The Oregon Option:Early Lessons from a Performance Partnership on Building Results-Driven Accountability

By Barbara Dyer, Partner
The Public's Work
for The Alliance for Redesigning Government
of the National Academy of Public Administation.

This research was supported by the U.S. Department of Health and Human Services, Department of Labor, and Department of Education. This report represents the views of the author and they should not be construed as representing the opinions or policy of any agency of the Federal Government or of the Academy as an institution.

July 1996

TABLE OF CONTENTS

Foreword

Table of Contents

CHAPTER 1: Introduction

The Challenge

The Oregon Approach

A New Partner

The Oregon Option

This Paper

A Caution

CHAPTER 2: Reaching Agreement

Achieving Clarity and Agreement on Results

All Parties Must be Ready

Parties Must Be Clear About the Results and the Process

The Workforce Cluster

The Family Stability Cluster

The Healthy Children Cluster

Some Lessons

CHAPTER 3: The Agreement: Who Does What?

Who Agrees to What?

Some Lessons

CHAPTER 4: Accountability for What? How?

An Emerging Concept of Accountability

What Results?

Results for Whom?

Deploying Which Programs?

Using What Performance Measures?

Can Fiscal Requirements be Aligned with Results?

With What Consequences?

Some Lessons

Chapter 5: Bridging Distrust

Building Trust in an Era of Distrust

Accountability Has Its Roots in Congress

A Final Thought

Officers of the Academy

Peter Szanton, Chair of the Board

C. William Fischer, Vice Chair

R. Scott Fosler, President

Feather O'Connor Houstoun, Secretary

Howard Messner, Treasurer

Alliance Advisory Board Executive Committee

David Osborne, Co-Chair

Gail C. Christopher, Co-Chair

Barbara Roberts, Co-Vice Chair

William Hudnut, Co-Vice Chair

R. Scott Fosler

Neal Peirce

Camile Cates Barnett

FOREWORD

Over the past 18 months, the Oregon Option brought together federal officials and their counterparts at the state and local level in Oregon to test new approaches to doing business in the intergovernmental system. The Academy's Alliance for Redesigning Government was engaged early on as a facilitator and constructive critic. From that vantage point, this report was developed to descibe the evolving concept of accountability for results that are crucial to the Oregon Option.

Clearly, the time is ripe to consider improvements to the current intergovernmental system. A 1995 study by the Advisory Commission on Intergovernmental Relations identified 618 intergovernmental categorical programs. As the number and complexity of these intergovernmental efforts have grown, so too has the list of requirements and restrictions imposed through statute, regulation, and custom. In this kind of centralized, prescriptive system, there is limited ability to experiment and to learn how to get the public's work done better, faster, and cheaper. What is needed is a more balanced approach that affords the system greater flexibility in return for accountability for achieving results. The Oregon Option is a laboratory for testing some of these ideas in a real-world context.

The Academy is pleased to have been asked to carry out this study. We hope that it will yield lessons useful to policymakers at all levels as they consider reforming and implementing new approaches to doing the public's work through our intergovernmental system. I commend consideration of this report to both policymakers and practitioners seeking to develop more results-oriented intergovernmental systems.

R. Scott Fosler

Academy President

CHAPTER 1

INTRODUCTION

...the question of a relation of the states to the federal government is the cardinal question of our constitutional system. It cannot be settled by one generation, because it is a question of growth, and every successive stage of our political and economic development gives it a new aspect, makes it a new question... Woodrow Wilson

The Challenge

The nation is once again engaged in an animated debate about the role of government and the appropriate balance of authority between levels of government. President Clinton proclaimed in his January 1996 State of the Union Address that the era of big government is over. In his FY 1996 budget, the president proposed several Performance Partnerships with state and local governments in the areas of public health, housing and environmental protection. Budget Director Alice Rivlin described these proposals as a "dramatic change in the way the Federal Government works with States to achieve the goals we share." The Republican Contract with America is an attempt to swing the pendulum from Washington to the states.

Few defend the status quo. Government's credibility is at a low point with the American people. Three out of four Americans mistrust government according to a recent survey conducted by the Washington Post, Harvard University, and the Kaiser Family Foundation. Government's ability to solve problems that matter to people seems to have given way to a litany of politician's unkept promises and a morass of bureaucratic road blocks. Meanwhile, people fear crime in the streets, worry about losing their jobs and health insurance coverage, and are concerned that the schools are not equipping children with the most basic skills.

The all-purpose elixir many leaders prescribe for what ails us is the Three D potion -- Downsize, Deregulate, and Devolve from Washington to the states. The debate about devolution tends to pull in two extreme directions. At one end are those who believe that Washington should be out of the social program business. Over the years Washington has created an entangled web of fragmented programs with cumbersome rules and regulations. States, they suggest, can do the job better and cheaper if given the money with no strings attached.

At the other extreme are those who fear that states, if given the chance, will abandon all responsibility for social welfare. They argue that states lack the capacity to assume all these new responsibilities and, with inevitable funding cuts down the road, the populations served by these programs will be the losers.

These debates about government's size or its structure and locus of authority beg a critical question:

What results do Americans want and expect from government and how do we support the proper intergovernmental alignment to achieve these results, effectively, and affordably?

There is a middle ground that gets more directly at this question. In this middle ground, federal, state, and local governments negotiate an agreement to achieve results. Each level of government shares responsibility for achievement, and each operates in a flexible environment that emphasizes ongoing learning and improvement. All are accountable for results. And progress is systematically measured and reported.

This middle ground -- we will call results-driven governance -- is being tested by local and state governments across the country, and through a growing number of federal, state, and local partnerships encouraged by President Clinton and Vice President Gore. One of the more far-reaching of these federal/state partnerships is with Oregon, known as the Oregon Option.

An Experiment in Results-Driven Governance: The Oregon Approach

Oregon is well along in a pioneering state and local effort -- the Oregon Benchmarks -- that focuses on results that matter to Oregonians. Through a participatory planning process beginning in 1987, the citizens of Oregon established a long-range vision of a better place. The Oregon Progress Board, a bipartisan citizen body was created by the legislature to translate this vision into a set of targets (Oregon Benchmarks) and measurable indicators. The Progress Board keeps track of achievement toward the benchmarks and reports this to the legislature and the public every 2 years.

Many local, county, and state government leaders use the benchmarks to guide them in allocating resources and designing services to help achieve results. Some county and local governments have replicated the benchmarking process for their own jurisdictions.

The Oregon Benchmarks cover issues as wide ranging as ecosystem protection, urban mobility, and industrial diversification. Human investment benchmarks focus on such outcomes as reduced pregnancy, diminished crime and recidivism, lower unemployment, higher per capita income, greater early childhood immunization, and stronger K-12 student achievement.

Through their 20-year strategic plan and the Oregon Benchmarks, the people of Oregon identified what they wanted to accomplish and they have begun to develop ways to pursue and measure their accomplishments. Several communities, nonprofit organizations, and businesses and civic groups are involved. State and substate government and nongovernmental leaders hope to replace an intergovernmental system entangled by the knots of mutual distrust with one that is aligned by a mutual commitment to results. And while this experiment in governance shows promise, there has been one critical missing link -- the federal government.

A New Partner

Federal grant and income transfer programs to state and local governments amount to over $237 billion in FY 1996. With over 600 grant programs, problems of overlap and inconsistency abound. In its 1993 report, the National Performance Review provides this harsh assessment:

Unfortunately, the myriad of federal mandates and regulations that accompany grant programs are cumbersome and very costly to administer, lack a coordinated implementation strategy between levels of government, and are not achieving the intended outcomes. Each separate program has its array of rules and regulations that must be observed, regardless of their impact on the effectiveness and quality of customer service. States and localities have limited ability to customize service delivery by integrating programs because of competing, often conflicting federal rules and requirements that accompany each program.

Despite problems in the intergovernmental system, Oregon governments have forged a number of promising innovations with federal agencies. Here are just two examples. In 1981, Oregon received a waiver from the Health Care Financing Administration to allow Medicaid funds formerly dedicated for nursing homes to be used for home and community-based care for the elderly and disabled. The Northwest Economic Adjustment Initiative established a new partnership between Oregon and the federal government to assist dislocated workers, businesses, and communities that must adjust to economic conditions and land management decisions which adversely affect the forest products economy.

These efforts are the foundation for a collaborative partnership with the federal government. Oregon and its federal partners have learned a great deal through these and other joint initiatives. But, important as these examples were, they remained ad hoc, operating on the edge of the larger system which continued to lumber along in its dysfunctional way.

Two years ago leaders in Oregon and the federal government committed themselves to testing a new kind of intergovernmental relationship -- a partnership that is driven by results.

The Oregon Option

This partnership, known as the Oregon Option, is testing the proposition that multiple levels of government can align their efforts to achieve results that matter to people. A Memorandum of Understanding (MOU) was signed by top-ranking officials in Washington and in Oregon committing them to work cooperatively to both determine the results to be achieved and to get the job done.

Among the signatories of the MOU were Vice President Al Gore, OMB Director Alice Rivlin, and Cabinet members Ron Brown, Henry Cisneros, Robert Reich, Janet Reno, Richard Riley, and Donna Shalala. From Oregon the signatories included several state and local officials such as then-Governor Barbara Roberts and Governor-elect John Kitzhaber, Multnomah County Chair Beverly Stein, Marion County Commissioner Randall Franke, Union County Commissioner John Howard, Lane County Commissioner Steve Cornacchia, Portland Mayor Vera Katz, Ashland Mayor Catherine Golden, and others.

The process formalized by the MOU has created the opportunity for local, county, state and federal officials to experiment with an intergovernmental relationship intended to improve Oregon's chance of making progress on their priorities.

This Paper: Learning to Build a Partnership for Results

This Alliance for Redesigning Government paper, supported by a grant from the United States Department of Health and Human Services (HHS) and the United States Department of Education, is intended for policymakers and practitioners attempting to develop the practice of performance partnerships. It is one of three papers focusing on the Oregon Option. The second paper is also sponsored by the Alliance and supported by the same grant -- Toward Results-Oriented Intergovernmental Systems: An Historical Look at the Development of the Oregon Option Benchmarks. It documents the development of the specific benchmarks and performance measures as they have evolved in the Oregon Option. The third paper, sponsored by the Annie E. Casey Foundation and written by Christina H. Macy, tracks its history and offers a broad perspective on this intergovernmental partnership.

The Oregon Option offers decisionmakers an opportunity to learn more about two often used and abused notions of governance. The first is the notion of results-driven accountability. Results-driven accountability means that leaders, managers, and staff are responsible for achieving specified outcomes. It is no longer sufficient to implement policy or operate programs by following procedures. They must also be able to demonstrate tangible progress toward the goals for which that policy or program was created. This notion is becoming more familiar in the public sector with a small but growing track-record. The Oregon Benchmarks provide the most developed statewide example of results-driven governance in the United States.

The second is the notion of intergovernmental partnership. The term partnership evokes pleasant images of hands grasping in shared commitment to a cause. But in practice, the term has euphemistically been applied to programs where one level of government uses its grantmaking or rulemaking power to achieve a purpose through another level of government. The Oregon Option is testing the former notion -- multiple levels of government mutually determining and achieving results.

While the effort builds on the particulars of the Oregon experience, it raises and illuminates some fundamental questions:

• How can multiple levels of government, and multiple entities within each level of government reach agreement on a desired set of results to be achieved within a state or substate region?
• Who agrees to what? What type of agreement is needed to define the partnership? What kind of structure is necessary to support the partnership?
• What accountability mechanisms will replace existing ones? How can equity be addressed in a flexible, results-driven system?
• Are adequate data available to measure results? How are results monitored across multiple providers and numerous funding sources?
• What if the results are not achieved? What are the most constructive responses from the many parties? How are distinctions made between "acceptable failure" and actual under performance? What are the learning opportunities? What are the consequences?
• How do we bridge distrust across branches and levels of government to achieve results?

In this paper we take stock of what we are learning from Oregon to begin addressing these questions. In the next section we focus on the challenge of defining results across levels of government. Oregon citizens had already engaged in a process to determine their priorities for a better future. In this initiative, the state was ready to negotiate with the federal government.

In the third section we look more closely at the agreements and the structure of the partnership. We examine the strengths and limits of the formal agreements and discuss the kind of support needed to carry out the work.

In the fourth section we focus on the emerging ideas about accountability for results that this experiment reflects. This is the central issue: what does it mean for multiple levels of government to actually be accountable for results? The Oregon Option is still in the trial and error stage on accountability, but some lessons are beginning to surface.

Section five touches on the crucial role of Congress if flexibility for results is to become a way of governing.

A Caution

Because this is an iterative process, the answers are in-the-making, not final. Some answers raise new questions. Others are disappointing. But on the whole, they begin to suggest a different kind of intergovernmental relationship.

Our current intergovernmental relationships are like the classical symphony orchestra. Every note is written. Every instrument has its part. Any departure from the score is problematic. The Oregon Option and other experiments like it are more like a jazz ensemble. The musicians share a common key, a basic tempo, but the details emerge from the playing.

CHAPTER 2

REACHING AGREEMENT

How can multiple levels of government, and multiple entities within each level of government reach agreement on a desired set of outcomes to be achieved within a state or substate region?

Achieving Clarity and Agreement on Results is Hard

Perhaps the most difficult, and certainly the most important ingredient in a results-driven partnership is developing a shared understanding among the partners of what results they seek to achieve. Reaching agreement on results is tough in the public sector for several reasons. First, the conversation leading to agreement on any results that matter to the public has to deal with deeply held and often conflicting values. There are strong advocates on all sides. Too often our approach to values conflicts in a the public arena is to ignore, simplify, and/or distort the values conflict. The legislative solution is to enact a little something for everyone in a compromise. Unfortunately, the consequence of good political compromise is a law that is fuzzy on results.

Second, we have had a long tradition of creating a new program for every new problem. As a consequence, we now have an entangled landscape of overlapping programs with at times inconsistent goals and rules. Or we have programs operating at cross-purposes. Another consequence is that many officials throughout the service delivery system view the world through their program lens. Programs are seen less as a means toward an end and more as an end in themselves. This makes it hard to discuss the underlying purpose for which a program is designed or whether that purpose is being achieved. The recent debate about welfare reform is a good example. Some citizens and policymakers might want to focus on the broad issues of wealth creation and reducing dependency among the poor. But the central public focus has been confined narrowly to reforming the specific rules and requirements within the welfare program.

Third, some results, particularly those related to human well-being, are elusive. They are difficult to measure and it is even more difficult to determine the underlying cause of success or failure. Many policymakers and program managers are reluctant to have decisions hinge on such imperfect proxies for cause and effect.

Fourth, many public employees are uncomfortable with the idea. They worry that all this talk about results is merely shorthand for eliminating their jobs. They worry about being held accountable for things beyond their control. And they are concerned that they will not have a say in the process.

Reaching agreement across governments has all of these challenges and more. Each government has its own constituents; social, financial and political circumstances; rules and governance structure; and traditions which influence the process. These need to be understood by all the parties. At the same time, there is rarely a clear governance structure with authority that bridges the multiple governments involved in the agreement process.

Governments have a tradition of relating to each other in ways that are often not conducive to collaboration. Local governments are suspicious of state governments. States are suspicious of the federal government. And the federal government is suspicious of the other two. Counties compete with other counties. States compete with other states. Some state and local leaders suspect that "results" in federal parlance is shorthand for more requirements. And federal officials suspect that state and local government are using results as shorthand for receiving money without accountability.

These challenges make achieving clarity and reaching agreement on results difficult but not impossible. There are thousands of state-local and regional collaborative efforts across the country, and local/state/federal performance partnerships are becoming more common as well. In this paper we focus specifically on the results-driven agreement among the federal government, a state, and its county and local governments. This paper will look more closely at the dynamics at play in the Oregon Option to draw some lessons about the process of reaching agreement.

All Parties Must Be Ready

There are several important factors which made it feasible for officials in Oregon and Washington, D.C. to consider an agreement committing themselves to results.

A Credible Statewide System Was In Place: Oregon had built a solid foundation of results-driven governance beginning with its strategic plan, Oregon Shines, in 1988. The system for defining benchmarks and measuring results was formalized in statute and had strong bipartisan legislative and gubernatorial leadership. The public had been engaged through community meetings and electronic voting in defining the benchmarks. The state had addressed many of the tough values issues discussed above in an open fashion. And the public dialogue was about a better future, not about a better government program.

A critical mass of Oregon employees and political leaders have bought-in to the process. Performance is measured and reported every 2 years. Now, 8 years after the initial strategic plan was developed, conditions have changed in Oregon. The benchmarking process enables leaders to track these changes, reexamine their assumptions, and revise their strategies. Governor Kitzhaber appointed a task force in April 1996 to assess Oregon's progress toward achieving its strategic vision. Oregon's form of results-driven governance is a dynamic approach that helps both to signal and adapt to changing conditions. Getting this far has not been quick or easy and the effort is far from complete. But it is steady in its evolution.

Catalytic State and Local Leadership Dedicated to Results: The Oregon Benchmark effort is not a project. It is a way of doing business. It did not happen overnight. It evolved with each successive year, and with each successive governor and legislative session. And it is still evolving. What is remarkable about this effort is not so much its logic -- it is basically applied common sense. What is remarkable is that this concept of governing is slowly becoming part of the political culture of the state.

Governor Kitzhaber is the third governor to lead this results-driven charge. Political leaders in America rarely embrace initiatives begun by their predecessors. Their election signifies a new beginning which they often interpret as a call to discard the old. By contrast, the Oregon Benchmark initiative is like a big mound of clay which is sculpted and refined by each new leader. Governor Goldschmidt formed the foundation. Governor Roberts solidified the process and communicated the vision. Now Governor Kitzhaber, in office only since 1995, is the sculptor.

Governor Kitzhaber is dedicated to a system of governance that can deliver meaningful results to citizens in an era of resource limits. He has a history of commitment to the principles of results. As president of the Oregon Senate, he was the architect of Oregon's health care reform and a champion of the legislation establishing the benchmarks and the Progress Board. Now as the state's chief executive, Governor Kitzhaber is still considering how to shape the clay. Some have expressed concern that he has not moved quickly enough or given sufficiently high priority to the Oregon Option. Others observe that this governor, thoughtful and deliberate in his approach, is considering how to sharpen the lines to deliver tangible results. In appointing the task force to review Oregon Shines, the governor emphasized his commitment to implementing a strategic plan that reflects current realities.

Several local governments also have adapted the strategic planning and benchmarking process to fit their circumstances. Progress Boards have been established in cities and counties such as Albany, Baker, Bend, Gresham, Clackamas County, Clatsop County, and Portland/Multnomah County among others.

The Political and Policy Environment in Washington was Right: The environment was right in Washington, D.C. for such a partnership. The Government Performance and Results Act had just been enacted. It directed the Office of Management and Budget (OMB) to administer several federal pilots which would pave the way for the federal government's transformation to results-driven governance. OMB Director Alice Rivlin was eager to attempt an intergovernmental pilot. Much of the domestic agenda the national government is asked to achieve can only be accomplished through the efforts of state and local government. A partnership with Oregon, while not a formal GPRA pilot, would be consistent with then-Director Rivlin's interest in pursuing an intergovernmental results-driven experiment.

President Clinton was committed to developing intergovernmental arrangements that would improve results on the ground. A host of initiatives such as the Empowerment Zones, Enterprise Communities Program, and the Education Flexibility Partnership Demonstration established by the Goals 2000 Act were pushing at the boundaries of intergovernmental relations.

The President's reinvention effort, The National Performance Review led by Vice President Gore, planted additional seeds for creating partnerships. The Vice President focused on the concept of granting administrative flexibility to state and local governments in exchange for results. In his first annual report of the National Performance Review, Vice President Gore emphasized several goals for changing the intergovernmental service delivery system:

• The number of categorical grants must be reduced.
• Governments at all levels must reduce the degree to which unfunded mandates are imposed on other levels of government.
• Program rules and regulations must be fundamentally rethought and their focus changed from compliance to outcomes, from sanctions to incentives.
• Federal interdepartmental and intergovernmental collaboration must be actively facilitated if any real improvement in government's credibility is to be successful.

In addition, the November 1994 elections brought about sweeping changes in the Congress which made some form of grant consolidation and devolution seem inevitable. The Oregon Option was a product of this environment. It provided a way to test the notion of flexibility in exchange for results as a variation on the devolution theme.

A Staff Structure Was Formed To Work-Out the Details: Vice President Gore formed an Interagency Action Team to conduct the planning and ensure overall coordination of the Oregon Option. The action team was chaired by a top-level staff person from the Vice President's National Performance Review. It included political and career staff from the President and Vice President's office, the Office of Management and Budget, the Census Bureau and the Departments of Health and Human Services, Education, Labor, Justice, and Agriculture.

From Oregon, participants included the director of the Oregon Progress Board and other key Progress Board staff, state agency heads, and county, regional, and local officials.

In July 1994, Secretary of Health and Human Services Donna Shalala chaired a meeting in Washington at which Oregon Governor Barbara Roberts, Multnomah County Chair Beverly Stein, and Portland Mayor Vera Katz presented a proposal for partnership. This meeting involved several Cabinet members and agency staff. It established high-level expectations for this partnership and prompted high-level commitment. The Interagency Action Team helped to organize this meeting and was able to provide the needed staff support to follow-through on decisions beyond this event.

Parties Must Be Clear About the Results and the Process

In September 1994, after several interagency action team meetings and conference calls, over 100 federal, state, county, and local officials gathered in Portland, Oregon to plan as a group. The purpose of the three-day meeting was to reach agreement among the participants regarding the what and how of the Oregon Option -- What benchmarks do we focus on? How do we move forward?

Start With Realistic, Shared Expectations: By all accounts, this was not an easy few days. Everyone came to the table with different expectations and much distrust. Several federal officials expected that Oregon would have identified the federal obstacles to progress that they wanted help on. Several Oregonians expected that the federal officials would have specific priorities that they had to focus on and a clear-cut path to eliminating administrative red tape. They all expected that in less than 3 days they would be able to sort out the issues and reach an agreement. They were wrong on all counts.

We have learned from these early meetings of the Oregon Option and other attempts at intergovernmental partnerships that public leaders' time horizons are driven by budget and elections cycles and crisis management. These condition public officials to expect quick action, quick results. Few are prepared for or tolerant of the time it takes to build a credible intergovernmental agreement. Fundamentally, these agreements work only if the parties take the time to build trust and capacity.

Despite their difficulty during this first encounter, the participants made discernable progress. By the end of the meeting they had begun to clarify and adjust expectations. They agreed on "clusters" of benchmarks to focus on, and they agreed on a set of principles to guide a memorandum of understanding between Oregon and the federal government.

Identify Clusters of Benchmarks of Mutual Interest: The benchmarks clusters helped the parties to frame their dialogue. The clusters they chose to work on were: (1) a strong workforce and education; (2) stable families; and (3) healthy children. They also formed a cluster to focus on cross-cutting data issues. Intergovernmental, interagency teams organized around each of these clusters. They were able to array all of Oregon's benchmarks relating to their cluster and debate which among them made sense to work on in partnership.

Define Terms and Establish Criteria To Guide Benchmark Selection: To reduce the confusion which invariably occurs when people try to agree upon results, it helps to define terms and establish selection criteria.

In Oregon, decisionmakers use "benchmark" to mean a measurably improved future condition or circumstance. Some terms used interchangeably in public parlance include outcome, milestone, result, or goal. A measure of progress toward achieving the benchmark is interchangeably referred to as a performance target, indicator, or measure.

The Oregon Benchmarks provided a menu of options. Each cluster had to determine what to work on. The federal partners did not want to impose their will on the state. They were respectful of the state's internal process and wanted to take their lead from Oregon. At the same time, they had to focus on benchmarks for which they had both the authority and capacity to act. Oregon had to consider which of their high priority benchmarks would best be achieved in partnership with the federal government. The cluster groups applied these general criteria to select benchmarks:

These criteria provide a framework for negotiating benchmarks across levels of government. Rather than one level of government imposing uniform benchmarks and performance measures on another level, each puts their own priorities on the table. They can identify common interests, pinpoint differences, and assess what's realistic to measure and/or achieve.

Determine The Process For Reaching Agreement: At its core, this is a negotiating process. Two or more parties determine what is to be achieved, under what terms, with what resources, and by whom. In the Oregon Option, cross-agency, intergovernmental staff groups formed to work on each cluster. Each took a different tack.

The Workforce Cluster -- A Continuous Improvement Partnership Process

The workforce and education cluster members included representatives of community colleges, local employment offices, local officials implementing the Job Training Partnership Act, Adult and Family Services, and private sector regional groups. State partners included the Department of Human Resources, JOBS and Vocational Rehabilitation, the Department of Education Office of Community College Services, and the Workforce Quality Council. Federal partners included central and regional office representatives from the Departments of Labor, Education, and Health and Human Services, the National Performance Review, and the Office of Management and Budget.

The participants used the criteria above for selecting benchmarks from the larger pool of Oregon workforce benchmarks. The federal Departments of Labor and Education analyzed each of the potential benchmarks to determine which matched the national interest and their organization's authority. A clear convergence of interest and authority was a necessary condition.

The process for reaching agreement that the workforce and education cluster uses is an iterative, continuous improvement model. They have conference calls every two weeks, they circulate an information fax weekly, and they have planning meetings as needed. They have a written agreement that specifies which benchmarks they are working on and what each level of government is accountable for.

This cluster is focusing on 6 benchmarks within 2 key segments of the population: the emerging workforce -- those young adults who are newly entering the workforce -- and the transitional workforce -- people who are unemployed or underemployed.

WORK FORCE CLUSTER BENCHMARKS

The Family Stability Cluster -- An Application and Approval Process

The family stability cluster included the Oregon Department of Human Resources, the Oregon Progress Board, the Workforce Quality Council, Adult and Family Services, Oregon Legal Services, business representatives, and community college representatives. The federal Department of Health and Human Services is the federal partner.

This group took a more prescribed path. They requested that the secretary of the Department of Health and Human Services grant a waiver of the law governing welfare as is her authority under section 1115 of the Social Security Act.

They followed this route for two reasons. First, Oregon had begun developing a welfare reform plan prior to the onset of the Oregon Option. In 1993 the state legislature adopted a proposal requiring that the existing public assistance system be replaced with a family support and work development system. The Oregon Progress Board and the Workforce Quality Council had the responsibility for developing the plan.

Second, the 1115 waiver process was in place. Since President Clinton took office, the administration has granted waivers in over 40 states to test different approaches to reforming the welfare system.

Because the waiver process is defined by a set protocol there was less of an opportunity for federal and state officials to develop a tailored partnership. In this approach the state submits a proposal and the federal government negotiates the terms and conditions of the waiver with the state. The secretary's decision is based on the merits of the state's proposal, the soundness of the proposal's evaluation design, its cost neutrality, and the limits of the secretary's waiver authority.

Though the waiver process is straightforward compared to the decision processes used by the other clusters, deliberations between state and federal officials leading to the waiver decision were not always smooth. Some in Oregon and some federal officials assumed that, since this was a route taken within the larger framework of the Oregon Option, there would be more give-and-take, the decision process would move more quickly, and Oregon would not be subject to the same quality control requirements as other waiver applicants. Others assumed that once Oregon had applied for a waiver they had to be treated like all other states subject to the same time lines and requirements. They essentially viewed this as operating outside the framework of the Oregon Option. All would have benefitted from better communication.

In addition, the deliberations focused very little on the specific set of results or benchmarks Oregon hopes to achieve, and more on procedural and cost issues. In part this is because of the prescribed nature of the waiver process, and in part this is due to the fact that this was a request for a change involving an uncapped entitlement program. Revenue neutrality is a requirement imposed by the Congress. If a state waiver contributes to a precipitous rise in the welfare rolls, the secretary of the Department of Health and Human Services will have to answer to Congress for the cost increase. This would not be popular at a time when government is trying to cut costs; and it is not allowed.

The federal government approved the Oregon Option Welfare Reform Waivers on March 28, 1996. Though the benchmarks were not the central focus of federal officials, Oregon leaders are dedicated to their achievement. They have targets for achievement within the first 2 years of implementation and targets for the year 2000.

FAMILY STABILITY CLUSTER BENCHMARKS

The Healthy Children Cluster -- An Early Win Process

The Child Health Cluster included county health commissioners, the Oregon Primary Care Association, the State Health Division and Department of Education, Oregon's Commission on Children and Families, and the Oregon Progress Board. The federal partners included the Vice President's National Performance Review, the Departments of Health and Human Services, Education and Agriculture, and the Office of Management and Budget.

Cluster members developed a hierarchy of priorities drawn from the Oregon Benchmarks on child and maternal health. They developed a timetable for achieving measurable progress. They wanted to show tangible results quickly in order to build confidence in the process and gain support for the longer-term challenges.

They identified benchmarks that they could achieve within the first two years of effort, in four to six years and in eight years. They concluded that they could achieve their stretch goals for immunizations, prenatal care and child care within the first two years. They determined that benchmarks for reducing HIV-infected babies, child abuse, and teen pregnancy could be achieved in four to six years.

Some Lessons

The process each cluster groups followed shaped relationships among the participants and defined the benchmarks they intended to focus on. The iterative process used by the workforce and education cluster has gone the farthest to test the notion of partnership. The child health cluster has achieved early, tangible results. The family stability cluster operated in the political spotlight of welfare reform and under the constraints of a more prescribed process. While the negotiating process did not shift the boundaries of federal/state relations, the benchmarks selected and the strategy for achieving these benchmarks can. The federal, state, and local players now have the opportunity to develop a real partnership around implementation.

In this section we've identified some important start-up lessons:

• The state must be sufficiently prepared with identified priorities, committed state and local leadership, and capacity to follow through.
• The parties involved in the agreement process must have the authority to act.
• The parties should have realistic expectations. Relationships take time to build. Results take time to achieve.
• The parties should have a clear understanding of and a commitment to the process they are using to reach agreement.
• The parties must clarify what results they are prepared to focus on and why.

Another observation is that each cluster focused on a few vital benchmarks. They did not develop all inclusive laundry lists. Too often in the public sector, whether in the spirit of compromise or of control, officials develop long and detailed lists of goals complete with specifications for their achievement. These have had the unintended consequence of deflecting attention from results to procedures. They also enable people to avoid making hard choices about priorities.

Now that we have looked at some of the start-up issues and lessons, let's look in the next section at ways to structure a credible results-driven partnership agreement.

CHAPTER 3

THE AGREEMENT: WHO DOES WHAT?

Who agrees to what? What type of agreement is needed to define the partnership? What kind of structure is necessary to support the partnership?

Who Agrees to What?

In something as complex as our federal system, many people must be involved in and committed to the results-driven agreement if it is going to work. Those at the highest level must give license and legitimacy to the agreement. They must make the effort a priority, define its general purpose and establish the broad principles to guide its implementation. Their top staff operate on their behalf as strategists. These staff keep the initiative moving and in focus. They provide broad direction and serve as a bridge to their bosses.

Those at the operational level are the tacticians. They must work out the details about how the agreement will be carried out within the boundaries of statutory and fiscal accountability.

Those With Authority: The Oregon Option is about improving results affecting people -- their families, their children, their education, and their jobs. The challenge the Oregon Option poses is whether the actions and programs of government can better align to enable more progress within the state. Consequently, people who have the authority to focus government actions and programs must be at the table.

At the federal level, those involved include the Vice President; the secretaries of the Departments of Health and Human Services, Education, Labor, Housing and Urban Development, Justice, Commerce, and Agriculture; the director of the Office of Management and Budget, and the assistant to the President for domestic policy. In Oregon the governor, and some county commissioners and mayors have been involved.

Ideally, legislative leaders would also be engaged. In the Oregon Option, Oregon Senator Hatfield played an active role in infusing the concepts of results-driven partnership into the congressional debates about flexibility and block grants. In a letter to Senator Nancy Kassebaum, he wrote:

The Oregon Option was conceived as a means to transform our current system to a model which is customer-centered and driven at the level closest to the customer, is focused on results, has relationships which are collaborative in nature, and ultimately is oriented toward self-sufficiency. In such a model, accountability is focused on the basis of fiscal and program concerns, and helps ensure that success is produced, not merely good explanations for why we've failed.

However, aside from Senator Hatfield, federal and state legislators have not been integrally involved in the Oregon Option. This is a weak link in the chain. Whatever flexibility the executive may have to waive a program's rules or redirect funds is defined by the program's enabling legislation. Many of the detailed reporting requirements to ensure programmatic and fiscal accountability have their roots in the Congress or in state legislatures. Results-driven intergovernmental partnerships are testing an approach which is fundamentally different than our current one and the learning must occur across branches of government as-well-as across levels of government.

Agree on Purpose, Principles, and Responsibilities: Federal and state leaders signed A Memorandum of Understanding (MOU) in December 1994 which officially launched the Oregon Option. The MOU establishes in broad terms, the purpose, principles, and responsibilities each of the parties agrees to.

Let's look at what this Memorandum of Understanding does and does not do. It does outline a set of governance ideas to test:

• the idea that we restructure the intergovernmental service delivery system to achieve measurable results for people within a state
• the idea of partnership -- shared investment, shared risk, and shared gain
• the idea that the system be managed and evaluated on the basis of results
• the idea of customer satisfaction
• the idea of decentralizing the responsibilities for service design and delivery and shifting more authority to the front-line
• the idea of streamlining and simplifying processes and eliminating barriers

It does affirm a special relationship between a state and the United States government. While the strengths and limitations of this relationship are not well defined, Congress has subsequently referred to "states having a Memorandum of Understanding with the United States government" in statutory language describing one condition necessary for possible Labor Department waivers.

It does establish high-level commitment and expectations using the important symbolism of an agreement signed by top leaders from each government.

It does express a strong hope that the lessons from this effort will serve to improve, if not help redesign, intergovernmental programs.

It does not grant any additional authority or resources to the effort. It assumes that there is a great deal of latitude within existing authorities to test these ideas. And there is. In fact, operating this demonstration within existing limits allows those working on the Oregon Option to gain a more refined understanding of what those limits actually are. However, it also does not reallocate any resources to ensure that the effort is adequately managed and staffed.

It does not hold anyone accountable for the success of the Oregon Option. The Memorandum of Understanding is useful symbolically, but it has no inherent muscle. It speaks about a results-driven service delivery system, but it does not define expectations of results for the partnership. It does not pave a clear path that would lead to larger systems redesign. There are no explicit responsibilities identified for the partners. It leaves open the question regarding who leads, how information is shared, and what reports the President, Vice President or governor should have. Consequently, the strength and staying-power of this experiment is dependent upon the personal dedication and perseverance of its partners.

In the Workforce Cluster, the team developed a written agreement which spells out in more detail expectations, roles, responsibilities for accountability, barrier removal, and data systems. While this agreement has no more authority than the Memorandum of Understanding, it does provide more guidance for implementation. Here is an excerpt from the agreement which deals specifically with the issue of removing barriers:

OREGON WORKFORCE OPTION -- BARRIER REMOVAL

Should the Memorandum of Understanding be more specific about roles, expectations, and results? Since the signing ceremony in December 1994, the signatories have not met as a group to review progress. At the federal level, there have been a few meetings at which cabinet officials have been briefed on the Oregon Option. Individually the federal signatories have had different levels of involvement based upon the extent to which the Option intersected with other priorities. At the state level there have been few if any meetings of state and local signatories as a group.

By contrast, those officials engaged in the Workforce Cluster meet regularly to work out problems and raise new questions. Is this because they have a more specific agreement?

Commitment from the top is an important ingredient for success. Having a signed agreement such as the MOU is of greatest value when the signatories make a conscious effort to incorporate its principles into the daily routine of governing. This may not require that there be more specificity written into the MOU, but it does require that the leaders establish an understanding about how they will implement this agreement.

Those Who Can Follow-Through: Agreement and commitment by top-level leaders is a necessary but not a sufficient guarantee of success. This form of partnership is dealing with existing resources, rules, and practices in new ways. At the operational or tactical level, the partnership needs people who grasp the concept of a results-driven intergovernmental system and who know the opportunities and limitations of the existing system. The tactical group must be able to anticipate issues, know where the levers are, and create options. This group must be able to answer, or get answers to, questions dealing with things as wide ranging as:

• What authority is granted to whom under the laws at issue in the partnership
• What requirements exist -- monitoring, reporting or auditing requirements -- and how these can be linked to the benchmarks
• Whether and how to blend separate funding streams
• What related initiatives can help inform this partnership
• Who else to pull into the group when they can't find the answers on their own

In the Oregon Option, the Interagency Action Team members are the strategists who form the bridge between the tacticians and the cabinet-level decision makers. The team, chaired by an official from the National Performance Review in Washington, and the director of the Oregon Progress Board, met first weekly and then at least monthly for the first eighteen months of this initiative. The team involved political and career officials from all of the partner federal agencies, OMB and the Domestic Policy Council. In Oregon, the team included state, county, and local officials.

The regular meetings provided the opportunity for tracking progress and trouble-shooting. These meetings were important to maintaining the momentum of the effort. In November 1995, the Action Team decided to meet on an as-needed basis, rather than monthly. The consequence is that they have not met as a full group since. Although the National Performance Review staff still provides assistance to the cluster groups, many of the Action Team members have lost track of the initiative.

Work Out The Details: The Cluster Groups are the real tacticians. They have to find ways to make it all work. The Cluster Group members are the What if? and How? people. What if we want to use childrens' Social Security numbers as our universal identifier for our statewide immunization registry? How do we accomplish this? What if we want to blend six separate funding streams all dealing with immunization into a single program? How can this be done? What if a consortium of high schools and secondary schools want to run a school-to-work program? How can we get the flexibility to allow them to apply for federal funds? The Cluster Group tacticians in the Oregon Option have had to work adeptly with what is in order to test what is possible.

For the tacticians to operate effectively, a number of elements must be in place.

Ongoing communication. The Workforce Cluster members place a high premium on communication. They convene every 2 weeks by phone to deal with whatever comes up. They issue a fax-update every Friday. And they have held face-to-face planning sessions in Oregon and in Washington, D.C.. This contributes to a growing level of trust among the intergovernmental partners, a safe place to test options, and a sense of order in a field of chaos.

Great expectations. The Child Health Cluster members established an ambitious goal and mobilized their forces to achieve it. They were determined to raise the immunization rate for young children from below 50% to 70% in less than 1 year.

The Cluster tacticians made it possible for Oregon to consolidate 6 separate funding streams, get permission from the Social Security Administration to use a child's social security number as a common identifier for the statewide immunization registry, gain cooperation from the Centers for Disease Control and Women, Infants and Children (WIC) to help develop a model electronic data system, and develop a data packet on health indicators comparing progress in every region of the state. The result -- they have raised the immunization rate of two-year-olds to 70% in less than 1 year.

Some Lessons

The experience of the Oregon Option provides some important lessons about how to structure an intergovernmental results-driven partnership.

• An effective partnership must have the commitment and involvement of people with authority, preferably from both the executive and legislative branches. These leaders establish the purpose and broad principles of the partnership. They also must establish clear expectations regarding what will be communicated to them by when. Leaders of the partnership should have scheduled briefings and opportunities to reflect upon the results as a group.
• An effective partnership must be supported by an infrastructure that includes strategists and tacticians. People who keep the big picture in focus, and people with the technical know-how to make that picture a reality.
• People must be in regular communication. Formal and informal channels of communication must be developed by the partners early and maintained over the life of the partnership.
• Written agreements should spell out principles, expectations of results as-well-as the roles and responsibilities of the partners.

In this section we focused on the infrastructure for partnership. But, no matter how well the partnership is designed, unless it actually achieves better results, it probably is not worth the effort. In the next section we will look at the question which is at the heart of performance partnerships -- How can multiple levels (and branches) of government really become accountable for results?

CHAPTER 4

ACCOUNTABILITY FOR WHAT? HOW?

An emerging concept. What results are we accountable for? For whom? Deploying which programs? Using what measures? Aligned with what audit specifications? With what consequences? What roles?

An Emerging Concept of Accountability

In their partnership agreement, the Oregon Workforce Cluster members refer to accountability as the responsibility accepted by each individual for promised results. They call accountability the glue that holds the system together. It provides cause for celebration when targets are met. And it offers an opportunity to reexamine assumptions and make course corrections when they are not met.

The Oregon Option reflects an emerging idea that accountability should be more about results and less about adherence to procedures. It should reflect the reality of equally strong parties sharing responsibility for improving something that the public cares about. It should replace the notion of one weaker party seeking permission from another stronger party. The process must enable its participants to track progress, learn from the data, and seek ways to improve continuously.

By contrast, we now rely on accountability mechanisms that tell us little about results. We can track the incidence of fraud and abuse. We can see that funds are distributed equitably. We can ensure that only those who are statutorily eligible for a program are served. We can make certain that rules are properly followed. But we cannot determine whether, as a function of our actions, children are healthier or better educated, or whether adults are better able to earn a living wage.

But what does it mean for governments to be accountable for results? The Oregon Option casts this question into high relief. What results? Is it possible for multiple governments to align policies, programs, and resources vertically and horizontally to achieve common objectives?

Who is to be held accountable for these results? In this partnership, can the federal government be held accountable along with Oregon for achieving the benchmarks? Can the federal Department of Labor in the Workforce Cluster share responsibility with the federal Department of Education and the state's governments for the school performance benchmarks? What measures reliably indicate progress? What are the rewards for success and the consequences for failure?

And what are the practical tasks and responsibilities of the partners in a shared accountability environment? What new roles and routines would replace existing ones? What kinds of skills are needed?

The Oregon Option serves to elevate these questions. It does not provide the answers but it does lend some depth and focus to the inquiry. The evolution of the Workforce Cluster, traced throughout this chapter, provides a good illustration of the complexity of these questions.

ACCOUNTABILITY: CURRENT AND EMERGING

What Results?

Accountability for results begins with all parties understanding and agreeing to a set of results. Because this is an emerging field, people often use different terms to mean the same thing. For example, "results" is used interchangeably with outcomes, benchmarks, goals, or milestones. To avoid confusion, the parties should choose a term, understand that it means an improved condition in the future, and move on.

Results are about ends, not means. They reflect an improved condition. They are cast in broad but measurable terms. They should specify what will change, for whom. In chapter two of this paper we focused on the challenge of defining results. Recall that the Workforce Cluster chose these benchmarks:

WORK FORCE CLUSTER BENCHMARKS

The focus at this stage should be on people and places, not programs and funding streams. Partners must be able to look through a wide-angle lens at issues and opportunities for results such as improved child health, safe and affordable housing or building a world-class workforce.

Results for Whom? The Equity Question

Modern American public policy is shaped by a deeply rooted value to ensure equity of opportunity and fairness in the distribution of resources. That value sometimes translates into a practice of sameness -- uniform standards, equal funding. Sometimes it translates into a practice of creating incentives or requirements to serve those who have historically been treated inequitably. We consider differential treatment to be inequitable. However, what about differential outcomes? How does this emphasis on results influence our ideas and practices to achieve equity?

A far-reaching court decision leading to the reform of Kentucky's education system defined equity in terms of results. The court ruled that the system is inherently inequitable if children are failing to learn. The significance here is that Kentucky was not forced to even out to the lowest common denominator. They were not told to spend education dollars equally. They were not required to ensure that targeted populations be granted equal access to special programs. Instead, they were required to achieve results for all of their students. They had to establish differential targets for each school to calibrate improvements to the individual school's baseline. The practice of sameness is, in this case, inconsistent with equity of outcomes.

In the Oregon Option, the Workforce Cluster is grappling with the equity question as they define performance measures. They are concerned that their choice of measures could create a perverse incentive for service providers to work with those most likely to succeed, leaving behind those hardest to serve. This is referred to as creaming. The cluster members developed a draft statement of principle as a way of keeping focused on the equity question:

All Oregonians have a right to expect and fully share in the success of our state. In measuring these workforce outcomes, we recognize that certain vulnerable Oregonians are proportionally under-represented in the achievement of the benchmarks. Therefore, those most vulnerable will have full access and receive equitable services. The workforce development system in Oregon will not be driven by lowest cost or quickest success, but rather by the value that all Oregonians will have the opportunity to be successful.

This statement combines the ideas of equity of results by pointing out that certain people are under-represented in achievement; equity of treatment by declaring that these people should receive equitable services; and equity of opportunity. The practical application of this statement of principles may prove difficult. Achieving results among more vulnerable populations equal to those of the rest of the population would likely require a heavier investment of resources for this population. In an environment of limited resources, policymakers are faced with a tough choice between serving fewer people intensively, or serving a larger number of people who can make it with just a little help. Equal results require unequal effort.

There are no easy answers to the equity question. An important first step is to unpack the issue and make clear decisions about both the policy intent and the practical strategies to be deployed. One advantage of a results-driven system is that decision makers can more closely monitor the consequences of their choices and make adjustments based upon data.

Deploying Which Programs?

The partners in a results-driven partnership must learn to view programs as a means to an end. Programs are pieces of a strategy. Several programs might work in tandem in order to achieve results such as decreasing teen pregnancy. For example, in one community a combination of teen mentoring, work apprenticeships, a school health clinic, a media/education campaign directed at teens and their parents, and after school recreation programs might be the best mix. The challenge is to create a portfolio of initiatives which, as a group, enhance the likelihood of success.

The Oregon Workforce Cluster is focusing on results which call into play several possible federal, state and local programs such as Carl Perkins, JTPA, JOBS, Wagner Peyser, Dislocated Workers, Adult Basic Education, Vocational Rehabilitation, and Higher Education. There are literally hundreds of programs affecting the results that the Oregon Option partners are trying to achieve. Many federal programs are undergoing modification in order to provide state and local governments more flexibility. Several are being considered for consolidation. This is a transition time.

Partnerships like the Oregon Option provide us the opportunity to discover how these programs can work together more effectively as pieces of a larger strategy. The partners must stay focused on results and assess which combination of programs best enable them to succeed. But this is not easy because it is a different starting point for most public officials. They view their program as the starting point largely because that is what they are employed to do and are currently held accountable for.

As a step toward aligning programs and resources vertically and horizontally, the partners should address these questions:

• What will it take to influence results? What combination of actions would likely make a difference?
• Which existing programs, on their own or in combination, have potential to influence the results we hope to achieve?
• What would enable us to deploy these programs more effectively as common elements of a larger strategy to achieve our results? Common eligibility requirements? Reporting requirements? Definitions? Planning cycles? Blended funds? Can these be achieved? Through what means?
• Who is responsible for implementing these programs and what authority do they have to operate more flexibly?

The answers to these questions are essential to building a strategy for results. They allow the partners to chart programs and resources in order to assess their strengths and identify gaps. And, depending upon what results the partnership is aiming toward, the answers will vary.

In this form of partnership, the decision to combine programs and blend funds would be determined by a pragmatic assessment of what's needed to achieve results. Once the partners identify the combination of programs, ideally, they would develop common procedures and requirements across programs. For example, if the strategy involves programs from the departments of Labor, Education and Health and Human Services, the partners would be responsible for developing common cross-agency, cross-program procedures.

Waiver processes might be better coordinated across departments, for example. Strategic plans might be cross-cutting rather than specific to individual programs. Reports and evaluations might take a broad view, encompassing several programs. This approach is not entirely possible now due to various statutory requirements of individual programs, administrative rules, and tradition. However, the Oregon Option experience suggests that greater alignment is possible even within current limits. But it requires that decisionmakers take time to develop acceptable cross-cutting arrangements and be willing to give up some control.

Using What Performance Measures?

How do we know we are making progress? What are the measures? Tracking performance is fundamental to a results-driven accountability system. The partners need the performance data in order to assess the strengths and weaknesses of their strategy, reinforce or redirect resources, and communicate progress to others.

The partners must agree upon the information they will track to determine whether they are making progress toward achieving the results. Ideally, for any strategy, the partners would define a few good measures which would be derived from their answers to questions like these:

• In the broadest sense, what would we want to observe to give us confidence that we are achieving our desired result? What would be different for people? In communities?
• Of that which we would like to observe, what can we realistically quantify? What is measurable?
• What are reasonable increments of time by which we should see measurable progress?
• What baseline can we establish by which to mark progress?
• What meaningful comparisons should we be making in order to better understand progress? Increments of change from our baseline? Comparison to an established standard? Increments of change compared to similar populations in other locations? Increments of change from one to another point in time?
• What data are we currently gathering and which measures most closely match our needs?
• What are the missing pieces and how do we compensate for inadequacies in data?

In choosing measures of progress, the partners should be clear about the different types of and purposes for measures.

A Note On Choosing Measures: A number of issues will arise when partners from several agencies and levels of government attempt to agree on a set of measures. First they will have to develop a collection of measures that are justifiably consistent with the purpose of each partner's organization. Second, they will have to analyze how the proposed set of measures compare with the definitions and requirements of the various programs at issue. They will have to determine which definitions they will use. If the various programs specify measures other than the ones the partners want to use, they will have to determine whether they have the authority to substitute their measures for those specified by the program, gather dual measures, or use the program's measures. These are not easy or obvious choices. Here is how the Workforce Cluster is proceeding.

Workforce Cluster members developed a matrix which displayed all the performance measures currently tracked under the existing programs. They discovered that for all of the programs they want to deploy as part of their strategy, there are approximately 144 performance measures currently required. Some overlap. Some collide. Some share common definitions, others do not.

Their next task was to develop a smaller number of cross-cutting, shared performance measures that met this test:

• As a group they should add up to achieving the benchmarks.
• They must be consistent with the statutory and administrative requirements of the various programs involved.
• They can be measured.

The Workforce Cluster members developed 5 broad categories of measures:

• Employment and Earnings
• Self Sufficiency
• Cost Effectiveness
• Customer Satisfaction
• Skill Gain

Within each broad category, they identified a few specific performance measures. In total, across these 5 categories, the cluster members identified approximately 15 performance measures. Ultimately their intent is to have these fifteen common measures replace the 144 disparate measures. But before they can be fully replaced, the partners must analyze the 144 measures. They need to understand the root of each measure -- who requires the measure and why it was established? What is involved in replacing the measure with a common, cross-program measure? And can they develop common definitions for each of these core measures? This is a messy and time-consuming process.

The Cluster's strategy is to view the five categories and the fifteen measures as their common framework. During the next six months they will vet these measures by asking a larger group of people from the partner organizations to review them. They will also attempt to assign numeric targets to these common performance measures. Data will continue to be gathered on the 144 measures until each is either eliminated or replaced.

A Note On The Disabilities of Data The logic of results-measurement is compelling, but our measurement systems may not be ready for the task. A reliable results measurement system requires several types of data:

• Small area data which would tell us about people in their communities. Census track data is too broad.
• Frequent data so that we can monitor changes in increments of one year or less. Five and ten year increments are not sufficient.
• Longitudinal data so that we can track the progress of a cohort over time to determine the long-term effect of a strategy or program.
• Customer satisfaction data in order to determine public perception of progress, their confidence in and satisfaction with services, and their overall sense of well-being.
• Effective practice data based more on evaluation research and less on anecdote, in order to understand the strengths and weaknesses of various interventions
• Common client data to understand the progress of people who are receiving multiple services delivered through multiple systems.

We are lacking in every category for at least two reasons. The first is cost. It is expensive to conduct population surveys frequently and over time. The second is complexity. Common client data poses a complex problem because each system uses its own definitions, tracks people in their own way, and people are reluctant to abandon their practice or share their information. Effective practice data is complex because the nature of the changes these programs are designed to influence makes it tough to determine cause and effect. Changes in people's lives occur for many reasons and are hard to attribute to one program or strategy.

People are working hard to address these disabilities across the country. At the federal level, all of the statistical agencies are working to develop solutions. Several state and local officials on their own or through their organizations have also been experimenting with new approaches to measurement. The Oregon Option and other pilots help to expose our measurement limitations and create a sense of urgency that we develop practical solutions.

Can Fiscal Requirements be Aligned with Results?

A results-driven partnership has little meaning if the standards by which state and local partners are audited do not reflect the partnership agreement. The federal Compliance Supplement to the Single Audit Act outlines the specific requirements of each federal program that grantees must follow. The Compliance Supplement is intended to ensure the fiscal integrity of federal grant programs by specifying what funds can and cannot be used for. Auditors focus on whether grantee spending adheres to the specifications of the Compliance Supplement.While federal grantees are subject to only one audit under the Single Audit Act, the auditor selected by the grantee must focus on the guidelines for each separate program detailed in the Compliance Supplement.

As discussed in the previous section on performance measures, in this model for accountability, the partners would develop a set of measures and standards consistent with the agreements they have reached. These measures and standards would extend to financial reporting requirements and ideally would replace those in the Compliance Supplement. The auditor would be directed to audit against these substitute standards.

This approach is technically and legally possible now. It is consistent with the philosophy of flexibility in exchange for results. But it goes against a tradition. Compliance monitoring deals principally with the question of whether dollars were spent as directed, not whether the dollars spent achieved a result. For example, under the Elementary and Secondary Education Act, a local education agency can use up to 5% of its funds under all programs to implement a coordinated, comprehensive services project. If the local agency develops a coordinated project and begins to document improved results, but uses 6% of its funds to accomplish this, the typical practice would be for the audit to focus on the 1% overage, not the accomplishment.

With What Consequences?

What happens if the effort is successful? If the partners achieve results on time and under budget, what is the reward? Conversely, what happens if they do not? What if little progress is made and the costs are high? What are the penalties? And, in a partnership model, who determines rewards and penalties?

Establishing Rewards and Penalties: We have not traditionally dealt well with rewards and penalties in government. While some programs grant authority to issue sanctions to grantees who do not comply with standards, these have rarely been issued. We do have a tradition of perverse consequences for operating efficiently. If we spend less to achieve a task, we get our budget cut next time around.

In this period of transition, policymakers are attempting to develop more effective rewards and penalties. For example, in Oregon's Welfare Waiver, they can keep the funds that they save from reducing their case loads to reinvest in their program. This is a revenue-neutral financial reward for success. Other types of financial incentives, recognition, increasing flexibility, and professional advancement are among the rewards that could be considered.

There are two types of failure. One is when partners fail to live up to their commitments, use poor judgement, or act dishonestly. For these there should be clear and immediate consequences such as leadership and personnel changes, redirecting or reducing funds, imposing more stringent requirements, or even terminating the partnership.

The other is failure to achieve a result due to setting overly ambitious goals, unforeseen changes in circumstances, or other conditions outside of the partners' span of influence. For these types of failure there should be opportunities to learn, make course corrections and improve.

Still to be resolved is the question of how intergovernmental partners share risk and rewards as equals. Because we have no precedent for rewards and penalties in a results-driven partnership, those involved in these early experiments must begin to develop the protocols.

Clarifying Roles and Responsibilities: While partners share accountability for results, each needs to be clear about expectations and consequences for their own performance. This is not a one-way street. State or local partners should be able hold federal partners to the agreement, and vice versa. But to do so the partners must define their mutual roles and responsibilities.

As a starting point, each partner should assess what they can do to contribute to success. At the federal level the roles might include taking responsibility to negotiate the agreement with the state and local partners, commitment to align practices and procedures across programs where possible, and working out alternative solutions where full alignment is not possible. The federal partners might also be responsible for ensuring that all federal parties necessary to the success of the partnership continue to be informed and engaged. They might be part of action teams working with state and local partners to identify and address barriers. Federal partners might devote attention to overcoming some of the generic data and measurement challenges, supporting research to better understand cause and effect, and brokering technical assistance for all the partners.

State and local partners might be responsible for strategic action plans for achieving results, negotiating agreements with each other and with the federal partner, developing data tracking and communication systems, identifying opportunities and problems, and helping to strengthen community capacity.

Partnerships challenge traditional intergovernmental patterns because they push the parties to be proactive rather than reactive. To succeed, each of the partners must continuously ask themselves what they can do to make it work. Then they must do what it takes.

Some Lessons

We have yet to define a common vocabulary or a set of practices that enable governments to be accountable to citizens for results that matter. Because this is an emerging approach to governing, there needs to be honest dialogue among leaders and practitioners about what this form of accountability means. The lessons from the Oregon Option about accountability are far from complete but we can extract a few:

• Partners must begin defining results by focusing a wide-angle lens on changes in peoples' lives. At this early stage partners must be able to step outside of their familiar programs and view conditions more broadly.
• Accountability for results, particularly in an environment of limited resources, forces policymakers to grapple with some tough issues around the question of equity. Equity of results may not be attainable. Equity of service levels may not be meaningful. Equity of access may not be sufficient.
• Programs must be seen as means to an end, not ends themselves. This is easier said than done.
• The process of jointly developing performance measures is messy and takes time, but it is a necessary element in a partnership. The process forces the partners to clarify their common purpose and justify or eliminate existing performance requirements.
• Reliable practices need to be developed for measuring results. Efficient, cost-effective data measurement systems must be created to capture real-time, small-area results.
• New protocols for rewards and penalties need to be developed to promote ongoing learning and improvement.
• All the partners must see themselves as accountable for the success of the partnership. People must learn new ways to operate on the job -- exercising judgement, learning from the information generated and continuously improving -- and their operating environment must support this approach to work.

This is an enormous challenge. Unfortunately, we don't have the luxury of working it all out in advance. We have to rebuild the race car while we speed down the track.

CHAPTER 5

BRIDGING DISTRUST

Building Trust in an Era of Distrust

Those working to develop the practice of results-driven accountability hope it will address our current inability to know whether policies and programs are making a difference. It is a logical response to the public's vocal discontent with government. It is straightforward, old fashioned common sense. Lots of local governments are doing it. So why is it so hard?

It's hard because our current approach to governance is based on mutual distrust. From top to bottom, every requirement derives from a deeply rooted presumption that left to their own devices, people in other branches or levels of government will not act honorably. This presumption is reflected in the way we draft laws, regulations and procedures, in the way we conduct audits and program reviews, and in the way we communicate.

A results-driven partnership, by contrast, must be based on the notion of trust but verify. Everyone in the process begins with the presumption that each party will act honorably and energetically to achieve mutually agreed upon goals. Consequently, this approach to accountability for results is not simply about gathering better outcomes-data (which is tough enough). It is about changing the tools of government to reflect this presumption of trust. We have yet to develop the replacement tools or the trust to move wholesale into a results-driven system of governance. But people across the governance spectrum are working to develop these tools and many are learning by doing.

Accountability Has Its Roots in Congress

Congress in its role as lawmaker has the responsibility to ensure that laws are implemented and achieving their goals as they were intended. Congress has the authority to define the objectives of executive action. In doing so, Congress sets the terms for accountability.

In the last several years critics have complained that Congressional oversight has led to excessive intervention in the day-to-day operations of the executive branch, imposing controls which severely limit administrative discretion. Moreover, critics argue that multiple controls, reporting requirements, funding calendars, eligibility requirements, cost categories, definitions and planning requirements imposed by Congress deflect the focus of the Congress and the Executive Branch away from results.

However, in recent years Congress has taken some significant steps which could strengthen the foundation for results-driven governance without compromising its oversight responsibility. The Government Performance and Results Act, for example, places a new emphasis on managing for results in federal agencies. Block grants are intended to consolidate funding streams and allow more flexibility in implementing programs. At its core, Congress must ensure an alignment of goals and expectations across related programs. Some suggestions for Congress which have emerged through the Oregon Option and elsewhere include:

• Establish how related programs with related statutes should coordinate.
• Align funding calendars, planning cycles and reporting systems across common or related programs.
• Establish common eligibility across any programs within similar purposes, allowing a small error rate and the ability to share data and collect common identifier (i.e. social security number).
• Establish common definitions for any common terms and concepts.
• Clarify broad goals and expectations of results but allow variation in methods of achievement.
• Clarify expectations to achieve results for populations disproportionately in need.
• Establish a single, streamlined auditing process for all the partners.
• Focus oversight agendas on issues related to results.

In a 1992 report of the National Academy of Public Administration entitled, Beyond Distrust, the panel wrote:

Effective governance under modern circumstances requires far greater emphasis on shared powers, negotiation, shared accurate information, and cooperative management than on classical separation of powers... It is time to find concepts and devices to enter a new phase of history... It is time to find within the constitutional framework enough flexibility, enough resilience, and enough acceptance of change to build bridges for effective and accountable government...

A Final Thought

Like pulling on a rubber band, these experiments stretch the fiber. If we let go too quickly the whole thing snaps back to its original form. If we pull too hard, it tears apart. But if we maintain the right tension, over time the band changes shape. Partnerships like the Oregon Option provide an opportunity to change the shape of American governance if we don't let go too soon, or pull too hard.