Two generations of performance evaluation and management system in Australia

Abstract

This paper describes and contrasts two generations of performance evaluation and management system in Australia over the past sixteen years. The first generation related to the decade from 1987 to 1997. It was centrally-driven, stressed formal requirements, and was principally concerned with ensuring evaluation findings were available to feed into the annual budget process. This system was successful in ensuring that evaluations were planned carefully, and that a growing 'library' of evaluation findings was available to support and to influence budget decision-making. However, this system had some weaknesses, including inadequate attention to the collection, use and reporting of performance information concerning government programs. The second generation system was introduced in 1997. It has entailed a highly devolved approach, stressing principles rather than formal requirements, and largely emphasizing the collection and publication of performance information, and its provision to the parliament. However, there is evidence that this information has proved insufficient to meet the needs of parliament, and that it is insufficient for purposes of sound management. Some departments continue to devote significant resources to the conduct of rigorous evaluations, in support of their ongoing management and for reporting to parliament and others. Finally, the paper offers lessons for efforts to develop performance evaluation and management systems in other countries. These include the pros and cons of a centrally-driven system, as compared with a more devolved approach. The critical role of powerful finance ministries is also discussed, together with the benefits from having a strong reform champion. The dangers of over-designing a performance evaluation system are also examined briefly.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 23 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.