In the wake of the current economic, health and climate crisis, governments face long-term and complex economic, social and environmental policy challenges. Asystemic understanding of what works is critical in addressing these difficulties. Supreme audit institutions (SAIs) have untapped potential to help governments meet these challenges. SAIs can provide a comprehensive, crosscutting view of how processes and programmes function across government. Through their objective analysis, they can strengthen the evidence base for policy decisions, ensuring citizens receive value-for-money.

Public policies that provide key services to citizens, such as education, health, welfare, infrastructure and sanitation, are increasingly developed and delivered involving different levels of government. The expectation is that this decentralisation will lead to better local public service delivery through greater citizen engagement and local accountability. For effective delivery of decentralised policies, adequate multi-level governance conditions are needed, such as a fiscal framework, regulatory processes and local capacity. The lack of effective multi-level governance can lead to gaps, duplication and overlap in policy delivery. This not only results in poor value for public money, but also opens the door to regional disparities in the delivery of public services and the potential undermining of citizens’ trust in government.

This chapter looks at the issues facing external audit in tracking crosscutting issues and policies that are delivered through different levels of government. It first lays the groundwork by defining multi-level governance and noting trends in decentralisation. Then, drawing on OECD studies and best practices, it takes up the various aspects involved in auditing policy delivery in a decentralised landscape, placing the focus squarely on Brazil with its exceptional degree of local autonomy. Touching on the subject of public accountability, the discussion goes on to clarify the distinction between centralised and decentralised external audit structures, and between exclusive and complementary/concurrent mandates. After highlighting external audit challenges specific to Brazil, the chapter concludes with a roadmap of key priority areas for improving the audit of decentralised policies in complex accountability structures, with the emphasis on co-ordination among audit bodies.

This chapter begins with a look at the application of risk assessment in audit selection generally and the corresponding need for sound evidence. The focus then shifts to Brazil and education, describing the complexity of auditing in a decentralised policy area. The chapter points to the existing wealth of high-quality education data in Brazil, and shows how a logic model can be applied to the sector to select significant indicators and thereby identify potential problems in policy implementation. Risk scenarios are then described, as is their application in six-step process for risk-based audit selection. The chapter concludes by noting how IT can be a useful tool in supporting audit selection in Brazil.

This chapter describes how the federal, state and municipal courts of accounts in Brazil can develop a multi-level governance (MLG) assessment framework for the audit of decentralised policies. It begins by proposing a four-step model that details the concrete actions that audit institutions should follow in developing a framework for specific policy sectors. It goes on to present a generic MLG assessment framework developed by the OECD for Brazil that can be used as a guide for developing sector-specific frameworks. The chapter then describes maturity indicators on key dimensions of the MLG system that should be assessed during the audit. Finally, it examines how this four-step model could be applied to the education sector in Brazil, taking as key input the generic model.

This chapter discusses how co-operation, co-ordination and collaboration among national and subnational audit bodies can lead to coherent, strategic auditing of decentralised policies. It begins by noting the key dimensions of collective action, and goes on to list three types of network governance structure, citing examples as well as the strengths and weaknesses of each. Key operational requirements for the audit network are explored next, with the emphasis on the need for collective understanding of, and agreement on, the network’s purpose. Challenges faced by these networks are then outlined, including information sharing versus confidentiality and privacy and differing audit standards and capacities. The chapter concludes by providing details on how an evidence-based risk assessment can be embedded in a collaborative process of audit selection within a network of external audit entities.