OECD Good Practices for Performance Budgeting

Introduction

Origins of the good practices

The OECD Senior Budget Officials (SBO) network has taken a strong interest in performance budgeting reforms for more than 14 years. In 2004 the SBOs set up a Performance and Results Network to increase understanding of these practices. Since then the OECD has conducted four surveys of performance budgeting practices (in 2007, 2011, 2016, and 2018), prepared country case studies and held regular meetings where experience was shared amongst member countries. Through these processes a substantial body of knowledge and experience has been accumulated. In addition, through in-depth reviews of budget practices, the OECD has had the opportunity to closely study performance budgeting practices and identify good practices.

Key aspects of the OECD’s understanding of performance budgeting are set out in the 2015 Recommendation on Budgetary Governance which includes, among its ten principles of modern budgeting, the principle that “performance, evaluation and value-for-money are integral to the budget process”. In particular, the recommendation calls on governments to “routinely present performance information in a way which informs, and provides useful context for, the financial allocations in the budget report; noting that such information should clarify, and not obscure or impede, accountability and oversight” and goes on to recommend "using performance information, therefore, which is (i) limited to a small number of relevant indicators for each policy programme or area; (ii) clear and easily understood; (iii) allows for tracking of results against targets and for comparison with international and other benchmarks; (iv) makes clear the link with government-wide strategic objectives." Moreover, the Good Practices take into account other recent studies of performance budgeting including the 2016 World Bank publication, Toward Next Generation Performance Budgeting, and the growing academic literature on the subject including the corpus of material published in the OECD Journal on Budgeting. The findings are also consistent with the latest summary of World Bank work on public sector performance (2018)

The Good Practices take the OECD Recommendation as a starting point and elaborate on how to apply these principles. The good practices examples should be understood as the best examples available, based on the OECD’s knowledge of current practices, including a number of examples volunteered by OECD countries. They do not necessarily represent ideal examples, since no country has an ideal system, and they need to be understood in relation to the specific context within which they were developed.

What is Performance Budgeting?

Performance budgeting is “the systematic use of performance information to inform budget decisions, either as a direct input to budget allocation decisions or as contextual information to inform budget planning, and to instil greater transparency and accountability throughout the budget process, by providing information to legislators and the public on the purposes of spending and the results achieved”. A shorter definition is “any budget that represents information on what agencies have done, or expect to do, with the money provided to them’’ (Schick).

The broader definition of performance budgeting includes spending reviews and programme evaluation as well as preparation of the annual budget using performance information. However, these Good Practices refer mainly to the core processes of budget formulation, budget execution, monitoring and reporting. Ex ante policy impacts assessment, spending reviews and ex post programme evaluations are discussed from the perspective of their relationship to performance budgeting. They do not include statements of good practices for spending reviews, ex ante or ex post programme evaluation. They are also mainly focused on performance approaches within the executive, which do not prejudge the role of supreme audit institutions in reviewing performance ex post.

The adoption of performance budgeting implies a shift in the focus of budgeting, away from management of inputs and towards a focus on the results of spending and the achievement of policy objectives. This represents a profound change in the character of the budget process, from a traditionally closed domain of budget specialists, focused on the numbers, to a more accessible, transparent and multi-disciplinary exercise, with greater involvement by the centre of government, line ministries, the legislature and citizens. One key consideration in many countries in this process is the link with the government “political programme”, the implementation of which is often coordinated through the centre of government. Aligning political priorities and financial means is essential for government to effectively fulfil its missions. Characteristic elements of a performance budgeting system are the following:

• In the preparation of the budget, spending decisions take account of priority policy objectives as well as past programme performance;

• The budget is presented to the legislature as a set of programmes. This may be in addition to, or replace, presentation based on administrative and economic classification;

• Non-financial performance indicators and targets, representing the planned objectives of spending, are presented in the budget and linked to spending programmes;

• During budget execution, budget managers have the autonomy to manage financial resources, balanced by the accountability for achieving results;

• Performance information is provided to managers together with budget execution data to help them monitor performance and improve the quality of spending;

• Reports to parliament on budget execution include performance information and a narrative explaining the reasons for under or over-performance.

In practice, the emphasis that countries have placed on performance, and the extent to which it has replaced more traditional forms of budgeting varies widely across OECD countries. While the first three features described above are common to most systems, fewer incorporate the managerial dimensions described in the fourth and fifth bullets. The OECD classifies the different approaches to performance budgeting seen in its member countries as follows:

Presentational Performance Budgeting

This shows outputs, outcomes and performance indicators separately from the main budget document. This is relatively easy to achieve and is appropriate where the objective is limited to demonstrating that budget allocations and actual expenditures are responsive to government’s strategic objectives and policy priorities. On the other hand by separating performance and budget data it is harder to relate the two.

Performance-Informed Budgeting

This includes performance metrics within the budget document and involves re-structuring of budget document on the basis of programmes. This approach requires considerable effort to achieve and is appropriate for governments that want to achieve more ambitious goals such as re-prioritisation of expenditure linked to performance and increased devolution of budget control to programme managers. This is the form of performance budgeting that many OECD countries have adopted.

Managerial Performance Budgeting

Managerial performance budgeting is a variant on performance informed budgeting. In this approach the focus is on managerial impacts and changes in organisational behaviour, achieved through combined use of budget and related performance information. This depends on a performance culture existing within government that will take time to establish if it does not already exist.

Direct Performance Budgeting

Direct performance budgeting establishes a direct link between results and resources, usually implying contractual type mechanisms that directly link budget allocations to the achievement of results, implying a budgetary response to over or under-achievement of performance objectives.

OECD countries that have implemented performance budgeting are divided approximately into equally proportions between the first three approaches (see Figure 1).

Figure 1. Performance budgeting approaches

Notes: Greece, Luxembourg and Portugal indicated not having a Performance Budgeting Framework in place. Data for Israel and the United States are not available.

Source: OECD (2018), OECD Performance Budgeting Survey.

No OECD country identifies itself as using direct performance budgeting as a general approach. However, funding for some types of services e.g. some payments to health providers are directly linked to performance.

Why is performance budgeting important?

Performance budgeting offers a wide range of potential benefits that are attractive to different stakeholders. For the centre of government, performance budgeting offers the chance to strengthen the alignment between decisions on budget allocation and government policy priorities, thereby boosting the chances that the government can deliver on important pledges. For parliaments, performance budgeting offers greater clarity on the purposes of spending and what goods and services will be delivered in return for the resources they have voted on, as well as a means of holding officials to account for the achievement of results. For Ministries of Finance, performance budgeting provides new types of information that help them make resource allocation decisions based on evidence of what works, plus tools to make line ministries more accountable for the effectiveness and efficiency of spending. For line ministries, performance budgeting provides tools to improve their internal decision making, and to make a stronger case to government in support of their budget proposals. It also helps programme managers to do their job, enabling them to track performance as well as spending. For citizens and civil society organisations, performance budgeting offers the prospect of greater transparency and accountability in respect of the purposes and results of public spending and, by opening up the ‘black box’ of the budget, it strengthens the basis for direct citizen engagement in the budget process.

Performance budgeting has been subject to continuous innovation. While many countries now report significant benefits, these benefits have often been realised only after a long period of experimentation and adaptation. For others the results still fall short of expectations. The logic has remained sufficiently compelling that, instead of abandoning the pursuit of performance, many of the early adopters have reoriented it from performance-based to performance-informed budgeting, or shifted the emphasis from central budget decision making to performance management in line ministries. Many countries have repeatedly modified the definition and classification of key concepts, such as outcomes and results, and their use in the submission and review of budget requests. Others have shifted from PB's original focus as an instrument for deciding the budget to a means of classifying or displaying decisions taken during budget formulation. For many countries, performance budgeting offers a useful tool to connecting the budget with broader notions of citizen wellbeing and the efficiency and effectiveness of government, in order to deliver better results for citizens.

Through a combination of moderated expectations and adaptation of approaches, OECD governments now see benefits emerging from these reforms. Looking at the various dimensions of effectiveness, OECD countries report the strongest results in the areas of improving the transparency and accountability of the budget, with weaker impact in the areas of resource allocation and moving towards a culture of performance (See Figure 2).

Figure 2. Ranking of different rationales for introducing performance budgeting and their effectiveness in practice

Note: Data for Israel and the United States are not available.

Source: OECD (2018), OECD Performance Budgeting Survey.

These modest averages hide a wide range of variation across countries in the perceived level of effectiveness of performance budgeting in meeting these goals. Some OECD countries have had considerable success with PB-type arrangements by recognising that performance has to be embedded in the culture and systems of public management, especially the strategic processes and goals of government, and that PB has a robust infrastructure across government to support the production and use of performance information.

Performance budgeting within a performance management system

Performance budgeting should be thought of as one component of a wider set of mutually supportive reforms aimed at creating a performance-oriented and accountable public sector. This could be described as the performance ecosystem. OECD experience suggests that performance budgeting is more likely to flourish and be effective when it forms part of such a broad-based performance management system. In order to fully fulfil its mission, such a performance management systems needs to be underpinned by proper management of evidence and systematic attempts to collect and interpret data. As an isolated initiative, performance budgeting is unlikely to lead to the changes in behaviour that are often anticipated. Therefore, in designing the approach to performance budgeting, close attention needs to be paid to the status of other performance related processes and reform initiatives, in particular:

• strategic planning at both national and sector levels

• Management of data and evidence, including data mining, mobilisation of administrative data and open data.

• medium-term budget frameworks

• spending reviews

• individual performance appraisal, as part of human resource management,

• performance-based contracting and payment systems,

• ex ante and ex post programme evaluation,

• performance audit.

Ideally the development of performance budgeting should be co-ordinated with, or at least informed by, these other performance oriented reforms.

Addressing different states of development of performance budgeting

Performance budgeting reforms have often been improvisational and evolutionary in nature. Almost every OECD country that has adopted performance budgeting has periodically revised its approach and, since few countries are fully satisfied with their systems, it can safely be assumed that they will continue to evolve. Performance budgeting may therefore be described as a journey rather than a destination. For countries that are at the start of this journey, the good practices cover the whole process from the design of a performance budgeting system through to system implementation, operation and oversight. For those countries that have already implemented performance budgeting, benchmarking against the good practices may offer some ideas for improvement.

Governments considering their approach to performance budgeting should pay careful attention to the balance between effort and reward. Introducing a new system of budgeting, often running in parallel with the existing system, imposes a substantial additional burden to the budget and staff. If this does not result in decisions being made in a different way the reform can become discredited. An incremental approach may be better than a “big bang” approach, especially where there are significant constraints, for example in staff skills or the quality of performance data. Incrementalism could mean initially piloting performance budgeting in a smaller group of ministries, before a phased expansion to cover the whole of central government, followed by agencies and local government gradually expanding or presenting the performance budget initially as supplementary information. It could also mean starting with a goal of implementing a presentational form of performance budgeting, before moving towards performance‑informed budgeting and including programme performance in the budget approved by the legislature. In a further phase of development ministries would make routine use of performance information for programme design, budgeting and management.