayor convince agency direc- tors and managers to buy into CitiStat? • Q: Do agency directors and managers buy A: By being persistent. • into CitiStat? If agency managers believe that CitiStat will • A: Now? Yes. disappear as soon as the next management fad • In Baltimore, an agency director or manager who materializes, they will wait it out. Indeed, agency does not embrace (or, at least, accommodate himself managers automatically assume that any new ini- • or herself to) CitiStat will soon be a former city tiative will soon disappear. And they have lots of • employee. At the beginning of CitiStat, agency direc- experience and evidence to support this assump- tors who fought the process were the targets of harsh tion. Consequently, a mayor has only one mecha- treatment at their biweekly meetings.Agency direc- nism to overcome this skepticism: to be consistent • tors who failed to respond to CitiStat's demands to and persistent. • produce results were removed. Agency directors • • 34 IBM Center for The Business of Government • • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • Q: What are the consequences of CitiStat employees and thus from their unions. Many performance for agency directors and union members had come to depend upon this • managers? extra income. Indeed, many had internalized it as part of• of A: Embarrassment or recognition and flexibility. their overtime as part of their informal contractconceivedly ontract with the • In an organization that explicitly and publicly city government and thus considered themselves • examines the performance of specific units—and entitled to this income.Thus, CitiStat quickly gener- thus the performance of the managers of those spe- ated union hostility. cific units—the penalty for poor performance can . be quite subtle yet quite significant: embarrassment Still, there was little the unions could do.The union • before peers. And, of course, managers who fail to contracts contained the usual provisions for over- perform up to expectations receive even more and time, but they did not guarantee such overtime. more careful scrutiny. Consequently, the unions possessed few ways, • within the context of the union contract, that they In contrast, when performance is examined could fight the mayor's efforts to control overtime. explicitly and publicly, managers who do live up to expectations get recognition that few will miss. The unions were further opposed to the implication Thus, by giving agency directors and managers the in CitiStat's push for better results and improved opportunity to accomplish something that is signifi- performance—the suggestion that city employees cant and recognized as such, CitiStat provides them should do more work for less pay. Indeed, many • with the opportunity to earn esteem—esteem from city employees view the mere existence of CitiStat • their peers as well as self-esteem.41 as maligning their work and effort. In addition, of course, managers who do produce Yet, the purpose underlying CitiStat—to produce • their targeted results will benefit from less scrutiny better results for citizens—is difficult to assail fron- • combined with more flexibility. tally. And, as Baltimore has improved performance in a number of visible dimensions, union opposition Agencies to CitiStat has primarily appeared to be self-serving. • What are the consequences of CitiStat • Q� q Q: Should union representatives be permitted performance for the agencies? to attend CitiStat meetings? A: The same as for their directors and • A: Baltimore said: "No." managers. Mayor O'Malley's position was consistent: If the City agencies that produce results that achieve their union lets us attend their strategy meetings, we will service-request targets and perform well in other less invite them to CitiStat. • quantifiable dimensions will earn more flexibility and less scrutiny. Agencies that consistently fail to meet their service-request targets will find them- City Employees selves subjected to much more frequent and much Q. How do the employees respond? more careful review and examination. Consistently poor performance can earn an agency pointed and A: Warily, antagonistically, appreciatively. penetrating interrogation. City employees are not uniform in their reaction to CitiStat. Some saw Mayor O'Malley and his CitiStat Unions primarily as a way of denying them overtime. Others saw it as a way for City Hall to force them to do more Q: How do the unions respond? work. Still others saw it as a way to demonstrate that A• Antagonistically,yet resignedly. they were actually doing something worthwhile. One of CitiStat's initial targets was overtime pay. Naturally, opinions of CitiStat vary among agencies. This quickly generated resistance from frontline With some, the reaction is more positive than • www.businessofgovernment.org 35 WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY others. And certainly it is fair to say that not every not attend their agency's biweekly CitiStat meeting. • city employee believes that CitiStat is wonderful; So they don't see the persistent questioning about nor does every city employee believe it is a disas- the same performance issues.They don't see the dis- • ter. As within any organization, there exists a range cussions of service-request data and performance • of views that reflects individual personalities, indi- against the SR targets.They don't see the body lan- vidual situations, and—certainly for CitiStat—indi- guage during these discussions or hear the tone of vidual performances. praise or reproof. • Q: What are the consequences of CitiStat Consequently, the mayor has to rely on agency • performance for frontline employees? directors and managers to convey the message. A: Not as significant as those for agency And, of course, agency directors can convey this • message through their persistence at their biweekly • directors or managers. AgencyStat sessions. Still, frontline workers can Frontline employees do not attend CitiStat meet- never be sure how much this persistence really ings—at least not regularly. Thus, they are subject comes from the mayor and how much this merely • to neither the praise nor the reproofs that the may- reflects their own agency director's idiosyncrasies. • or's staff regularly distributes (if only in subtle and indirect ways).The performance of specific frontline The leadership of any organization always has a diffi- units is often explicitly examined at CitiStat meet- cult time getting its message to frontline employees. • ings and compared with the performance of identi- Still, when trying to do so, persistence does count. • cal or similar units, yet the frontline supervisors of these units may be blissfully unaware of much of Citizens these discussions. And yet it is through these face- • to-face discussions at the regular CitiStat meetings Q: What do citizens think about the CitiStat • that the mayor's leadership team motivates improved strategy? performance. Consequently, the motivational impact A• Very little. of CitiStat is much less direct or forceful for those • who are not regular participants (or, at least, regular If you walked through downtown Baltimore and observers) in the CitiStat process. asked individual citizens, "What do you think of CitiStat?" the most honest answer you would get is a Q: Do frontline employees buy into the Huh? Citizens don't pay attention to governments • CitiStat concept? management strategies. But they care about the results of those strategies. • A: Some do; some don't. • Some accept it as just another reality of city employ- Consequently, a much more relevant question ment. Some believe in it. Others hate it. Certainly would be: "How satisfied are you with the city's many city employees accept CitiStat, but that atti- response to your service requests?"And, if you sur- • tude is not unanimous. veyed citizens who had called 311 with a specific • service request, you might discover that (although they have never heard of CitiStat or CitiTrack) they Q: How does a mayor convince frontline employees to buy into CitiStat? were happy with what the city actually did. • • A: Again, by being persistent. journalists • To convince city employees of the value of CitiStat, Q. How much attention do journalists pay to • the mayor, his leadership team, and the CitiStat CitiStat? staff have to take the same approach that they do • with agency directors and managers: They have to A: A little more than citizens. • be persistent. Over the years, Baltimore's CitiStat has made a • Unfortunately, the mayor's persistence is less visible good, occasional feature story for newspapers. to frontline employees. After all, most of them do Yet, over the past two years, even The Baltimore Sun 36 IBM Center for The Business of Government • • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • only mentioned it about once a month. Nationally, Baltimore's CitiStat gets mentioned in a newspaper • article about five times a month.42 Other Stakeholders • Q: Do any other stakeholders care about . CitiStat? • A: The Baltimore Neighborhood Indicators Alliance cares. • Although CitiStat is the city's internal management strategy, it has also built a reputation among those • who are working to create indicators for jurisdic- tions and communities. For example, the mayor's office has built relationships with the Baltimore Neighborhood Indicators Alliance (BNIA).43 BNIA • describes itself as "committed to promoting, sup- porting, and helping people make better decisions using accurate, reliable, and accessible data and • indicators to improve the quality of life in Baltimore • City neighborhoods." BNIA has created a set of 40 • outcome indicators for 55 different neighborhoods in the city—it calls them "The Vital Signs"—that are • designed to "take the pulse" of city "neighborhoods • by measuring progress toward a shared vision and desired results for strong neighborhoods, good quality of life, and a thriving, vital city over time.1144 • BNIA even makes use of some of CitiStat's data. For example, one of BNIA's indicators is for "dirty streets and alleys"; for this indicator, BNIA uses the . number of "incidents of dirty streets and alleys" reported to Baltimore's 311 phone system per • 1,000 people.41 • The reputation of Baltimore's CitiStat in the commu- nity indicators movement extends beyond the city limits.The Community Indicators Consortium (CIC) is a network of people and organizations (such as • the Baltimore Neighborhood Indicators Alliance) that focus on the use of various measures of condi- tions, trends, and the impact of government on communities. In 2007, CitiStat won one of the • CIC's "Innovation Awards," which are sponsored • by the Urban Markets Initiative at the Brookings Institution.46 • www.businessofgovernment.org I 37 • • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • • • • • • • What Did Baltimore Accomplish .? • • Mayor O'Malley created a room, collected data, As a key component of the $350 million in financial • analyzed the numbers, held meetings, asked lots of savings, Baltimore frequently features the $30.9 mil- • questions of middle managers, and pushed agency lion that it has accumulated through reduced over- directors for new strategies.Through all of this, the time (see Table 1).48 mayor sought better performance. But what, exactly, • did Baltimore accomplish? Q: What else did CitiStat produce? What were • Results some performance improvements? • A: It filled potholes—lots of potholes, and • Q: What results did CitiStat produce? very quickly. . A• It saved Baltimore money. Among the multiple targets that Baltimore set for completing various service requests, none was In March 2007, at a conference organized by the more visible than Mayor O'Malley's "48-Hour • Community Indicators Consortium a CitiStat staff Pothole Guarantee." In fact, filling potholes was • member delivered a presentation emphasizing two one of the most important SRs for the Department • important impacts of CitiStat: of Transportation. If a citizen called 311 to request that the city fill a pothole, the operator would take • • "Through improved accountability on overtime the data and declare that it would be completed • spending, absenteeism, and managed contracts, within two days. Consequently, an examination of • the program has demonstrated cumulative posi- the city's data for this one priority SR reveals the tive financial impacts of over $350 million in its kind of progress that the city made in tracking its • seven years of existence.This does not include data, in encouraging citizens to request services, • service improvements benefits. and in fulfilling these requests. • "This allowed reinvestment of $54 million in the • previous two fiscal years in children's programs, In the middle of Fiscal Year 2002, Baltimore's • including $25 million in school construction CitiTrack system began collecting and reporting • and renovation."47 biweekly data on the Department of Transportation's • Table 1: Cost Savings from the Reduction and Control of the Use of Overtime, FY 2001—FY 2006 • (in millions of dollars) • FY 2000 FY FY FY FY FY FY FY Total: FY • Fiscal Year Baseline 2001 2002 2003 2004 2005 2006 2007 2001-2007 Overtime 24.1 18.3 15.2 20.2 20.5 20.9 19.9 22.8 137.8 • cost • Savings from n/a 5.8 8.9 3.9 3.6 3.2 4.2 1.3 30.9 baseline • • 38 IBM Center for The Business of Government • • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • performance on various SRs, including its SR for the department took, on average, over seven • filling potholes. An examination of the 14 weeks days to fill a pothole. In mid-June (June 8-21), • (of 26) in FY 2002 for which CitiTrack data are it filled only three potholes—none of them • available (see Table 2) reveals a number of aspects within the two-day target. of the department's performance: • Contrast this with the department's performance • • The number of pothole SRs was relatively mod- four years later. As Table 3 on page 40 illustrates, est for most of the winter, not rising over 100 by Fiscal Year 2006 (the last year for which com- until late March. plete data are available), the department's ability to • Through most of the spring, the number of pot- fill potholes had improved significantly. hole SRs during any two-week period remained • The total number of pothole SRs received during • above 100, although it never went over 200. the year was over 10,000. • The total number of pothole SRs received during . For no two-week period was the total less the 14 (of 26) weeks for which data are avail- than 200. • able was 1,231. • On average, for all but one of the 26 two-week On average, the Department of Transportation periods, the department took less than a day to • achieved its two-day target for most of the period. fulfill these SRs. . • In June, CitiTrack began reporting data not only . On average, for the entire year, the department • on the average time to fill a pothole but also on completed these SR requests in less than the percentage of pothole SRs completed within 0.7 days.49 the two-day target. • Although the department failed to meet the • Performance was particularly poor for two two- two-day SR target 540 times, it did achieve this • week periods. For December 22 to January 4 objective in 94.9 percent of these SRs.50 Table 2: Baltimore Department of Transportation Performance in Fulfilling Pothole Service Requests, • FY 2002 (November 24, 2001 to June 30, 2002) Number of Average Days Percent Filled Number Not Two Weeks Ending New Potholes to Fill in Time Filled in Time • Dec 7 40 0.6 n/a n/a Dec 21 n/a n/a n/a n/a Jan 4 30 7.4 n/a n/a Jan 27 48 1.0 n/a n/a Feb 1 73 1.1 n/a n/a Feb 15 74 0.8 n/a n/a • Mar 1 48 2.0 n/a n/a Mar 15 54 0.4 n/a n/a Mar 29 101 0.9 n/a n/a • Apr 12 138 0.7 n/a n/a Apr 26 123 0.7 n/a n/a May 10 153 0.7 n/a n/a May 24 183 0.4 n/a n/a . Jun 7 165 0.4 95.8 7 Jun 21 1 16.4 0 3 Note:Because CitiStat did not begin recording data until halfway through FY 2002,data is not available for July 1 through November 23. . Sources: 14 CitiTrack Statistical Reports,created for the Department of Transportation from December 12,2001 to June 22,2002. www.businessofgovernment.org 39 WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY Table 3: Baltimore Department of Transportation Performance in Fulfilling Pothole Service Requests, • FY 2006 (July 1, 2005 to June 30, 2006) Number of Average Days Percent Filled Number Not • Two Weeks Ending New Potholes to Fill in Time Filled in Time Jul 15 312 0.7 92.2 25 • Jul 29 288 1.2 90.1 28 • Aug 12 289 0.4 96.0 12 Aug 26 261 0.5 95.5 12 • Sep 9 234 0.9 96.2 9 • Sep 23 252 0.5 96.4 9 Oct 7 212 0.5 98.1 4 Oct 21 246 0.9 94.1 15 Nov 4 286 0.5 98.6 4 • I', Nov 18 268 0.4 95.2 13 • Dec 2 268 0.4 98.2 5 • Dec 16 223 0.9 85.5 32 • Dec 30 315 0.5 99.1 3 Jan 13 510 0.5 99.0 5 Jan 27 514 0.6 96.5 18 • Feb 10 519 0.9 97.5 13 • Feb 24 588 0.8 92.0 47 • Mar 10 454 0.5 98.7 6 • Mar 24 486 0.5 98.6 7 Apr 7 449 0.4 99.8 1 Apr 21 596 0.7 93.6 39 May 5 553 0.5 97.0 17 S May 19 519 0.8 89.6 54 • Jun 2 408 0.6 96.5 15 Jun 16 649 0.9 85.6 93 Jun 30 416 0.9 j 87.1 54 Sources:26 CitiTrack Statistical Reports,created for the Department of Transportation from July 16,2005 to July 1,2006. • The department met the two-day SR target less that citizens had learned that calling 311 to request than 90 percent of the time during only four of that the city fill a pothole actually got the pothole • the 26 two-week periods (one in December, one filled. Indeed, despite the increase in the number of in May, and two in June). pothole SRs, the city was, on average, certainly filling The most interesting observation is simply that in potholes quicker in FY 2006 than at the end of FY • four years the number of citizens calling 311 to 2002. And, although the data required to make this request that the city fill a pothole essentially quadru- calculation are not available, it would also appear pled-from perhaps 2,500 (or less) in FY 2002 to that the city was also fulfilling more of these SRs over 10,000 in FY 2006.This could be because the (on both an absolute and percentage basis) within A weather in Baltimore was considerably worse in its two-day target. FY 2006 than in FY 2002 or because the streets of the Still, in FY 2006, the department did not live up to the • city had deteriorated significantly in the intervening mayor's 48-hour pothole "guarantee."Although it kept three years. But another reasonable explanation is • its average completion time under one day for all but . 40 IBM Center for The Business of Government • • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY one of its 26 reporting periods (and that was during For example, in January 2002, the Bureau of Water • the summer), it nevertheless missed the 48-hour target and Wastewater had 36 SRs; by June, the number • for 540 requests, or 5.3 percent of the time. had increased by 50 percent to 55. Of these 55, the city had classified nine of them as "Priority • Q• How long did it take to get real results? Service Requests," often referred to as the "Magic • 9" or the "Priority 9."When the Bureau of Water A: Too long—and yet quicker than in a lot of and Wastewater appeared for its biweekly CitiStat other cities. session, these nine SRs were often on the agenda. Baltimore launched CitiStat in the summer of 2000. Two years later, Baltimore's approach to producing Over the years, Baltimore modified these priority . results was only beginning to collect real data on SRs in two significant ways. First, it tightened its time • results and only beginning to have an impact on its commitment for many of them. Second, it expanded Department of Transportation's ability to fill potholes the priority list. (See Table 4 on page 42.) quickly.This, however, is not surprising. After all, changing the behavior of any large organization For example, the target time to clean up "Rip Rap" (public, nonprofit, or for-profit) is very difficult. (debris left over at the end of a construction project) Nevertheless, within two years, Baltimore's CitiStat was cut in October 2003 from 14 days to seven • was beginning to have the desired impact. Six years days. Then, in January 2006, the target time was • later, the impact was significantly bigger. cut further from seven to four days. How many cities can accurately report today that, At the same time, when the Bureau was unable to • in the past year, it filled 9,575 potholes in less than meet a target consistently, the mayor's office relaxed • two days? How many cities can simply report with it, only later to tighten this target again. For example, any reasonable level of accuracy how many pot- in February 2002, when Baltimore created a target holes it filled in the past year in less than two days? for fixing low water pressure, it began at 10 days. How many cities can simply report with any reason- But a month later, with only a few such service • able level of accuracy how many potholes it filled in requests, this target appeared to be too long and was the past year within any specific target period? reduced to two days. By April, as the number of these service requests increased significantly, the two-day • New York City can. It reports that, in fiscal year target seemed too short, and it was moved up to 2006, it received 45,228 calls to 311 about pot- seven days. A year later, in March 2003, it was holes, and the city's Department of Transportation relaxed further to 14 days.The Bureau was, how- repaired approximately 99 percent of them within ever, slowly improving its performance; thus, in 30 days of notification. Indeed, New York boasts that March 2005, Baltimore cut back this target to 10 this was "the highest rate ever reported." During days; and, in January 2006, it made this target even FY 2006, New York City reports, it repaired a total tighter: five days. • of 179,728 potholes." • Indeed, Baltimore was constantly evaluating these Q: Did Baltimore do anything else besides fill targets. If an agency is consistently fulfilling an SR in • potholes? less than the target time, the mayor's office will ask • whether it could be lowered. Indeed, agency direc- A: Yes, it improved its performance—and tors know they can get such a request. Consequently, tightened its targets—for a number of they too are thinking about what targets they should • priority service requests. consider reducing and to what new level.An agency SWhenever a citizen calls 311 with a service request, director does not want to get surprised with such a • the operator gives the caller both a service-request request; thus, as performance improves, the agency's number and a target time (in days) in which the city director and managers will discuss what is possible commits to fulfilling the request. Over the years, the and what they should propose. • city has added SRs and tightened its target times for numerous SRs. www.businessofgovernment.org I 41 • • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • low Table 4: Baltimore Bureau of Water and Wastewater Target Resolution Times in Days for Priority • Service Requests, 2001-2006 • Target Resolution Times in Days • Jan 7 Jul 1 Jul 1 Jul 1 Jul 1 Jul 1 Priority Service Requests 2002* 2002 2003 2004 2005 2006 The Original Sewer Overflow 1 1 1 1 1 "Priority 91 Sewer Water in 3 1 1 1 1 1 ' Basement • Rip Rapt 14 14 14 7 7 4 Storm Inlet Choke 30 14 14 10 10 7 • Discolored Water 2 7 7 7 3 Exterior Water Leak 15 5 7 4 4 2 • Low Water Pressure 7 14 14 10 5 No Water 1 3 1 1 1 Water in Basement 3 3 5 2 2 2 Three additional SRs Sidewalk Repairs' 30 30 30 30 30 • added to the priority list to create the Asphalt Paving 15 15 15 15 15 Re airs' current 'Priority 12`4 p • Taste & Odor 10 10 10 10 10 Complaints* • *This is the earliest date for which a CitiTrack Statistical Report is available. • t"Rip Rap"is the debris left on the street at the end of a construction project. • f On September 25,2004,the original list of nine Priority Service Requests was expanded to 12. • Source:CitiTrack Statistical Reports,Bureau of Water and Wastewater(created on:01-07-2002,07-06-2002,07-05-2003, • 07-17-2004,07-16-2005,07-15-2006)and CitiStat staff. • Q: How good is Baltimore's city government for every 100 citizens who want a pothole filled, • at producing results? five citizens are disappointed. Or are they? Do A• Quite good—but it would not claim to be these citizens really get out their stopwatches when . perfect. they call 311? Do they carefully write down the • time when they call and then check it 48 hours Was the Baltimore Department of Transportation's later? Or were 95 of them pleasantly surprised pothole-filling performance inadequate, adequate, when they discovered that their pothole had been • successful, or triumphant?To answer that question, filled within less than 48 hours—indeed, within one must first answer the more fundamental question: less than 24? "Compared with what?" Compared with past perfor- • mance, Baltimore certainly improved. Moreover, I When Federal Express delivers your package at know of no other city that is prepared to claim that 10:31, FedEx counts it as late. But do you really care? it fills 95 percent of its potholes within 48 hours of receiving a request. Of course, for a city to make Q: Is service-delivery time the best way to this claim, it must first have a mechanism (311 or measure Baltimore's performance? S something equivalent) for soliciting such service requests and collecting such data. A: Mayor O'Malley thought so. What about the quality? Did the city fill a pothole Still, for every 100 pothole SRs it received, Balti- only to have to refill it two months, or two weeks, more does miss its target on five of them. That is, 42 IBM Center for The Business of Government • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • or two days later? How many of the 10,115 pot- In reality, any change in the behavior and perfor- holes were repeat offenders?The city is not yet mance of a public agency is the result of the inter- keeping such recidivism data. active effect of numerous causes—including both managerial initiatives and environmental factors. What about the city's transportation system? Does After all, Mayor O'Malley's decision to focus on • filling potholes really improve transportation in the improving the performance of city agencies as • city? Should the mayor attempt to develop a new measured by the time it takes them to fulfill citizens' kind of street surface that would eliminate potholes service requests reflects his own judgment about and thus the need for filling them? Should the mayor what Baltimoreans needed their city government • assign the city's smart analytical staff not to tracking to do. Other mayors at other times or in other cities, • potholes but to figuring out what kind of transporta- facing different environmental factors, or even tion system the city needs in the future? another mayor of Baltimore in 2009, could easily choose other priorities. • These are, of course, questions of policy priorities. • And certainly all species of elected officials—which In Baltimore, in 2000-2006, the external circum- includes all mayors—would like to be known for stances in which Mayor O'Malley found the city their sagacious, big-think ideas. Still, mayors of very contributed not only to his decision to create . populous, centuries-old, extremely dense cities have CitiStat but also to how city agencies and city • a more basic priority.They can, perhaps, get away employees behaved (both because of CitiStat and with some big-think pretensions, but only after they independent of it). In fact, in 2000-2006, public ihave demonstrated that they can improve the deliv- managers and public employees in Baltimore might ery of basic city services. have recognized—if only implicitly—that the perfor- mance of city agencies had deteriorated significantly By filling lots of potholes quickly, a mayor does not and that they needed to begin producing results again. win the Nobel Prize in physics. But by fulfilling this • and a variety of other service-delivery requests from Moreover, Baltimore did not make just one simple, citizens, a mayor can win re-election. isolated change. Baltimore's CitiStat approach is not just one tactic but a comprehensive and com- Cause and Effect plex strategy with many features and qualities. • These include the data, the meetings, and the fol- Q: Can Baltimore"prove" that CitiStat was low-up.They include analytical talent, managerial the cause? effectiveness, and mayoral leadership. Some of these features are easier to specify than others. For exam- ple, it might be possible to define how much more Any change in the results produced by a public Baltimore used data in 2006 compared with what it • agency has many causes. Rarely does a public used in 1996 or what St. Louis used in 2006. But • agency take only one action while carefully holding what indicator could categorize the level of mayoral the rest of its behavior faithfully constant.And even leadership in Baltimore in 2006 and compare it with if, to examine the impact of this one action, the the city's mayoral leadership in 1996 or with New • agency tried and was able to do so for a long Orleans' leadership in 2006?And even if all this • enough period of time (years? decades?), there were possible, how much of any improvement in would still exist a variety of external factors that the performance of the Department of Transportation are constantly changing and which do—or, at least, can we attribute to the data? How much to leader- might—have an impact on the results. And it is ship? How much to the impact of sunspots? It is • difficult to rule out any impact from many of these impossible to say. potential causes. After all, for any improvement in • any kind of performance, one possible explanation For public managers, it is almost impossible to • is always regression towards the mean.52 And who prove cause and effect." can conclusively eliminate the possible effects of the alien spacecraft that crashed in Roswell, New Nevertheless, Baltimore tells a very plausible story. • Mexico, on July 4, 1947? And the story is not just about potholes. It is a story . www.businessofgovernment.org 43 WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • about improvements in numerous SRs in numerous • city departments. Baltimore has not morphed into nirvana, nor has its Department of Transportation • transformed itself into an organization that Fred • Taylor, Luther Gulick, Herb Simon, and Peter Drucker would all seek to immortalize. Still, Baltimore has made progress—indeed, given • where it started, significant progress. And it is hard to deny that its CitiStat strategy has been one conse- • quential cause of this progress. • Q: What results can other mayors in other • cities expect CitiStat to cause? A: A lot, or maybe a little, or maybe nothing. Another mayor in another city cannot simply copy Baltimore's CitiStat. Any mayor has to adapt its core concepts to his or her city's current needs. After all, during Martin O'Malley's seven years as mayor, Baltimore's CitiStat was not fixed, but frequently changing. Indeed, the same is true for Baltimore's current mayor, Sheila Dixon; she will have to do the • same thing. She too must continually adapt the use of the CitiStat strategy to Baltimore's current needs and her own priorities. • A mayor can go through the CitiStat motions— creating a fancy new room, generating lots of numbers, and holding impressive meetings—and • produce no real improvement in city-agency perfor- • mance. Another mayor can employ this performance strategy in a way that hardly looks like Baltimore's , CitiStat—and yet ratchet up performance significantly. • CitiStat guarantees nothing. 44 IBM Center for The Business of Government • • • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • • • • • • • • What Is the Future of CitiStat? • • • Is CitiStat just another fad soon to be superseded by Indeed, within New York, a number of city agencies • the next big management thing imported from New have adapted the CompStat concept.The city's • York, New Zealand, or New Pig54? Or is the CitiStat Human Resources Administration has JobStat and performance strategy sufficiently robust that it can VendorStat.The Administration for Children's Ser- • be adapted by many public executives, even if they vices has ChildStat.The Department of Correction • give it their own configuration and brand name? has TEAMS (which stands for Total Efficiency Accountability Management System, but which is • Uniqueness essentially CorrectionStat).The Department of Pro- bation has STARS (which stands for Statistical Track- Q: Could the CitiStat performance strategy be ing, Analysis & Reporting System, but which is effective in a jurisdiction that did not have essentially ProbationStat). Even the New York City • a strong-mayor form of government? Off-Track Betting Corporation has BET-STAT (which • A: Sure. stands for "Branch Efficiency Through STATistics") iCitiStat requires leadership—not a particular organi- Moreover, in Baltimore, almost every city agency • zational structure. No characteristic of the CitiStat that regularly appears at CitiStat has created its own iapproach restricts its effectiveness to a city with a internal "AgencyStat."These are usually scheduled a strong mayor. It could easily be employed in a day or two before the agency's CitiStat session and • municipality with a council and city-manager form were originally designed to help the agency director of government." predict and prepare for the questions that the agency • would be asked at CitiStat. Most AgencyStat sessions Moreover, CitiStat is not uniquely useful for a munici- have, however, moved beyond merely being a kind • pality. It could be employed at the state level. Indeed, of debate-preparation exercise. Now, in Baltimore, iin 2005, Governor Christine Gregoire of Washington most agency directors use their AgencyStat internally created her own version, called GMAP (for Govern- to run their organization. Indeed, the Health • ment Management,Accountability, and Performance). Department has multiple AgencyStats; in addition to • And in 2007, Martin O'Malley, after he was elected the department-wide HealthStat, it also employs • governor of Maryland, created StateStat. LeadStat, KidStat, BabyStat, and DrugStat. • Q: Could the CitiStat performance strategy Evolution • work within an agency itself? • Q: Did Baltimore's CitiStat change over time? A: It has. A: Certainly. After all, the original version of this performance • strategy was CompStat, developed by the New York CitiStat began using the available data—primarily • financial and personnel data. And, as a result, much City Police Department. CitiStat is an adaptation of of the initial focus was on such administrative issues CompStat. • as controlling the use of overtime and unscheduled • www.businessofgovernment.org I 45 i WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • leave.Then, as Baltimore created its CitiTrack data- (specifically Michael Enright, his first deputy base and linked it to its 311 phone system, it began mayor, and Matthew Gallagher, the director to shift its focus to the delivery of specific service of CitiStat"), Dixon promoted Christopher • requests and the achievement of specific targets for Thomaskutty (a deputy director of CitiStat under • them. Finally, as performance against these targets O'Malley) to head her CitiStat operation. improved, the city shortened the time. Q: What kind of approach will increase the Q: How should a CitiStat performance probability that the next mayor will con- strategy evolve over time? tinue using CitiStat to manage a city? • A: By responding to new performance deficits A: Produce real, visible results! and mayoral priorities. A CitiStat-style performance strategy is of no value • CitiStat should not be a rigid process.Yet, after a unless it helps government produce better results. few years, everyone may have settled into a routine. If a mayor effectively employs CitiStat to ratchet up Every two weeks, the data are analyzed and prob- the performance of city agencies, the succeeding lems identified. Every two weeks, the meetings are mayor is likely to conclude that it is useful. If a • held, progress reported, problems debated, solutions mayor creates a CitiStat but is unable to show that it suggested, decisions made, and follow-up sent. can indeed have an impact on agency performance, Every two weeks, the routine starts all over again. the successor will quickly discard it. • This routinization is very predictable and very reason- Thus, sustainability depends on two factors: able. A city is a very operational unit of government. It never solves the pothole problem or the crime • Real, better, visible results • problem or the restaurant-health problem. It cannot A credible explanation of why and how the • forget about these recurring problems.After a suc- cessful year, it cannot take the next one off. It cannot strategy worked • relax. Neither the mayor, the CitiStat staff, nor the agencies can assume that they have fixed any of Certainly, the new chief executive will need to assert that he or she has made "sign if icant —even essen- these problems. Even If a particular measure has tial"—upgrades to the strategy.The new executive • revealed some significant improvements, and even if • will want to claim that some critical deficiencies the actions of the relevant agencies have contributed have been fixed.The new chief executive may give directly and unquestionably to this improvement, nei- ther the mayor nor the agency can slow down. If the the strategy a new and flashier name. Still, any chief • executive who inherits a performance strategy that is agency, the mayor, and the CitiStat staff fail to be • vigilant, the successes can quickly disappear. working—and is widely believed to be working— will want to continue (but, of course, improve) the • sustainability use of that strategy. • The new chief executive could try to create his or Q: Is it possible to ensure the continuity of her own performance strategy. But this is costly— executives? • a CitiStat at strategy over a change in chief costly in dollars, costly in personal time, costly in staff time, and costly in the inevitable setbacks and • A• Sure. It already has. mistakes. Only if the new chief executive believes that there exists something that is obviously and sig- nificantly superior—whether it comes from New governor of Maryland. On January 17, 2007, York or New Zealand or New Pig—will the old O'Malley was inaugurated as governor, and Sheila performance strategy be replaced. • Dixon, president of the Baltimore City Council, took • the oath of office as Baltimore's mayor. She has In Baltimore, Mayor O'Malley created the 48-hour enthusiastically continued O'Malley's CitiStat strat- pothole guarantee. If a citizen calls 311 with a egy. And although O'Malley took several of his service request for filling a pothole at a specific • staff that were central to the operation of CitiStat 46 IBM Center for The Business of Government • • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • location, O'Malley guaranteed that it would be filled within 48 hours. What are subsequent mayors • going to do? Create the 72-hour pothole guarantee? • And if future mayors conclude that CitiStat helps to achieve the 48-hour pothole target, they are unlikely • to cavalierly discard CitiStat. • To predict whether a CitiStat will be sustained, sim- ply check to see if it is clearly producing visible iresults that are important to citizens. i www.businessofgovernment.org I 47 • • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • • • • • • • Endnotes • • 1. In November 2006, Martin O'Malley was elected Center)and by the State of Washington (which calls governor of Maryland. its adaptation GMAP, for Government Management, • 2. Christopher Swope, "Restless for Results: Accountability, and Performance). • Baltimore Mayor Martin O'Malley is tracking perfor- 6. In this report, I use the word agency to refer to mance on a scale never seen before in local government. any unit within city government that reports at a CitiStat • He wants change, in a hurry," Governing, April 2001, meeting(regardless of whether its official name is depart- • pp. 20-23. Ellen Perlman, " 'Stat' Fever:The practice ment, division, bureau, section, office, or agency). I use • of collecting data to monitor and improve government the phrase agency director to refer to the top executive performance continues to gain momentum and evolve," official of this agency.And I use the phrase agency man- • Governing,January 2007, pp. 48-50. agers to refer to those executive officials who manage a • 3. Lenneal J. Henderson, The Baltimore CitiStat subunit within that agency, a subunit that is big enough or Program:Performance and Accountability(Washington, important enough so that its manager consistently attends D.C.: IBM Center for The Business of Government, 2003). CitiStat meetings. • Worcester Regional Research Bureau, CompStat and 7. Fortunately,there exists no International • CitiStat:Should Worcester Adopt These Management Organization for Standardization (ISO)that certifies whether Techniques, Report No. 03-01,Worcester, Mass.: February a city has implemented an acceptable, accredited version of • 18, 2003. Cathy Sharp,Jocelyn Jones, and Alison M. CitiStat. Still,you better hurry up and get your city's CitiStat Smith, What Do We Measure and Why?:An Evaluation up and running,and recognized as effective, before some • of the CitiStat Model of Performance Management and guardian of a new ISO standard of which you have never its Applicability to the Scottish Public Sector(Edinburgh, heard knocks on your door and asserts that you can't claim • Scotland: Scottish Executive Social Research, 2006). to be a recognized CitiStat unless you comply with their • 4. See the Government Innovators Network,The Ash regulations,answer their questions,jump through their • Institute for Democratic Governance and Innovation,John hoops, and qualify to hang their certificate on your wall. F. Kennedy School of Government, Harvard University: 8. CitiStat could also be employed in a city-manager • http://www.innovations.harvard.edu/awards.html?id=3638. form of government. In such a city, the strategy would be • 5. Although this report focuses on CitiStat as a implemented by the city manager, but the specific results leadership strategy that mayors can use to improve the to be achieved would need to be worked out and agreed • performance of city agencies, the key components of the to by the city manager and the city council. • strategy can be employed by elected chief executives 9. Time is an even scarcer resource than money. • at any level of government—and, indeed, by the heads Indeed, when it comes to time, no human is richer of public agencies(whether they are political appoin- than any other. Every one of us has exactly 168 hours • tees or civil servants).After all, CitiStat is an adaptation in a week. Some may need to sleep less than others. • of CompStat, which was created by the New York City Nevertheless, every individual has only 168 hours to • Police Department(and which has also been adapted by allocate among sleeping, eating, working, and playing. a variety of public agencies in New York City). Moreover, Moreover, you cannot save any time in a "clock bank" • the CitiStat-CompStat approach has also been adapted by for next week.You cannot borrow any time from someone • state governments—by the Ohio Department of Job and else. Once the week is over, you have spent every one of Family Services (which calls its approach the Performance your 168 hours, and you can't get any of them back. • • 48 IBM Center for The Business of Government • • • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • 10. Overtime is a serious problem in lots of govern- (Washington, D.C.: IBM Center for The Business of • ment jurisdictions.This is because each decision to pay Government, 2004), p. 23. overtime is made by a middle manager or frontline super- 16. http://www.ci.baltimore.md.us/news/citistat/ visor who has an immediate problem: Something—a job, index.html. responsibility, duty, or task—needs to be covered, and 17. For additional information on Baltimore's 311 • none of those on regular duty are available to cover it. system, see: http://www.ci.baltimore.md.us/news/311.html • But a member of the existing staff—someone already on and http://www.baltimorecity.gov/news//press/020328.htmi. the payroll—can (and will)cover it—but only if he or she 18. Weick, "Small Wins." • is paid overtime. In the immediate short run, this is (for 19. For a discussion of this concept, see: Mark H. • the middle manager or frontline supervisor)a very logi- Moore, Creating Public Value:Strategic Management in • cal decision. (This is particularly true if the organization's Government(Cambridge, Mass.: Harvard University personnel budget contains no slack—if, that is, the orga- Press, 1995). • nization staffing levels are based on the assumptions that 20. In the public sector, the best direct connection • no one is ever sick and that no emergencies ever arise.) with which I am familiar between an agency output and • Indeed, only if the organization undertakes the necessary the desired outcomes is the immunization of children planning and scheduling to ensure that all but the rarest of against measles.As one study reveals, the production of • jobs, responsibilities, duties, and tasks can be covered by the output has a greater than 99 percent chance of pro- s • the regular staff will the pressure to employ overtime to fix ducing the desired outcome. "Studies indicate that more immediate short-term problems disappear. than 99%of persons who receive two doses of measles Of course, many employees prize overtime oppor- vaccine(with the first dose administered no earlier than • tunities. Indeed, many employees view their overtime pay the first birthday) develop serologic evidence of measles . as an entitlement—no different from their regular salary. In immunity."William L.Atkinson, Charles Wolfe, Sharon addition, middle managers and frontline supervisors may G. Humiston, Rick Nelson (eds.), Epidemiology and • view it as a form of patronage.They can offer overtime to Prevention of Vaccine-Preventable Diseases, 1 Oth edi- • individuals on their staff as a reward for good work or as a tion (Atlanta, Ga:The Centers for Disease Control and • quid pro quo for something else. Prevention, 2007), p. 139. Consequently, controlling overtime is not only 21. In New York City, the Fund for the City of New • an administrative task; it is also a personnel challenge. York working with the Mayor's Office of Operation cre- • In many government jurisdictions, management's efforts ated the Street Cleanliness Scorecard.The purpose was to change overtime practices have created bitter fights in to shift the evaluation of the Sanitation Department—and both the legislature and the courts. thus its efforts—from merely focusing on the output of its • 11. Robert D. Behn, "Management by Groping work(the trash collected)to the outcomes(how clean the Along," The Journal of Policy Analysis and Management, street actually is). For an explanation of these scorecards Vol. 7, No. 4 (Fall 1988), pp. 643-663.Also, Robert D. (including examples of the various scores and citywide • Behn, Leadership Counts:Lessons for Public Managers scores from FY 1975 through FY 2006), go to: http://www. • (Cambridge, Mass.: Harvard University Press, 1991), nyc.gov/html/ops/downloads/pdf/Scorecard/about— Chapter 7, "Management by Groping Along," scorecard.pdf. pp. 127-150. For scorecards for recent years (by borough and by • 12. Karl E.Weick, "Small Wins: Redefining the Scale community board), go to: http://www.nyc.gov/html/ops/ • of Social Problems,"American Psychologist,Vol. 39, html/scorecard/scorecard.shtml. No. 1 (January 1984), pp. 40-49. 22. BOMFOG was the acronym that journalists gave 13. Louis Henri Sullivan, "The Tall Office Building to a phrase that Governor Nelson Rockefeller repeat- Artistically Considered," Lippincott's Magazine, edly used in his political speeches: "the brotherhood of March 1896. man and the fatherhood of God."When Rockefeller got 14. For an examination of why even a mayor will to this phrase,journalists knew he was into his conclud- benefit significantly by consulting a variety of other indi- ing oratorical flourish and they could safely leave to start • viduals-maven individuals who are obviously not nearly as writing their story.Today, of course, no politician uses the • clever or discerning as the mayor—see:James Surowiecki, BOMFOG phrase. It is hardly politically correct. Still, they The Wisdom of Crowds (New York:Anchor Books, 2004). each have their own boilerplate rhetoric. • 15. Robert D. Behn, Performance Leadership: 11 23. Why? Because the outgoing mayor ven an out- Better Practices That Can Ratchet Up Performance going mayor from the same political party—wants to leave • www.businessofgovernment.org 49 • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • office with as many friends as possible.And one way to Hall. In particular, Baltimore adopted the idea that there • make friends is to shower money upon them. should be a podium behind which stands the commander • 24. This is not a hypothetical example. I know of at of the precinct being examined. Baltimore also adopted least one mayor who did this. Hiding the CitiStat meetings the concept that everyone else in the room should be sit- lowered the cost of failure, but it also increased the prob- ting.And it adopted the idea that the overall leaders of the ability of failure. Indeed, this CitiStat disappeared (though organization (for example, the chief of department and the • it is actually hard to say that it disappeared since it never chief of patrol, or the first deputy mayor and the director • really "appeared"). of CitiStat)sit at a table facing the podium. • 25. This too is not a hypothetical example. I know of 33. The web address for 311 City Call is: at least one mayor who did this, while also asserting that https://baltimore.CustomerServiceRequest.org. Here, from • this CitiStat was indeed a high priority. a drop-down menu, a citizen can select from 70 different • 26. 1 did not understand this until I visited Baltimore types of service requests or enter a keyword. • for the second time Qanuary 2004). Fortunately, the sec- 34. More information about Motorola's 311 and CRM and set of CitiStat sessions that I observed were for the technology can be obtained from: Motorola, Inc., 1301 • same agencies that I had watched three months earlier East Algonquin Road, Schaumburg, IL 60196; 1-888-567- • (October 2003).Thus, I could quickly see the clear con- 7347, http://www.motorola.com/publicservice. • tinuity of the performance issues being examined.Yet 35. http://www.ci.baltimore.md.us/news/citistat/ most public officials who seek to understand how CitiStat index.html. • works visit Baltimore for only one day. 36. Okay. I accept that this is an exaggeration. • 27. For a discussion of the concept of a government Staffers in any budget office—or, I hope, at least some agency's "performance deficit," see: Robert D. Behn, "On staffers—also want to ensure that the money is spent not • why public managers need to focus on their: Performance just according to the rules but also productively, and, Deficit," Bob Behn's Public Management Report,Vol. 4, if the money is not being spent productively, to modify • No. 1 (September 2006); Robert D. Behn, Performance how the money is spent to improve how productively it is Leadership: I I Better Practices That Can Ratchet Up spent. Nevertheless, such purposes are usually subservient • Performance(Washington, D.C.: IBM Center for The to saving money. • Business of Government, 2004), pp. 10-11. 37. Actually, because CitiStat quickly developed an 28. In terms of Howard Gardner's multiple intel- international reputation, almost every CitiStat session ligences, the CitiStat staffer needs both "logical-math- includes a visitor from some city, other jurisdiction, or S ematical intelligence"and "the personal intelligences." public agency. • Howard Gardner, Frames of Mind:The Theory of Multiple 38. As Robert Axelrod explains in The Evolution of Intelligences(New York: Basic Books, 1983), particularly Cooperation (New York: Basic Books, 1984), people who • chapters 7 and 10. know that they will work with each other in the future • 29. Whenever current data have varied far from have an incentive to maintain good, professional relations. • the norm, regression towards the mean (a reversion to 39. For information on HousingStat, see: traditional behavior) is always a legitimate prediction. http://www.baltimorehousing.org/index/ps—housingstat.asp. • See Francis Calton, "Regression Towards Mediocrity in 40. In Maryland, the City of Baltimore provides not • Hereditary Stature, The Journal of the Anthropological only traditional city services but also the services that Institute of Great Britain and Ireland,Vol. 15 (1886), would otherwise be provided by a county.The City of . pp. 246-263. Baltimore is almost entirely surrounded by Baltimore • 30. This is only roughly true.CitiStat staffers often County, a separate jurisdiction with no responsibilities . move on to other jobs (or other lives), which occasion- inside the city limits. ally leaves some vacancies.And qualified analysts are 41. Robert D. Behn, "On the value of creating: Esteem • not instantaneously available, leaving some holes that the Opportunities," Bob Behn's Public Management Report, remaining staff must cover. Vol. 1, No. 9(May 2004). 31. Jack Maple, The Crime Fighter(New York: Random 42. On June 1, 2007, 1 did a LexisNexis search for House 1999), p. 33. "CitiStat" and "Baltimore." Here are the number of articles • 32. Baltimore borrowed the general framework of that I found over the past two years with these two words • the room from the New York City Police Department, somewhere in the full text. modifying the arrangement of the furniture and technol- ogy to fit within the one room that was available in City • 50 IBM Center for The Business of Government • i WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • Donald T. Campbell and H. Laurence Ross, "The Type of Publication Number of Articles Connecticut Crackdown on Speeding:Time-Series Data • The Baltimore Sun 24 in Quasi-Experimental Analysis," Law&Society Review, Vol. 3, No. 1 (August 1968), pp. 33-54. Maryland newspapers 53 Note that regression towards the mean can explain an improvement in performance from an unusual low or a Southeast newspapers 84 deterioration in performance from an unusual high. • Northeast newspapers 34 53. For a discussion of the difficulty of employing the program-evaluators "gold standard"—the double- Midwest newspapers 2 blind, controlled experiment—to determine what impact iWestern newspapers 1 any management initiative has had, see Robert D. • Major (national) news- Behn, Leadership Counts:Lessons for Public Managers papers 81 (Cambridge, Mass.: Harvard University Press, 1991), ichapters 8 and 9. Magazines and journals 6 54. "New Pig Corporation manufactures the world's largest selection of industrial absorbents for oil spill • Google Scholar found 108 different publications cleanup." See http://www.newpig.com. containing the word "CitiStat."43. For a discussion of the relationship between 55. In fact, the city of Palm Bay, Florida, which has a • city-manager form of government, has created PalmStat. CitiStat and BNIA, see: Marsha R. B. Schachtel, "CitiStat 56. Honest. Check out: http://www.nycotb.com/view- Using the Baltimore Neighborhood Indicators Alliance: Page.cfm?pageld=25. Using Information to Improve Communication and 57. Governor O'Malley chose Enright to be his chief • Community," National Civic Review,Vol. 90, No. 3 (Fall of staff, and Gallagher to be a deputy chief of staff and • 2001), pp. 253-265. director of his StateStat initiative. 44. http://www.ubalt.edu/bnia/about/index.html. 45. http://www.ubalt.edu/bnia/indicators/Expl- San itation-DirtyStreets-VS3.html. i46. http://www.communityindicators.net/. 47. http://www.communityindicators.net/ • documents/CICCitiStatPresentation.pdf. • 48. These are the savings from eight departments: • Fire, General Services, Health, Housing and Community Development, Recreation and Parks, Solid Waste, • Transportation, and Water and Wastewater. • 49. Eyeballing the Average-days-to-fill row suggests • that this number is indeed about 0.7.A calculation from the underlying data reveals that the average for the entire • year is 0.66 days. . 50. Again, this number is not in the table. It can, how- ever, be easily calculated from the Number-not-filled-in- time row. • 51. The Mayor's Management Report, Fiscal • 2006 (New York: City of New York, Mayor's Office of Operations, September 2006), pp. 60-61. • 52. There are two classic references for this phenom- enon: Galton's original 1886 article describing it, and • Campbell and Ross's investigation of how it explained the results of one public-policy initiative. . Francis Galton, "Regression towards Mediocrity • in Hereditary Stature,"Journal of the Anthropological Institute of Great Britain and Ireland, Vol. 15 0 886), • pp. 246-63. • www.businessofgovernment.org 51 • WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • • • • ABOUT THE AUTHOR � Robert D. Behn is a lecturer at Harvard University's John F. Kennedy School of Government and the faculty chair of the School's executive-education program on "Driving Government Performance: Leadership Strategies that . Produce Results." He specializes in governance, leadership, and the manage- ment of large public agencies, and is working on a book on "PerformanceStat." Dr. Behn served on the staff of Governor Francis W. Sargent of Massachusetts, as a scholar in residence with the Council for Excellence in Government, and on the faculty of the Harvard Business School and of Duke University's . Terry Sanford Institute of Public Policy, where he was director of its Governors Center. He has led retreats for gubernatorial offices and cabinets in several states. . He has conducted executive-education seminars in Bangkok, Berlin, Copenhagen, Lisbon, Monterrey, Oslo, Reykjavik, and Sydney, as well as for the United Nations, govern- • mental associations, and federal and state agencies. • Dr. Behn publishes the monthly Bob Behn's Public Management Report. He is the author of Rethinking Democratic Accountability (Brookings, 2001), and he is also the author of Leadership Counts: Lessons for • Public Managers (Harvard University Press, 1991). • His articles include "Why Measure Performance? Different Purposes Require Different Measures," Public • Administration Review(2003); "Rethinking Accountability in Education," International Public Management • Journal, (2003); "The Psychological Barriers to Performance Management: Or Why Isn't Everyone Jumping on the Performance-Management Bandwagon?" Public Performance& Management Review(2002); "Strategies • for Avoiding the Pitfalls of Performance Contracting," Public Productivity and Management Review(1999, • with Peter A. Kant); "Branch Rickey as a Public Manager,"Journal of Public Administration Research and Theory(1997); and "Management By Groping Along," The Journal of Policy Analysis and Management(1988). Dr. Behn's "Cutback Budgeting" won the Raymond Vernon Prize for the article in Volume 4 (1985) of The • Journal of Policy Analysis and Management that"makes the greatest contribution to intellectual discourse on public policy and management." His "Management and the Neutrino:The Search for Meaningful Metaphors" • won the Marshall E. Dimock Award for the best lead article in Public Administration Review in 1992. • Dr. Behn earned a B.S. in physics from Worcester Polytechnic Institute, and a Ph.D. in decision and control • from Harvard University. . He grew up a fan of the Brooklyn Dodgers. At the end of the 1967 season, however, he went to Fenway • Park for the Boston Red Sox' last game. (If you don't understand the significance of this, he will explain it • to you in more detail than you want to know.) • 52 IBM Center for The Business of Government . WHAT ALL MAYORS WOULD LIKE TO KNOW ABOUT BALTIMORE'S CITISTAT PERFORMANCE STRATEGY • • • • KEY CONTACT INFORMATION To contact the author: . Robert D. Behn John F. Kennedy School of Government • Harvard University 79 John F. Kennedy Street Cambridge, MA 02138 (617) 495-9874 • fax: (617) 496-1722 e-mail: redsox@ksg.harvard.edu • website: http://www.ksg.harvard.edu/TheBehnReport i • www.businessofgovernment.org 53 • 0 CENTER REPORTS AVAILABLE COMPETITION, CHOICE, Implementing Alternative Sourcing Federal Intranet Work Sites: An • AND INCENTIVES Strategies: Four Case Studies (2004) Interim Assessment(2002) Edited by Jacques S. Gansler and Julianne G. Mahler and Priscilla M. Determining a Level Playing Field for William Lucyshyn Regan • Public-Private Competition (1999) Designing Competitive Bidding for The State of Federal Websites:The Lawrence L. Martin Medicare(2004) Pursuit of Excellence(2002) • Managing for Outcomes: Milestone John Cawley and Andrew B.Whitford Genie N. L. Stowers • Contracting in Oklahoma(2001) International Experience Using State Government E-Procurement . Peter Frumkin Outsourcing, Public-Private in the Information Age: Issues, A Vision of the Government Partnerships,and Vouchers(2005) Practices, and Trends (2002) • as a World-Class Buyer: Major Jon R. B16ndal M.Jae Moon Procurement Issues for the Coming Effectively Managing Professional Preparing for Wireless and Mobile Decade(2002) Services Contracts: 12 Best Practices Technologies in Government(2002) • Jacques S. Gansler (2006) Ai-Mei Chang and P. K. Kannan Contracting for the 21st Century: Sandra L. Fisher, Michael E. • A Partnership Model (2002) Wasserman, and Paige P.Wolf Public-Sector Information Security: • A Call to Action for Public-Sector Wendell C. Lawther CIOs (2002, 2nd ed.) Franchise Funds in the Federal Don Heiman Government: Ending the Monopoly E-GOVERNMENT The Auction Model: How the Public in Service Provision (2002) Sector Can Leverage the Power of John J. Callahan Supercharging the Employment g Agency: An Investigation of the Use E-Commerce Through Dynamic Making Performance-Based of Information and Communication Pricing(2002, 2nd ed.) • Contracting Perform: What the Technology to Improve the Service David C.Wyld Federal Government Can Learn from of State Employment Agencies The Promise of E-Learning in Africa: State and Local Governments (2002, (2000) The Potential for Public-Private • 2nd ed.) Anthony M.Townsend Partnerships(2003) Lawrence L. Martin , Norman LaRoc ue and Michael Latham • Assessing a State s Readiness for q Moving to Public-Private Global Electronic Commerce: Lessons Using Technology to Increase Citizen Partnerships: Learning from from the Ohio Experience(2001) Participation in Government:The • Experience around the World (2003) J. Pari Sabety and Steven I. Gordon Use of Models and Simulation (2003) • Trefor P.Williams Privacy Strategies for Electronic John O Looney IT Outsourcing: A Primer for Public Government(2001) Seaport: Charting a New Course for Managers (2003) Janine S. Hiller and France Belanger Professional Services Acquisition for • Yu-Che Chen and James Perry Commerce Comes to Government America's Navy(2003) The Procurement Partnership on the Desktop: E-Commerce David C.Wyld Model: Moving to a Team-Based Applications in the Public Sector E-Reporting:Strengthening • Approach (2003) (2001) Democratic Accountability(2004) Kathryn G. Denhardt Genie N. L. Stowers y Mordecai Lee • Moving Toward Market-Based The Use of the Internet in Understanding Electronic Signatures: • Government:The Changing Role of Government Service Delivery(2001) The Key to E-Government(2004) • Government as the Provider(2004, Steven Cohen and William Eimicke Stephen H. Holden 2nd ed.) Jacques S. Gansler State Web Portals: Delivering and Measuring the Performance of Financing E-Service(2002) E-Government(2004) . Transborder Service Systems: Diana Burley Gant,Jon P. Gant, Genie N. L. Stowers Pathways for Innovation or Threats and Craig L.Johnson to Accountability? (2004) Restoring Trust in Government: • Alasdair Roberts Internet Voting: Bringing Elections to The Potential of Digital Citizen the Desktop (2002) Participation (2004) Competitive Sourcing:What Happens Robert S. Done Marc Holzer,James Melitski, Seun • to Federal Employees?(2004) Yong Rho, and Richard Schwester g • Jacques S. Gansler and William Leveraging Technology in the Lucyshyn Service of Diplomacy: Innovation in • the Department of State(2002) Barry Fulton 54 To download or order a copy of a report,visit the IBM Center for The Business of Government website at:www.businessof9overnment.org • From E-Government to M- An Introduction to Financial Risk A Learning-Based Approach to Government? Emerging Practices Management in Government(2001) Leading Change(2000) in the Use of Mobile Technology by Richard J. Buttimer,Jr. Barry Sugarman State Governments(2004) M.Jae Moon Understanding Federal Asset Toward a 21st Century Public Management:An Agenda for Reform Service: Reports from Four Forums Government Garage Sales: (2003) (2001) Online Auctions as Tools for Asset Thomas H. Stanton Mark A.Abramson • Management(2004) David C.Wyld Efficiency Counts: Developing the Labor-Management Partnerships: • Capacity to Manage Costs at Air A New Approach to Collaborative • Innovation in E-Procurement: Force Materiel Command (2003) Management(2001) The Italian Experience(2004) Michael Barzelay and Fred Barry Rubin and Richard Rubin • Mita Marra Thompson Winning the Best and Brightest: • Computerisation and E-Government Federal Credit Programs: Managing Increasing the Attraction of Public in Social Security:A Comparative Risk in the Information Age(2005) Service(2001) . International Study(2005) Thomas H. Stanton Carol Chetkovich Michael Adler and Paul Henman Grants Management in the 21st A Weapon in the War for Talent: The Next Big Election Challenge: Century: Three Innovative Policy Using Special Authorities to Recruit iDeveloping Electronic Data Responses(2005) Crucial Personnel (2001) Transaction Standards for Election Timothy J. Conlan Hal G. Rainey Administration (2005) R. Michael Alvarez and Thad E. Hall Performance Budgeting: How NASA A Changing Workforce: and SBA Link Costs and Performance Understanding Diversity Programs in Assessing the Impact of IT-Driven (2006) the Federal Government(2001) Education in K-12 Schools(2005) Lloyd A. Blanchard Katherine C. Naff and J. Edward Ganesh D. Bhatt. Kellough Transforming Federal Property The Blogging Revolution: Management:A Case for Public- Life after Civil Service Reform: Government in the Age of Web 2.0 Private Partnerships (2007) The Texas, Georgia, and Florida • (2007) Judith Grant Long Experiences (2002) David C.Wyld Jonathan Walters • Bridging the Digital Divide for The Defense Leadership and Hard-to-Reach Groups(2007) HUMAN CAPITAL Management Program:Taking Career Heike Boeltzig and Doria Pilling MANAGEMENT Development Seriously(2002) Can Governments Create Universal Joseph A. Ferrara and Mark C. Rom • Internet Access? The Philadelphia Results of the Government The Influence of Organizational Municipal Wireless Network Story Leadership Survey(1999) Commitment on Officer Retention: • (2007) Mark A.Abramson A 12-Year Study of U.S.Army Abhijit Jain, Munir Mandviwalla, Profiles in Excellence: Conversations Officers(2002) and Rajiv D. Banker with the Best of America's Career Stephanie C. Payne,Ann H. Huffman, • Executive Service (1999) and Trueman R.Tremble,Jr. • Mark W. Huddleston Human Capital Reform:21 st FINANCIAL Reflections on Mobility:Case Century Requirements for the United International Int MANAGEMENT Studies of Six Federal Executives Development States Agency for for Int . (2000) Anthony C. E. Quainton and Credit Scoring and Loan Scoring: Michael D. Serlin Amanda M. Fulmer • Tools for Improved Management of Federal Credit Programs (1999) Managing Telecommuting in the Modernizing Human Resource • Thomas H. Stanton Federal Government:An Interim Management in the Federal Report(2000) Government: The IRS Model (2003) . Using Activity-Based Costing to Gina Vega and Louis Brennan • Manage More Effectively(2000) James R.Thompson and Hal G. Rainey Michael H. Granof, David E. Platt, Using Virtual Teams to Manage Mediation at Work:Transforming and Igor Vaysman Complex Projects: A Case Study of Workplace Conflict at the United • the Radioactive Waste Management States Postal Service(2003) Audited Financial Statements: Project(2000) Lisa B. Bingham Getting and Sustaining "Clean" Samuel M. DeMarie Opinions(2001) Douglas A. Brook . To download or order a copy of a report,visit the IBM Center for The Business of Government website at:www.businessofgovernment.org I 55 CENTER REPORTS AVAILABLE �► Growing Leaders for Public Service San Diego County's Innovation The Challenge of Developing Cross- (2004, 2nd ed.) Program: Using Competition and a Agency Measures: A Case Study of Ray Blunt Whole Lot More to Improve Public the Office of National Drug Control Services(2000) Policy(2001) Pay for Performance:A Guide for William B. Eimicke Patrick J. Murphy and John Carnevale Federal Managers(2004) • Howard Risher Innovation in the Administration of The Potential of the Government Public Airports(2000) Performance and Results Act The Blended Workforce: Maximizing Scott E.Tarry as a Tool to Manage Third-Party • Agi I ity Through Nonstandard Work Government(2001) • Arrangements(2005) Entrepreneurial Government: David G. Frederickson James R.Thompson and Sharon H. Bureaucrats as Businesspeople(2000) • Mastracci Anne Laurent Using Performance Data for The Transformation of the Rethinking U.S. Environmental Accountability: The New York City • Government Accountability Office: Protection Policy: Management Police Department's CompStat Y •Using Human Capital to Drive Change Challenges for a New Administration Model of Police Management(2001) (2005) (2000) Paul E. O'Connell • Jonathan Walters and Charles Dennis A. Rondinelli Moving Toward More Capable Thompson Government:A Guide to Creating a Culture of Innovation: onal Design (2002) Designing and Implementing 10 Lessons from America's Best Run Thomas Organizational Stanton • Performance-Oriented Payband City(2001) Systems(2007) Janet Vinzant Denhardt and Robert B. The Baltimore CitiStat Program: James R.Thompson Denhardt Performance and Accountability Managing for Better Performance: Understanding Innovation:What (2003) g g P Enhancing Federal Performance Inspires It?What Makes It Lenneal J. Henderson Management Practices(2007) Successful? (2001) Strategies for Using State Information: . Howard Risher and Charles H. Fay Jonathan Walters Measuring and Improving Program Government Management of Performance(2003) Seven Steps of Effective Workforce g Planning(2007) Information Mega-Technology: Shelley H. Metzenbaum Ann Cotten Lessons from the Internal Revenue Linking Performance and Budgeting: Service's Tax Systems Modernization Opportunities in the Federal Budget • (2002) Process(2004, 2nd ed.) Barry Bozeman Philip G.Joyce INNOVATION AdvancingEnd Computing:High g P g How Federal Programs Use Outcome Managing Workfare:The Case of the Linking to National Goals (2003) Information: Opportunities for Work Experience Program in the New Juan D. Rogers and Barry Bozeman Federal Managers (2004, 2nd ed.) , York City Parks Department(1999) The Challenge of Innovating in Harry P. Hatry, Elaine Morley, Shelli B. • Steven Cohen Government(2006, 2nd ed.) Rossman, and Joseph S.Wholey New Tools for Improving Sandford Borins Performance Management for Government Regulation:An A Model for Increasing Innovation Career Executives:A"Start Where • Assessment of Emissions Trading Adoption: Lessons Learned from the You Are, Use What You Have" Guide and Other Market-Based Regulatory IRS e-file Program(2006) (2004, 2nd ed.) Tools (1999) Stephen H. Holden Chris Wye Gary C. Bryner • Staying the Course:The Use of g Religious Organizations,Anti-Poverty Performance Measurement in State g Relief,and Charitable Choice:A Governments (2004) Feasibility Study of Faith-Based MANAGING FOR Julia Melkers and Katherine • Welfare Reform in Mississippi (1999) PERFORMANCE AND Willoughby • John P. Bartkowski and Helen A. Regis RESULTS Moving from Outputs to Outcomes: Business Improvement Districts and Practical Advice from Governments Innovative Service Delivery(1999) Using Evaluation to Support Around the World (2006) Performance Management: A Guide • Jerry Mitchell for Federal Executives(2001) Burt Perrin An Assessment of Brownfield Kathryn Newcomer and Mary Ann Redevelopment Policies:The Scheirer • Michigan Experience(1999) • Richard C. Hula 56 To download or order a copy of a report,visit the IBM Center for The Business of Government website at:www.businessofgovernment.org • Using the Balanced Scorecard: Managing Across Boundaries: A Public Deliberation:A Manager's Lessons Learned from the U.S. Postal Case Study of Dr. Helene Gayle and Guide to Citizen Engagement(2006) • Service and the Defense Finance the AIDS Epidemic (2002) Carolyn J. Lukensmeyer and Lars and Accounting Service(2006) Norma M. Riccucci Hasselblad Torres • Nicholas J. Mathys and Kenneth R. Thompson Managing"Big Science":A Case A Manager's Guide to Choosing • Study of the Human Genome Project and Using Collaborative Networks Performance Leadership: 11 Better (2002) (2006) >. Practices That Can Ratchet Up W. Henry Lambright H. Brinton Milward and Keith G. . Performance(2006, 2nd ed.) Provan Robert D. Behn Managing the New Multipurpose, Multidiscipline University Research The E-Government Collaboration Performance Accountability: Centers: Institutional Innovation in Challenge: Lessons from Five Case The Five Building Blocks and Six the Academic Community(2003) Studies (2006) Essential Practices(2006) Barry Bozeman and P.Craig Boardman Jane Fedorowicz,Janis L. Gogan, • Shelley H. Metzenbaum and Christine B.Williams Assessing Partnerships: New Forms • Implementing OMB's Program of Collaboration (2003) From Forest Fires to Hurricane Assessment Rating Tool (PART): Robert Klitgaard and Gregory F. Katrina:Case Studies of Incident • Meeting the Challenges of Integrating Treverton Command Systems (2007) Budget and Performance(2006) Donald P. Moynihan John B. Gilmour Leveraging Networks:A Guide for Public Managers Working across • The Philadelphia SchoolStat Model Organizations(2003) (2007) Robert Agranoff PRESIDENTIAL Christopher l ley y, Leigh Botwinik, and Mary Shelley ExtraordinaryResults on National TRANSITIONS Goals: Networks and Partnerships in • What All Mayors Would Like to the Bureau of Primary Health Care's The President's Management Know About Baltimore's CitiStat 1000/6/0 Campaign (2003) Council:An Important Management • Performance Strategy(2007) John Scanlon Innovation (2000) Robert D. Behn Margaret L.Yao Public-Private Strategic Partnerships: Engaging Citizens in Measuring and The U.S. Postal Service-Federal Government Reorganization: • Reporting Community Conditions: Express Alliance(2003) Strategies and Tools to Get It Done A Manager's Guide(2007) Oded Shenkar (2004) Alfred T. Ho Hannah Sistare The Challenge of Coordinating"Big • Science"(2003) Performance Management for W. Henry Lambright Political Executives:A"Start Where • NETWORKS, You Are, Use What You Have" Guide Communities of Practice:A New (2004) • COLLABORATION, Tool for Government Managers(2003) Chris Wye AND PARTNERSHIPS William M. Snyder and Xavier de Souza Briggs Becoming an Effective Political The Challenge of Managing Across Executive: 7 Lessons from Boundaries: The Case of the Collaboration and Performance Experienced Appointees (2005, Office of the Secretary in the U.S. Management in Network Settings: 2nd ed.) Department of Health and Human Lessons from Three Watershed Judith E. Michaels Services(2000) Governance Efforts(2004) Mark T. Imperial Getting to Know You: Rules of • Beryl A. RaclinEngagement for Political Appointees Leveraging Networks to Meet The Quest to Become"One": An and Career Executives (2005) National Goals: FEMA and the Safe Approach to Internal Collaboration Joseph A. Ferrara and Lynn C. Ross Construction Networks(2002) (2005) W Six Trends Transforming Government William L.Waugh,Jr. Russ Linden (2006) Applying 21st-Century Government Cooperation Between Social Security Mark A.Abramson,Jonathan D. to the Challenge of Homeland and Tax Agencies in Europe(2005) Breul, and John M. Kamensky • Security (2002) Bernhard Zaglmayer, Paul • Schoukens, and Danny Pieters Reflections on 21st Century Elaine C. Kamarck Government Management(2007) Leveraging Collaborative Networks Donald F. Kettl and Steven Kelman in Infrequent Emergency Situations . (2005) Donald P. Moynihan . To download or order a copy of a report,visit the IBM Center for The Business of Government website at:www.businessofgovernment.org 57 • CENTER REPORTS AVAILABLE The Management of Regulation Transforming Government: The SUPPLY CHAIN • Development:Out of the Shadows Revitalization of the Veterans Health MANAGEMENT • (2007) Administration (2000) Cornelius M. Kerwin Gary J.Young Digitally Integrating the Government • Transforming Government: Dan Supply Chain: E-Procurement, Goldin and the Remaking of NASA E-Finance, and E-Logistics(2003) SOCIAL SERVICES (2001) Jacques S. Gansler,William a W. Henry Lambright Lucyshyn, and Kimberly M. Ross Implementing State Contracts for Enhancing Security Throughout Social Services: An Assessment of The Power of Frontline Workers the n g Churnhain Thro g the Kansas Experience(2000) in Transforming Government: Supply 04) The Upstate New York Veterans David J. Closs and Edmund F. Jocelyn M.Johnston and Barbara S. Healthcare Network(2003) McGarrell Romzek Timothy J. Hoff Investing in Supply Chain Security: • Delivery of Benefits in an Emergency: Makin Public Sector MergersCollateral Benefits(2005, 2nd ed.) Lessons from Hurricane Katrina(2007) Work:g Lessons Learned (203) James B. Rice,Jr.,and Philip W. Spayd • Thomas H. Stanton Peter Frumkin RFID:The Right Frequency for Efficiency Counts: Developing the Government(2005) • Capacity to Manage Costs at Air David C.Wyld STRATEGY AND Force Materiel Command (2003) Ben • TRANSFORMATION Michael Barzela and Fred Thompson Practices in ng Procurement y p Practices in Higher Education The Importance of Leadership:The The Transformation of the (2007) Role of School Principals(1999) Government Accountability Office: Richard R.Young, Kusumal Paul Teske and Mark Schneider Using Human Capital to Drive Ruamsook, and Susan B. Purdum Change(2005) • Leadership for Change: Case Studies Jonathan Walters and Charles in American Local Government(1999) Thompson • Robert B. Denhardt and Janet Vinzant Denhardt Transforming the Intelligence • Community: Improving the Managing Decentralized Collection and Management of Departments: The Case of the U.S. Information (2005) Department of Health and Human Elaine C. Kamarck Services (1999) Beryl A. Radin Executive Response to Changing Fortune: Sean O'Keefe as NASA • Corporate Strategic Planning in Administrator(2005) Government: Lessons from the W. Henry Lambright • United States Air Force(2000) Colin Campbell Ramping Up Large, Non-Routine Projects: Lessons for Federal Transforming Government:The Managers from the Successful 2000 Renewal and Revitalization of the Census(2005) Federal Emergency Management Nancy A. Potok and William G. Agency (2000) Barron,Jr. R. Steven Daniels and Carolyn L. Clark-Daniels The Next Government of the United States: Challenges for Performance Transforming Government: Creating in the 21 st Century(2005) the New Defense Procurement Donald F. Kettl System (2000) S Kimberly A. Harokopus Reforming the Federal Aviation Administration: Lessons from Trans-Atlantic Experiences in Health Canada and the United Kingdom Reform: The United Kingdom's (2006) . National Health Service and the Clinton V. Oster,Jr. United States Veterans Health Administration (2000) Improving Service Delivery in Marilyn A. DeLuca Government with Lean Six Sigma (2007) S John Maleyeff 58 To download or order a copy of a report,visit the IBM Center for The Business of Government website at:www.businessof9overnment.org . • BOOKS Collaboration:Using Networks and Managing for Results 2002 Partnerships (Rowman & Littlefield Publishers, • (Rowman & Littlefield Publishers, Inc., 2001) Inc., 2004) Mark A.Abramson and John M. John M. Kamensky and Thomas J. Kamensky, editors • Burlin, editors Managing for Results 2005 Competition,Choice,and Incentives (Rowman & Littlefield Publishers, • in Government Programs Inc., 2004) (Rowman & Littlefield Publishers, John M. Kamensky and Albert Inc., 2006) Morales, editors John M. Kamensky and Albert Morales, editors Memos to the President: Management Advice from the E-Government 2001 Nation's Top CEOs • (Rowman & Littlefield Publishers, (John Wiley& Sons, Inc., 2000) Inc., 2001) James J. Schiro, editor • Mark A.Abramson and Grady E. Means, editors Memos to the President: Management Advice from the E-Government 2003 Nation's Top Public Administrators i (Rowman & Littlefield Publishers, (Rowman & Littlefield Publishers, • Inc., 2002) Inc., 2001) Mark A.Abramson and Therese L. Mark A.Abramson, editor Morin, editors New Ways of Doing Business . Human Capital 2002 (Rowman & Littlefield Publishers, (Rowman & Littlefield Publishers, Inc., 2003) Inc., 2002) Mark A.Abramson and Ann M. Mark A.Abramson and Nicole Kieffaber, editors Willenz Gardner, editors The Procurement Revolution Human Capital 2004 (Rowman & Littlefield Publishers, (Rowman & Littlefield Publishers, Inc., 2003) Inc., 2004) Mark A.Abramson and Roland S. Jonathan D. Breul and Nicole Harris III, editors Willenz Gardner, editors • Transforming Government Supply Innovation Chain Management (Rowman & Littlefield Publishers, (Rowman & Littlefield Publishers, Inc., 2002) Inc., 2003) • Mark A.Abramson and Ian Littman, Jacques S. Gansler and Robert E. editors Luby,Jr., editors Integrating Performance and Transforming Organizations • Budgets:The Budget Office of (Rowman & Littlefield Publishers, • Tomorrow Inc., 2001) (Rowman & Littlefield Publishers, Mark A.Abramson and Paul R. Inc., 2006) Lawrence, editors Jonathan D. Breul and Carl Moravitz, editors Leaders (Rowman & Littlefield Publishers, • Inc., 2002) Mark A.Abramson and Kevin M. Bacon, editors Learning the Ropes:Insights for Political Appointees • (Rowman & Littlefield Publishers, Note: Rowman & Littlefield books Inc., 2005) are available at bookstores,online Mark A.Abramson and Paul R. booksellers,and from the publisher Lawrence, editors (www.rowmanlittlefield.com or 800- 462-6420). . I 59 About the IBM Center for The Business of Government • Through research stipends and events, the IBM Center for The Business of Government stimulates research and facilitates discussion of new approaches to improving the effectiveness of government at the federal, state, local, and international levels. The Center is one of the ways that IBM seeks to advance • knowledge on how to improve public sector effectiveness. The IBM Center focuses on the future of the operation and • management of the public sector. About IBM Global Business Services , With consultants and professional staff in more than 160 countries globally, IBM Global Business Services is the world's largest consulting services organization. IBM Global Business Services provides clients with business process and industry expertise, . a deep understanding of technology solutions that address specific industry issues, and the ability to design, build and run those solutions in a way that delivers bottom-line business • value. For more information visit www.ibm.com. For additional information, contact: Jonathan D. Breul • Executive Director IBM Center for The Business of Government 1301 K Street, NW • Fourth Floor,West Tower • Washington, DC 20005 (202) 515-4504, fax: (202) 515-4375 • e-mail: businessofgovernment@us.ibm.com • website: www.businessofgovernment.org • • • caa�ecra�� ^a�.uxRs t 1 � Faster, Cheaper • • HOME NEWS&TOPICSs • BOOKS • CURRENT BETTER, • March 7, 2011 • Iowa Charter Agencies: Did they Work? • By Jim Chrisinger and Cynthia Eisenhauer • Iowa's auditor recently issued a report characterizing Iowa's award-winning Charter Agencies • program as having "flunked its goals." S As key participants in that effort, we believe the auditor missed the point. The Charter Agencies program was an initiative under former Democratic Gov. Tom Vilsack, which • granted six agencies greater operating flexibility in return for better results and cost savings. • • Rather than look at the overall impact of this initiative, state Auditor David Vaudt, a Republican, • attempted to determine what savings could be directly linked to the agency's enhanced flexibility. • His own office acknowledges the challenge of this approach. "It is extremely difficult, if it not • impossible, to tell how much it saved," Warren Jenkins, chief deputy auditor told the Des Moines • Register. "There was no verification that what they reported resulted from the initiative or if it came • from any other reason." • We believe this is the wrong question to be asking. Who cares if the cost savings realized can be traced to the different rules?The actual goal of the Charter Agencies program was to get better • results for less, and every charter agency achieved that goal. • Nothing in the initiative required or even sought cause-effect linkages. Charter Agencies sought to holistically improve public value, not mechanically improve the workings of current activities. •• It worked. The Charter Agencies program showed improved results. A few examples: • Human Services reduced child welfare shelter stays by 20 percent, or 10 days, and 34 percent • more low-income children received access to health-care coverage. • Corrections reduced probationers'failure rate by 17 percent and increased the amount of • restitution paid to victims by 23 percent. • • The Department of Revenue (DOR) improved the rate of individual income tax refunds issued within 45 days from 75 percent to 94 percent. • • Charter Agencies also raised millions of dollars in new revenues, and not by simply charging new • fees or raising rates. DOR collected more of the tax money it was owed. Alcoholic Beverages • • Better Faster Cheaper • • • adopted a more business-like approach to its pricing and internal operations. Natural Resources sold Iowa conservation-related hats, sweatshirts and other merchandise. Charter Agencies also "saved" money, but not by shaving this line item or that line item. Instead, • money was "taken off the top" at the beginning of the fiscal year. This approach says, "If we have • this much money, less than before, how can we best spend it to achieve our results?" • The difference may appear to be subtle, but its impact can be large. More than $1.5 million was taken off the top from three of the Charter Agencies in the first year. While that's not a huge sum, this feature did at least pioneer the concept of a money-for-flexibilities trade-off in Iowa state government. The six Charter Agencies generated over$22 million in combined savings and entrepreneurial revenue in the first year, easily exceeding the $15 million target. We do not claim these results were directly linked to specific flexibilities awarded to Charter Agencies. We do claim that Charter Agencies helped achieve them, and point the way toward a better way of "doing government" in the 21st century. We will not overcome the current financial crisis, improve American competitiveness and close government's credibility gap with the public through incremental, mechanical improvement. . The most important question here was, "Can Charter Agencies create more public value and help close the budget gap better than 'regular' agencies?" We believe the results are a resounding "Yes." Jim Chrisinger Jim Chrisinger is a GOVERNING contributor. He is a Senior Partner at the Public Strategies Group. • E-mail:jim ftsg.us Cynthia Eisenhauer Cynthia Eisenhauer is a GOVERNING contributor. She completed 34 years in public service in January 2007,most • recently as Chief of Staff for Iowa Governor Tom Vilsack and Lt. Governor Sally Pederson. E-mail: Cynthia@psg.us This column originally posted at Governing.com - Better, Faster and Cheaper, March 7, 2011 and reprinted with permission. • Better Paster Cheaper • • i • By abandoning the traditional annual budget in favor of a biennial budget framework, one small city has saved staff time, improved its strategic planning capabilities, and enhanced its overall financial condition. • • Taking the Plunge: The Conversion • to Multimvear Budgeting • By Andrea Jackson Multi-year budgeting offers a number of potential advan- review did little to facilitate long-term financial planning. tages over annual budgeting, including decreased staff The improved long-range planning potential of multi-year bud- time,enhanced long-range planning, and improved pro- geting offered another important benefit to city officials:enhanced • gram evaluation.Multi-year budgeting facilitates the integration of fund balance position and bond rating. In the early 1990s, financial and strategic planning, replacing incrementalism with a Auburn's general fund balance had dipped to precariously low lev- • more strategic approach to resource allocation.As more and more els,having been tapped for a series of capital improvement projects. governments have become aware of these and other benefits,many By the end of the decade,the general fund balance had been signifi- L • jurisdictions have converted to multi-year budgeting,usually on a cantly improved thanks to a renewed emphasis on careful budget- biennial basis. Like many management reforms, however, imple- ing and close monitoring. Both the city manager and the finance • mentation of multi-year budgeting requires careful planning in director included fund balance maintenance as a strategic goal of order to realize the desired benefits.This article describes one city's biennial budgeting. conversion from an annual budget to a biennial budget,emphasiz- ing the most important considerations for governments pondering Of Form and Function similar initiatives. Local governments that budget on a multi-year basis typically employ one of the following variations of biennial budgeting: Bien-The Case for Change nial financial plan,rolling biennial budget,or classic biennial bud- The City of Auburn,Alabama,is a council-manager government get. The biennial financial plan is characterized by an annual • with a population of 43,000 and a general fund budget of$35 mil- appropriation linked to a tentative, unenforceable spending plan lion. The city has a well-established history of innovation, long- for the following year. The rolling biennial budget is a two-year • range planning and goal-oriented stewardship of public resources. spending plan comprised of two one-year appropriations that are As a result, launching a multi-year budgeting effort was a logical adjusted annually. The classic, or traditional, biennial budget is a • extension of Auburn's approach to government.Auburn is believed two-year spending plan in which 24 months worth of expenditures to be the only local government in Alabama engaged in multi-year are approved simultaneously.The distribution of these three types • budgeting. of biennial budgets has been almost even among governments that In the spring of 2000,the city manager asked the finance director practice multi-year budgeting.' • to consider preparing a biennial budget with the intention of reduc- In deciding which type of biennial budget to use, governments ing the amount of staff time devoted to budget preparation and need to consider applicable state laws,as well as the characteristics • publication and to the mid-,year review process. Because Auburn of the various types of budgets. For Auburn, the classic biennial does not have a separate budget office or even one staff position budget made the most sense. The Alabama State Code requires • dedicated to budget administration,finance staff were perpetually municipalities to adopt an annual budget and to have an annual involved in some aspect of the budget.The budget also consumed a financial audit. However, there is no apparent legal prohibition • great deal of time for line department staff.A reduction in the num- against adopting two annual budgets at one time.In the city's view, ber of staff hours devoted to the budget represented significant sav- the rolling biennial budget did not offer the time savings of the clas- • ings that could be channeled into other important activities. sic biennial budget,and the biennial financial plan did not facilitate The city manager's other motivation for investigating biennial long-term planning as well as the classic biennial budget. Given • budgeting was to extend the staff's planning and management hori- state legal requirements and the city's major objectives in convert- zon.The city's annual budget process had been goal-oriented since ing to multi-year budgeting, Auburn opted in favor of the classic • the institution of the annual citizen survey,the results of which were biennial budget. used by the City Council to establish priorities for the upcoming • budget period. These priorities then served as the basis for the Winning Organizational Support preparation of departmental goals and budget requests.Despite its Introducing change is never easy in the public sector,especially • success in linking goals and priorities to resource allocation, when it involves something as fundamental as the budget process. Auburn's annual budget appropriation and subsequent mid-year To be successful,the conversion to multi-year budgeting must have • • 24 AOGUST2002 GOVERNMENTFINAWF.REVIEw • • • 1Y1.. _L�mufi `"- y,�,.A R�OkPdiK6,..�Hil lll ix:�1 z . IR,a.�:.cc= .: EtA4�evMd'It ,4tS' ._e.. NSS.�.�cv�Y221U .R�1Poii IYd... il@pk9i • the support of the governing body,the chief executive,and depart- Adding to Human Resource's burden,Finance proposed and the ment directors. Winning the support of these key stakeholders city manager approved an additional change to the personnel bud- • requires that finance officers be able to clearly explain the rationale get spreadsheets.It was decided to calculate salaries so as to include behind multi-year budgeting and to identify the benefits thereof. the amount of the cost-of-living salary increase to be proposed to The process of implementing biennial budgeting in Auburn the City Council. In the past, cost-of-living increases were not began with a meeting between the finance director and her staff included in the proposed budget. Rather, the money required to • about the city manager's proposal and the goals of biennial budget- fund these raises was included separately, as one of the Council's ing.The response from finance staff was positive and supportive, "key decisions." This approach presented a disadvantage in that • which encouraged the city manager to take his idea to the other the effect of the cost-of-living increases on fund balance was not department directors.The potential of biennial budgeting to reduce represented in the proposed budget numbers. To maintain the the amount of staff time required for budget preparation earned the desired level of the general fund balance,the city determined that immediate support of these important stakeholders. The depart- the effect of any proposed cost-of-living increase would be included • ment directors had ample experience in implementing innovative in the proposed budget document. improvements, so they enthusiastically accepted the challenge of Given that salaries and benefits comprise the largest single com- . preparing a two-year budget despite the increased workload that ponent of the city's budget,the accuracy of the personnel budget was inevitable during the conversion. spreadsheets is always critical to the overall budget process. . Once the city manager had secured departmental support for Because of the increased workload necessitated by the conversion biennial budgeting, he took the recommendation to the City process,the Finance Department was unable to include the person- • Council for approval. Given the city's generally favorable disposi- nel spreadsheets in the department budget packets distributed at the tion toward innovation and change,the Council was not surprised budget kickoff meeting.Thanks to a significant overtime effort by by the proposal and threw its support behind the effort to convert the Human Resources Department,however,the personnel spread- to biennial budgeting.Although multi-year budgeting is sometimes sheets for both years—including the proposed cost-of-living viewed as a threat to the governing body's budgetary oversight,the increase—were distributed to the departments shortly thereafter. City Council did not express any concerns to this effect. The city manager partially mitigated the potential for concern by assuring Forecasting Revenues council members that they would continue to receive monthly bud- Multi-year budgeting complicates revenue forecasting by extend- get reports and that any increases would require City Council ing the time horizon from one year to two or more years.The fur- approval via ordinance. ther projections are extended, the less accurate they become. As envisioned by the city manager, biennial budgeting would Fundamental changes in a government's revenue mix or significant reduce the City Council's workload with respect to the budget. deviations from expected revenues following the adoption of a • Midway through the first biennium,the administration would com- multi-year budget may require revisions to revenue estimates or pare actual to projected revenues.As long as actual revenue collec- adjustments to appropriations. tions to date and projected revenues through the end of the fiscal year Through the years,the Finance Department has endeavored to were sufficient to support appropriations and maintain desired fund produce realistically conservative revenue projections for the city balance levels,no additional mid-year review would be required. manager's consideration. For each major revenue source, and for most of the minor ones,Finance maintains a month-by-month rev- • Conversion Challenges enue history going back 15 years or more.Staff carefully monitors Having secured the support of the key stakeholders in city gov- national, state, and local economic conditions, and analyzes the • ernment,the Finance Department turned its attention to the logis- revenue implications of annexations,population growth, building tics of developing the proposed biennial budget for fiscal years permit data,and developments at Auburn University.Because the 2001 and 2002.The CityCouncil did n r of approve the conversion city collects its own sales taxes, Auburn has a valuable source of • to biennial budgeting until late April. Under the previous budget information about the vitality of the local economy.However,the calendar, finance staff already would have begun assembling the city's revenue forecasts are complicated by the fact that it relies on spreadsheets,forms,and instructions to be used by the departments several state-shared revenues that cannot be reliably projected. • in submitting their budget requests.Already behind schedule,staff Forecasting revenues for Auburn's first biennial budget was a real decided to use essentially the same spreadsheets and forms as challenge.Although the city used the same methodology it had used . before,using color coding to differentiate materials for the second to project revenues for the annual budget, it was somewhat more year of the biennium.This section describes three major challenges conservative with the second year's revenues because of the extend- • of Auburn's conversion to biennial budgeting:estimating personnel ed timeframe. The fact that 60 percent of the city's four major needs,forecasting revenues,and balancing the budget. sources of general fund revenue are received by mid-year helps alle- • viate some of the uncertainty.It is also helpful in identifying any sig- Estimating Personnel Needs nificant deviations from expected revenues so that the city manager Multi-year budgeting forces governments to assess future staffing can make any necessary adjustments sooner rather than later. levels and their effect on the operating budget. In Auburn, the Human Resource Department is responsible for preparing person- Reviewing Departmental Budget Requests nel budget spreadsheets for each department. These spreadsheets Following the submission of the departmental budget requests • provide detailed information for each authorized position,includ- for the first biennium,the city manager and Finance staff faced the ing the name of the employee occupying the position,the employ- ever-daunting task of balancing the budget requests to the projected ee's salary grade and steps,and the employee's salary for the next revenues while maintaining appropriate fund balance levels. budget period. Auburn has more than 300 regular full-time Although the city adhered to the same procedures it had used to employees,a dozen regular part-time employees,and approximate- balance the annual budget, the process was complicated by the ly 150 temporary employees,which increases significantly on a sea- additional year.The city manager reviewed all capital outlay items sonal basis.Calculating salary information for all of these employ- and capital projects,making cuts and deferrals as he deemed neces- ees and estimating future personnel needs for two years instead of sary.All departments were given a target budget amount for each • one posed what seemed like a monumental obstacle for the Human budget year that was a specific percentage less than their original Resources Department. budget requests. Department heads were allowed to change line • Gov FRNMENTFINANCF.REVIEv, AUGUST2002 25 • • • COMPARISON OF BIENNIAL BUDGETCALENDAR TO ANNUAL BUDGETCALENDAR • Biennial Budget Anual Budget Jun Departments'goalsdue Jun 1 Departments'goalsdue • Budget forms issued Budget forms issued • Jun 30 Revenue projections due Jun 30 Revenue projections due Departments'budget requests due Departments'budget requests due • Jul 1-30 Review and analysis Jul 1-30 Review and analysis Publish proposed budget Publish proposed budget Aug Budget hearings Aug 1 Budget hearings • Aug15 Key decisions Aug15 Keydedsions Sep 10 Budget adoption Sep 10 Budget adoption • Sep 20 Current years end of year budget Sep 20 Current year's end of year budget adjustment adoption adjustment adoption • Sep 30 Working document due Sep 30 Working document due Dec 30 GFOA document due Dec 30 GFCA document due Secondyear Secondyear May I Mid-year budget forms issued • May 15 Mid-yearreview of revenue May 15 Mid-year revenue revisions due Expenditure budget Cuts,ifneeded Departments'budget revisions due • May 15-30 Mid-year review and analysis Publish mid-year budget Jun 1 Departments',goal revisions due Jun 1 Departments'tgoals due • Mid-biennium budget forms issued Budget forms issued Jun 5 Mid-year review Jun10 Mid-year budget amendment adopted • Jun 30 Revenue projections due Jun 30 Revenue projections due Departmente budget requests due Departments'budget requests due Jul 1-30 Review and analysis Jul 1-30 Review and analysis • Publish proposed budget Publish proposed budget Aug 1 Budget hearings Aug 1 Budgethearings Aug'15 Keydeasions Aug 15 Keydecislons' • Sep 10 Budget amendment adopted Sep 10 ! Budget adopter! Sep 20 Currentyear's end of yea r budget Sep 20 Current year's end of year budget adjustment adoption adjustment adoption • Sep 30 Working document due Sep 30 Working document due Dec 30 GFOA document due • item amounts as long as the target budget amount was achieved. ment of the Office of the City Manager and the Finance • The city made provisions to borrow money for capital projects that Department.Department heads and their staffs did not participate could not be accommodated by the projected available resources as they had in previous years.And as the end of the first fiscal year and that could not be cut or deferred because of Council priorities. approached, staff had only to review and revise the second year's numbers instead of building a new budget from scratch.This saved • Results substantial time for the City Council and all of the city employees As Auburn geared up for the preparation of its second biennial involved in the budget process. budget,the citywide consensus was that multi-year budgeting had City officials scheduled preparation of the second biennium to lived up to its promises. The management and staff of all city begin two months earlier than the last time.This allowed the city to departments agreed that biennial budgeting had reduced the incorporate the mid-year review for the second year of the first amount of staff time devoted to budget preparation, shifting biennium into the preparation of the proposed budget for the sec- resources toward the achievement of other goals. During the first and biennium.Another factor in the early scheduling was that City biennium, the mid-year review required only the limited involve- Council election activities were to begin in July for the August elec- • • 26 -AuGUST2002 GOVERNMENTFINANCEREVIEw • i tion.As such,the mayor and City Council members,all of whom budget packets distributed to the departments for preparation of • were up for election,could devote appropriate attention to the pro- their budget requests. posed budget before the campaign hit full stride. •Budgeting for the salaries and wages of temporary employees has Auburn also achieved its other goal with respect to biennial bud- been a perennial problem for the city.Auburn has never been able geting—to extend the city's planning horizon.Since the conversion to precisely budget the personnel costs of temporary employees to biennial budgeting,the city manager and finance director have whose pay fluctuates by season and whose work schedules are witnessed a paradigm shift in both the department heads and mid- unpredictable. Departments like Parks and Recreation that dle management.The notion of"fixing things at mid-year"is now employ many such employees have traditionally budgeted for • obsolete,having been replaced by a forward-facing perspective that these employees as a group, using a single lump sum amount. takes into consideration the long-term financial and operating However,the annual audited financial statements always showed impacts of spending recommendations.The mid-year review is now significant budget balances for these departments because of over just a mechanism for assessing the adequacy of revenues. budgeting for temporary employees. Thanks to the accounting During the initial biennium, Auburn experienced firsthand the change describe above,the city determined that the actual expen- increased uncertainty that accompanies multi-year revenue fore- ditures for these employees was approximately$500,000 less that . casting. In July 2001, sales tax revenue decreased from the prior what had been budgeted. year for the first time in more than a year.Sales tax revenue is the •Capital outlay and debt service expenditures are budgeted by • single largest component of the city's general fund budget,so nega- department. Although this approach does not vield a GAAP- tive fluctuations have serious budgetary implications. During the based budget,it does facilitate accountability for and comparabil- • mid-biennium review, Auburn had to reduce its original revenue ity of departmental operating costs. For the second biennium, projections for the biennium by a net amount of approximately Finance entered the debt service amounts directly into the depart- • $500,000 (although projected sales tax revenue was off by more mental budget spreadsheets instead of generating a separate list of than$1 million for the biennium,other revenue sources combined debt service requirements by department.This change is intended • to make up for half of this amount).To offset the decrease in bud- to prevent math errors and omissions for departmental debt ser- getary resources,the city opted to finance a property purchase of vice expenditures. • approximately$1 million,thereby maintaining the desired ending general fund balance. Conclusion The economic uncertainties of projecting revenues for a longer The consensus opinion among Auburn's senior management time horizon provided the impetus for a recommendation to estab- team is that the city's first biennial budgeting experience was an • lish a"permanent"reserve of a portion of general fund net assets. unqualified success.Auburn reduced the staff required to prepare As part of the mid-biennium review,the city manager recommend- and review the budget, extended its planning and management ed establishing a$4 million permanent reserve in the general fund. horizon,and maintained major fund balances.The city also intro- This amount represented approximately 13 percent of the budgeted duced a number of process improvements that have further simpli- • general fund expenditures.The City Council approved this recom- fied budget preparation and resulted in additional time savings. mendation,and included a provision in the mid-biennium amend- Although the conversion to biennial budgeting was not without • ing ordinance to establish this reserve for use only in times of natur- challenges,the benefits realized so far have been well worth the fini- al disaster or economic downturn. tial effort. By thinking through the conversion issues identified in • this article,other governments can successfully transition to multi- Process Improvements year budgeting and enhance their financial and strategic manage- • The Finance and Human Resources departments learned a great ment.1CtI deal from Auburn's first biennial budget experience.As a result,the • city has introduced several improvements to its budget process that NOTE will save additional time and money and enhance overall financial B.Blom and S.Guajardo,"Multi-Year Budgeting:A Primer for Finance • management.These improvements are discussed below. Officers,"Government Finance Review 16(2000):39-43. •Maintaining separate spreadsheets for each year of the biennium • proved to be too confusing for city departments and finance staff. ANDREA JACxsON,CPA,CGFM,is finance director for the City of Auburn, The simplest process improvement was to include both years of Alabama.Sbe has more than 17 years of experience in governmental • the biennium on the same spreadsheet. This seemingly minor accounting and auditing,and holds an MBA from Auburn University. adjustment greatly facilitated the analysis of departmental budget • requests and the preparation of the budget document. •Given the critical importance of an accurate personnel budget,the city improved its accounting for salaries and benefits by establish- ing separate salary accounts for the various classes of employees (regular full-time, regular part-time, temporary part-time, etc.). • This enhanced level of detail facilitates the analysis of departmen- • tal budget requests, providing better information on the fiscal impact of proposed increases in personnel costs. •Heading into the city's second biennium, the human resources director questioned the wisdom of preparing detailed personnel • budget spreadsheets for the second year of the biennium.Instead, it was suggested that the amounts budgeted for salaries and bene- fits the first year be increased by a flat percentage accommodating both merit pay increases and proposed cost-of-living increases. • The second-year amounts would then be updated position by position during the mid-biennium review to ensure their accuracy. • This change made it possible to include personnel figures in the • GOVERNMENTFINANCEREVIEW AUGUST2002 2.7 • • • • :slative Analyst Report - Biennial Budgeting (File No. 021309) • (OLA#024-02) W:Members of the Board of Supervisors Worn:Adam Van de Water,Office of the Legislative Analyst *ate:October 1,2002 O:Biennial Budgeting(File#021309) * Summary and Scone of Request Opervisor Maxwell,through the Board of Supervisors,requested that the Office of the Legislative Analyst(OLA)research the potential advantages and &advantages of enacting a biennial budget,provide a list of local governments that currently enact biennial budgets,and make policy recommendations Othe City and County of San Francisco. * Executive Summary_ terest in biennial budget cycles has grown in recent years in response to several factors,including increasing time spent preparing and reviewing udgets and more restrictions on and uncertainty of future revenues.Most governments that have enacted a biennial budget cycle pass two-year ending plans but continue to appropriate funds on an annual basis(a 'rolling"biennial budget).A less common form of the biennial budget is the One"biennial budget,which either appropriates funds for two-year time periods or two consecutive but non-transferable time periods. !cording to recent surveys,12 states,nine of the largest 35 U.S.counties,and 25 California cities currently have some form of a rolling biennial budget *d an additional nine states,two major counties and 18 California cities enact'true"biennial budgets. Wer discussing the six primary advantages and disadvantages of biennial budget cycles,this report looks at the relevant language contained in the le arter and the Administrative Code governing how it could work in San Francisco.While the City Charter currently requires the Mayor to submit and Board to approve an annual budget,the Board could enact a non-binding two-year spending plan by ordinance.Additionally,while the ministrative Code contains some provisions requiring long range planning,not all these provisions are currently exercised.The OLA therefore raises *yen key issues for the Board"s consideration should they wish to pursue the enactment of a biennial budget cycle in San Francisco.These policy epsiderations provide the foundation for any successful transition to a new budget process and would need to be answered by the Board before enacting �ew budget cycle. • Biennial Budgeting:An Overview ennial budgeting has many variants and there are almost as many nuanced forms as there are governments that enact them.Most commonly, vernments enact a rolling biennial budget,where the legislative body passes a biennial financial plan but continues to appropriate the funds annually. cording to recent surveys!,12 states,nine of the 35 largest U.S.counties,and 25 California cities currently enact rolling biennial budgets(see Appendix Oor a list of states,major counties and California cities that enact biennial budgets).The more restrictive,and therefore more rare, 'true"biennial *dget passes both a biennial financial plan and a biennial appropriation every two years.In these nine additional states,two major counties and 17 �lifornia cities,departmental budgets are appropriated for the full two years with a limited mid-cycle or off-year review to adjust for unforeseen changes :tererevenues or expenditures. st in biennial budgeting has grown in recent years,due in part to growing time spent preparing and reviewing budget documents,recent attention at the federal level,and local government funding uncertainty as a result of state restrictions on property taxes.Proponents of biennial budget cycles contend that reduced Board time spent preparing,reviewing,and approving annual budgets provides more focused time for improved financial management and departmental oversight while forcing policymakers to make financial decisions over longer timeframes. • It was this experience in Texas that brought the issue to the federal level with the Bush administration and prompted reports by the General Accounting • Office,Congressional Budget Office(CBO),Office of Management and Budget,Center on Budget and Policy Priorities,and several congressional committees and advocacy groups.In California,the impetus for cities to adopt multi-year budgets came in 1978 with the passage of Proposition 13,which i restricted annual property tax rate increases to one percent and created uncertain funding futures for local governments.With approximately ten • percent of all California cities enacting some form of a biennial budget,California now has more cities with biennial budget cycles(43 by recent surveys) • than any other state. • Opponents of biennial budget cycles counter that funding uncertainty,coupled with a limited ability to predict future economic conditions as well as the potential for reduced departmental accountability,is exactly why most government agencies continue to have annual budget cycles. Advantages of Biennial Budgets • Advocates of biennial budgets point to the following six advantages of converting from an annual to a biennial budget cycle: 1.Long range-planning-Biennial budgeting can improve long-range and strategic planning,as it requires forecasting expenditures and revenues up to • twenty-eight months in advance.This provides longer time horizons to department heads and policymakers as they allocate resources across programs and anticipate future needs.Because the timeline is extended from one year to two,budgeting can be better coordinated with capital improvements,fleet • management changes,or other major municipal expenditures.For these reasons,the National Advisory Council on State and Local Budgeting, • recommends long-range planning as one of the five essential features of good budgeting3. 2.0pportunities for staff rede>l to ent-Biennial budgeting frees some department officials and Board and Mayor"s budget staff from annually preparing budget documents,time that could be spent improving financial management,conducting audits,and/or analyzing program effectiveness.The • Board could also use the time in mid-cycle budget years to gather information,formulate policy,and test and evaluate programs. • 3.Policy em hasi -Biennial budget cycles could move the Board from line-item consideration of the budget to a longer-range,more policy-driven approach.According to the City of Renton,WA Finance Administrator Victoria Runkle,Washington State officials that recently converted to biennial • budgeting found that"the legislative bodies focus on the outcomes of the programs,and less on the actual ways the program was managed."However,as • CBO Director Dan Crippen points out,this policy-based approach assumes that Board actions can be separated into issues with and without fiscal . impacts,which"could be confusing and might create new difficulties." 4.Department/contractor flexibility_-Two-year appropriations provide departments and private and not-for-profit contractors with more certainty in • funding over the longer 2-year period and,in some cases,can allow them to adjust spending levels across years within this larger two-year window. • 5.Improved cross-year comparisons-In the absence of zero-based budgeting,budgeting over two years could make cross-year comparisons and • associated policy decisions easier for the Board of Supervisors to do as specific funding allocations would be projected for two years instead of the current one year. • 6.Minimal transition issues-Officials in Arizona and Connecticut,the only states that have converted to biennial budgeting in the last 10 years,say that . they did not experience significant transition issues or technical difficulties in shifting to biennial budgeting. • Disadvantages of Biennial Budgc1gg • Opponents of biennial budgets point to the following six disadvantages of converting to a biennial budget cycle: • 1.Departmental oversight-Departments and contractors may be less responsive to the Board if their budgets were protected for two years.The absence • of annual budgetary reviews may further increase the authority of the Mayor"s Office,which is responsible for departmental oversight and program implementation.In Ohio and Connecticut-two states that recently converted to a biennial budget cycle but did not expressly design new oversight • functions-oversight did not increase in the off year as some proponents had hoped. • 2.State and federal funding uncertainty_-According to the Mayor"s FY 2002/2003 Proposed Balanced Budget,20.9%,or$1.03 billion,of the City and • County of San Francisco"s FY2002-2003$4.9 billion budget comes from state,federal and other governmental sources.As these entities operate on an • annual budget cycle,it is especially difficult for counties(which typically receive a higher percentage of state and federal funding than do cities4)to • predict these funding levels as much as 28 months in advance.Further complicating matters is the fact that the federal fiscal year begins October 1,three • months after the Charter requires the City and County to pass its budget. 3.Unforeseen events-The limited ability to project future economic and/or programmatic conditions and the inevitability of unforeseen events may lead • to a budget process that is biennial in name only.According to the Controller"s Office,biennial budgets would only increase the risk of forecasting future • revenue sources-such as hotel,property and sales taxes,a risk already present in the creation of annual budgets.A review of biennial budgets by the • Government Finance Officer"s Association,found that biennial budgets assume stability and therefore work best in times of economic growth or • certainty.Most California cities that currently operate on biennial budget cycles still have annual appropriation cycles for this reason and nearly 90 percent of all California cities continue to budget on an annual basis. • .Time savings-Depending on the strength of restrictions or willingness to avoid making significant technical or policy changes in the off-year,biennial ;udgeting may not lead to appreciable time savings for the Mayor,Board,or legislative staff.While survey evidence from finance directors in smaller ities suggests that it takes no more time and frequently takes less time to create a biennial budget,there is no guarantee that converting to a biennial *udget cycle will reduce the burden on elected officials or staff.For example,even in the absence of an annual budget process,departments must still *recast expenditures and maintain annual budgets to manage cash-flow,the Board must still review departmental performances,the Controller"s Office �ust still complete the Comprehensive Annual Financial Report,the 6-,9-,and 36-month budget projections and all state requirements. Workload-Biennial budgets may only serve to raise the stakes of budget negotiations-and hence the stress involved and how long they take-during he first of the two years as well as lead to less involvement by the Board in off-years. 0 Software and Accounting Changes-Converting to a biennial budget may require changes to the City and County"s budgeting and accounting practices, dding potential additional costs to the City and County. SBiennial Budgets and San Francisco Law Oamual Budgets Required by Charter 0e city charter currently mandates the submission of an annual budget and all appropriation ordinances and provides the Board of Supervisors with the Wthority to require the creation of non-binding multi-year budget plans.Article IX,Section 9.1oi of the San Francisco Charter requires the Mayor to Submit to the Board of Supervisors each year an annual proposed budget." While appropriations must still be made annually,Section 9.1oi further states that"the Board of Supervisors by ordinance may require multi-year dget plans and other budget planning strategies to be performed by the several departments and offices of the City and County."The Board of upervisors could therefore,by ordinance,change to a rolling biennial budget cycle and would need to pass a voter-approved charter amendment to ange to a true biennial budget cycle. Ourrent Long Range Planning Provisions Are Limited On Francisco currently has three provisions in place to project revenue and expenditure levels into the future and a fourth will become effective ginning in 2003.While these provisions are meant to assist the Mayor and the Board with their long-range funding decisions,only one-the joint *port issued by the Mayor"s Budget Office,Controller,and Budget Analyst-is currently presented to the Board to help guide their multi-year decision- 40aking.Full implementation of the authority granted in the following sections could obviate the need for and provide longer-term projections than nactment of a biennial budget. Joint Report 0irst,Section 3.6 of the San Francisco Administrative Code provides that the Mayor,Budget Analyst and Controller jointly prepare a three-year 0timated summary budget for the City and County,providing a snapshot of the City"s financial condition over the next three years.5 This approximately *-page report summarizes preliminary projections of future surpluses and shortfalls and is currently the only financial projection available to assist *e Board in making multi-year funding decisions.6 While an invaluable tool for estimating the impacts of major decisions and changes in the current Sar on conditions in subsequent years,the joint report is in no way intended to express the Mayor or Board"s funding priorities and cannot be used to Soject spending for individual departments or programs. 46 Four Year Projections 0cond,Section 9.1oi of the San Francisco Charter requires the Mayor"s annual proposed balanced budget to include a summary containing,"a 1scussion of trends and projections of revenues and expenditures of the City and County for the subsequent four years."The Mayor"s Proposed Budget 002-2003 issued on June 1 contains cursory mention of trends and projections but no funding estimates for any year beyond the budget yearn.Rather San projecting these figures for the subsequent four years,more often they summarize these figures for the prior three years. .Three-Year Forward Plans Oird,Section 3.5(b)of the San Francisco Administrative Code requires that, • Commencing with fiscal year 1998-99,each department,board,commission and agency shall develop and annually review a strategic • plan which contains at least a three-year forwardplan to reflectpolicy outcomes from the operations of the respective department, . board,commission,or agency consistent with the then-approved budget. ,bile it is unknown which of the approximately 6o City and County departments currently develop these forward plans,no department formally *esents them to the Board. *Efficiency Plans ally,the Performance and Review Ordinance of 1999 added Section 88.4 to the San Francisco Administrative Code,which states, * Beginning 20o3 and each year thereafter,the head of each department shall prepare and submit to the Mayor by October ist and to the . Board of Supervisors by November lsi a departmental efficiency plan.Each plan shall include a customer service element,a strategic * planning element,an annual performance element,and a performance evaluation element for the previous fiscal year,as set forth more fully below[see Sec.88.4 for more details].The plan shall cover a period of not less than three years forward from the fiscal year in which . it is submitted. In 200o and 20oi,a pilot group of departments(including the Board of Supervisors,Department of Public Works,Recreation and Parks Department, • Elections Department and a few others)began to develop departmental efficiency plans with mixed success.While some elements of these efficiency plans are already practiced citywide(for example,the inclusion of performance measures),many of the details of these efficiency plans will need to be developed for the first time over the course of the next fourteen months.The long-term success of these efficiency plans will depend on each • department"s acceptance of them and the support they receive in developing them. • Implementation Considerations for San Francisco • Should the Board of Supervisors wish to consider making the transition to a biennial budget in San Francisco,they must first address the following • considerations in greater detail.These questions,which are presented here in no particular order,are critical to the long-term success of any multi-year budget in San Francisco. •Level of departmental oversight-How accountable and in what way should departments be to the Board of Supervisors?Will guaranteeing two years of funding to departments erode their accountability and responsiveness to the Board?Is the same true for a rolling biennial budget where funds are still appropriated annually?Should funding be connected to program performance or another measure of accountability? • •Political authority and organizational structure-Does the often-adversarial nature of San Francisco's executive/legislative relationship make enacting a • biennial budget cycle in San Francisco necessarily more difficult?Does the unique nature of San Francisco"s organizational structure(with a strong Mayor and county Board of Supervisors)present difficulties above and beyond those faced by a City Manager or County Chief Administrative Officer • form of government?Are there compromises-such as the Board relinquishing some line-item managerial control of department budgets in return for increased program performance and accountability from departments-that could be worked out to improve the nature of this relationship?Would an • initial budget retreat between the Mayor and the Board,as practiced in numerous other local governments,prove helpful? • •Role of Key Budget Players-How could the Board utilize the Controller"s Office,Budget Analyst"s Office,or Legislative Analyst"s Office(OLA)more • efficiently to advance the long-range planning goals contained in a biennial budget cycle?Should the Budget Analyst continue to look at line-item details • or is there a need to amend their contract to look at longer term and/or broader issues?Does the Board plan to change the add-back process and how • would this annual process work with a biennial budget? •True Biennial Budget-If the Board wished to consider enacting a true biennial budget,how feasible would it be to amend the charter?What restrictions,if any,would the Board place on off-year amendments? i •Rolling Biennial Budget-If the Board wished to consider enacting a rolling biennial budget,how binding,if at all,would the goals and projections • contained in the two-year financial plan be to passage of the Annual Appropriation and Annual Salary Ordinances?What process would govern the mid- • cycle review?How would this process differ from fully implementing those long-range provisions currently contained in the Charter and Administrative • Code? •Transition-How would the Board transition to a biennial budget cycle?Would the first year of the biennial cycle impact the entire budget and all • departments or are there reasons to implement gradually and/or only for selected departments? • •Timing-How would a biennial budget be best coordinated with election cycles,union Memorandum of Understanding negotiations,capital projects and utility or rate tax changes?One possibility(sketched out in Appendix B in calendar form and based on the City of Oakland"s model)would make next • fiscal year(2003-2004)a transitional year,with the first year of the biennial cycle beginning the following year(2004-2005)after the election of a new • mayor and as many as six new supervisors. Conclusion and Recommendation • Enacting a biennial budget cycle in San Francisco is a policy matter for the Board of Supervisors. • If created carefully,biennial budgets can act as the catalyst to move the Board away from line-item consideration of the budget and toward increased • long-range planning while decreasing the annual time spent reviewing budget documents.However,there is nothing inherent in biennial budgeting that assures this transition will take place.Successful implementation of a biennial budget requires careful consideration of many related elements(raised as . unanswered questions in the prior section)as well as widespread commitment to the new goals and processes it creates.Unless the Board,Mayor and all • departments can clearly foresee the advantages of transitioning to a biennial budget and have the appropriate resources to work toward that goal,the creation of a biennial budget in San Francisco will only establish new requirements for finance managers that do not advance the policy goals of long- • range strategic planning. Should the Board of Supervisors wish to discuss enacting a biennial budget in San Francisco,the Office of the Legislative Analyst recommends calling a • hearing to discuss the issue in more detail with key budget players such as the Mayor"s Budget Office,the Office of the Legislative Analyst,the Budget Analyst"s Office,members of the Board of Supervisors and representatives from several city departments. • Based on the preliminary research contained above,the OLA finds that the easiest form of a biennial budget to implement and the most appropriate for • the City and County is the rolling biennial budget beginning in fiscal year 2004-2005.As they still appropriate funds annually,rolling biennial budgets �ovide departments and city contractors with increased future funding stability while maintaining their annual accountability to the Board.The Board �uld transition to a rolling biennial budget cycle through passage of an ordinance and adoption of a new budget process.If based on the process urrently in place in Oakland,selected key budget-related activities might roughly follow the calendar provided in Appendix B. Ah i.ppendix A:States,Major U.S.Counties and California Cities that Enact Biennial Budgets dVIPORTANTNOTE:Within the categories 'rolling"and 'true"listed above,there is significant variety.Some cities with rolling biennial budgets ve non-binding two-year spending plans and continue to appropriate funds annually while some cities with true biennial budgets appropriate funds iennially but make budget adjustments as frequently as every six months.The important distinction made here is that,with the exception of annual r semi-annual budget adjustments,funds are largely appropriated either every year or every other year. Qine States Enact'True"Biennial Budgets* Indiana Minnesota New Hampshire North Carolina *North Dakota Oregon Texas QWashington *Wyoming States Enact 'Rolling"Biennial Budge 0 Arizona Arkansas Connecticut Hawaii Kentucky 4PMaine Montana Nebraska �.Nevada �.Ohio �.Virginia �.Wisconsin Is (Source:National Conference of State Legislatures,1999) 0o of the Largest 35 U.S.Counties Enact 'True"Biennial Budgets* 0 *Oakland County,MI 40 Orange County,CA We of the Largest 35 U.S.Counties Enact'Rolling"Biennial Budgets 0 Cook County,IL Cuyahoga County,OH San Diego County,CA King County,WA *Dallas County,TX (Source:1993 survey by Victoria Runkle,Renton,WA Finance Administrator• 18 California Cities Enact'True"Biennial Budgets* 1. 2.Barstow • 3.Berkeley • 4.Claremont • 5.El Centro • 6.Emeryville • 7.Hemet 8.Highland . 9.Livermore . io.Lodi • ii.Los Altos 12.Oakland • 13.Paso Robles • 14.Pleasanton 15.San Carlos . 16.Santa Maria 17.Selma • 18.Watsonville 19.Yorba Linda • 20. • 25 California Cities Enact'Rolling"Biennial Budgets • 2.Alameda • 3.Calabasas 4.Colton • 5.Daly City • 6.Danville • 7.Del Mar . 8.Fullerton • 9.Glendale io.Hayward ii.Huntington Beach . 12.Laguna Hills . 13.Mission Viejo 14.Moreno Valley • 15.Palo Alto • 16.Palos Verdes Estates • 17.Rancho Palos Verdes • 18.Redding • 19.Redwood City • 20.San Luis Obispo 21.San Rafael �.Santa Barbara �.Santa Rosa .Santee �.Sunnyvale �.West Hollywood 0ources:California Society of Municipal Finance Officers 6117199 Two Year BudgetAppropriation Survey;Government Finance Officers Association • BudgetAwards;September OLA Telephone Interviews) Many of these cities have annual spending plans for cash flow management purposes. • ational Conference of State Legislatures,1999;1993 survey by Victoria Runkle,Renton,WA Finance Administrator;California Society of Municipal inance Officers 6/17/99 Two Year Budget Appropriation Survey;Government Finance Officers Association Budget Awards;September Telephone terviews. According to the Government Finance Officers Association,as a partial response to this loss of millions in property tax revenues,California cities Weloped and adopted multi-year budgets to mitigate the long-term fiscal impact of Proposition 13 and the future loss of property tax revenue. 101A good budget process is characterized by several essential features.A good budget process:incorporates a long-term perspective,establishes linkages 41 broad organizational goals,focuses budget decisions on results and outcomes,involves and promotes effective communication with stakeholders,and *ovide incentives to government management and employees."P.3 Recommended Budget Practices:A Framework for Improved State and Local Wvernment Budgeting.1998• We htt�//www.lao.ca.gQv 2000_reWrts/calfacts/2000_calfacts_state-local.html. on 2002,the Mayor"s Budget Office,Budget Analyst,and Controller issued their three-year projection for fiscal years 2002/2003 through 2004/2005 the Mayor and the Board on March 14 and presented their findings to the five-member Budget Committee March 19.At the request of Budget mmittee Chair Maxwell and Vice-Chair Peskin,a preliminary outlook for 2003/2004 was subsequently presented to the Mayor and Board on August 2002 to reflect known changes since the March report. The Controller"s six and nine-month reports project financial conditions for the start of the next fiscal year but do not project conditions any further 14to the future. The budget year,as opposed to the current year,is the next fiscal year for which the Mayor and the Board propose and appropriate funds respectively. • Appendix B:Potential Rolling Biennial Budget Calendar? • Selected Key Activities Based on Oakland,CA Next Fiscal July-Aug Sept-Oct Nov-Dec Jan Mar-April May-June • Year= - �ransition Year Feb (FY 03-04) •ayor/Board Pass FY03-04 Budget Retreat: Mayor"s Proposed • Budget,AAO, Review Perf. Balanced Budget i ASO Measures & Submittal; Budget • Objectives Deliberations Wtaff Review Details/ Procedures of Budget Budget Support Transition to Biennial Budget Instructions to Depts *Public Public Budgetary Election: Public participation • Participation Focus Mayor, 6 Groups or Supervisors Surveys ist Yr.Of July-Aug Sept-Oct Nov-Dec Jan Mar-April May-June • Biennial(FY _ 04-05) Feb • • Mayor/Board Pass FY04- Receive Proposed end Mayor"s Proposed • 05/05-o6 year of Approved Balanced Budget • Budget, FY04-05 FY04-05/05-o6 Submittal; Budget • AAO,ASO Budget Deliberations . Staff Supplemental Audits, Review of Budget Support Budget Programs, Financial Instructions to Mgmt Depts Public Public Budgetary • g rY Public participation S Participation Focus • Groups or Surveys end Yr.Of July-Aug Sept-Oct Nov-Dec Jan Mar-April May-June Biennial(FY _ • 05-o6) Feb • Mayor/Board Pass FY05-o6 Budget Retreat: Mayor"s Proposed AAO,ASO Review Per£ Balanced Budget Measures & Submittal; Budget Objectives Deliberations Staff � Review Details/ Procedures of Budget Budget Support • Biennial Budget Instructions to • Depts • Public Public Budgetary Public participation Participation Focus Groups or Surveys Return to Legislative Analyst Reports • • • • • •