public financial management (IMF 2001; World Bank 2002). The 2000 fiscal responsibility law disciplined federal finances and, in 2001, the budgetary system was further rationalized. Moreover, since the late 1990s, planning, management and budgeting are becoming more integrated. Budget reform has gradually become the key lever for reforming the state, disciplining public finances and improving public management. The TCU has thus benefited from significant improvements in the budgetary and accounting systems. This contrasts with the experience of Argentina where the failings of external auditing are compounded by weaknesses in public budgeting and accounting.
Despite important improvements in the institutional environment, effective changes in the quality of the scrutiny and oversight of federal accounts have been slow (Pessanha 1999). The effectiveness of external auditing remains inhibited by important dysfunctions in the institutional linkages between the TCU and the other components of the system of fiscal control. Although the TCU follows the court model and has quasi-judicial powers, its insertion into the judicial system is questioned and contested. At the same time, its relations with the legislature, its ultimate principal, exhibit important shortcomings as a result of skewed political incentives. As a result, proposals to reform the TCU continue to abound.
In terms of the relations between the TCU and the judiciary, there is controversy over the judicial nature of the TCU, the legal status of its decisions, and the enforceability of its audit rulings. While TCU follows the court model, its status as a quasi-judicial administrative court is contested. While its decision-making process mirrors a judicial process, its linkages with the judiciary are ambiguous. As a result, the enforcement of audit recommendations and the imposition of sanctions for non-compliance suffer serious shortcomings.
TCU considers itself a quasi-judicial organ as an administrative court. As Speck (2000:210) argues the ‘procedural structure of the TCU is currently guided by the logic of procedural legitimacy, which is a characteristic trait of the judicial process.’ However, the quasi-judicial powers of the TCU have been contested from the onset by the courts, which consider the TCU as an administrative rather than a judicial body. A key unresolved controversy is whether audit rulings are subject to judicial review (Rocha 2003). The courts question the judicial nature of audit rulings based on the legal principle that there cannot exist administrative acts shielded from judicial review. Furthermore, the Brazilian judicial system does not establish the legal principle of stare decisis (‘stand by things decided’) and each case is therefore evaluated on its own merits. This allows the courts to re-examine each TCU ruling independently of any legal precedent, preventing the gradual emergence of a body of audit jurisprudence. Moreover, the TCU does not have the right to coactive execution, which is enforced by the public prosecutor’s office.
The Supreme Court has ruled on repeated occasions to circumscribe the competences of the TCU, for example excluding the state oil company Petrobras from its purview in 2002 or contesting its oversight of regulatory agencies and public companies. It has also challenged the judicial nature of audit rulings, reducing them to simple recommendations to administrative authorities (Silvera e Silva 2002). In response to the courts’ challenge, the TCU has staunchly defended the judicial nature of audit rulings as ‘coisa juzgada administrativa’ that can only be appealed in the courts in case of procedural irregularities (Guimarães Souto 2002:228; Affonso 1997, 1996), but to no avail. As Rocha (2003:15) remarks:
‘the more the TCU insists on the jurisdictional nature of its decisions, the more it is exposed to the criticism of the judiciary, which, attentive to defending its own prerogatives, has imposed restrictions on the TCU’s competence.’