2 parts disc and paper EssaysGuru ONLY

Oregon Progress Board, call - (503) 986-0039, write: Oregon Progress Board, 775 Summer Street N.E., Salem, Oregon 97310, web: http://www.com.state.or.opb

Schick, Alan (1966).“The Road to PPB: The Stages of Budget Reform”, Public Administration Review 26 (December): 243-258 in Albert Hyde, Government Budgeting: Theory, Process and Politics, Brooks Cole: Pacific Grove, CA, 1992, pp. 46-60.

Wildavsky, Aaron (1992). The New Politics of the Budgetary Process, second edition, Harper Collins, New York, Chapter 10, pp. 415-433.

Wildavsky, Aaron (1984), The Politics of the Budgetary Process, fourth edition. Little Brown: Boston, MA, Prologue, Chapter 4, pp. 135-144.

Chapter 7

ZERO-BASED BUDGETING

Origin and Purpose

Peter Phyrr at Texas Instruments invented zero-based budgeting (ZBB) in the late 1960’s. His book (Zero-Base Budgeting,1973) remains the most definitive and comprehensive study of this approach to budgeting.

Phyrr’s goal was to create a decision-making mechanism that would force Texas Instruments to remain competitive in a rapidly changing set of market conditions. He believed this could best be done by putting managers in the position of constantly asking why we are doing what we are doing. Should we be using our resources to do something else? One of the best ways of putting this question at the center of attention is to begin each budget year without assuming that you will have what you received last year. What if you had to start over from scratch each year and justify everything you were doing from the ground up? In short, why not create a level playing field for everyone and assume that you could stop what you have been doing and use the resources to do something else instead? Wouldn’t this generate more innovation and new product lines, thus ensuring the company of constant market place preeminence? Peter Phyrr thought so and created ZBB to achieve this goal.

The philosophy of always questioning why we should continue doing what we have been doing in the past is the heart and soul of zero-based budgeting. The philosophy and the system that Phyrr created for Texas Instruments was adopted by Governor Jimmy Carter in 1971 when he was governor of Georgia. Like all other budgeting formats, ZBB should be viewed as a way of structuring the universe of budget discourse. By altering the rules of evidence and the kinds of questions that get asked, ZBB assumed that budgeting could be used as a driving force to shape public policy and force more systematic planning.

In practice, ZBB only exists in modified form in the public sector, usually as “decision” or “add packages” that require managers to submit discrete activities or programs for funding above a predetermined reduced level that is specified in the beginning year’s budget. In general, ZBB asks managers to reduce their budget to a 70-85% level from the previous year and then add back requests that they want approved for funding. This approach is commonly called "modified ZBB". While it encourages managers to put some existing programs in competition with new programs, it saves them from the severe strain of justifying all programs from the ground up.

The key elements of the ZBB format include Base Budget, Decision Units, Reduced Level Budget (RLB), and Decision Packages that are priority ranked. The Base Budget is defined as the expenditure level necessary to maintain last year’s service level at next year’s prices. In short, it increases the existing budget by the rate of inflation. The Reduced Level Budget is the percentage that the budget is required to be reduced below the base budget, as specified in the budget instructions. The reduced level budget includes services considered critical or essential and may include new or existing programs. Decision Packages are discrete additions to the reduced level budget ranked in priority order to maintain existing programs, serve increased workloads, or add new programs.

Zero-based Budget Example

The following five pages provides a tabular overview of zero-based budgeting. Form 7: 1 summarizes the reduced level budget for a community corrections program. The reduced level budget is followed by three decision packages, each summarized on a separate page. The final form (7:2) provides a summary overview of the reduced level budget and each of the decision packages that are rank ordered. These forms are similar to the ones used by the State of Oregon when modified zero-based budgeting was phased in over a three-year period, starting in 1978.

ZERO BASED BUDGET: FORM 7-1

Dept. ___Community Corrections__ Reduced Level Budget ____X_____

Org. Unit/Code _FUND BO_ Decision Package Number ______

Program Unit/Code _______

Description: Community Corrections

80% Reduced Level Budget

ZERO BASED BUDGET: FORM 7-1

Dept. ___Community Corrections__ Reduced Level Budget ____X_____

Org. Unit/Code _242_ Decision Package Number __1__

Program Unit/Code _______

Description

Restitution Center

ZERO BASED BUDGET: FORM 7-1

Dept. ___Community Corrections__ Reduced Level Budget _________

Org. Unit/Code _242_ Decision Package Number __2____

Program Unit/Code __2421___

Description:

Job Development Coordination

ZERO BASED BUDGET: FORM 7-1

Dept. ___Community Corrections__ Reduced Level Budget _________

Org. Unit/Code _FUND BO_ Decision Package Number __3____

Program Unit/Code _______

Description:

Probation/Parole Bracelet Monitoring

123

Steps in Doing ZBB

1. Any department or work unit can, if necessary, identify a reduced budget level, which still allows it to achieve the most important objectives for which it is responsible.

2. For budgeting purposes, the department/agencies are divided into a series of decision units. These units may coincide with the formal organization or they may coincide with programs that cut across organizational lines.

A decision unit should meet the following criteria:

· The decision unit's responsibilities can be reflected through measurable accomplishments.

· There is one individual or a set of individuals who are accountable for the accomplishments of the unit.

· There must be specific fiscal and accomplishments information, both historical and projected, available for the unit.

· The decision unit is at a level where significant policy decisions can be made-neither too large to bury major issues or too small to be insignificant in its own right.

3. Initial budget and program priorities should be developed by the manager(s) directly responsible for the decision unit. These priorities should then be reviewed at each higher level of management and integrated with priorities from other decision units in the agency.

4. After developing the reduced level budget, the unit manager develops specific incremental budget requests which may restore existing services not included in the reduced level budget, add new services, or reflect alternative to present methods. These decision packages are supposed to be independent of one another and, if approved, allow the decision unit to accomplish a specified set objective for a known cost.

5. After the decision packages have been developed, they are ranked in priority order by a process determined by the decision unit manager and forwarded to the next higher level of review. Decision package priorities from multiple decision units are integrated so ultimately there is one department-wide list of prioritized decision packages.

6. If the process is done effectively, top management of an organization will have gained both a planning and a budget management tool. Decision unit managers will have had an opportunity to express their priorities, and middle and upper management will have integrated those individual lists to meet overall organization objectives. Top management will have some assurance that, if the lowest priority decision packages are not financed, the least important contributions to achieving the organization's objectives have been deleted. There also will be a clear statement of what will be accomplished if a decision package is adopted, or should not be expected of the agency/program if it is deleted.

A. Selection of Decision Units

As a first step in the budgeting process, the chief executive officer traditionally asks each department to identify and submit for approval the program structure proposed for use in presenting the budget request. A similar request applies to the decision unit selection. Once a department has developed a proposed list of decision units, it is submitted to the budget analyst for review and approval.

Normally, decision units simply correspond to the formal organizational structure, even though one of the purposes of ZBB is to stimulate new thinking about the way in which creative work might get done. Small departments, which have previously presented their budgets as a single program, may wish to identify only a single decision unit. Larger departments may wish to use the existing organizational structure or may alter the number or size of decision units in order to stimulate creative thinking about how to get better results out of existing resources. Selection of a large number of units greatly increases the paperwork and detailed supporting data that is required. Selection of too few may compromise the purpose of the unit by combining such dissimilar responsibilities that establishment of a clear set of policy issues is virtually impossible.

The following example displays the alternatives available:

Level 1: Executive Department

Level 2: Budget and Management Personnel Accounting

Division Division Division

Level 3: Budget Analysis Personnel Accounting

Services Control

Management Analysis Training Accounting

Services

Employee Payroll

Relations

The Executive Department could conveniently settle on its Level 2 divisions as the appropriate decision units. Each has a identifiable manager, measurable accomplishments, and a readily determined fiscal history.

The Department might wish to go to Level 3 to allow for more specific budget presentations and priority development. Each of the Level 3 subprograms has a manager and a set of identifiable activities. Some of them, however, may lack an easily determinable fiscal history, since they may not have previously been budgetarily independent. If this is the case, management must decide whether it is worth the time and effort to gather the necessary data for purposes of undertaking a Level 3 budget formulation.

B. Selection of Decision Unit Managers

This is a key decision to be made early in the budgeting process. These managers are expected to perform several of the essential functions of the ZBB system. While it is desirable to have one individual responsible for each decision unit, some find that a participatory management approach is most effective. The manager (or decision-making process) should be able to perform the following services:

· Assess the decision unit's area of responsibility and determine which of the activities are most essential to meeting assigned objectives.

· Develop a summary of measurable accomplishments that could be achieved at reduced or alternative budget levels.

· Prepare, in coordination with the department budget staff, the fiscal data for different budget levels.

· Prepare incremental decision packages beyond the reduced level budget.

· Rank these decision packages in priority order.

While each decision unit should have a single manager, a manager may be responsible for more than one decision unit.

C. The Reduced Level Budget

One of the basic advantages of ZBB is that an organization can identify a reduced budget level, which still allows it to achieve the most important objectives for which it is responsible. In short, ZBB forces managers and organizational units to determine how far cuts can be made without fatally compromising the original and lawful purpose of the organizational unit.

One of the myths surrounding ZBB is that each department must submit a budget that justifies, in detail and in priority order, every dollar proposed for expenditures. This is not the case. For example, a department cannot pay less than its existing obligation for debt service. In addition, some expenditures are not limited in the budget. Since these expenditures are not subject to conventional budget constraints, there is little reason to calculate a reduced level budget for them.

The phrase "zero base budgeting" has been interpreted by some to mean that every department must start from zero each budget period. This is not the case. It is true that each decision unit manager is expected to assess current activities within the unit to determine which are the most important to fulfill unit responsibilities. All activities are supposed to be considered expendable (reduced to zero) until priorities are established. Realistically, the budget request from each decision unit should consist of a reduced level budget and incremental decision packages to arrive at a total request for the budget period. This interpretation and use of ZBB is called "modified zero base budgeting", to distinguish it from an approach that starts from a base of zero.

The ZBB process establishes an upper limit on the aggregate total of all decision units' reduced level budgets. The total of all reduced level budgets may not exceed the specified adjusted budget, (usually established in the 70-90% range).

The adjusted budget represents the total financing needed to maintain existing approved programs through the next budget year. The adjusted budget includes:

1. Legislatively mandated programs and activities.

2. Less all costs of excluded categories, such as:

Debt Service

Capital Construction and Capital Improvement Projects

Restrictive grants

3. Full financing of salary levels at next year’s salary rates.

4. Documented inflationary and other cost increases for Materials and Services and Capital Outlay

After calculating the Adjusted Level Budget, the next step is to multiply the adjusted budget by the amount specified in the budget instructions. (We will use 80 percent for illustrative purposes in the discussion that follows). The product is the reduced level budget (RLB) limit for the department. This is a department-wide control figure, not a budget plan. Individual reduced level budgets for each decision unit do not have to be under 80% of the adjusted budget for that decision unit. The control figure applies only to the total department budget. Thus, one decision unit may have a RLB at 90% or 100% of its adjusted budget. This is acceptable if other decision units' RLB are sufficiently reduced so that the total agency RLB is no greater than the hypothetically established 80% dollar level.

The purpose of the department-wide reduced level budget limit is to determine the activities and related costs that the agency would consider of highest priority if the funds available would allow for only 80% of the purchasing power projected for next year.

The RLB for next year may differ substantially from the budget approved for last year. This is fully allowable as long as the total department RLB, from all funds, is less than or equal to the hypothetical 80% of the adjusted budget.

Multi-funded departments may be tempted to reduce other or federally funded programs while "protecting" General Fund programs. This is not likely to pass the scrutiny of the analysts in the central budget office. As stated above, the reduced level budget figure is a control figure only. Within that figure, departments may:

· include new programs or program enrichments at the expense of existing program;

· increase expenses for one category (Personal Services, Services and Materials, etc.) at the expense of another;

· shift expenditures from one decision unit to another.

D. Development of Decision Packages

Following the creation of a reduced level budget for each decision unit, decision unit managers should develop incremental decision packages to be requested for their programs.

· Use of incremental decision packages is the method by which a decision unit manager describes the difference between the unit's reduced level budget and the total budget request.

· Each decision package should represent a single, clear policy decision. It should include the specific accomplishments to be achieved and all costs to be incurred during the year.

· The focus of a decision package is on the program impact proposed. A package should not normally be prepared for a single position, one piece of equipment, or one item of expenditure for Services and Supplies. Instead, all related costs necessary to accomplish a program objective should be identified.

· In some cases a decision package should include estimates of the fiscal and program impact of the proposal beyond next year. This is particularly true if the proposal represents a new program or if a planned change in the financing base will take place in the future.

· The package may deal with restoration of current activities not included in the reduced level budget. It may also represent a program expansion based on anticipated workload demands or program enrichment to improve the quality of service.

· The description of a decision package should include a discussion of potential impact on department revenues. If the package would produce substantial matching revenue from other jurisdictions or would generate other revenue, this should be explained in some detail. This is particularly true if failure to approve the package would eliminate revenue that has been received by the department during the previous year. In addition, it should be noted whether the revenues involved are available only for the purposes described in the package, or could be used to offset costs in other programs.

E. The Ranking Process

Ranking produces a listing of decision packages in order of decreasing importance or effectiveness. For the process to be effective, department management must develop a specific method to be used. Agency and program objectives must be identified. Decision unit managers must develop packages that relate accomplishments to objectives. Finally, management must be able to judge the relative importance of decision packages of not only like, but also dissimilar functions.

The ranking process should start with the decision unit manager. This allows the individual who developed the decision packages (and is accountable for the proposed accomplishments) to do the initial ranking of the relative importance of the proposals.

The ranking process above the decision unit level will depend on department program and/or organization structure. In many cases program decisions will follow organization lines. In multi-program settings and complex organizations, program decisions may cross organization lines.

Decision packages for a medium-sized department with five decision units and two major programs is depicted below:

A

28 Packages

B1 B2

10 Packages 18 Packages

C1 C2 C3 C4 C5

4 Packages 6 Packages 3 Packages 9 Packages 6 Packages

At decision unit level C, five managers must rank from three to nine packages within decision units. At level B, program managers must rank 10 and 18 packages, respectively. Finally, the administrator (A) must rank 28 packages involving two programs and five decision units (seven if B1 and B2 are treated as independent decision units from those at Level C).

Although the administrator may be able knowledgeably to evaluate the number of packages in this example, in large organizations the volume may be too great. Therefore, some form of consolidation may be necessary to focus management's attention on the most critical decisions.

Ranking Approaches

Ranking must be performed with a clear understanding as to who is responsible for decisions and what criteria are to be used to judge among packages. Ranking may be done by individual managers or by some form of management committee. A committee allows for more than one perspective to be used in the ranking of decision packages. If a committee is used, it should include individuals who are knowledgeable about the operations to be judged and, if possible, include the decision unit manager who developed the initial package.

The following ranking approaches are most commonly used.

Single standard - The single standard approach is the simplest and best for dealing with a single program or several programs. Under this method all decision packages are evaluated against a single criterion. This could be dollar savings, cost benefit ratio, increased effectiveness, or known public demand for the services. The key element is that all packages must be considered in terms of their common characteristic. For example, a road maintenance department my decide on repaving priorities on the single standard of age.

Weighting - Because many programs are complex and a single standard may not provide enough information for effective ranking, a more sophisticated method called "weighting" may be more useful. Under this approach, several criteria are used, with each receiving a different value. Thus, a self-supporting activity might be worth four points while a program enrichment would receive one. The "point counts" of the packages are then reviewed, differences are resolved, and a consensus rank is passed up the next level of management. Either individuals or committees can use the weighting system.

Major categories - Another form of the voting method is to organize major categories into which decision packages may be grouped. This places packages from throughout the organization into large categories for ranking. Major categories could be:

· Efforts which generate revenues that exceed costs;

· Efforts which pay for themselves in the first year;

· Efforts of substantial long-term economic benefit;

· All other efforts.

Multiple standards – A multiple standard approach is also a variation of the weighting system, but is more complex in that it focuses on management considerations as well as economic merits. Under this approach the decision to implement a package may depend on answers to the following questions:

· Is there a demonstrated demand for the service by the public?

· Do we have the necessary technical skills?

· Will line management accept and execute the program?

· Can we afford not to act?

· What are the political consequences of acting? not acting?

Decisions on what kind of ranking system should be used will, of course, depend on the needs of the department and should be kept as flexible as possible. Statutory authority to perform certain activities is not, in itself, a desirable criterion for ranking priorities. The department's responsibility is to define for the executive and legislative bodies the activities that are most critical in achieving the department's overall responsibilities.

Departments should not devote an inordinate amount of time determining whether package number 4 is more important than package number 5, but should be concerned that packages ranked 4 and 5 are more important than package 9 and that package 9 is more important than package 14.

Example of Weighted Ranking: City of Portland Water Bureau

The following method is used by the City of Portland Water Bureau to rank capital projects for funding and scheduling priority. It is a multi-step process that is intended to fulfill identified needs of the water system and the community it serves while taking project cost into consideration.

STEP 1

Projects are identified, defined, and costs and time to complete are estimated. A Bureau-wide representative committee reviews these projects and selects those suitable for further consideration.

STEP 2

Projects are rated against seven needs that the Water Bureau has identified to be met by the funded CIP projects. Each project is rated from 3 to 0 depending upon how well it fulfills each need. The following guidelines are used to ensure a consistent and uniform rating of projects:

Once completed, the ratings in turn are multiplied by a weighing factor established to weight higher value needs more than those considered less important. The weighing factors reflect the Bureau's interpretation of its primary mission. The resulting ratings for each need are then summed to make a final benefit rating score. The formula for the benefit rating score is then:

The seven needs identified by the Bureau to be met by the funded CIP projects are:

1. Public Health

The water system needs to prevent the spread through the water system of any disease or harmful quantities of any contaminant. Projects that address this need could prevent water quality degradation by correcting deficiencies in the distribution, storage or supply systems. Projects may also contribute to regulatory rule making or be intended to allow the system to meet current or identified future rules. The public health need is considered the highest rated need and is assigned a weighing factor of 5. Following are examples of how project benefits are rated as meeting the public health needs:

Rating Typical Project for Listed Rating

3 -- Reconstruction of disinfection system to comply with existing state regulations.

2 -- Improvement to distribution system circulation system to eliminate chlorine deficient areas.

1 -- Study to provide data for comment to proposed rule making.

0 -- Project that provides no significant benefit to meet this need.

2. Public and Employee Safety

The water system needs to prevent injury to the public, City employees, and public and private property from failure of the water system, insufficient water system capacity, or activities of Water Bureau employees and contractors. Projects included in this category include studies to identify natural and man-made hazards, enhancements to improve fire protection provided to the City and improvements in water supply and distribution system redundancy and reliability. The need to promote public and employee safety is considered a high priority and is assigned a weighing factor of 4. Following are examples of how project benefits are rated as meeting the public and employee safety needs:

Rating Typical Project for Listed Rating

3 -- Fencing project to keep children from drowning in the Roslyn Lake discharge pipe or a project to construct new ladders on elevated tanks to meet OSHA standards.

2 -- Security lighting to prevent use of system property for criminal access to private property.

1 -- Earthquake hazard analysis of bureau facilities.

0 -- Project provides no significant benefit to meet this need.

3. Maintain System

The water system needs adequate maintenance to protect the asset value of the system and ensure the system can continue to provide water in sufficient quantities and pressure, continuously, for existing residential, commercial, industrial and irrigation uses. Projects and studies that meet this need are those related to the location and replacement of deteriorated or outdated system assets, such as pipelines, buildings, and pumping systems. Also meeting this need are projects and studies to establish construction, operation, and maintenance standards and programs for the water system.

Fulfillment of the need to maintain the water system assets is essential to the long-term provision of water service and is therefore assigned a rating factor of 4. Following are examples of how project benefits are rated as meeting systems maintenance needs:

Rating Typical Project for Listed Rating

3 -- Replacement of leaking water mains.

2 -- Project to add a corrosion protection system to pipelines.

1 -- Development of construction standards for distribution pipelines.

0 -- Project provides no significant benefit to meet this need.

4. Environment and Conservation

The water system needs to protect and enhance the natural resources required for the water system to function, to assess and respond to the public's expectations for environmental and conservation programs of public utilities, and to utilize water system assets to enhance the environment of the system's customers where such use does not deter from the provision of quality water service. Projects that fulfill this need are those that protect the quality and quantity of ground and surface water, raw water sources, water and electrical conservation studies and projects, assistance to others in maintenance of air, land, and transportation resources, and improvements and maintenance to Water Bureau assets that allows secondary, non-water service, uses that enhance the public's living environment.

To fulfill the environment and conservation need is important to the long-term provision of water service and to obtain customer approval of water system operation and is assigned a rating of 3. Following are examples of how project benefits are rated as meeting the need to protect the environment and to promote conservation.

Rating Typical Project for Listed Rating

3 -- Projects to protect the well field from existing contamination.

2 -- Water leak detection projects.

1 -- Project to enhance public use of Dodge Park.

0 -- Project provides no significant benefit to meet this need.

5. Economic Viability

Water service needs to be available at a cost that will provide this essential human service at affordable prices that are competitive within the region so that the existing economy is stable and planned growth and development within the service area is encouraged. Projects that fulfill this need include those that insure water meter accuracy, water rate studies, improvements in the efficiency of the operation of the water system, and projects to enhance water service to growing areas of economic development or redevelopment. Maintaining the economic viability of the water system and the region's economy is an important need of the water system and is assigned a weighing factor of 3. Following are examples of how project benefits are rated as meeting the need for economic viability :

Rating Typical Project for Listed Rating

3 -- Project to construct a water main to a PDC redevelopment area.

2 -- Project to replace water meters known to be under-registering.

1 -- Project to revise rate setting model or capital construction that reduces long-term maintenance costs.

0 -- Project provides no significant benefit to meet this need.

6. Respond to Growth

The water system needs to identify future system requirements, develop management plans to accommodate system growth, and design and construct improvements to the supply, storage, and distribution systems to meet new demands placed on those facilities. Projects that fulfill this need are short and long-term planning studies, water supply, storage, and piping that add to the water system's capacity and area of service, enhance water reuse and alternate water use, and production of development and system operation/maintenance standards to accommodate system growth. Meeting the growth needs of the water system is important to the economy and livability of the region served by the water system and is assigned a weighing factor of 3. Following are examples of how project benefits are rated as meeting the growth needs of the City and region:

Rating Typical Project for Listed Rating

3 -- Development of water supplies (including interties, replacement of leaking mains, reuse projects) which would fulfill projected demands in a significant way.

2 -- Regional water supply plan.

1 -- Oversizing water main to allow service to area of potential future growth.

0 -- Project provides no significant benefit to meet this need.

7. Degree System Affected

The water system needs to provide equivalent service to all of its customers and to utilize its available resources to benefit the largest number of customers possible. Projects and studies that fulfill this need the best are those that raise the level of service to all customers or significantly improve service to customers receiving water service below that received by the majority of customers. Fair and equitable water service is important to the public's approval of water system operations and is assigned a weighing factor of 2. Following are examples of how project benefits are rated as meeting customer needs:

Rating Typical Project for Listed Rating

3 -- Headworks water treatment improvements or a project to increase pressure in an area where existing pressures are less than 40 psi.

2 -- Projects that improve service to a large service area.

1 -- Projects that improve service to a pressure zone within a service area.

0 -- Project provides no significant benefit to the water system.

STEP 3

Projects are placed in groups under each need category (except Degree System Affected) by selecting the need that each project best fulfills. Degree System Affected is not used because it is considered more of a measure of amount of benefit than a distinct benefit of its own. Each project can only be listed under one need. Projects that are mandatory are removed and placed in a separate group. A "Mandatory" project is one which the Bureau has no option other than to fund the project, such as one that is required by existing contract, by law, or by state and Federal regulations. Placing first the projects with the highest total rating score then places the projects under each need category in rank order.

STEP 4

The Water Bureau sets the funding available for the CIP. The cost of the Mandatory projects is deducted from the total CIP funding available. One half of the remaining CIP funds is then divided among and Allocated to the six need categories by selecting a minimum benefit rating such that funding all projects with that or a higher rating will consume the now Allocated portion of the available budget. The full listing of projects is then divided into two groups - those that will be funded by the Allocated CIP funds and those that must compete for the Non-Allocated CIP funds. At least the highest ranked project in each need category shall be included in the allocated funds group.

STEP 5

Each of the final two groups of projects is placed in rank order by their total benefit rating score (highest score is #1) and by their cost (lowest cost is #1). Adding each project’s rank number for benefit rating score and cost, and then placing as first priority the projects with the lowest combined rank number then determine a final rank

order within each group. Only the first five years of costs are shown for ongoing projects. An example of this process is illustrated on the following page.

Step 5 Ranking Illustration

Allocated Fund Group

Benefit Rating Cost Final Rank Order

Project Combined Within Full

Name Score Rank Estimate Rank Total Group CIP

A 23 1 $10,000 1 2 1 1

B 16 2 40,000 4 6 3 3

C 14 3 12,000 2 5 2 2

D 10 4 20,000 3 7 4

Non-Allocated Funds Group

Rating Cost Final Rank Order

Project Combined Within Full

Name Score Rank Estimate Rank Total Group CIP

E 15 2 $10,000 1 3 1 5

F 16 1 35,000 3 4 2 6

G 12 3 80,000 4 7 4 8

H 10 4 30,000 2 6 3 7

STEP 6

All projects in the Allocated Funds group are funded as their schedules allow. Projects are funded in the Non-Allocated Funds group in their rank order until the non-allocated CIP funds are consumed. Those projects not funded are then scheduled for funding in future years and for reconsideration in the following year's CIP.

STEP 7

The Bureau-wide representative committee reviews the initial prioritized schedule developed in Step 6 and adjusts project schedules to accommodate inter-dependent projects.

STEP 8

After the appropriate adjustments are made, the CIP document is prepared and submitted to the Portland City Council for approval. This final document rank orders as many as 200 separate water projects.

Summary: Strengths and Weaknesses of ZBB

Zero-based budgeting, like program and performance budgeting, is intended to introduce a more systematic planning and analytic perspective into the decision making process for the allocation of scarce resources. Instead of focusing on performance measures (as does performance budgeting measures) or the achievement of objectives (as does PPBS), ZBB structures the process so that participants are forced to consider whether current resources might be better spent on other activities.

How successful has the ZBB reform initiative been? One way of measuring the success of ZBB is to see how much of it has survived since it was first introduced into the public sector in 1971. Perhaps the most lasting legacy of the ZBB reform initiative is the notion of “reduced level budget” and decision packages (sometimes-called “add” or “cut” packages). Many jurisdictions have incorporated these techniques as common operating procedures within their budget preparation process. This has been especially useful during periods of scarce revenue by engaging program managers in an on-going conversation with both their subordinates and superiors about how to get “the most bang for the buck”. Whether this has produced the kind of creative approaches to solving public problems is less clear. In the case of welfare reform, most of the incentive to change came from outside the system, although career administrators on the inside crafted the creative solutions.

After walking through the multiple steps that have to be carried out in order to put ZBB successfully into practice, you can better appreciate the following summary of potential strengths and weaknesses of ZBB:

Summary Strengths of ZBB

· ZBB requires managers and employees to reexamine why programs and activities are being funded in lieu of other possibilities.

· First line supervisors and street level bureaucrats are required to be involved in the process of assessing existing work and redesigning future allocation of resources.

· More members of the organization are knowledgeable of the budgeting process, especially the interplay between managing programs and people within a fixed set of resources.

· More planning tends to get done before budget allocation than after, thus increasing the clarity of goals and objectives.

· There is less likelihood that the momentum of the past will secure the future of programs after they have lost their purpose for existing.

Summary Weaknesses of ZBB

· Since more people need to be involved at lower levels of the organization in preparing and executing the budget, the process is more complex, time-consuming and costly than traditional object code budgeting.

· By focusing attention on the ends, purposes, and new directions for the future, the process increases the possibilities for conflict and the need for additional time to resolve this conflict.

· Many programs, like a suit of clothes, are interactive with other programs and activities. The creation of independent and self-standing decision packages burdens those responsible for these synergistic and interdependent kinds of activities.

· The systematic weighting and ranking of decision packages provides an opportunity for ZBB to appear far more rational than it may actually operate in practice. The need to choose criteria and then to decide on what the weighting should be opens up opportunities for considerable subjective judgment.

ZBB Exercise

1. Prepare a zero-based budget for your work unit, using no more than four decision packages if your department has three or more divisions. Use the forms on the following page to prepare your decision packages and to summarize the results of your work. Use Form 7:1 for both your reduced level budget and your decision packages. Be certain to capture this information by properly labeling the summary information at the top of Form 7:1.

2. Provide a brief explanation of the purpose or function of the decision package in the space provided. The Reduced Level Budget for the department as a whole should be treated for this purpose as a separate decision package.

3. Explain how the function accomplishes its tasks in the space titled "Methods".

4. List any alternatives to provide this function in the space provided.

5. Analyze and briefly explain the impact of both retaining or eliminating this function from next year’s budget in the space titled "Impact".

6. Using the information you developed in the previous section on performance measures, list quantifiable workload or performance measures in the space provided.

7. Under the budget section of the form, list next year’s cost of this decision package by major budget category. Indicate the full-time equivalent positions (FTE) affected and place the total cost of the package in the box provided.

8. Rank the decision packages by priority order with the first priority for funding as #1.

9. Using the Summary Form 7:2 list the reduced level budget first, then list each decision package in ranked order. Indicate the ranking in the far left-hand column.

10. Record the costs of the reduced level budget and each decision package in the "amount" columns. For purpose of this exercise, costs should be distributed by funding source using the General Fund, Federal Funds and Other Funds as broad categories.

11. Record the FTE's in the appropriate column.

12. Calculate the cumulative costs in the columns provided.

13. Be prepared to discuss and explain your department's reduced level budget and decision packages in the meeting with other department heads to balance next year’s budget.

144

ZERO BASED BUDGET: FORM 7-1

Dept. ___________________ Reduced Level Budget ___________

Org. Unit/Code __________ Decision Package Number ________

Program Unit/Code _______

Description:

14.

10

146

Study Questions

1. What is the rationale behind the adoption of a ZBB budgeting system?

2. To what extent does this rationale make sense in the light of the conditions that prevail in your agency?

3. What are the essential differences between the assumptions of performance budgeting, PPBS and ZBB-based budgeting systems?

4. What conditions are necessary for the implementation of ZBB?

5. Who tends to gain and who tends to lose in adopting ZBB?

6. How does the role of the chief executive officer and the legislature differ under a ZBB format in comparison either to line item or PPBS formats?

7. What has been the experience of those agencies, departments, etc. that have adopted ZBB?

8. What are the overall strengths and weaknesses of ZBB in comparison to other budgeting alternatives?

9. What are the most difficult practical problems in constructing a zero-based budget?

References and Additional Resources

Lee, Robert & Ronald Johnson. (1998). Public Budgeting Systems, sixth edition, Aspen, pp. 113-114, 219-220.

Lyden, Fremont J. and Mark Lindenberg. (1983). Public Budgeting in Theory and Practice, Longman: New York, Chapter 6.

Peter Phyrr. (1974). Zero-Base Budgeting. John Wiley and Sons: New York.

Schick, Allen. (1978). "The Road From ZBB”, Public Administration Review (March/April) pp. 177-180.

Wildavsky, Aaron (1984). The Politics of the Budgetary Process, fourth edition. Little Brown: Boston, MA, Prologue, Chapter 6, pp. 202-221., 278.

Conclusions to Section II

Use of Budget Formats by the State of Oregon*

As with any government system, the Oregon budget system has evolved to meet the changing needs of its actors and constituents. This evolution reflects the blending of budget and administrative theory with the demands and experiences of practice. Thus, this summary is a snapshot of the features of a dynamic governance system. In this summary we will describe this interplay among the four budget formats in Oregon’s budgeting process.

Oregon has traditionally followed the practice of most states in using line item budgeting as the central framework for organizing and allocating state resources. As we have seen, the line item budget has the great virtue of emphasizing fiscal control and accountability. However, over the decade of the 1990’s, the Oregon Department of Administrative Services (DAS) and the Oregon legislature have implemented a strategic shift in budgeting emphasis from line item and object control to a more program, zero base and performance oriented budgeting.

Currently, Oregon uses an amalgam of all four budgeting formats, each having a distinct part to play at different stages of the budgeting process. Although it remains strongly grounded in the line item format, program, performance and modified zero-base techniques are increasingly being used in the development and presentation stages of the budget process. This is a result of an increased demand by decision makers for more information, which in turn reflects an increased level of concern by the taxpaying public that its governing agents are providing them with the best return on their hard-earned tax dollars.

One of the implications of this hybrid system of budgeting is that agency budget analysts are constantly preparing different presentations of the budget for different stages of the budget process. For example, each agency initially builds an Agency Request Budget in conformance with the biennial Budget and Legislative Concept Instructions. (Oregon Department of Administrative Services, 1998a). These instructions call for a comprehensive budget that must demonstrate analysis in the line

147

* Kent Robinson took responsibility for preparing much of the initial draft material for this section

item, program, and zero-base formats. Building the agency request focuses managers on agency planning and program development needs. The agency request also meets central budget office (DAS) needs for financial accuracy and consistency through the line item format. The zero-base contribution to the agency request provides alternative program spending and policy options to the Governor.

In contrast, the official Governor’s Recommended Budget document, (Kitzhaber, 1998) reduces each agency’s request to a brief standard line item table and a short summary narrative of program and policy changes. While the content and form of the Governor’s recommended budget responds to statutory requirements about what must be included in the Governor’s report to the legislature, it also reflects the Governors personal views of how the budget document should be used for policy and management purposes. For example, Governor Kitzhaber used the Recommended Budget to de-emphasize spending and to focus legislative and public attention on policy issues important to the Governor’s agenda. But there is still an important emphasis on illustrating incremental changes in program levels of operation from year to year which can be easily spotted by comparing proposed current year levels of program expenditures with previous budgets. Also by using a standard table with dollar amounts, readers are able to make quick comparisons of expenditures between programs. Finally, the summary table also provides a communications tool for sharing proposed budget changes with the public.

In contrast to the technical detail of the Agency Request Budget, the members of the legislative Joint Ways and Means Committee demand a complete, yet understandable budget presentation. Each state agency responds to this need by preparing a legislative budget presentation. Clarity of communication becomes a primary goal of this budget preparation. While simplified line item tables are most often used to accomplish this goal, the detail provided by program and performance formats play a much larger role for the committee staff who use this information to make budget recommendations to members of the Joint Ways and Means Committee. This detailed information has become increasingly important to elected officials as they look for demonstrations of program need and agency accomplishment to justify agency and program appropriations.

Following its review, the Legislature codifies its budget decisions into a Legislatively Adopted Budget. From this adopted budget, agencies again reformat their budgets into tools that can be used for managerial and financial control. The line item budget format plays a critical role during the next phase of budget implementation.

Line Item Budgeting – The central goal of line item budgeting is to maintain control over the amount of dollars legislatively approved for expenditures. This principle is closely adhered to in implementing and tracking Oregon’s budget after it has been approved by the Legislature and signed by the Governor (the Legislatively Adopted Budget, or the Legislatively Approved Budget after “E” Board action). The funding allocations and revenue levels in these statutes are promptly entered into the state’s R*Stars accounting system (Figure 7-3). Once this data entry is approved by the Department of Administrative Services Budget and Management Division, it becomes the operating budget and legal control for spending by an agency.

R*Stars provides the detailed, routine reporting to agencies to ensure that spending rates and totals remain within legislatively-set and calendar limits. In addition, the R*Stars system is used by the State Controller’s Division to produce the annual and special financial reports required by federal mandate, by professional accounting standards, and by generally accepted accounting principles.

For budget construction purposes, however, the agencies and DAS rely on the Automated Budget Information System (ABIS). This separate database links the following three sets of information: revenues, expenditures, and staff position data using a line item organization. This data is then compiled, maintained and reported for budget preparation and execution. The ABIS User’s Manual provides the documentation on the ABIS system (Oregon Department of Administrative Services, 1998b).

ABIS, like R*Stars, is a subsystem of the larger Oregon Statewide Financial Management Systems (SFMS) (Figure 7-3). This larger electronic system integrates the state budgeting information in ABIS with the other major administrative systems and functions of: Personnel, Position Control, Payroll, Accounting, Purchasing, Reporting, and Treasury. The Oregon State Controller’s Division provides a useful summary of the Oregon state financial management systems (Oregon Department of Administrative Services, 1999)

The overall goal of ABIS and SFMS is to maintain a workable balance between two competing needs. On the one hand, there is the need for centralized control to meet the requirements of the Oregon Constitution, state statutes, and public financial administrative policies and practices. On the other hand, there is a need to have discretionary control for managers in order to make the highest and best uses of taxpayer dollars to carry out the legislative mandates of their agencies and program. This discretionary need is accommodated in several ways. First, while ABIS has the

Figure 7-3

Integration of ABIS and R*Stars With

The Oregon State Financial System (SFMS)

SCD

Datamart

Treasury

Accounting

R*Stars

Pos/Control

PICS

Reporting

SFR

Budgeting

ABIS

Payroll

OSPS

Purchasing

ADPICS

Personnel

PPDB

163

capability to track very detailed line item expenditures, object codes are grouped and presented in the following eight general categories:

· Personal services;

· Services and supplies;

· Capital outlay;

· Special payments;

· Debt service

· Capital improvement;

· Capital construction (Major construction/acquisition); and

· Nonlimited expenditures.

In addition to the advantage of having broad object code categories, managers are administratively given considerable discretionary authority in the transfer of funds within categories. While this discretion varies by department and division, in general, there is a desire to push more accountability and discretion downward to ever-lower levels of administrative responsibility.

In addition to organizing expenditures into broad categories, ABIS similarly aggregates revenues into four broad fund sources of General Fund, Lottery Funds, Federal Funds and Other Funds. These categories form the basis for line item budgeting and reporting.

While ABIS may encourage managers to think more broadly than may be the case in other jurisdictions with larger numbers of object code categories, the analytic framework is still primarily incremental. This is because ABIS arrays the line items and revenue categories in comparison to historical and current funding levels. This arrangement encourages analysts and decision-makers to think in terms of making small, if any, changes from the previous year’s levels. The following types of information produced by ABIS illustrate this point.

· the actual revenues and expenditures from the past biennium;

· the estimated actual revenues and expenditures for the current biennium;

· Emergency Board actions for the current biennium up to a mid-biennium cutoff date.

· The above information, reported in columnar form, is often compared against a “current service level” budget for the coming biennium.

As of this writing, the DAS Budget and Management Division has begun to develop the Oregon Budget Information Tracking System (ORBITS) as a replacement for ABIS (State of Oregon, 2000). Implementation of the ORBITS will meet several administrative goals including: (1) providing more flexible support for budget decision making; (2) enhancing connections with the R*Stars accounting system; (3) tracking lottery revenues as a fund source; (4) establishing better links with the budget systems used by individual state agencies and commissions; (5) automating manual budget processes; (6) extending central policies and procedures to ensure statewide consistency; (7) presenting the system for a windows desktop computer. Current plans call for preparing the 2003-2005 biennial budget with the ORBITS system. Currently however, ABIS remains the Oregon state budget sub-system.

Program Budgeting - Over the course of the last decade, budgeting in Oregon state government has shifted to reflect an increasing desire to understand the programmatic consequences of public expenditures. With this change, the basic unit of analysis in the budget process has shifted from the line item to the “program unit.”

The shift to a programmatic approach is reflected throughout Oregon’s budget process. For example, the biennial Budget and Legislative Concept Instructions, reflect the themes of program budgeting. Under the instructions each agency must include in its budget submission:

· its agency mission statement and statutory authority;

· the environmental factors that affect service needs and program design;

· agency strategies for meeting needs;

· program descriptions;

· links to the Oregon benchmarks; and

· program results;

In practice, the agency mission statement, the environmental and clientele factors, the agency goals and strategies for meeting needs, and the general program descriptions all point toward the hierarchical organization of program budgeting. While the information on agency strategies may address both long-term goals and shorter-term program and action initiatives, the current instructions have yet to evolve into a full application of program budgeting. The current instructions do not require the explicit description of program-level objectives with measurable actions and time frames for accomplishment. However, many agency budget presentations to the legislative Ways and Means committees include descriptions of program level goals, activities, and responses.

The agency information on clientele, service needs, and the external environment summarizes the foundation for the planning each agency must accomplish to support effective programs and activities. Independent of, or in complement to, the state budgeting process, many state agencies conduct extensive strategic planning initiatives. The agency or its appointed citizen board or commission may actually conduct the planning exercise. Critical to the success of strategic planning is the ability to translate strategic goals and values into operational objectives and programs. This remains a challenge for even the most experienced Oregon state agencies. As an example, the Oregon State Department of Forestry builds its biennial budget submission; organizes its issues, policies and programs; sets its annual work plans; and measures outcome accomplishment from a continuing series of strategic plans developed by the State Board of Forestry (Oregon State Board of Forestry, 1995).

Performance Budgeting - Reflecting the intent and initiatives of the Oregon Progress Board and the legislature, the Oregon state budget process is increasingly embracing the themes and measures of performance budgeting. In coordination with efforts of the Progress Board, state agencies have begun to develop performance measures that link agency goals and programs with broad public needs and conditions. However, many agencies and the budget decision making process have yet to fully use performance information in the allocation of budget resources. There are a variety of reasons for this, including the complexity of the task for some programs and agencies, fear that the information might be used to defund programs, the shortage of staff time, etc.

The biennial Budget and Legislative Concept Instructions request that agencies continue to track and use performance and other outcome measures. The instructions direct each agency to update the linkages between its agency goals and program outcomes, and the appropriate measures in the Oregon benchmarks. Agencies can include both quantitative and qualitative performance measures in their budget request. For example, in the Agency Overview of their 1999-2001 legislative budget presentation the Oregon Youth Authority included a cost per day comparison of its different treatment services and treatment streams. This cost data was then placed in context with average costs per day from other states.

The Youth Authority then described several other applications of performance measures. The description listed the primary and secondary Oregon benchmarks addressed by the agency and its programs. Figure 7-4 on the following page summarizes these performance benchmarks. These primary and secondary benchmarks provide strategic outcome goals for the agency and its performance. At the program objective level, the agency set eight agency-level performance measures related to recidivism, custody security, youth education, youth earnings and restitution, staff training, agency affirmative action, and workplace safety. The description of these performance measures related on-going program-level performance to numerical indicators of outcomes over a time period. Finally, because of the Youth Authority’s involvement in a national project on juvenile facility security, the agency presented a detailed and refined set of performance measures related to the operation of juvenile detention facilities.

To meet the test of a full scale performance budget, the data would need to provide decision makers with an understanding of how additional increments of spending will affect performance outcomes. This is a high bar that most agencies are unlikely to meet in the foreseeable future, for all of the reasons outlined earlier in the chapter on performance budgeting.

In summary, the Oregon state budget process is moving agencies toward a performance budgeting framework. Many of the elements of performance budgeting

Figure 7-4:

OREGON YOUTH AUTHORITY

are evident in the budget preparation process. However, the links between program objectives and performance measures remains unclear, and the relationship of costs to program effort and resource inputs to outcomes similarly remains elusive. Recognizing this weakness, the 1999 regular session of the legislature passed HB 3182. In addition to revising the agency budget preparation instructions, this bill established a performance budgeting pilot project. Under the pilot project, several state agencies will prepare their 2001-2003 budget using enhanced performance budgeting.

In the 2001 legislative session, HB 3358 took the concern for performance budgeting an additional step by significantly increasing the authority of the Progress Board to establish “goals for Oregon’s progress, including measurable indicators of the achievement of those goals and [assisting] state agencies and their partners in developing performance measures that provide linkages to the measurable indicators” (ORS, Chapter 582). In addition to increasing the Progress Board’s legal authority, HB 3358 increased its political authority by adding legislative members to the Advisory Committee and making the Governor the head of the Progress Board. These changes increase the possibility that state agencies will make greater use of performance measures in their future budgeting processes.

Zero-based Budgeting - The current Oregon state budget format reflects a strong bias in favor of modified zero-base budgeting. While line item budgeting, program budgeting, and more recently, performance budgeting remain evident in agency budget analyses and presentations, zero-base budgeting exerts a major influence on budget analysis and preparation. As described above, zero-base budgeting does not mean building a program and agency budget from scratch with each new budget cycle. Similarly, the state of Oregon does not justify each program and agency budget from a complete zero base each biennium. Instead, the state follows a more practical modified zero-base approach of the kind outlined in the previous chapter. This includes the development of a base budget, the development of a current services level, the development of two different 10 percent reduction packages, and the development and selection of policy or decision packages.

Oregon, like most jurisdictions using a modified zero-based budgeting system, organizes decision packages around existing “program units.” Agencies begin with the development of a base budget for each agency and its program units. The base budget is the most recent Legislatively Adopted Budget, plus any budget actions taken by the Legislative Emergency Board, and any adjustments in the cost of personal services as projected by the state’s Position Information Control System (PICS). The “current service level” is then built from the base budget.

The “current service level” represents the cost of continuing the current legislatively approved package of program services and activities into the future. The current service level reflects an agency’s base budget with adjustments from a series of standardized “essential packages.” The biennial Budget and Legislative Concept Instructions assigns a standard code number to each essential package that adjusts the base level of spending. As defined in the 1999-2001 instructions, the current service level reflects the adjustments of several standard decision packages:

· plus personnel vacancy and position adjustments (Package #010);

· plus new program phase-in costs (Package #020);

· plus one-time charges (Package #020);

· less discontinued program phase-out costs (Package #020).

· plus inflation and other cost adjustments for purchases (Package #030);

· plus mandatory caseload increases (Package #040);

· plus fund shifts or revenue reductions (Package #050)

Using the current service level as a foundation, agencies may then propose and develop a set of “policy packages.” Policy packages are equivalent to the decision packages of zero-base budgeting. Added together, the base budget, the essential packages, and a justified selection of policy packages represents an Agency Budget Request. Policy packages may accomplish several tasks:

· Form new programs and initiatives, or expand current ones;

· Reduce or end programs;

· Implement interagency partnerships;

· Transfer programs between agencies;

· Shift from one fund type to another;

· Establish or increase fees;

· Implement reorganization and reinvention proposals;

· Fund legislative concepts considered by the Legislature;

· Present Capital Construction projects.

The combination of an inflation-adjusted current service level and a set of policy packages usually pushes the Agency Budget Request above past spending levels.

Following the intent of zero-base budgeting, each agency must then develop two 10 percent reduction options. The first option models a 10 percent reduction in General Fund and Lottery Fund resources. The biennial budget instructions direct agencies to break the 10 percent reduction into two five percent increments. Building these increments requires that agencies select a set of policy packages to meet a 10 percent reduction, and then select a set of policy packages to restore a five- percent increment of activity.

Staff analysts for BAM carefully analyze agency packages, including current service level packages (CSL) as well as agency policy packages. Based on this analysis BAM makes recommendations to the Governor. BAM staff assumes that there is no “entitlement” to any piece of the budget. Analysts frequently make recommendations to cut an agency’s base program rather than cutting an agency’s 10% “cut package”.

Building from the 10 percent reduction option, the DAS Budget and Management Division analysts and the Governor restore combinations of decision/ policy packages. The restoration of a particular combination of policy packages reflects the Governor’s policy agenda, and the relative efficiency and effectiveness of each program or activity. The Governor’s Recommended Budget for each agency emerges from this “add back” process. The agency works very closely with the DAS analysts and the Governor to structure and analyze the impacts of desired combinations of policy packages. This intense period of proposal and analysis is similar to the federal budget “pass back” following OMB review and adjustment of federal department and agency budgets.

The second reduction option reflects the requirements of HB 3182 passed during the 1999 regular legislative session. For the Governor’s Recommended Budget proposal, this option requests an analysis of a 10 percent reduction in all funds from all sources. The programs and activities eliminated by the reduction must then be ranked on a cost-benefit basis as a criterion for restoration toward the Governor’s proposal.

220

Following the intent of zero-based budgeting, the development of the two reduction options provides a decision context for the agency, DAS Budget and Management analysts, the Governor, and the Legislature. The reduction exercises also force agency managers to reconsider the relative importance of all the activities and programs conducted by the agency.

Closing Reflections On Oregon's Use of Budget Formats – Each of the four budget formats—line item, program, performance, and zero-based budgeting—offers unique strength to the Oregon budget process. Budget formats filter, structure and focus the information used by a variety of decision makers in the budget process. Thus, the conscious selection and melding of the formats at each stage of the budget process represents a decision by the process managers to influence behavior and process outcomes. An adroit blending of budget formats can focus the budget debate away from contention and toward agreement. But in the end, budget formats present a fundamental tool for directing the policies and behavior of state agencies.

Formats and Democratic Governance: Summary Conclusions on Uses and Comparative Strengths of Budget Formats

Since the creation of line-item budgeting at the turn of the 20th century, there have been repeated attempts to supplement object code budgeting with other budgeting formats. As we have seen in the previous chapters in this section, line item budgeting does not provide information that enables us to answer the following questions: 1. What are we achieving with the expenditure of tax dollars (PPBS/Program)? 2. How efficient and cost effective are we in the expenditure of these resources (Performance)? and 3. Should we be doing this at all? Or should we spend tax dollars on X rather than Y (ZBB)? Each of the four budgeting formats we have examined in this section attempt to bring these questions to the center of the budget deliberations by restructuring the decision rules and what counts for evidence in the budget allocation process.

Table 7-5 summarizes the comparative characteristics of the budgeting formats we have examined in this section. One of the purposes of Table 7-5 is to illustrate the extent to which budget reform is really part of a larger set of analytic techniques that are designed to introduce more systematic analysis and evaluation to public decision making at all levels. These techniques are all designed to correct the defects of the incremental consequences of the political process, which tends to rely on "satisficing" rather than "optimizing as the standard for deciding what to do.

TABLE 7-5

SUMMARY CHARACTERISTICS OF BUDGET REFORMS

A. Purpose / Types of Budget Reforms

1 2

Goal Setting Budget Tools: Management Budget Tools:

Problem/ Solution Identification Implementation and Control

______________________ _________________________________

-PPBS -MBO

-Strategic Planning -Performance budgeting

-ZBB

-Cost / benefit analysis

-operations research

-critical path analysis

B. Assumptions Shared in Common by Budget Reforms

1. Linearity - Goal should precede action.

2. Causality -alternatives should be systematically assessed.

3. Controllability - The best alternative should be chosen to achieve the desired outcome.

As Table 7-5 indicates, all of the budget reforms introduced to correct the defects of this "satisficing" bias built into line item budgeting have the following characteristics in common:

· They assume the world and its processes can be made to conform to the fundamental principles of rationality. These assumptions include the ability to separate causes from effects, the capacity to structure means in a linear fashion to achieve targeted ends, and ability to control the flow of decision making.

· They seek to introduce more systematic analysis to decision making.

· They seek to introduce greater control over both the decision making and management processes.

Some of the budgeting reforms, such as PPBS and strategic planning, are intended to help focus attention on problem identification and solution setting. Others, like MBO and Performance Budgeting, are managerial centered and intended to be performance oriented. Others, like ZBB and cost/benefit analysis, are used both for planning purposes and managerial control.

Why has there been such a compelling need to restructure the rules of the budget game over the past fifty years? Many argue that it is the only way to force participants to confront overlap and duplication, unmet needs that aren't adequately represented in the political process, and to engage in some long term planning to solve problems that are not likely to get adequately addressed in the political process. But what does this mean in practice? In practice it means that power and influence get shifted in the budgeting process to the experts who gather the information, analyze it, and try to make sense out of it. Table 7-6 below seeks to capture this inherent bias that is built into the budget reform formats we have discussed in this section.

As Table 7-6 suggests, the rules of the game by which budget reformers wish to play shifts the locus of budget decision making away from elected officials to managers and analysts, in short, to career administrators. It is no accident, then, that the most budget reforms have been initiated by the executive branch of government, not by the legislative branch. It is also no accident that these reforms work best in local governments with strong city managers and with weak part-time city councils (O'Toole and Stipak, 1991).

Table 7-7 reformulates Table 7-5 by isolating the differences among the various budgeting formats with respect to their reliance on information, expertise, and potential

for managing conflict in the budgeting process. The differences between object-code budgeting and the other three formats are striking.

Table 7-6

Budget Format Comparisons

	Purpose? Central Question? Problem to be Solved?	Chief Characteristics	Central Actors	Conditions for Success	Strengths	Weaknesses
Line Item/ Object Code/ Incrementalism	Financial accountability Is the money being spent according to intention? Preventing mis-appropriation of funds.	Focus is on past, with last year as the “base” Use of formulas, like “fair share” to balance the budget	Accountants and central office budget managers interested in balancing the budget. Constituency advocates in the legislature	Minimum amount of staff time and expertise is needed to create and track the budget	Works well when time is short, there is considerable complexity, multiple stakeholders, and high potential for conflict	Not good for dealing with questions of efficiency, effectiveness, future &/or neglected concerns
Program Budgeting (PPBS)	Program accountability Is the program achieving its goals and objectives? Program effectiveness	Focus is on program plans, goals and objectives. Requires budgeting to a plan	Program managers & program analysts	Requires the resources necessary to develop program plans, goals and objectives	Provides clear linkage between program activities and budget allocation Links parts to whole and present to future	Requires considerable investment of resources High potential for conflict

TABLE 7-6 continued

Budget Format Comparisons continued

As Tables 7-7 and 7-8 illustrate, object code budgeting has high value for those in the political arena who have to reach decisions relatively quickly on complex issues that may frequently be conflict-ridden. Tables 7-7 and 7-8 provide an evaluative summary of the capacity of the differing budget formats discussed in Section II of this Handbook to operate successfully under different conditions of programmatic complexity/interdependence, potential for conflict, informational/analytic capacity, and the amount/kind of expertise available during the budget process. Tables 7-7 and 7-8 remind us that line item budgeting is quite good in operating in environments characterized by high levels of programmatic complexity/interdependence, high potential for conflict, and a shortage of time and information. By focusing on dollars, coupled with the infinite ability to divide dollars into ever more finite increments, issues of conflict, interdependence, and analytic sophistication get obfuscated with line item budgeting. The other formats put considerable pressure on analysis, focus attention on important programmatic consequences of budget increases/decreases, and require considerable attention to program complexities and interdependencies. There is considerable irony in the fact that increased complexity has spawned the need for additional information which the line item format does not provide. But once created, these performance, program, and program budgeting formats add layers of complexity to the budgeting process, which frequently create incentives to fall back onto line item budgeting in order to make the process more manageable. You can use Tables 7-7 and 7-8 to assess the conditions that exist within your program/agency and the consequent likelihood of success for each of the respective budgeting formats.

The history of efforts to implement budgeting reform ultimately turns on who should rule? How much influence should be lodged with budget analysts and program managers by comparison with the influence lodged with elected officials and citizens? Answers to these questions ultimately turn on one's theory of democratic governance. We will return to this centrally important issue in the concluding section of this Handbook (p. 206 ff.).

PA 503NFP: Public Budgeting and Finance Douglas Morgan

Lewis and Clark College

PA 503NFP: Public Budgeting and Finance Douglas Morgan

Lewis and Clark College

6

168

					TABLE 7-7
			CHARACTERISTICS OF BUDGETING SYSTEMS
		Low									High
		1	2	3	4	5	6	7	8	9	10
Ability to Manage	ZBB								Line Item
Program Complexities/	Program
Interdependencies		Performance
Ability To Manage	ZBB								Line Item
Conflict		Program
		Performance
Informational/	Line Item									ZBB
Analytic Demands									Program
									Performance
Need for Specialized	Line Item									ZBB
Expertise										Program
									Performance

TABLE 7-8

CHARACTERISTICS OF BUDGETING ENVIRONMENT

Low High

1 2 3 4 5 6 7 8 9 1 0

Program Complexity

Potential for Conflict

Informational/ Analytic Capacity

Amount & Kind of Expertise

HIGH CONFLICT &	LINE ITEM	ZBB PROGRAM
PROGRAM COMPLEXITY LOW	PERFORMANCE	ZBB PROGRAM PERFORMANCE

LOW INFORMATION/EXPERTISE HIGH

OREGON REFERENCES AND ADDITIONAL RESOURCES

Kitzhaber, John. (1998). 1999-2001 Governor’s Budget and Tax Expenditure Report. Salem, Oregon: State of Oregon Department of Administrative Services. 2 vols.

Oregon Department of Administrative Services. (2000). “ORBITS PROJECT.” Salem, Oregon: State of Oregon Department of Administrative Services, Budget and Management Division. Also, see Sept. 29,2000 version at <http://www.bam.das.state.or.us/budproj/orbproj.htm>

Oregon Department of Administrative Services. (1999). “State Finance 101: State of Oregon Financial Structure.” Salem, Oregon: State of Oregon Department of Administrative Services, State Controller’s Division. Also, see <http://scd.das.state.or.us/lfo.htm>

Oregon Department of Administrative Services. (1998a). 1999-2001 Budget and Legislative Concept Instructions. Salem, Oregon: State of Oregon Department of Administrative Services (DAS), Budget and Management Division. These instructions are published biennially in the spring of even numbered years at the beginning of each budget preparation cycle. Also see DAS home page at <http://www.bam.das.state.or.us/>

Oregon Department of Administrative Services. (1998b). ABIS Users Manual. Salem, Oregon: State of Oregon Department of Administrative Services, Budget and Management Division. This manual is updated in March of even numbered years to support preparation of the biennial budget.

Oregon State Board of Forestry. (1995). FPFO Forestry Program for Oregon. Salem, Oregon: Oregon Board of Forestry and the Oregon Department of Forestry.

Zero Based Budget Reference and Additional Resources

Lyden, Fremont J. and Mark Lindenberg (1983). Public Budgeting in Theory and Practice, Longman: New York, chapters 1 and 2.

Lee, Robert & Ronald Johnson. (1998). Public Budgeting Systems, sixth edition, Aspen. Chapters 1, 2, 15.

Rubin, Irene. (2000). The Politics of Public Budgeting: Getting, Spending, Borrowing and Balancing, 4th edition. Chatham House: New York, Chapter 1.

Schick, Allen. (1983). "Incremental Budgeting in a Decremental Age", Policy Sciences. (Amsterdam: Elsevier Science Publishers B.V.,), pp. 1-25.

Schick, Alan. (1966). “The Road to PPB: The Stages of Budget Reform.” Public Administration Review 26 (December): 243-258. Reprinted in Albert Hyde, Government Budgeting: Theory, Process and Politics, Brooks Cole: Pacific Grove, CA, 1992, pp. 46-60.

Wildavsky, Aaron. (1992). The New Politics of the Budgetary Process, second edition. Harper Collins, New York, Chapter 1, 10, pp. 430-440.

Wildavsky, Aaron. (1984). The Politics of the Budgetary Process, fourth edition. Little Brown: Boston, MA. Prologue and Chapter 1.

Wildavsky, Aaron. (1978). “A Budget for All Seasons.” Public Administration Review, (November/December) pp. 501-509.

Chapter 8

Revenues: Establishing the Budget Parameters

Public budgeting is a political and technical process of matching and allocating monetary resources against needs. These resources include revenues from a wide variety of sources, including: taxes, fees for service, debt instruments, and funds from other levels of government. Forecasting the sources, timing and level of the revenues that will be available for expenditure during the budget period is a critical task of the budget preparation process. In fact, revenue forecasting and estimation is the fundamental driver in shaping the budget formation process.

The forecasted level of revenues sets the upper limit for the proposed level of spending in the budget. Under-estimation of revenues leads to reduced spending that may fail to meet critical needs, while the over-estimation of revenues leads to program cutbacks part way through the budget cycle. Developing and agreeing upon a revenue estimate requires both technical skill and political agreement.

Despite the importance of revenue forecasts in establishing clearly defined sidebars for budget development, public budgeting often separates the development of plans for program spending from revenue generation. This separation is especially true at the federal level where deficit spending is legally permissible. The sheer size and complexity of the federal budget, combined with a congressional committee structure that divides responsibility for revenue collection from budget development and appropriations spending, further reinforces a separation of revenue planning from budget expenditures. But even in many local jurisdictions where deficit spending is illegal, separate departments frequently undertake revenue forecasting and budget development. This separation sometimes causes many department and agency budget analysts to place relatively little emphasis on the revenue context surrounding their spending estimations. This problem has decreased significantly in recent years as jurisdictions have had to deal with the dual consequences of a declining economy and an electorate that is more inclined to jealously guard its pocket-book

169

* Kent Robinson took responsibility for preparing much of the initial draft material for this chapter with assistance from Jon Yunker.

This handbook chapter reviews the importance of revenue forecasting to the public budgeting process with a focus on the following topics that are of general importance to all public agencies.

A. Budgeting as a Revenue Forecast;

B. Types of Revenues in Public Budgeting;

C. Taxation as a Revenue Source;

D. The Oregon Tax Revenue Situation in Context;

E. Tax Expenditures: Lost Revenues or Purposeful Policy Decisions?;

F. Debt Instruments as a Revenue Source;

G. Organization of Revenues;

H. Key Sources of Information.

As an organizational aid to the reader, Chapter 9 provides a more in depth discussion of revenue forecasting techniques. This includes a description of the analytic and economic techniques used by the state of Oregon and local governments in revenue forecasting.

A. Revenue Forecasting in the Budget Process

Building a public budget requires an assessment of needs and program resources. More importantly, however, it is essential to know what financial resources are available to support these needs. The budget process provides an opportunity to adjust the levels of spending so that they will match the revenues that are available. This contrasts to the private sector, where revenues are first adjusted to meet expenditure needs. If a private company projects revenues that are below anticipated expenditure levels, the first goal is to generate more money through sales and advertising before resorting to laying people off or cutting expenditures. For the public sector, the opposite strategy is followed. First, revenue estimates are undertaken, which then serve as the basis for deciding the levels of expenditure and program service .

While the development of a revenue forecast rests heavily on statistical, economic and financial techniques, revenue forecasting is as much a political as it is a technical undertaking. No matter how technically accurate the work of revenue forecasters, if this work does not have the faith and confidence of the budget actors, it will be ignored. For this reason, revenue forecasters have to be mindful of the larger political backdrop that includes policy visions, political agreements, and citizen support. Revenue forecasters play a major role in building and maintaining the trust and support of elected officials and taxpayers in the budget process.

The revenue forecast is a public document, which means that the budget actors must find the forecast process credible and the revenue estimates accurate. While technical committees of revenue specialists usually perform the actual detailed forecasting work, a broader review by stakeholder groups is usually necessary to build the larger credibility necessary for this technical work to have its maximum impact. Both the technical and the review process must be open to the media and to the general public to ensure that no one sector, group or individual monopolizes control of the process or the forecast.

Second, to be credible, all of the budget actors must understand the forecast, including the procedures used in developing the forecast. The involved elected officials must understand the basic assumptions and the reasons behind the forecast. All of the key actors must understand the upside and downside of the forecast. Revenue forecasters must be able to explain the probability and impacts of any significant reductions in revenues. For example, the inability to explain a 10% reduction for a given category of revenue usually begins to sow the seeds of lost credibility.

Third, to reinforce credibility, a revenue forecast must be consistent with the forecasts of neighboring states or jurisdictions. While revenue forecasts for the state or local governments may be marginally different, any major differences in outcomes must be consistent with differences in the techniques and assumptions that are used in producing the forecast.

Beyond the revenue forecasting process, revenue estimates must have sufficient accuracy. Accuracy is often measured in terms of some kind of percentage deviation of actual revenues from the forecast. A forecast that predicts revenue within two percentage points of anticipated expenditures is considered within the acceptable margin of error. But, revenue forecasting is an art as much as it is a science. No forecaster or forecasting group knows for certain what revenues will be one - two years into the future. Because of this, consistent matching of forecasts and actual revenues rarely happens. For this reason, some “cushion” always needs to be built into the revenue and expenditure balancing equation.

B. Types of Revenues in Public Budgeting

A variety of revenue sources contribute to the public budgeting process. States or local governments may generate their own revenues through taxes and fees. Some revenues originate from other levels of government as direct grants or on a matching basis. For some states, Lottery funds have become an important source of revenue, which is not available to local governments and special districts. Revenues and rents from the sale and rental of resources provide still another source of revenue. Various borrowing and debt instruments are used to generate revenue to finance capital spending, such as water/sewer treatment facilities, transportation infrastructure, public buildings, etc.. In many jurisdictions unspent funds may be carried forward for spending in the new fiscal year.

Ideally, each of the above sources of revenue will provide a balanced, stable flow of resources. An effective blend of revenue sources allows sufficient stability to build the organization, develop and retain staff expertise, and develop the program activities necessary to effectively address public problems and issues. Under a balanced blend of revenues, a temporary increase in one source of revenue will help offset a cyclical or temporary decline in another source. For example, during a period of economic downturn revenue generated by business licenses, and taxes for liquor, cigarettes, gaming and gasoline do not decline as rapidly as individual and business income taxes. The comparative strengths and weaknesses of these various revenue sources will be discussed further in the sections that follow.

1. Intergovernmental Transfers: These are revenues provided to a jurisdiction by other levels of government or other agencies. Intergovernmental may result from any of the following combination of circumstances:

· direct grants without restrictions

· direct grants for the performance of service

· matching funds at a given ratio as a condition of expenditure

· reimbursement of local jurisdictions for the loss of tax revenue

· resources funneled through a government to another level of government

For example, federal transfers fund a major portion of the health and social service programs administered by the Oregon Department of Human Services. Accessing and using many of these federal grants requires the spending of state matching funds. In a similar set of payments, the state of Oregon transfers millions of dollars each biennium to counties for the local delivery of social services. Counties often must contribute a local match to use these funds.

Other federal transfers are tied to federal land ownership or to a percentage of the revenue generated by federal resource management activities. In the past, these transfer payments provided a major source of income to many western states. With the recent severe reduction in federal timber sale programs, these revenues have been declining. However, in the last legislative session Representative Defazio from Eugene played a key role in brokering a Congressional agreement that will increase payments in 2002 to rural counties in Oregon, totaling more than 230 million dollars a year. The counties where the federal ownership of lands is located or where the federal activity took place becomes the beneficiary of these intergovernmental transfers. The most important of these transfers include the following:

· Payments in Lieu of Taxes (PILT) (compensation for lost property tax revenues because of federal land holdings);

· U.S. Forest Service 25 Percent Fund payments (wealth transfer and compensation for lost property tax revenues because of federal land holdings. Federal law earmarks these funds for county road maintenance and schools);

· Taylor Grazing Act payments (compensation for lost property tax revenue because of federal land holdings);

2. Taxes: Most states collect a personal and corporate income tax as well as a sales tax. Property taxes provide the predominant source of revenues for special purpose districts and a significant portion of revenue for counties, cities, and general purpose local governments. Recent ballot measures have limited the property tax rate increases in many parts of the United States, particularly in California and Oregon (see pp. 172) In addition to income taxes, other major tax sources include: a per gallon motor fuels tax, a weight mile tax on heavy trucks, a cigarette and tobacco products tax, forest severance taxes, insurance taxes, and gift and inheritance taxes.

3. Fees for Uses and Services: Increasingly, governmental jurisdictions at all levels are looking for new ways of generating revenues by charging for the services they may provide to the public, to other agencies, to local governments, or any other entity that may have a demand for a service or special expertise. Many state agencies generate some form of revenue from these kinds of fees and charges. For example, most states charge corporation fees, higher education tuition and fees, commercial fishing licenses, state court fees, drivers license fees, vehicle title fees, state park recreation fees and revenues for hunting and fishing licenses. At the local level, jurisdictions charge fees to corporations for the privilege of having access to county and city residents, while city or water district residents pay water and sewer fees to cover infrastructure and water/sewer provision. The most significant characteristic of “fees for service” is that that they cannot be used to fund general government services. They are considered “dedicated” funds and can only be used to support the activity for which the fee is collected.

4. Sale of Products and Assets; Rents: A number of state programs generate revenues through the sales of products and commodities. Other programs generate rents through the lease of state properties. For example, a percentage of state-controlled sales of alcoholic beverages contribute to the state general fund. In some western states, the sale of state timber provides revenues to help support K-12 education and some counties generate revenues through the sale of forest products from their county lands. Rents from state facilities such as docks and utility corridor right-of-ways also generate rental income.

5. Debt Instruments: Borrowing funds through the sale of bonds or other financial instruments is an important source of revenue for state and local governments. While borrowing opens governments to debt liability, long-term financing provides an effective means of funding capital projects and other major purchases. For example, some states use various forms of debt financing to support higher education, forest rehabilitation, pollution control, highway construction, housing, and mass transit construction.

6. Lottery and Other Gaming Revenues: Until 1964 lotteries were illegal in every state in the country. Today lotteries operate in 37 states and the District of Columbia. Net proceeds from lotteries are used to support job creation and economic development , public education, protection of the environment, public health and a multitude of other public activities. Table 8-1 shows the dramatic growth in the revenue generating capacity of legalized lottery and gaming activities. However, the lotteries’ contributions to state budgets are modest, as shown by Table 8-2 on the following page.

Table 8 - 1

Growth and Change in U.S. Lottery Sales, 1973, 1987, and 1997

(Sales totals in billions of 1997 dollars)

1973 1987 1997

Number of state lotteries 7 23 38

Sales by type of game

Passive drawings 1,946 1 0

Instant games 24 4,196 14,435

Daily numbers 0 6,418 7,731

Lotto 0 6,937 10,088

Keno 0 0 1,791

Other 0 0 34

Total without VLT's 1,970 17,553 34,079

VLT's 0 0 2,318

Per capita sales without VLT's 35 127 150

Source: Clotfelter and Cook, "Redefining 'Success' in the State Lottery Business,"

Journal of Policy Analysis and Management 9 (1990), Table 1, p. 100; International Gaming &

Wagering Business, April 1998, p. 44; June 1998, pp. 4849; cited in Charles T. Clotfelter, Philip J. Cook, Julie A. Edell, "State Lotteries at the Turn of the Century" Report to the National Gambling Impact Study Commission www.pubpol.duke.edu/people/faculty/clotfelter/lottrep.pdf

.

In 1997, total revenues from the 38 state lotteries amounted to only about 2.2 percent of the own-source general revenue for those same states. By contrast, state general sales taxes and income taxes each averaged one quarter of all own-source general sales taxes and income taxes each averaged one quarter of all own-source general revenue collected by states.

196

Table 8 - 2

State Ownsource Revenues and Lottery Revenues, 1997

(amounts in millions of dollars)

Ownsource Lottery Lottery

general revenue (b) as percent

State revenue (a) of total

Arizona 8,262 79.7 0.96%

California 73,584 711.9 0.97%

Colorado 7,349 92.7 1.26%

Connecticut 10,071 252.6 2.51

DC* 2,986 34.2 1.14%

Delaware 2,797 66.7 2.38%

Florida 25,984 802.4 3.09%

Georgia 13,707 558.5 4.07%

Idaho 2,552 17.7 0.70%

Illinois 23,355 571.2 2.45%

Indiana 12,132 172.0 1.42%

Iowa 6,352 42.5 0.67%

Kansas 5,425 55.9 1.03%

Kentucky 8,967 152.0 1.70%

Louisiana 9,200 98.4 1.07%

Maine 2,760 40.0 1.45%

Maryland 11,388 392.3 3.44%

Massachusetts 18,002 696.0 3.87%

Michigan 25,590 563.4 2.20%

Minnesota 13,581 84.1 0.62%

Missouri 10,054 131.3 1.31 %

Montana 1,920 6.3 0.33%

Nebraska 3,576 19.5 0.54%

New Hampshire 1,807 51.3 2.84%

New Jersey 20,600 612.4 2.97%

New Mexico 4,990 20.5 0.41

New York 44,912 1,530.6 3.41

Ohio 21,799 750.4 3.44%

Oregon 7,768 72.8 0.94%

Pennsylvania 25,792 691.0 2.68%

Rhode Island 2,392 42.2 1.76%

South Dakota 1,247 5.9 0.47%

Texas 31,746 1,161.1 3.66%

Vermont 1,386 23.6 1.70%

Virginia 14,545 331.1 2.28%

Washington 14,101 94.2 0.67%

West Virginia 3,998 43.9 1.10%

Wisconsin 13,012 132.6 1.02%

TOTAL 509,691 11,205 2.20%

(a) General revenues minus intergovernmental transfers. (b) Salesprizesoperating expenses. See Table 2.

Source: Charles T. Clotfelter, Philip J. Cook, Julie A. Edell, "State Lotteries at the Turn of the Century" Report to the National Gambling Impact Study Commission www.pubpol.duke.edu/people/faculty/clotfelter/lottrep.pdf

7. Balance Carried Forward: At the state level, all General Fund moneys must be spent by the close of the current fiscal year or biennium. Most routine agency program and activity spending falls within the biennium time frame. At the end of the biennium or fiscal year after all appropriations and adjustments, a net balance remains in the General Fund. Any positive net balance may be carried forward into the next biennium as available resources. Local governments may label this balance as the beginning balance or the available fund balance. While governments as a matter of policy try to minimize this carry forward balance, it may reach sizable amounts, especially when it is used as a source of revenue for meeting financial obligations prior to the receipt of property tax revenues.

From these major categories of revenues, a blend of revenues emerges to fund each jurisdiction. Figure 8-3 presents the blend of revenues used to support the state’s General Fund in the 1999-2001 Governor’s Recommended Budget (Kitzhaber, 1998). As a contrast, Figure 8-4 presents the revenue blend for Washington County, Oregon for its 1997-1998 fiscal year (Washington County, Oregon, 1997).

C. Taxation as a Revenue Source:

For better or worse, taxation binds together all citizens of a nation, state or jurisdiction. The contribution of personal resources toward the common good represents a major element of citizenship. Beyond a contribution to the general welfare, taxation provides a powerful tool for the achievement of public policy. Taxation also provides a means to redistribute income across society, to discourage or encourage desired behavior, or to redress the hardship of a particular industry, class or group. For some citizens, taxation represents an intrusion of the community into the development of private wealth and success.

Since the application of a tax has obvious political, policy, economic and administrative ramifications, imposing and raising taxes presents a major political challenge. Many citizens and politicians are inclined to view such requests as the product of government inefficiencies or the unquenchable thirst of career administrators and clients for new government resources. As a result, the burden of proof is always

State of Oregon Governor's Recommended Budget: General Fund Resources –

1999-2001

Figure 8- 3

GF Beginning Balance

5%

3%

3%

Revenues

All Other GF

8%

Tax

Corporate Income

Lottery

1%

Tax

Cigarette/Tobacco

1%

Insurance

79%

Personal Income Tax

Figure 8- 4

0%

Fines and Penalties

2%

Licenses and Permits

30%

Balances

Beginning Fund

20%

Revenues

Intergovernmental

20%

Taxes

9%

Charges for Service

8%

Revenues

Miscellaneous

6%

Funds

Transfers from Other

5%

Interfund Revenues

Washington County, Oregon 1997-1998 All Fund Resources

Total Resources = $363.0 Million

on the proponents of proposed tax increases to convincingly demonstrate that they are justified by: a widely communicated, clear demonstration of need; a guarantee of a lack of waste by government; and often the linking of revenues to specific projects or purposes. In short, increasing taxes requires a tremendous political effort and the spending of scarce political capital.

1. Tax System Design Principles: As a policy instrument, taxation should serve to encourage desired citizen behavior. Without careful design and forethought, taxation may have numerous unintended consequences. Tax systems may be designed to shift or retain wealth in a given group. Such shifts are measures of tax burden equity. Tax systems may be:

· Regressive: places the greatest tax burden on those with the least wealth to spend beyond necessities and the least ability to pay;

· Proportional: the tax burden affects the same percentage of income or wealth across all groups;

· Progressive: places the greatest tax burden on those with the greatest wealth and ability to pay.

Whether regressive, proportional, or progressive, each tax applies a burden. The burden is frequently calculated as the effective rate of the tax payments as a percentage of income.

In many policy situations, the “ability to pay principle” becomes an important tax system design criterion. Under this principle, regardless of the recipient of the benefit, those with the greatest ability to pay contribute the most tax support. In contrast, the adoption of a “benefits received principle” places the greatest tax burden on those receiving the greatest program benefits, regardless of income. For example, under the “benefits received principle” citizens with school-aged children would be required to carry a greater load of education related taxes.

As a governmental entity designs a tax system, it must settle on a balance of the above principles. The design will result in a relatively regressive, proportional, or progressive tax. A number of other taxation principles often guide tax system design:

· Individual Equity: Depending on the equity criteria selected by the jurisdiction, the tax burden should be equitably distributed;

· System Equity: The application of a tax to achieve a policy goal should have a minimal impact on the overall equity of the larger tax system;

· Economic Growth: The tax system should support economic growth and stabilization;

· Relative Tax Burden: Designers of a tax system must understand how the actual tax burden is shifted through the economic system. For example, a corporation may decide to deal with an increase in its corporate income tax by passing on the increased costs of doing business through higher prices to consumers. This may have a differential consequence for consumers. For those on fixed incomes, the net result of higher prices is to reduce disposable income and to increase the proportion of disposable income spent to meet ones personal income tax obligation. However, for individuals who have incomes that are indexed to “cost-of-living” increases, they may be able to absorb the price increases without necessarily having to spend more of their disposable income on meeting their personal income tax obligations. But in both cases, the increase in corporate income tax may in reality shift the actual tax burden from corporations to individuals.

In addition to the political and policy aspects of taxation, the design of a taxation system has economic ramifications. The imposition of a tax adds an increase in costs to the price of a service or good, thus raising costs for everyone participating in the market. Consequently, tax systems in the United States are almost always designed to minimize burdens on effective markets.

Finally, the administration of a tax system carries costs. The tax must be applied in a non-arbitrary, fair manner. Citizens in similar circumstances must receive similar treatment under the tax policy. Administering a tax system to meet these criteria requires record keeping, processing, mailing, and enforcement. Different tax systems carry a different proportion of administrative costs. Ultimately, the goal is to keep administrative costs as low a possible. In addition, to reassure citizens of the fairness in tax system administration, jurisdictions and revenue collection agencies attempt to make the tax system understandable to citizens.

2. Major Types of Taxes: Generating revenues through taxation rests on three different tax bases. A tax base is an identified form of income, wealth, or consumption. Each taxing jurisdiction must carefully delineate the content of the tax base or bases it will rely upon for revenue generation. In order ensure a stable, consistent, and continuing flow of revenues, states and jurisdictions must blend sources of tax revenue. In blending taxes, the state or jurisdiction is actually blending tax bases and the ability of citizens within those bases to contribute revenues. Following is a summary of the major types of taxes, their bases, and their uses (Lyden and Lindenberg, 1983):

a. Income tax - Income taxes are designed to apply to citizens and to corporations. For citizens, the income tax base includes salary and wages, but may, or may not include interest income, income from investments (capital gains), and income from other sources. A series of adjustments, exemptions, and deductions refine the income tax burden to each citizen’s situation. The income tax rate structure determines the actual burden carried by each payer.

The income tax can effectively combine all forms of income under a tax with efficient administration. An income tax may be applied to large areas or jurisdictions. By requiring employers to administer the tax withholding, tremendous effectiveness is achieved in the administration of the tax.

Income tax revenues are subject to the cycles of the economy. In a strong economy, income tax revenues often exceed forecasts, while in a declining economy, income tax revenues drop. In inflationary times, taxpayers may be pushed into higher tax brackets before government can make adjustments. Indexing for inflation or changes to deductions can help to compensate for inflation driven bracket creep.

The federal and state governments make extensive use of the income tax for purposes of revenue generation. Assuming it is legal under state law, cities may also levy an income tax to ensure that workers who live in other jurisdictions but work within city boundaries contribute to city services.

The Oregon Constitution was amended in 1917 to allow for the collection of a state income tax (Oregon Legislative Revenue Office, 1999). The tax was collected for one year in 1923 and then repealed by the voters a year later. After three additional initiative petitions and legislative referrals, the voters approved a state income tax in 1930. By 1938, the state property tax was completely offset by the income tax. Except for 1940, no state-level property tax has been collected since.

Over the years, the Oregon income tax code has undergone many revisions and adjustments to interface with the federal tax code and the federal definition of taxable income. For example, the 1997 Oregon legislative session permanently connected the Oregon income tax to future federal changes to taxable income. That same session also adopted an earned income tax credit for low-income taxpayers. The state income tax brackets have been indexed for inflation since 1993. Oregon state income tax rates range from 5 to 9 percent. After deductions and credits the average effective tax rate in 2001 was about 5.5 percent. Importantly, personal income tax collections comprise approximately 85 percent of the state’s general fund revenues.

b. Corporate income taxes - The tax base for a corporate income tax is often defined as corporate net earnings. Refinements to this general base include deductions for capital losses, depreciation of capital investments, and spending on research and development. In practice, corporations can pass the costs of an income tax on to consumers through higher prices, reduced wages to workers, or to reduced dividends to shareholders. The corporate tax structure and corporate tax rates may encourage or discourage corporations from locating in a state or jurisdiction.

According to the Oregon Legislative Revenue Office (1999), the Oregon corporate income tax has a corporate income tax component and an excise component. Corporate net income attributable to Oregon is taxed under a corporate excise tax. Oregon uses the federal definition of corporate taxable income with some modifications as its tax base. A three-factor formula considering property, payroll and sales is used to determine Oregon income.

c. Property taxes - Property taxes rely on a tax base of accumulated wealth. Generally, whenever local governments improve the level and quality of services it provides to citizens (i.e., water/sewer service, transportation, housing, parks, etc.), property values increase. The property tax captures this increased value of wealth.

Property taxes apply to both citizens and corporations. Personal property, monetary or financial assets, and equipment may be subject to property taxes. For citizens, property taxes often apply to the real property of land and improvements, such as homes. In some states a property tax also is applied to personal property, including: automobiles, trailers, and other vehicles, furniture, clothing and jewelry. Corporations often pay property taxes on land, buildings and equipment.

The property tax rate is usually expressed in “mills” or one-tenth of one percent of the real market value or the assessed value. A one mill rate yields $1 dollar of revenue on $1,000 of assessed valuation. Often, property is assessed by several districts for varying services. These “overlapping districts” include county, city, school, and special districts (i.e., fire, hospital, library, water/sewer, vector control, etc.) among others. In most states the county assessor summarizes these multiple assessments into a single tax bill.

County and local governments rely heavily on the property tax. Insulated from economic cycles, the property tax can provide a stable stream of revenue under most economic conditions. Unlike income taxes and sales taxes, decision-makers can precisely set the property tax rate to meet proposed spending requirements, as long as the maximum limit set by law is not exceeded.

But, property taxes are difficult to administer. A property tax system requires assessors, collectors, and administrators. Keeping up with the changing value of taxable wealth is difficult. In many cases, the tax base may not match the actual use or value of the asset. Farmland within an urban area provides a common example. Similarly, identifying financial and other intangible assets may be especially difficult in an age of computerized, global transactions.

One of the earliest forms of taxation in the United States, property taxes are not popular with citizens. Inequities in valuation are common and annual tax bills demand a single, lump sum payment. Payment of property taxes as part of a monthly mortgage payment helps to hide the property tax and to smooth out its burden. Yet, the property tax is generally seen as regressive in its impact. Lower income citizens who do own property and wealth will pay a larger percentage of their available resources in a property tax than citizens in a higher income bracket. Taxable thresholds, exemptions and financial payment strategies are used to help citizens on fixed incomes to comply with property tax assessments. Property used for religious and charitable purposes is commonly exempted from property taxes altogether.

In the 1980’s Oregon imposed a high property tax burden. Recently, a number of ballot measure initiatives have severely reduced this burden and restructured property tax collection in the state. The property tax was first imposed through the Oregon State constitution in 1917. Taxable property includes real property, mobile homes, and some tangible personal property used by business. Under the original property tax system, each jurisdiction adopted a tax levy. The total value of the levy was legally permitted to increase in value by up 6 percent each year. Jurisdictions were left to set their tax rate based on their base levy and some percentage increase that was needed to offset any increases in local spending. Jurisdictions with rapid growth usually approved the full 6 percent increase permissible under the law in order to acquire the resources to support the infrastructure expansion necessary to support the increased growth. But, Measure 5 adopted by Oregon voters in 1990 changed all of this.

Measure 5 set rate limits on individual parcels of property. Property valuation continued to be determined on the basis of real market values, but under this measure, non-school taxes on any property parcel cannot exceed $10 per $1,000; school taxes cannot exceed $5 per $1,000 (Oregon Legislative Revenue Office, 1999). These limits set a total maximum tax payment for each property owner. However, Measure 5 did not impose tax limits on levies backing the sale of bonds.

In November 1996, voters adopted Measure 47 as a second limit on property taxes. Because of confusion over its intent, Measure 47 was clarified by the 1997 Legislature. Voters approved this interpretation as Measure 50 in May 1997. Measure 50 made major changes to the Oregon property tax system by establishing two forms of limits. Measure 50 shifted Oregon property taxes from a levy-based system to a rate-based system. The measure:

· established a fixed, permanent tax rate for government operations. Voters may approve local option levies for 5 years to augment funding for operations. These levies must be approved at a general election or any election at which at least 50 percent of the eligible voter cast ballots.

· established an assessed property value based on the 1995-1996 value less 10 percent. This maximum value may increase by 3 percent annually up to real market value.

· reaffirmed Measure 5, which fixed limits of $10 per $1,000 of assessed valuation for non-school taxes and $5 per $1,000 of assessed valuation for school taxes.

The combined result of these tax initiatives has been to strongly limit property taxes. As recently as 1992-93 Oregon ranked 14th among all states and the District of Columbia in property taxes as a percent of personal income. By 1999-2000, after these initiatives took effect, Oregon ranked 33rd among all states (see Table 8- 5).

d. Sales Tax - Sales taxes derive revenues from consumption. The general sales tax has become a primary vehicle for revenue generation at the state level. Local jurisdictions may piggyback on a state sales tax and define the same set of sales transactions in its tax base. However, the location of the stores that generate sales may be distant from the areas in need of government services. Applying the tax at a state level helps to improve this mismatch, however, commercial centers near state boundaries present special collection problems (i.e., Washington State with a state general sales tax, and Oregon without a sales tax). It is quite easy to escape the effects of a sales tax by simply crossing the border to make purchases. This is much easier than moving your domicile or relocating your employment in order to escape the effects property or income taxes.

In most cases, the general sales tax is levied at a single point in the economic exchange process, usually at the retail level between the retailer and the consumer. In contrast to the lump sum annual payment of a property tax, a general sales tax applies a small incremental rate at each transaction. Once begun, the incremental nature of a general sales tax may make it more acceptable to taxpayers.

The transactions covered by a sales tax vary. In some states all transactions are taxed. In many states food and prescription drugs are exempted from the tax. To compensate for the regressive nature of a general sales tax, state legislatures may provide refunds for low-income residents.

Sales taxes are relatively easy to administer. The burden of collection, however, falls to merchants. Computerized checkout equipment that can quickly identify exempted items has eased implementation of the sales tax. Collecting taxes on the barter and exchange economy is difficult. The emergence of E-Commerce (the practice of selling through the internet) has posed new problems for imposing and collecting sales taxes.

Special purpose sales taxes, or excise taxes, include: gasoline and diesel fuel taxes; tire and air travel excise taxes; and taxes on cigarettes, tobacco products and alcoholic beverages. Oregon was one of the first states to impose a gasoline tax, yet its citizens have consistently rejected a state general sales tax. Voters rejected Measure 7, which would have collected a 5 percent sales tax in 1986. In a second attempt, voters rejected Measure 1 in 1993. Measure 1 would have imposed a 5 percent sales tax, sales tax refunds for low income citizens, sped up Measure 5 property tax relief, set a state spending limit, and shifted state tax revenues to support education.

Most states rely on a combination of income, property and sales taxes to generate a stable, consistent stream of revenues. Local governments rely on property taxes, general sales taxes, fees for services, and intergovernmental transfers to generate a similar balance in revenue flow. Sales and income taxes reflect the fluctuations of the state economy. Based on accumulated wealth, property taxes are relatively insulated from economic cycles.

D. The Oregon Tax Revenue Situation in Context

While the above discussion briefly describes the major taxes used in Oregon, ranking Oregon against national averages provides a better context for assessing Oregon’s overall revenue situation. Table 8-5 below shows the sources of income the state has selected, and how the various tax burdens compare to the national averages. As of 1999-2000 Oregon ranked as follows:

Table 8-5

Oregon’s National Tax Ranking: 1999 - 2000

Sources: http://www.taxadmin.org/fta/rate/99taxbur.html

*http://www.census.gov/govs/statetax/00staxrank.html, May 2001

** http://www.taxadmin.org/fta/rate/tax_stru.html The percentage in parenthesis

represents the share of total state and local revenue collected from this source.

As the comparative state summary demonstrates, Oregon places great reliance on the personal income tax. This reliance provides strong revenues in times of economic growth. With a weakened economy, however, income tax revenues and state revenues may quickly decline. Corporate income taxes rank below the national average. Following the passage of property tax Measure 5 and Measures 47/50 in 1995-1996, reliance on the personal property has now fallen below the national average. However, Oregon’s lack of a sales tax places a greater burden on the remaining forms of taxation and fees imposed for government services.

Without a sales tax Oregon lacks a stabilizing element in its revenue stream. Oregon counties and local jurisdictions face a similar instability. Recent property tax limitations have increased the importance of the state personal income tax, especially for education funding. This shift has increased the susceptibility of school districts to the fluctuations of the state economy. In 1990, the State funded 30 percent of the K-12 school expenditures through the state income tax; property tax covered the remainder. By 1999 state support for schools had risen to 70 percent of school expenditures. With the shift, the funding of K-12 education now competes with programs historically funded at the state level, such as human service programs, public safety and higher education. On the positive, Oregon’s strong reliance on the income tax results in a generally progressive tax climate for low and middle income citizens.

E. Tax Expenditures: Lost Revenues or Purposeful Policy Decision?

The structure and implementation of a tax code at any level of government serves not only to raise the necessary revenues, but also to effect public policy decisions and to influence citizen and corporate behavior. Exceptions to a tax code provide relief from taxes to certain segments of society. Such targeted relief from taxes is known as a tax expenditure. For the recipient beneficiary, not collecting tax revenue is equivalent to a direct government spending for their benefit—hence the name "tax expenditure". Not collecting taxes serves to encourage a particular behavior by corporations or citizens.

Granting a tax expenditure signals a policy decision to either shift the remaining tax burden to non-target groups, or for government to broadly forego revenue. Tax expenditures take a variety of forms. These include: deductions, exclusions, deferrals, preferential tax rates, and tax credits. Some tax expenditures are quickly and widely recognized. For example, the federal income tax mortgage interest deduction on a primary residence also applies to Oregon homeowners (Kitzhaber, 1998). The behavioral incentives in this expenditure serve to encourage home ownership and the stimulation of the banking, real estate, forest products, home construction and home furnishings industry.

Tax expenditures are frequently criticized as a matter of public policy for a combination of the following reasons. First, many expenditures are not widely debated and aid a limited group of beneficiaries. Establishing a tax expenditure may be a well-examined policy decision, but more frequently it is a quiet legislative action that is buried in the arcane details of the tax code. Second, the policy outcomes, impacts and effectiveness of many tax expenditures are rarely reviewed or debated for broad public benefit. Third, tax expenditures tend to erode broad general taxes. Once one group receives relief, other groups desire equal treatment. Fourth, tax expenditures are often liberally granted during times of economic stress without regard for their impacts on government revenues or policy implications.

Beginning in 1995, Oregon joined 32 other states in reporting tax expenditures. The Oregon Budget Accountability Act directed the governor to identify and summarize the impact of all state tax expenditures as part of the preparation of the governor’s biennial budget recommendation. For the 1999-2001 biennium, the governor identified 293 tax expenditures.

Of the 293 total expenditures, 101 were related to local property taxes, and 152 were related to personal and corporate income taxes. The remaining 40 expenditures were related to various state taxes. Importantly, 87 of the 152 income tax expenditures were tied to Oregon’s relationship to the federal tax code. While the connection with the federal tax code may cost the state revenues, it makes citizen and corporate compliance with the state income tax code much easier. State income tax collection and administrative costs are accordingly reduced.

Effectively valuing tax expenditures requires making many assumptions. However, the Governor’s 1999-2001 Tax Expenditure Report indicates that “spending” through tax expenditures over the biennium will total about $19.9 billion. This compares to the roughly $16 billion that the State of Oregon and local tax districts will actually collect in taxes over the same period.

F. Debt Instruments as a Revenue Source

Borrowing funds through the sale of bonds and other financial instruments presents an important source of revenues for state and local governments. While borrowing opens governments to debt liability, long-term financing provides an effective means of funding capital projects and major purchases. In emergency situations, governments can use short-term borrowing to smooth out gaps in cash flows. However, extended and excessive reliance on borrowing reduces the jurisdiction’s credit rating and increases borrowing costs. One of the central principles of public debt management is that governments should not use borrowing to finance routine spending for operations.

The financial community has developed two primary forms of long-term debt financing for the public sector (Lyden and Lindenberg, 1983; and Lynch, 1990): guaranteed debt and “nonguaranteed debt”. Guaranteed debt commonly takes one of two forms. One form consists of “full faith and credit” obligations packaged as “general obligation bonds”, which receive the full backing of all the revenues and resources of a state or local jurisdiction. A second form of guaranteed debt is backed by more narrowly defined revenue sources. These are called “revenue bonds” and are backed by a specific stream of revenues and resources generated by a particular project, program or new facility. Revenue bond debt liability is limited to the particular program and not the revenues and resources of the entire jurisdiction. In contrast to these long-term forms, short-term borrowing is often undertaken with the expectation that the debt will be paid by anticipated tax revenues. This type of borrowing is known as “tax or revenue anticipation borrowing.”

Nonguaranteed debt is financed similar to the way we finance our homes or cars. Instead of backing our loan with either a stream or a fixed pot of revenue, we are charged interest at a rate which represents the amount of risk the lender is taking. This kind of borrowing in the public sector takes the form of bonds, which can be issued either as term or serial bonds. With term bonds, the issuing jurisdiction makes periodic interest payments on a regular basis and must pay off the principal in a lump sum after a set period of years. With serial bonds, the jurisdiction makes periodic interest and principal payments on a regular basis from the time of issue to repayment. Many jurisdictions budget a “debt service fund” or a “sinking fund” to hold payment reserves for, and to organize, debt payments.

Determining the limits of borrowing by a state or jurisdiction is a complex task. Debt is primarily used to finance capital projects and major purchases. Capital projects are usually understood to be relatively expensive, have long-term returns and a permanent, physical presence. Capital investments include buildings and facilities, and large equipment such as fire engines. For some smaller jurisdictions the purchase of police cars would be considered a capital investment. Traditional accounting practices require that the useful life of a purchase should exceed the debt repayment period.

Setting the level of borrowing by the state or a local jurisdiction reflects a balance among the following set of considerations: (1) community needs, (2) repayment ability reflecting recent revenue flows, operating expenses and borrowing, (3) the strength of the economy, (4) the value of the real estate property tax base, (5) public opinion, (6) a willingness to accept higher interest costs because of a reduced credit rating, and (7) the essential purpose of the capital project. The construction of a public budget with a capital investment schedule provides the forum for making conscious decisions on these tradeoffs. The governor or local executive plays a key role in framing this debate by including debt proposals in the recommended budget. The credit rating granted to a state or jurisdiction may reflect conscious decisions by the state governor and legislature, or the executive and the local commission/ council.

Unfortunately, the politics of budgeting often push the categorization of expenses away from operating expenses and routine purchases and toward capital purchases. These purchases can then be covered under the sale of bonds rather than under taxes and other generated revenues. By using this technique, limited general fund resources can be extended without immediate political pain. News of such borrowing ultimately reaches the financial community and affects the state or jurisdiction’s credit rating.

At both the state and local government levels, public budget documents clearly separate capital expenditures and their financing from operating expenditures. This separation reinforces the difference between long-term capital expenditures and short-term operating needs. However, the capital budget should integrate proposed debt into a debt service account. Effective analysis of future facilities operations and maintenance costs should accompany each major capital expenditure. A complete public budget should recognize both the differences of capital expenditures and their integration into other governmental functions. The Oregon Governor’s Recommended Budget dedicates an appendix with detailed listings to capital budgeting, debt management and capital construction. This arrangement clearly separates capital purchases from operating expenses.

In Oregon state government, the State Debt Policy Advisory Commission advises the Governor and the Legislature on how to best manage the state’s debt, borrowing costs, and credit rating. The Office of the State Treasurer manages the state’s borrowing activities. Oregon sets its general obligation borrowing authority based on the True Cash Value of taxable property in the state. As of July 1997 this value totaled $209.2 billion. Based on this value, the state’s total borrowing authority is $30.7 billion. As of June 1998, $2.9 billion, or 9.6 percent of this total borrowing authority had been utilized. Repayment of most of this general obligation debt is tied to specific program revenues with little or no contribution from the state general fund.

Each legislative session determines the volume of bonds and the financing agreements that the state may issue during the coming biennium. State borrowing falls into the following categories:

· General Obligation Bonds supported by the General Fund—Bonds issued under this authority support higher education, forest rehabilitation and reforestation, and pollution control.

· General Obligation Bonds supported by program revenues—Bonds issued under these authorities include: Veterans home loans, higher education building projects, water resources, highways, elderly and disabled housing, and pollution control.

· Revenue Bonds tied directly to program revenues—Bonds issued under this authority include: housing, highway user tax, lottery revenue bonds, mass transit financing, economic development, state scholarship commission, and capital improvements for the state fair.

· Conduit Revenue Bonds—Under this authority, the state issues bonds for local governments and businesses. The local governments and businesses are responsible for repayment and the debt liability. Bonds of this type support economic development, health, housing, education and cultural programs in local jurisdictions.

· Certificates of Participation—Certificates include long term financing agreements for major equipment purchases such as: installment sales agreements, lease purchase agreements, and loan agreements.

At the local level, general obligation bonds are often used for the construction of facilities. Bancroft Improvement Bonds are often sold to finance local improvement districts. The Oregon State Treasurer may provide assistance to Oregon counties and local governments during the preparation and sale of bonds and other debt instruments.

G. Organization of Revenues

Each source of revenue used by a government or jurisdiction brings a different degree of flexibility in its use and in accountability requirements. Following accounting traditions, a public budgeting process must effectively separate revenue sources and revenue uses. This separation into “funds” ensures that spending meets the intent and requirements set by law or by the granting source. The budgeting process and the subsequent accounting during budget execution must support effective reporting on how funds were used.

Public budgeting recognizes seven major types of revenue funds: (1) governmental or general funds; (2) special revenue funds; (3) capital project funds; (4) debt service funds; (5) enterprise funds; (6) internal service funds; and (7) fiduciary funds including trust and agency funds (Lee and Johnson, 1989). Special funds outside of the general fund are sometimes recognized as “off budget” because of their limitations.

A general fund gathers all revenues without spending or accounting restrictions. General funds may be used by the entity for any legal purpose. Because of this flexibility general funds are classified as discretionary funds. General funds provide the funding for large portions of general government operations. General fund revenues are often used to provide the local supplement or dollar matching to leverage revenues in more restricted fund types. Revenues raised through general taxes, fees generated for services provided, and fines often provide unencumbered revenues to the general fund.

In contrast to the wide discretion of general fund revenues, special revenue funds are collected for a specific purpose or program support. These funds carry restrictions on use and accounting that require separation from other resources. These funds are often categorized as dedicated funds. At the federal level the Federal Highway Trust Fund, supported by federal motor fuel taxes, provides an example of a special revenue fund. Similarly, at the state level, revenues from gasoline taxes are dedicated through a special revenue fund for highway construction and repair activities.

Capital project funds are dedicated revenues collected and managed for the construction of buildings and facilities, or for the purchase of major pieces of equipment. Revenues for capital project funds may originate as a transfer from the general fund, or from bond sales paid by general fund revenues.

In a similar use, debt service funds provide the dedicated resources to meet interest and principal payments on general long-term debt. Again, general fund revenues may be transferred to this fund to make payments and to meet accounting needs.

Proprietary or enterprise funds operate in a business-like manner separately from the rest of the government activities and accounting. Enterprise funds separate these funds from other funds and government functions in the public budget. Revenues collected from water and sewer services, public electricity provision, road and bridge tolls, and the sale of timber from state or county lands are often managed under enterprise funds. The use of federal funds may require the establishment of a proprietary fund to ensure dollar matching and effective accounting during spending.

In a related use, internal service funds set up accounts that describe a “business” function within government. For example, a general motor pool, a common equipment pool, human resource services, or public affairs services may operate as an internal business with each operational department paying general fund revenues for services. Payments for these services would be recognized as an inter-fund transfer in a public budget.

Fiduciary funds provide a means to separate unique assets from the general functions of government operations. In using these funds governments act as fiduciary agents for other government programs or agencies, or client groups. At the federal level, the Social Security Trust Fund provides a major example. At the state level, the retirement funds, such as the Oregon Public Employees Retirement System funds, act as a fiduciary fund. In several other examples, revenues from the sale of timber and forage, and rents from the use of state lands, are funneled into a Common School Fund. Interest earned by the Common School Fund is distributed to counties based on the number of school-aged children in the county. Most of the earnings from the Oregon Lottery contribute to special funds for education, economic development, and parks and natural resource programs.

Revenue categorization by funds reflects both the sources of the revenues and the revenue’s potential use in government spending. While many special revenue funds and fiduciary funds exclusively receive revenues from a dedicated source, many funds receive revenues from several sources. A general fund may receive any unencumbered revenues. General funds or lottery revenues often supplement or leverage special revenue funds, capital project funds, and debt service funds. Some federal revenues have dedicated sources, uses and fund designations. Some other federal funds are relatively unencumbered and in practice replace general fund dollars.

The state of Oregon budget recognizes four categories of revenues: the state General Fund, lottery funds, federal funds, and other funds (Oregon Department. of Administrative Services, 2000). Each of the four funds provides varying levels of flexibility in the use of revenues:

· General fund (GF): available for any discretionary use.

· Lottery funds (LF): generally dedicated to economic development and education. Within these two uses, these funds are “fungible” with general fund moneys. Some lottery funds are specifically dedicated to state parks and stream rehabilitation, or to the education endowment fund.

· Other funds (OF): Many other funds are dedicated by law to a specific purpose. In use, however, some other funds may replace (are fungible with) other general fund resources, as in higher education tuition and park user fees. Some federal funds are listed as other funds, and some other funds are really general fund and federal moneys spent for a second time (used in an internal service fund for central services).

· Federal funds (FF): Federal funds generally are dedicated by federal law to specific purposes. These funds may require state dollar matching, or some state “maintenance” to demonstrate a state contribution to the program. Much of these funds (75 percent) support Human Services programs. Within the specific use, some of these federal funds may be fungible with and replace general funds.

H. Forecasting Revenue for Budgeting

We began this chapter with a review of revenue forecasting for budget purposes. After describing the major types of revenues, the chapter described the major forms of tax revenues. Tax expenditures provide a policy tool for encouraging behavior and supporting desired industries and economic development. Public budgeting and accountability require a clear organization of revenues into funds. Special capital budgeting analyses guide the use of borrowing as a revenue generation tool. With this completed overview, we now turn in Chapter 9 to some of the mathematical and statistical techniques used to forecast revenues.

Study Questions

1. Consider the impacts of an inaccurate revenue forecast in the development of a budget for your agency or program. How would your agency respond if the actual revenues generated were 2 percent below forecast levels after the first year of the biennium? How would it respond with a 5 percent shortfall in actual revenues?

2. Budget revenue estimates in effect set the spending levels in a public budget. Revenue forecasts often contain conservative estimates. For your agency or program, consider the added increment of program services the agency could provide with a 3 percent greater revenue forecast. How might you justify the added increment of program and revenues to the legislative Ways and Means Committee?

3. Which revenue sources fund the operations of your agency or program? If you work for an Oregon state agency, what is the blend of General Fund, other and federal revenues that support your program?

4. Does your program generate any revenues? What types of revenues? Are these fees for services? dedicated taxes? sales of products or rents? Are these revenues retained by the agency in a special revenue fund, or returned to the general fund?

5. How does your agency organize its revenues for spending and accounting purposes? Does your agency use any special revenue funds, debt service funds, enterprise funds, internal service funds, or fiduciary funds?

6. Oregonians have consistently refused a state sales tax. How has this shifted reliance on other forms of taxation?

7. Are state and local gasoline taxes a progressive, proportional, or regressive tax?

8. Consider the clients served by your agency or program. Design an income or property tax expenditure that would encourage behavior your agency would desire.

9. Does your agency or program make capital expenditures? Do these purchases meet the traditional test of a capital purchase?

10. Does your agency rely on bonding or another debt instrument to finance these capital purchases? How does this debt compare to annual tax revenues collect in your state or jurisdiction?

197

CHAPTER References and ADDITIONAL REsources

Clotfelter, Charles T. and Philip J. Cook, Julie A. Edell, "State Lotteries at the Turn of

the Century" Report to the National Gambling Impact Study Commission, June 1,

1999. www.pubpol.duke.edu/people/faculty/clotfelter/lottrep.pdf

Kitzhaber, John. (1998). State of Oregon 1999-2001 Tax Expenditure Report. Salem, OR: State of Oregon, Department of Administrative Services. p. 82. The most recent Governor’s budget provides a catalogue of proposed state spending, revenue sources, capital projects and debt. The companion tax expenditure volume catalogues all the tax expenditures in the Oregon tax code. Contact the Oregon Department of Administrative Services for both the budget and expenditure documents.

Lee, Robert D. and Ronald W. Johnson. (1989). Public Budgeting Systems. 4th ed. Rockville, Maryland: Aspen Publishers. pp. 209-210 on Fund Accounting.

Lyden, Fremont, J. and Marc Lindenberg. (1983). Public Budgeting in Theory and Practice. New York: Longman. Especially Chapter 8 on Revenue Systems.

Lynch, Thomas, D. (1990). Public Budgeting in America. 3rd ed. Englewood Cliffs, New Jersey. Especially Chapter 8 on Capital Budgeting and Debt Administration.

Oregon Department of Administrative Services. (2000). “Oregon Budget 101.” Salem, OR: State of Oregon, Department of Administrative Services, Budget and Management Division. An informal two-page summary of the key terms and processes used to build the Governor’s Recommended budget.

Oregon Department of Administrative Services. (1999). “State Finance 101.” Salem, OR: State of Oregon, Department of Administrative Services, State Controller’s Division. A summary of the Oregon state financial systems. Summarizes how revenue sources in the budget are channeled through the State Treasury for program and agency use. Also available at: http://scd.das.state.or.us/lfo.htm

Oregon Legislative Revenue Office. (1999). Basic Tax Packet. Salem, OR: State of Oregon, Legislative Revenue Office. p. C-1. A summary of the Oregon state tax system. Provides concise summaries of the major taxes that contribute revenue to the Oregon state government. Prepared as a backgrounder for state legislators. Contact the Legislative Revenue Office in the state capital building to obtain the most recent version. The Legislative Revenue Office prepares and offers a wide range of analyses on tax policy and the impacts of ballot measures.

Pruett, Joseph F. and Ernest Enke. (1983). “Administering Revenue Systems” in Handbook of Public Budgeting, edited by Thomas Lynch, and Jack ?? New York: Marcel Dekker, Inc.

Schuster, Ervin, G. and Krista M. Gebert. (2001). “Property Tax Equivalency on Federal Resource Management Lands” Journal of Forestry, Vol. 99, No. 5, pp. 30-35.

U.S. Dept. of Commerce. (1995). “Government Finances 1992-1993” and “Government Finances 1995-1996”. Washington DC: U.S. Department of Commerce, Census Bureau. 1992-1993 statistics as formatted by the Center for Budget and Policy Priorities (CBPP). Washington DC (1996). 1995-1996 statistics as formatted by the Oregon Legislative Revenue Office (1999) above.

Washington County, Oregon. (1997). Proposed Budget FY 1997-1998. Hillsboro, Oregon. pp. 310 and glossary.

199

Chapter 9*

Revenue Forecasting Techniques With Oregon Examples

Revenue forecasting forms the foundation of a public budget. The identification of the types, availability, and timing of revenue resources dictates the levels of spending possible in a budget. Revenues generated by taxation, licenses and fees for service, and lottery games require accurate forecasting techniques, most of which are grounded in statistics and economics. The application of these quantitative techniques is the primary topic of this chapter.

Even though revenue estimation relies heavily on quantitative, statistical and economic techniques, there are two caveats that need to be mentioned before discussing these in more detail. First, sophisticated techniques are less important than building trust in the process. We have already discussed the importance of creating a revenue forecasting process that has the confidence of the participants (pp. 157 ff.). With the large number of participants with different priorities, it is unlikely that the process will produce an easy consensus. While this may occur, it can't be assumed. Like any contest where there are likely to be deep differences of opinion, it is important that the parties enter into the debate with some agreement on the rules of the game. In this case, agreement on the size of the pot of money to be divided up is terribly important before debate begins on how the division should be made.

This trust-building dimension to revenue forecasting is especially the case when the techniques for assembling revenue estimates are partly science and partly artful judgment. The techniques that may feel familiar and solid to the revenue forecaster may easily appear as a mysterious “black box” to the general public and elected officials. Revenue forecasters must therefore make a special effort to translate their approaches and processes into terms that are easily understood and trusted by the various attentive publics who get involved in the process as it moves through the various stages of review and finalization. Along the way, program managers, budget analysts, elected officials, and organized interest groups will bring different expectations and levels of

sophistication to the work that revenue forecasters undertake.

200

* Kent Robinson took responsibility for preparing the initial draft of this chapter based on material initially written by Drew Barden, City of Portland Economist.

A second caveat to the technical-centered nature of revenue estimation is a recognition that there are too many revenue sources and too little time and money to be equally thorough and systematic in making estimates for each revenue source. Even small jurisdictions must track several sources of tax revenues, intergovernmental transfer revenues, revenues from fees for service, licensing revenues, and opening balances. Each source of revenue represents a different level of importance to the overall revenue picture. Accordingly, a key concept of revenue forecasting is to focus time and attention on those revenue sources with the greatest contribution and impact to the budget.

Primary sources of revenues, such as property taxes for local governments, or income taxes for state governments, demand the greatest attention and the greatest rigor in forecasting techniques. More minor sources of revenue, especially those with little variation from year to year deserve adequate, but limited attention. Forecasters should not spend extensive time or develop complex techniques on very minor revenue sources with little historical data, or with little annual variation.

Effective revenue forecasting is both an art and a science. The art of forecasting reflects the application of judgment, both in the selection of forecasting techniques and tools, and in the selection of a forecast result that can provide the basis for political agreement. As noted above, the selection of forecasting tools and techniques reflects the importance of the revenue source, the time available, the data available, and the variation in the data.

Even with ample time and data, a forecaster can choose among several forecasting techniques. Each technique will yield a slightly different revenue estimate. The forecaster’s sense of context, understanding of the larger economy and revenue situation, and intuition supports his or her selection of an estimate. And, where a simpler, more easily understood technique provides acceptable results, building credibility may call for its selection and use. Once a technique has been chosen, an estimate has been made and a supporting rationale developed, forecasters can sometimes be placed under pressure by various political forces to revise estimates to meet the needs of the moment. For example, David Stockman, Director of the United States Office of Management and Budget during the Reagan administration, was placed under considerable pressure by the President to rerun computer estimates with more optimistic assumptions about revenue. This pressure was the result of the need to keep political promises to multiple constituency groups (Greider, 1981). While this is a dramatic and rare occurrence, officials are elected to keep promises to their constituents. Since there are usually far more collective promises in the political arena than revenue to support them, there is always pressure on those responsible for the revenue stream to “squeeze blood from turnips”.

In addition to working against the constant political backdrop of “finding additional revenue”, most analysts operate in a world of process deadlines, special last minute requests, and heavy workloads. This pressure often leads analysts to quickly extrapolate from past comprehensive forecasts and to present quick-and-dirty “good enough” estimates. The luxury of a complete formal analysis with full data is a distant ideal. Yet, in some cases, these quick, but knowledgeable estimates may provide sufficient information for political debates. Here, the relative or magnitude differences in revenue estimates may be much more important than the absolute level of estimates themselves.

In the end, the final selection of a revenue estimation may fall across a range of potential alternatives. Selecting the most conservative estimate may under-estimate revenues. Selecting more conservative estimates also opens the estimates to political scrutiny and mistrust. Erring in the opposite direction of overly-optimistic revenue estimates leads to political agreements built on extreme optimism. Optimistic forecasts may lead to belt-tightening later. The most useful estimate may be the “most likely” estimate. Most likely, or baseline, revenue estimates are neither optimistic nor pessimistic. Ideally, the actual collected or transferred revenues come in at, or slightly above, the most likely estimate. Most likely estimates may, in fact, be slightly pessimistic (Barden, 2000).

Thus, the main task of the forecast group is to generate a set of numbers that work. These numbers get the budget and the agency through the immediate fiscal year or biennium without program cutbacks. Additionally, the revenue forecast should also not create financial problems for the agency in the out-years. Optimistic revenue estimates for the current year should not result in reduced revenues in future years. Front loading revenue support only leads to reduced levels of services and dashed expectations in future years. The most effective revenue estimates are multi-year. For local governments multi-year may equal five years. For jurisdictions like the state of Oregon which operate on a biennial budget cycle, revenue estimates may extend two or three bienniums into the future.

Several additional principles help to inform the judgment of analysts in the estimation and selection of revenue estimates.

· A forecast needs to be generated with a systematic and coherent set of underlying assumptions and conditions so that it is reproducible. That is, the same initial conditions and assumptions always produce the same results.

· For all but the newest revenue sources, a previous forecast should always exist. The previous forecast might be the second year estimate of last year’s five-year forecast. The new forecast should always be compared to the old forecast so that elected officials and managers can see how and why resource estimates have changed.

· All revenue estimates need to be rationally developed out of past history; this is particularly true for a revenue estimate for a new revenue source.

· Initial conditions are extremely important. Next year’s forecast needs to be continuously conditioned by both the current year and history. Any forecast should be accompanied by a history, estimates for the current year, and a multi-year resource forecast.

· Forecast results must be presented in a coherent document that clearly conveys the estimates to elected officials and managers. The forecaster needs to develop a core set of tables, text discussion, and graphics that appear in regular format, even as the content changes.

This last point stresses the need for analysts and forecasters to effectively explain and communicate their forecast results, assumptions, limitations, and processes. In the heat of debate elected officials will often leave the assumptions and limitations behind as confusing details. Briefing materials should take this behavior into account and package assumptions and limitations in concise easily presented concepts. In communicating the results and details of a revenue forecast, “less is more.”

A. Revenue Forecasting Techniques

Revenue forecasting techniques range from the simplest use of educated estimation to the application of complex systems of statistical and mathematical models. The application of a forecasting technique reflects the importance of the revenue source, the time available, the data available, and the variation in the data that a technique must describe or model.

This handbook section provides a brief summary of the range of revenue forecasting techniques. The intent of this overview is not to teach the technique, but to develop an appreciation for the variety and utility of some of the more commonly used approaches to revenue forecasting. The discussion of each technique provides an overview of the approach, and then gives examples from local governments in Oregon, or from the state revenue forecasting process. The techniques are loosely arrayed from the most simple to the more complex.

· Educated estimation;

· Different forms of averaging;

· Trending methods;

· Simple regression analysis;

· Systems of regression equations and complex modeling;

· Deterministic spreadsheets.

1. Educated estimation. Educated estimation may provide the most appropriate estimation technique for minor revenue sources. The approach is especially applicable when revenue levels vary little from year to year, and when the forecaster is knowledgeable of the history and context for the source. In another appropriate application, the forecaster may have knowledge of a one-time, isolated spike in a normally stable flow of revenues. An analyst may be forced into an educated estimate when historical data are very limited or unavailable.

In developing an educated estimate a forecaster considers recent revenue levels and any changes that would affect the revenue stream. Changes affecting the future revenue stream might include, demographic trends, revisions to the state or federal tax regulations, changes in tax enforcement, or economic cycles. Importantly, forecasters using educated estimation should carefully document their assumptions. Accumulating estimates with comparable assumptions through time allows the development of a rudimentary data set. This data set can ultimately be used to support the forecasting of future estimates, and the ready development of a forecast using more refined techniques.

Examples of revenue forecasts that use educated estimates appear in the City of Portland and State of Oregon forecasts. The City of Portland uses educated estimation to forecast the money recovered annually through audits, and to forecast a small revenue category of Miscellaneous Revenue Refunds. The State of Oregon forecasters use educated estimates to forecast annual revenues from the state inheritance tax (Parker, 2000).

2. Different forms of averaging. When a data set of historical revenue information shows a single steady visible trend, or shows constant explainable variation, averaging may provide the most appropriate forecasting tool. As with educated estimation, averaging is appropriate for relatively minor revenue streams. In developing a forecast using averaging, an analyst should consider any changes in the legal, economic, demographic or organizational environment that could produce changes in past trends or introduce new variation into the forecast estimate.

A forecaster using revenue averaging can follow several strategies. Where no trend appears in the data, the familiar arithmetic average of all available past per period revenue levels is normally used. If a simple trend appears in the historic revenue data, a rolling average that uses only the most recent two or three data points may provide an effective estimate. However, computing an estimate using two, three, or four rolling data points may result in a prediction above or below the trend in the data. The analyst needs to exercise judgment in selecting the number of data points to include in the rolling averaging estimate. Many local jurisdictions use averaging to forecast in-lieu property tax revenues and the catchall category of Miscellaneous Revenues.

3. Trending methods. Where historical data of past actual revenues and the forecasted estimates for those revenues are available, a trending technique, such as exponential smoothing, may provide an appropriate forecast estimate (see Ullmann, 1976). If sufficient variation appears in the actual revenues, a trend becomes difficult to identify and the technique has limited effectiveness. The technique is appropriate for data with limited variation and without spikes.

The trending technique of exponential smoothing estimates the future revenue for the next year based on a relationship between the actual revenues for past years and their comparable forecast estimates. A weighting factor between 0 and 1 sets the relative balance between the actual and forecast estimates in the future prediction. While computer packages can perform the calculations to select the weighting factor and to make the future estimation, the analyst must carefully review the selection for the weighting factor. Too low a weighting factor dampens the effect of the actual revenues, while too high a weighting limits the value of past estimates.

Local jurisdictions frequently rely on trending techniques to forecast telephone utility license fees and state revenue sharing for cigarettes and alcoholic beverages.

4. Linear regression analysis. The next step up in complexity of forecasting techniques is the statistical technique of linear regression analysis. Linear regression uses historical data to first build a relationship between past revenue levels and one or more economic, social, or demographic explanatory factors. A regression data set usually pairs the level of actual revenue collections with numerical data on each different explanatory factor. The data is often collected on an annual, quarterly or monthly basis. Using the computed regression relationship, and recent levels of explanatory factors, forecasters can then estimate an expected future revenue level. Using linear regression techniques to forecast future revenues assumes that the conditions affecting the explanatory factors, and their relationship to revenue generation remain unchanged from the past. The regression relationship is expressed as a mathematical equation. Each explanatory factor serves as a variable in the regression equation.

In contrast to the relatively simple data requirements of the previous techniques, linear regression has more comprehensive data needs. Gathering the wide variety of data measures and indicators from which to select out a final set of explanatory factors is often an expensive and a long-term process. Buying sets of data sometimes provides an alternative to the data collection process. Once the data is available, building a refined regression model to predict revenues requires professional abilities and a computerized statistical package. As with the other forecasting techniques, the amount of time and resources invested in the estimation effort should be proportional to the importance of the revenue source. Even with a long-term, comprehensive data set and a high level of talent, regression models may generate a revenue estimate of only limited quality. When this occurs a critical explanatory factor has not been identified, or is missing from the regression equation.

Linear regression models consisting of one relationship (one regression equation) find many applications in revenue forecasting. The State of Oregon uses single equation regression models to predict revenues from various Oregon lottery games. Reflecting the historic data that influence lottery ticket sales, future revenues from a game may be estimated. Important explanatory factors that affect lottery ticket sales and revenues include variables considered “controlled” by the Lottery, as well as impacts associated with seasonality. Lottery controlled variables include promotions, advertising campaigns, game changes, accounting changes, and others. Seasonal variables can include week of the month, month or season of the year, and even certain holiday related relationships.

In a second example, the City of Portland uses a regression equation model to forecast annual revenues from business licenses. This revenue source is too large and important to rely on trending. It is also highly affected by the business cycle and the national economy. After an analysis and selection of explanatory variables, the final regression equation contains three variables to explain revenues: the annual unemployment rate for the Portland metropolitan area, national corporate profits, and the license revenues from the preceding year. Including the level of revenue collected in the previous year as an additional explanatory variable may greatly improve an estimate when the other explanatory factors are weak.

5. Systems of regression equations and complex modeling. As budget revenue sources grow in terms of absolute size and contribution, the application of more complex forecasting techniques may become justified. In an increasing step in forecasting complexity, several linear regression equations may be linked together into a system to forecast a revenue stream. The most important revenue sources, such as the state personal and corporate income taxes, demand these more sophisticated forms of forecasting techniques. At this level, complex computerized models that can describe the state’s entire economy become the most effective forecasting tool. This type of forecasting model must incorporate key information from regional and national economic conditions, as well as the specific data related to Oregon tax collections.

Moving toward increasingly complex techniques implies a commitment on the part of the agency or government to ensure high levels of credibility in its revenue estimates. Regardless of how sophisticated the model is, without clear presentations, constant communication, and an open process, public trust in revenue forecasts may be quickly lost. In short, forecasters and budget process managers must work aggressively to prevent the attentive public from treating the projection as a mysterious “black box”.

The data needs for complex forecasting models grow in relative size and complexity with the models. The cost of purchasing or gathering the necessary data and historical information may become a major expense. Similarly, building, maintaining, operating, and interpreting the output from complex forecasting systems takes an extremely high level of professional ability. Most states and large cities reserve a professional analyst position or even a small staff of professionals for the revenue-forecasting task. Individual operating agencies may dedicate an analyst position if the size and importance of a revenue stream is sufficient.

Effectively interpreting the forecast estimates from a complex system requires considering both the estimates generated and the error that surrounds the estimates. In many cases the error that surrounds an estimate is a product of the statistical and mathematical procedures used in the estimation process. The error surrounding an estimate provides a range of probability and a quality context for the estimate.

The error may originate from several sources: (1) a variety of structural assumptions built into the model; (2) the interaction of several statistical estimates and regression equations; or (3) the variation in the data used to run the model. In the case of a forecast from a single linear regression equation, statistics from the regression calculation will provide error and quality measurements. In systems of several regression equations, and in larger complex systems, sensitivity analysis procedures are used to develop a range of possible revenue estimates under different data conditions and structural assumptions. Professional review of both forecast estimates and any sensitivity or error analysis is critical to ensuring public trust in the forecasting effort.

The application of systems of regression equations and complex modeling in revenue forecasting are many. In a relatively simple example, the revenue estimation model for the Oregon Lottery Megabucks game links together two regression equations and the effect of a statistical distribution. The two regression equations use explanatory variables such as: jackpot size, the jackpot size of other lottery games, the number of tickets sold, the day of the week, and the month of the year. The statistical distribution captures that random probability of purchasing a ticket with a winning number. With continued refinement, the model for Megabucks has become very effective in predicting expected revenues (Oregon Office of Economic Analysis, 1998). In contrast to this relatively simple application, the model used to forecast state personal and income tax revenues uses more than one hundred equations. The major elements of this model are described in the following section B, "Revenue Forecasting Techniques Used in the State of Oregon".

6. Deterministic spreadsheet. In contrast to the cyclic or random nature of many revenue streams, the property tax system in Oregon is highly defined by statute. For example, Measure 50 (see Chapter 8, pp. 170 ff.) established a fixed permanent tax rate for each jurisdiction and limited growth on assessed values to three percent per year. New construction is assessed at a ratio of assessed to real market value for the area where construction is located. As a final definition, Measure 50 limited the total local government assessment to $10 per $1,000 of value multiplied by the property’s real market value. This assessment limit excluded school districts as part of the tax limit formula. With these clear definitions, effective spreadsheet relationships can be developed. This accuracy is very important for local jurisdictions that rely so heavily on the property tax as the major source of income (i.e., 51% for the City of Portland, excluding federal and state grants).

Summary: Selecting a Revenue Estimate. This section of the chapter has outlined the major methods for forecasting current year and future year revenues. Each technique has data needs, and brings strengths and weaknesses to a forecast estimate. Selecting the appropriate tool, the appropriate assumptions, and the most appropriate estimate are left to the forecaster’s judgment. The use of judgment is really the art of revenue forecasting. A revenue estimate must have a sound technical basis, be repeatable under similar assumptions, and gain the support of elected officials.

B. Revenue Forecasting Techniques Used by the State of Oregon.

The State of Oregon categorizes its budget revenue sources into a General Fund, Other Funds, and Federal Funds. Forecasting the revenues for each fund is critical to the development of the Governors Recommended Budget and the Legislatively Approved Budget. Forecasting the revenue levels for the General Fund falls mainly to the State Economist and the staff of the Office of Economic Analysis in the Department of Administrative Services. This process is summarized in Table 9-1 on the next page (Gates, 1999). Some of the forecasts for general fund revenue is completed directly by the relevant agencies. These include: liquor apportionment by the Liquor Control Commission, Insurance Taxes by the Department of Consumer and Business Services, State Court Fees by the Judicial Branch, etc.

Forecasting the revenues for the many different sources of Other Funds falls to the agency responsible for the particular revenue program. For example, the Oregon Department of Transportation is responsible for the forecasting related to the motor fuels tax and the truck weight/ mile tax. Similarly, forecasting the expected levels of federal funding potentially available to the state is the responsibility of each recipient state agency. For detailed information on the forecast techniques used for each “Other Fund” revenue source, contact the Office of Economic Analysis or the responsible agency.

Forecasting General Fund revenues requires the Office of Economic Analysis to estimate revenues from several taxes, licenses and fees, charges for service, sales income, interest income, and other miscellaneous sources. Of this list, the personal and corporate income taxes are by far the most important sources of revenue. Personal

210

Figure 9-1

Economic and Revenue Models[footnoteRef:1] [1: Gates, S. 1999. Oregon revenue models annual review of methodology. State of Oregon, Department of Administrative Services, Office of Economic Analysis. p. 1, figure 1. ]

Governors Council of Economic Advisors

Reviews Process and Model

DRI U.S. Economic Model

U.S. Economic

and History Forecast

Oregon Preliminary Economic Model

Oregon Economic

History

Review by:

Governors Council of

Economic Advisors and

DAS Advisory Committee

Oregon Final Economic Model

Oregon Oregon

Personal Income Tax Models

Corporate Income Tax Models

Personal Corporate

Income Tax Income Tax

History History

Forecaster Judgment

Legislative & Policy Changes

Final Revenue Estimate

211

income tax revenues account for about 85 percent of all General Fund revenues. Corporate income tax revenues account for about 7 percent of all General Fund revenues. The Office of Economic Analysis also forecasts revenues transferred to the state Economic Development Fund from the Oregon Lottery.

To forecast General Fund revenues from personal and corporate income tax revenues, the Office of Economic Analysis relies on a complex economic model of the Oregon state economy. Using the model, the Office of Economic Analysis prepares an economic and revenue forecast on a quarterly basis. Updated forecasts are available in March, June, September, and December (Oregon Department of Administrative Services, 2000). The timing of these forecasts becomes critical in years of budget preparation and legislative review. In years with legislative sessions, the March and

June (released on May 15th) forecasts set the revenue estimates from which the legislature builds its Adopted Budget.

Importantly, preparation of the quarterly forecasts is an open process with peer review at key decision points by the Governor’s Council of Economic Advisors. An internal Department of Administrative Services advisory committee also reviews the forecasting procedures at key junctures. The legislative branch is fully represented in the process through the participation of the Legislative Revenue Office.

The Oregon economic model was developed in 1980 by State Economist Chang Mook Sohn and several consultants from the Data Resources Incorporated (DRI) consulting firm. The model has been updated through the years as the state’s economy, and tax structure and code have changed, but the general structure has remained constant. The model currently relies on 68 equations and economic identities. The Oregon revenue models translate the state economic forecast into personal and corporate income tax revenues (Griffiths and Warner, 1996).

The forecast of income tax revenues begins with the outputs of a national economic model (see Figure 9-1 on previous page). Oregon currently uses a DRI model of the national economy but reviews this model every several years. The Governors Council of Economic Advisors reviews the assumptions built into the national model and the selection of its possible output scenarios. An output of a national economic activity baseline, or possibly a selected alternative scenario of economic activity is selected to feed the Oregon economic model. This Oregon-specific model attempts to describe the state’s economy in detail. The DAS Advisory Committee and the Governor's Council review the results of a preliminary run of the Oregon economic model. After review of the data, assumptions, error ranges, and alternative possible economic scenarios, a final output scenario from the Oregon economic model is developed and selected.

With a model of the recent Oregon economy developed, economists can then forecast income tax revenues. Two income tax models, one for personal income, the other for corporate income, translate the model of the Oregon economy into income tax revenues. Reflecting very new economic signals and trends, forecasters may then exercise personal judgment to revise the income tax revenue estimates. Finally, the estimates are revised to reflect any recent policy or legislative changes in order to produce a final set of estimates. For the 1999-2001 biennium, forecasts estimated personal income tax revenues of over $8.4 billion. For the same period, corporate income tax revenues were forecast at about $837 million (Kitzhaber, 1998).

The major components to the income tax models include: a personal income distribution model, a tax calculator, an individual tax liability model, a personal collections model, a corporate liability model, and a corporate collections model (Gates, 1999). Importantly, it takes three years to gather the data necessary to model a just-closed tax year. Further, the forecasters can only make assumptions about very recent tax law and policy changes. Collecting the data to confirm or disprove these assumptions takes several years following the change in tax regulation or statute.

Revenues from the Oregon Lottery provide the third major category of state revenues. In contrast to the complex economic models used to forecast income tax revenues, forecasting lottery revenues relies on a much simpler set of models (Oregon Office of Economic Analysis, 1998). Economists usually forecast the revenues expected from each lottery game using linear regression techniques. For some games, the regression equations are coupled with probability techniques to capture the random aspect of the lottery. For the 1999-2001 biennium, forecasters estimated lottery revenues transfers of $580.4 million (Legislative Adopted Budget). While lottery funds are allocated and earmarked for education, natural resources and economic development purposes, they can, and are used, to replace state General Funds.

The motor fuels tax and the weight mile taxes on trucks provide the fourth major source of revenues to the state. The fuels tax and the weight mile tax revenues are treated as “dedicated funds” and thereby designated to be used for highway maintenance and construction. However, these taxes are considered as Other Funds in the state budget. For the 1999-2001 biennium, forecasters with the Oregon Department of Transportation (ODOT) forecast motor fuels tax revenues (gasoline and diesel) of about $812 million (Kitzhaber, 1998). The department also forecast the revenues from the weight/mile tax on heavy trucks at about $443 million.

ODOT forecasters rely on a complex modeling system to estimate future motor fuel and weight/ mile tax revenues (Malik, 2000). The model relies on inputs of historic and recent transportation data, data on the Oregon economy, Oregon population and demographic changes, and recent forecasts of national economic activity. Real personal income is also an important predictive factor. The modeling system uses about 220 explanatory factors (independent variables) to estimate tax revenues. Seasonal influences are also considered. The assumptions underlying the system conform to the assumptions used by the State Economist to forecast the General Fund revenues. As an example, the system employs a regression equation to relate explanatory factors such as:

· the fuel efficiency of cars

· the price of gasoline and general Oregon prices

· employment participation

· Oregon population

· change in Oregon personal income in relationship to motor vehicle fuel consumption in the state.