Chapter 22
Performance Budgeting
Alfred Tat-Kei Ho
Many countries around the world, both developed and developing economies, have some form of performance budgeting (Andrews, 2005; Curristine, 2005; Organization for Economic Cooperation and Development, 2007; 2013; Gupta, 2010; Raudla, 2012). Advocates of these reforms suggest that performance budgeting can help enhance public accountability and effectiveness of governmental programs and spending. This trend signals that many policymakers in different countries want to use budgets not only as a tool to control and allocate resources, but also as a platform to support strategic planning, enhance agency performance, and strengthen public accountability. Nonetheless, enthusiasts of performance budgeting reform in many parts of the world, even in some developed economies that have been trying to use the tool for decades, often face skepticism and disappointing results. Many studies have shown that policymakers often ignore performance information when they make appropriation decisions and that other considerations, such as partisan politics, priorities of elected officials, and bureaucratic routines, still dictate budgetary decision making (Andrews, 2005; Bourdeaux, 2008a; Gilmour & Lewis, 2006; Joyce, 2008; Lauth, 1987; Raudla, 2012; Schick, 1990; Willoughby & Melkers, 2000).
The primary goals of the chapter are to reexamine the dilemma that performance budgeting reformers face and suggest how reform goals and strategies can be established more realistically in different institutional contexts. The chapter offers a brief review of performance budgeting practices by different countries, discusses the factors that influence the adoption of these practices and the key challenges different countries face, and considers the implications for the goals and implementation strategies of performance budgeting systems.
Performance Budgeting Practices around the World
The idea of integrating departmental and program performance information with budgeting and resource allocation is not recent. In the United States, the federal government's effort to integrate spending decisions with performance information dates back to the 1950s, and many subsequent reforms, such as the Planning-Programming-Budgeting System in the late 1960s, the zero-base budgeting reform in the 1970s, and Management by Objectives in the 1980s, were introduced to make budgetary considerations more rational and performance oriented (Kelly, 2003; Schick, 1966; US General Accounting Office, 1997). Many US state and local governments also began to experiment with performance budgeting concurrent with federal reforms. For example, Sunnyvale, California, and Milwaukee, Wisconsin, were performance budgeting pioneers in the 1970s (Hatry, 2006). In the United Kingdom, selected departments also adopted program performance budgeting in the 1970s (Rose, 2003).
In the early years, reforms were primarily spearheaded by managerial interests, and performance measurement targeted mostly cost-efficiency, output, and workload concerns (Sterck & Scheers, 2006). However, policymakers and the general public tended not to be highly interested in operational issues, and traditional performance budgeting did not get much legislative and public interest (US General Accounting Office, 1997). In the 1990s and early 2000s, however, a new wave of performance budgeting reforms diffused through government. Influenced by the “reinventing government” movement (Osborne & Gaebler, 1992), the focus of new performance budgeting reforms shifted to outcomes and results about which major stakeholders were more concerned. For example, Canada introduced outcome-oriented management in 1994 and required departments to submit plans and spending priorities to Parliament in the budgetary process. New Zealand also shifted its performance budgeting exercise from the traditional output and operational focus to a strategic, outcome-oriented focus in 2001 by requiring departments to develop statements of intent and outcome and output goals (Baehler, 2003). Similar shifts to results-oriented budgeting were pursued by the United Kingdom and the Netherlands in the late 1990s and by France and Japan in the early 2000s (Rose, 2003). In the United States, the Performance Assessment Rating Tool (PART), established by the George W. Bush administration, had a similar emphasis on outcome-oriented planning and results demonstration (Posner & Fantone, 2007). At the subnational levels, many states, municipalities, and counties in the United States also adopted similar results-oriented budgeting reforms in the 1990s and 2000s, and many professional organizations, including the Government Finance Officers Association (1998), the Governmental Accounting Standards Board (1994), and a consortium of state and local professional associations have advocated strongly for these reforms (National Performance Management Advisory Commission, 2010).
The practice of results-oriented budgeting has even spread among developing countries. Despite these countries' administrative capacity constraints and cultural and political context, attempts to integrate performance information into the budgetary process have been introduced with the hope that resource allocation decisions can become more rational and public programs can be managed more cost-effectively and accountably. For example, local governments in China have begun many performance-based budgeting pilot reforms since 2003 (Niu, Ho, & Ma, 2006; Chen, 2011). In 2003, Colombia also tried to link performance information with government spending and national policy objectives more closely through the introduction of its Results-Based Investment Budget initiative (Castro, 2009), and in 2005, India launched a similar initiative of outcome-oriented budgeting, trying to tie government spending more closely to policy outcomes that the public was concerned about (Government of India, 2005). Late in the first decade of the new century, the Budget Directorate of the Ministry of Finance in Chile also introduced performance budgeting reform, requiring agencies to specify their mission, objectives, strategic products, users, and beneficiaries, as well as performance targets and indicators (Hawkesworth, Melchor, & Robinson, 2013).
Despite the global momentum toward performance budgeting or outcome-oriented budgeting, how the idea is actually adapted and implemented may differ according to institutional and cultural contexts. Based on a framework suggested by the Organization for Economic Cooperation and Development (2007; Curristine, 2005), the worldwide practices of performance budgeting can be categorized into three types: presentation performance budgeting, performance-informed budgeting, and direct performance budgeting. Under presentational performance budgeting, performance information, such as performance goals and program measures, is integrated into the official budgetary documents and presented to major stakeholders. However, it is not a significant input in the budgetary process. Under performance-informed budgeting, performance information is not only presented but also integrated more systematically into the budgetary process, from budget preparation and appropriation to budget execution. Of course, stakeholders and participants in the process generally consider performance information an important input, but it is not the only factor used to determine funding and resource allocation. Finally, under direct performance budgeting, performance results and budgetary decisions are directly linked, and appropriations or program allocation are based on results accomplished. Formal evaluation or formula-driven allocation is used to link performance measurement and budgeting systematically, and clear incentives are provided to make departments and programs more performance conscious.
This framework is very helpful for understanding the practice of performance budgeting, especially when one tries to look beyond the rhetoric of “results-oriented budgeting” or “budgeting by outcomes” in different countries' contexts. It should also be noted that within any particular government, all three types of performance budgeting can be used, depending on the goals and institutional context of the reform, the program types, and the stages of the budgetary process. For example, in the budget preparatory stage, the budget agency of the executive branch may use performance-informed budgeting to conduct conversations with departments about their budget needs and program accomplishments. In the appropriation stage, departments may rely on presentational performance budgeting for some programs and performance-informed budgeting for others if certain legislative committees are interested in knowing how tax money has been used to accomplish policy results. For some programs, direct performance budgeting may also be mandated by past legislation requiring the legislature or specific departments to allocate a certain level of funding to programs based on past program results and anticipated future needs. Once funding is appropriated, program managers within each department may also use different types of performance budgeting to allocate the appropriated amount of resources to subprograms. Hence, there is no one-size-fits-all approach to performance budgeting. The reality of performance budgeting may be a mosaic of practices among programs and departments, and the picture may evolve over time, depending on the changing institutional contexts of budgeting.
Institutional and Organizational Constraints on Performance Budgeting Practices
If performance budgeting practices are diverse and constantly evolving, what are the key factors influencing the change and adoption of practices? Past studies have pointed out at least four factors that should be considered carefully: the capacity constraints of the government, the role of leadership, the degree of politicization and stakeholder involvement, and the program type of spending.
Capacity Constraints
The administrative capacity of the budget office and departments often dictates what performance budgeting can accomplish realistically. To make performance budgeting work, the budget office and departments need the capacity not only to measure and report performance but also to analyze and use the information effectively. Studies have shown that insufficient staff capacity to evaluate programs (US General Accounting Office, 2003); inability to develop and keep track of valid and reliable performance measures, especially cost-efficiency and outcome measures (Breul, 2007; Ho & Ni, 2005; Jordan & Hackbart, 2005; Melkers & Willoughby, 2005; Hatry, 2006; Willoughby, 2004; US General Accounting Office, 2003); and inadequate information technology and data management support (Dull, 2006; Jordan & Hackbart, 2005; McNab & Melese, 2003) are some of the major hurdles to performance budgeting reform. If these barriers exist in the United States and other developed countries that have tried to implement performance budgeting for decades, the challenges can be expected to be even more severe among developing countries that have just begun to initiate these reforms.
Need for Leadership Support
Strong leadership support may help overcome some of the capacity constraints in performance budgeting. Studies have shown that if elected officials, especially the chief executive such as the governor, the mayor, or the president, are interested in using performance information or even in championing the reform, they will rally the necessary political support to help agencies overcome some of the resource barriers (Ho, 2003; Dull, 2006). Also, with their support, performance information is more likely to be taken seriously by agencies and is more likely to have an impact on budgetary decision making, communication between the executive branch and the legislature, or communication between the budget office and agencies (Ho, 2006; Melkers & Willoughby, 2001 2005). In addition, the experience of the Government Performance and Results Act of the US federal government and many state reform experiences in the United States show that if the pressure of legislative oversight is strong, particularly if the legislature mandates the presentation of performance information in the budgetary process, the executive branch is more likely to take performance budgeting seriously (Bourdeaux & Chikoto, 2008; Melkers & Willoughby, 2001 2005; Yi, Willoughby, & Arnett, 2009).
Barriers Created by the Political Environment
At the same time, the legislative political environment may create barriers to the functioning of performance budgeting. For example, if there is significant distrust between the legislature and the executive branch, legislators are unlikely to trust the performance information provided by agencies. As a result, performance budgeting is likely to be presentational only and is unlikely to have any effect on the legislature's appropriation considerations (Dull, 2006; Gilmour & Lewis, 2006; Ho, 2003; Moynihan & Andrews, 2010; Raudla, 2012).
In response to this politicization problem, some observers of performance budgeting reforms suggest that it is important to engage elected officials early in the design process to get their buy-in, especially those in the legislature who are responsible for appropriation and program oversight (Breul, 2007; Posner & Fantone, 2007). However, legislative involvement, though desirable and important, often runs into two potential problems. First, politicians themselves sometimes hesitate to define program outcomes and performance expectations transparently and clearly. This is especially true when the budgetary environment is highly politicized and fragmented due to diverse interests and values, when performance measurement information can be used unfavorably by the media to criticize the ruling administration or the legislature, or when there is no political consensus among the legislative leadership or the major parties on how agency or program performance should be defined and measured. In such a political environment, ambiguity about program outcomes is preferred by policymakers, and they may have little interest in specifying too openly how performance measurement results should be used and then linked to funding decisions (Radin, 2000 2006; Ho, 2003).
Second, legislative engagement in performance budgeting may create an accountability dilemma. A key premise of new public management and the fundamental logic of results-oriented budgeting is that for results-oriented budgeting to work, agencies often need some flexibility and discretion in management so that resources can be optimized to achieve the desired outcomes of citizens and elected officials. If legislators are heavily engaged in the design and execution of performance budgeting, such as determining what input, output, and outcome measures should be used, how program structure should be defined, and what performance-improving strategies should be pursued, they are likely to micromanage the process of performance budgeting and may shift the focus of budgeting from results-oriented control back to the traditional process control. This accountability dilemma presents a challenge for performance budgeting: Should legislators exercise more oversight of agencies' resource allocation process to ensure greater transparency and accountability, or should they give up some of their monitoring authority and let managers manage with greater discretion as long as they can accomplish the expected outcomes and policy goals (Bourdeaux, 2008b)? There is no easy answer to this question, and the balance between the two, which is reflected in the design of performance budgeting, may depend on the level of trust between different branches of the government, its institutional history and tradition, leadership style, and their governing philosophies (Dull, 2006). For example, after decades of performance budgeting reforms and the push to make agencies focus more on strategic planning and service outcomes through the Government Performance Results Act of 1990, the US Congress still requires agencies to submit detailed line-item budgets in the appropriation process because under the constitutional principle of separation of powers, Congress has the exclusive right to appropriate funds to agencies, and the budget authorization and appropriation processes are crucial checks and balances for the executive power.
The Dutch parliament, in contrast, decided to trim many input performance measures and line-item spending controls, such as personnel, capital, and goods and services, in the course of their performance budgeting reform in the 2000s. The structure and content of the Dutch performance budgeting reform has shifted legislative attention to broad policy goals and outcomes and high-level operating output information. As a result, agencies are given more discretion in program design and resource allocation and can have more freedom to use performance information for strategic planning, organizational learning, and improvement of operational efficiency and effectiveness (de Jong, Van Beek, & Posthumus, 2013). The difference in the strategies and content of performance budgeting by the Dutch and US legislatures shows how each country's reform path is often framed and constrained by its history, political structure, and institutional norms.
Constraints Caused by Program Type
Besides politics, the program type may also influence how policymakers decide whether performance budgeting should be presentational, informational, or direct (Radin, 2000 2006). For certain intergovernmental programs, funding eligibility and amounts may be linked specifically to past performance. For example, some of the recent education reforms in the United States and in other parts of the world are trying to move in this direction by linking student performance to school funding decisions. For other programs, performance can be completely irrelevant since funding is nondiscretionary and is dictated by legislative mandate. Many entitlement spending programs for individuals, such as Social Security and Medicare, are examples of this program type. For some other programs, direct performance budgeting may even be logically difficult and undesirable. For example, funding for basic research and development may not yield any visible and measurable outcomes until years or decades later. Also, it is hard to measure the outcomes of regulatory programs, such as environmental compliance or health and safety inspection programs. For these programs, process or output measures, such as how many cases are examined within a certain time frame, are easy and plentiful, but measuring the outcomes of these programs can be tricky, since they are designed to prevent certain undesirable events from happening rather than to produce measurable goods or services. In the absence of disasters or undesirable happenings such as disease outbreak or crime, policymakers and appropriators have to wrestle with the question of whether the funding of these regulatory programs can be reduced since there seems to be less need for them, or whether it is a sign of effective enforcement, thus justifying continuous or more funding. The ambiguity of outcomes and the seemingly unclear causal linkage between program activities and results make the application of direct performance budgeting problematic for this type of program.
Hence, many administrative and political factors influence the practice of performance budgeting. Agencies' capacity to produce valid and reliable performance data, political trust between the executive branch and the legislature, partisan politics and fragmentation of interest within the legislature, institutional history, leadership support, legislative mandates, and program type may all play a role in determining whether presentational, performance-informed, or direct performance budgeting should be used. Since these factors have unique implications for each stage of the budgetary process, the practice of performance budgeting may vary from the executive preparation stage, to the legislative appropriation stage, and then to the budget execution stage. For example, while performance-informed budgeting may be practiced between the budget office and departments in the budget preparation and execution stages, presentational performance budgeting may be more likely in the legislative appropriation phase due to the political and institutional factors discussed. Hence, there may not be a single performance budgeting approach for any government. What is more likely is a mosaic of presentational, performance-informed, and direct performance budgeting practices at different stages of the budgetary process and by different departments and programs.
Implications for Performance Budgeting Design and Practices
Given an understanding of the constraints discussed, students and practitioners of performance budgeting should resist a common but erroneous assertion: that performance budgeting is about inserting more economic rationality into the legislature's appropriation decision so that public spending is done more accountably and effectively. Performance budgeting no doubt should be concerned with the legislative phase of public budgeting. The appropriations process, however, is only one of the many possible linkages between performance results and budgeting, and economic rationality alone should not dictate the agenda of a multifaceted and diverse process.
In addition, policymakers should resist the simplistic logic that if performance budgeting is to work, the legislature should appropriate more funding to programs with greater performance and should eliminate programs or reduce spending in programs with poorer performance. Sometimes programs with poor performance may need more public investment so that agencies have the necessary capacity to handle service planning and delivery challenges and achieve the desired results. In other circumstances, well-performing programs may be eliminated because they are no longer needed due to changes in the socioeconomic contexts and public service demand. Hence, performance results and funding decisions should not be linked mechanically (US General Accounting Office, 2003). Performance budgeting provides a structured process for policymakers, managers, and stakeholders to consider the goals and desired outcomes of government spending. It may also help agencies become accountable and inform key stakeholders about what results have been achieved and what value and net social benefits public spending has brought to society. However, program performance should not be used to dictate how policy goals should be prioritized and how funding should be allocated.
Moreover, policymakers should resist the expectation that performance budgeting will ultimately lead to lower government spending as public programs become more efficient and cost-effective. As already noted, not all public programs are or should be governed by the cost-effectiveness logic in funding decisions. Legislative mandates, politics, and other policy priorities such as equal access to services are equally important. Moreover, because of constituency demands, policymakers seldom ask agencies to do less and spend less when performance has been improved. Instead, agencies are often asked to do more with the same funding level or even to do more with more when they demonstrate greater cost-efficiency and effectiveness. As a result, any attempt to link performance results to the appropriation amount can be a futile exercise, and any conclusion drawn from the relationship between spending and performance without sufficient understanding and control of program outcomes and output can be misleading. Indeed, studies have found that the level of spending is not necessarily lower among governments that practice performance budgeting (Klase & Dougherty, 2008). Even if program performance is linked to funding requests in the budget preparation stage, the relationship is often weak, since many other factors, especially political considerations and policy priorities of the ruling party, are more important (Gilmour & Lewis, 2006).
So how should performance budgeting practices be improved? Studies suggest that the key focus should be on the organizational learning process rather than on budgeting strategies and spending amounts. Performance information needs to be more integrated into a structured process to enhance inter- and intra-agency communication and managers' focus on accountability and result improvement (Melkers & Willoughby, 2001 2005; Moynihan & Andrews, 2010; US General Accounting Office, 2003). Even if the spending level is not changed due to political constraints and legislative mandates, structured dialogue between the budget office and departments about performance can still help redefine outcomes or output measures, change intraprogram resource allocation to realign priorities and activities, and enhance program performance and public satisfaction with the results (Ho, 2011).
To facilitate structured dialogue on performance, it is important to focus on building the capacity of the central budget office and departmental budget offices. Performance budgeting is a resource-intensive exercise and requires significant investment in the analytical capacity of the staff, data management capacity, and organizational learning. If performance budgeting is not so much about rationalizing appropriation decisions but is more about public accountability and organizational learning to improve planning, budget execution, and operational efficiency and effectiveness (US General Accounting Office, 2003; de Jong et al., 2013; Ho, 2011), agencies and the budget office need to be equipped with staff capacity, training systems, and analytical tools for program evaluation, quality control, and impact analysis so that agency managers and budget reviewers can be intelligent users of performance information and evidence-based evaluation results (Newcomer & Scheirer, 2002; Heinrich, 2007). Also, lack of cost and outcome data and poor quality of data can be major hurdles to building the reliability and credibility of the performance budgeting system (Breul, 2007; Melkers & Willoughby, 2001 2005; Ho & Ni, 2005). Hence, investment in information technologies, such as cost accounting and data management systems, is important. With better technological support, the budget office and departments can collect, process, store, analyze, and report performance measures more cost-effectively. Also, the systems may help reduce human errors and frauds during the data collection process. From the staff perspective, good technological support can reduce the paperwork and time pressure, common causes of staff resistance and anxiety about performance budgeting.
Given that many governments, including those with a long history of performance budgeting reforms, are still struggling with data availability and capacity-building issues and still trying to establish the appropriate institutional logic and roles of performance information in the budgetary process, legislators and implementers of performance budgeting should be cautious about using punitive incentives, such as budget cuts, personnel demotion, or program elimination, to give the tool more impact. Studies show that if the stakes for failing or succeeding are too high, agencies under pressure will try to game the system by manipulating the data, lowering the performance targets (Bevan & Hood, 2006). They may also try to enhance their performance by creaming or skimming the service recipients who are easier to serve or are more likely to succeed, and ignore the original mission or legislative intent of a program or policy (Courty & Marschke, 2007; Heckman, Heinrich, & Smith, 2002). These behaviors certainly distort the results of performance budgeting, which is to help agencies become more accountable, cost-effective, and responsive to their service mission and goals. Hence, dialogue between the budget office and departments in the process of performance budgeting should be framed constructively, and agencies should be allowed to learn from past mistakes but are encouraged to show improvement over time. The power of the purse to cut agency funding or restructure programs may be used as the ultimate threat, but other incentives, such as giving agencies more administrative discretion and public recognition, empowering them with better training, and emphasizing the norm of public service motivation, are probably more effective and less distortionary (Heinrich & Marschke, 2010).
Finally, rallying leadership support behind a performance budgeting reform is critical, and such support may come from several possible sources. First, leadership can be found among peer organizations or neighboring jurisdictions, which can provide models for best practices. Because many jurisdictions and organizations like to benchmark their success against others, the pressure of isomorphism should not be underestimated and can be an important force of change to push performance budgeting forward. Hence, if a jurisdiction can find or join a peer group that can learn from each other and recognize significant achievements during the reform implementation process, they are more likely to succeed in the reform process.
At the same time, leadership support within an organization is also needed. Support from the top leadership is important to help set a clear vision of the reform, articulate its rationales, rally support and resources for the reform, and set the tone of a results-oriented culture. With strong top leadership support, performance budgeting is more likely to find grassroots leadership. Performance budgeting needs staff ownership of the reform and bottom-up ideas of innovative changes since performance information is most likely to be used in the stage of budget execution and resource allocation at the program management level. Hence, how to get the buy-in of program managers, empower them in the implementation process, and let them see the value of performance budgeting are some of the key strategies in sustaining the success of performance budgeting.
Summary
The idea of integrating performance information into the budgetary process has been advocated for decades. These ideas and practices have evolved, and many studies have been conducted to analyze their success and associated implementation challenges. Based on the review of the literature, a few key lessons emerge:
- Performance budgeting is a multifaceted process that is important in different stages of the budgetary process, but based on various countries' experiences, it is more likely to have a substantive impact on budget planning, strategic goal setting, budgetary reporting and communication, and program-level budget execution and management.
- Each organization is likely to have a mosaic of presentational, performance-informed, and direct performance budgeting practices in different stages of budgeting and in different programs. There is no one-size-fits-all structure because the design and implementation strategies of performance budgeting are contingent on the influence of legislative politics, the relationship between the executive and legislative branches of the government, the power and style of executive leadership, the administrative capacity of agencies, and the nature of the program.
- Performance results are seldom, and should not be, linked mechanically to funding decisions.
- Rather than focusing on the punitive incentive of the power of the purse, performance budgeting should focus on constructive dialogues between the budget office and departmental managers about continuous program improvement and organizational learning. They should also use positive reinforcement and capacity-building strategies to maximize potential benefits for program execution and service delivery.
- Articulating a clear vision and the goals of performance budgeting, listening to key stakeholders about their priorities and concerns, and rallying top leadership support are also important to the launch of a performance budgeting reform.
- Getting staff buy-in at the program management level and empowering them with information technology support and training are critical to ensure that performance budgeting can make substantive positive impacts on resource allocation, program management, and accountability reporting.
As many countries continue to struggle with high public debt, the growing pressure of social spending and the need to keep taxes low to stay economically competitive in the global economy, determining how to prioritize public spending and ask government agencies to do more with less, and the quest for greater transparency, efficiency, and cost-effectiveness in public spending will remain top concerns for many policymakers around the world. From this perspective, performance budgeting initiatives are unlikely to wane despite the rising politicization of the public budgetary process. New efforts and innovative practices to integrate budgeting with performance measurement, strategic planning, goal setting, program management, and public reporting will continue to emerge given the policy challenges and many practical needs of the foreseeable fiscal environment.