Does performance auditing matter? The government administration’s response to the performance auditing of the Norwegian Office of the Auditor General
MetadataShow full item record
The point of departure for this thesis is Michael Power’s provocative proposition in his 1997 book The Audit Society (Power, 1997). There he states that auditing produces only reassurance. This thesis is an empirical investigation into the influence of the Norwegian Supreme Audit Institution to see if this assertion is true.
The Supreme Audit Institution (SAI) can be described as a tool of the parliament for use in exercising control over the government administration. In this thesis, I concentrate on one particular method of the SAI, namely “performance auditing”. Performance auditing is similar to evaluation but is conducted within the context of institutional control. This method has become increasingly important with the advent of new management principles and techniques in the public sector from the 1980s onwards. The new public management (NPM) system was partly inspired by management techniques in the private sector and is focused on the efficient and effective use of resources. Interest in the performance of public entities, however, predates NPM by many decades.
Accountability mechanisms are needed to prevent misuse of power. For this reason, the role of external control institutions is rarely questioned. Some scholars nevertheless accuse such institutions of providing assessments that are too detailed and narrow. At the same time, others describe how SAIs are at risk of becoming political and running into conflicts with the executive.
The goal of this thesis is to investigate how auditees and members of the parliament (MPs) respond to performance auditing. It is their reactions and opinions that are the dependent variable. In this thesis, a mixed-method approach – including a questionnaire for auditees, interviews and document analysis – was adopted.
Contrary to Power’s assertions, my study shows that performance auditing does have an impact. A majority of civil servants find performance auditing helpful. Their perception of its usefulness depends on the extent to which their comments were taken into account in the performance auditing process, whether they considered the performance audit report to be of high quality, whether the performance audit contributed to improvements and how they perceived the SAI as an institution.
There are nevertheless differences in how respondents perceive performance audit reports. In general, civil servants at ministries and top executives tend to be more negative. This suggests that the offices that are responsible find it uncomfortable to be held to account.
Civil servants more exposed to performance auditing are more negative too. This may indicate control overload. Alternatively, it may indicate discomfort linked to justified control.
In one-third of cases, the performance audit is used to hold a minister to account. This tendency increases when the media or other actors take an interest in the report. Pressure from political opponents and demands for concrete measures from the parliamentary Standing Committee on Scrutiny and Constitutional Affairs (control committee) strengthen this tendency in equal measure.
Auditees use different strategies when they are subject to performance auditing. Their willingness to make changes depends on pressure from and sanctions imposed by the parliamentary control committee, as well as on their own perception of the report. If they disagree with the report’s findings and sanctions are weak, they will resist change. If they agree with the findings and sanctions are tough, they will implement change. At the same time, apparent indifference towards the report does not necessarily mean less impact. When the findings of a report are important, the ministries will use them to make improvements, regardless of whether the report triggers debate.
Performance audit reports have a moderate impact on the public debate. Thus the SAI’s direct dialogue with the ministries is important. At the same time, disagreements based on different values and logic can impede the SAI’s influence on the audited entities.