Lessons from the national school build program

The Australian National Audit Office's report into the operations of the National school build program makes for fascinating reading for someone like me interested in the details of management. While the report does not provide the type of smoking gun that the Australian opposition would have liked, it does provide an insight into the problems created by the Rudd Government policy approach and associated administrative structures and styles.

Although no-one so far has looked at the connections, this is in fact the second report into the operations of what are called National Partnership Agreements. The first came with the release by Indigenous Affairs Minister Jenny Macklin at the end of August 2009 of a report reviewing the initial operations of the $672m Strategic Indigenous Housing and Infrastructure Program. This program became part of the National Partnership Agreement on Remote Indigenous Housing and was intended to deliver 750 new houses by 2013 in Northern Territory remote indigenous communities. My then analysis of the report can be found here.

The ANOA report describes the overarching framework that is meant to guide Commonwealth-State arrangements in these terms:

Delivery of programs that span Commonwealth, state and territory
jurisdictions has been the subject of recent Council of Australian Government
(COAG) reforms. The reforms aimed to enhance public accountability for
service delivery by clarifying roles and responsibilities between levels of
government and improving collaboration. Rather than dictating how things
should be done, the new framework focuses on the achievement of mutually
agreed outputs and outcomes, providing the states and territories with
increased flexibility in the way they deliver services to the Australian people.

National Partnership Agreements are one element in this. These generally involve:

1. Negotiation between Commonwealth and States of a framework agreement setting out objectives and broad approaches.
2. Development of an agreed implementation plan setting what will be done and how, along with reporting arrangements on progress.
3. Both agreements and implementation plans contain provisions for variation.  

On the surface, this all seems very sensible. However, problems can arise where:

1. The Commonwealth as funder actually dictates the detail of what is to be included in the partnership on a like-it or lump-it basis. The problem is compounded where terms dictated ignore regional diversity, with a one size fits all approach. The only thing that the states can do in these cases is to try to negotiate on points of detail.
2. The Commonwealth essentially dictates what must go into implementation plans. These may include, for example, requirements that require Commonwealth approval of particular and detailed expenditure plans before any action can commence. They may also include a plethora of requirements and sub-objectives.
3. The Commonwealth is inflexible in implementation.

Problems are compounded because all states and territories have their own rules and accountabilities regarding expenditure that must be complied with, including formal legal requirements. These National Partnerships often involve very large sums of money.  The central coordinating agencies in the states and territories are concerned not just with efficiency and probity issues, but also with the management of cash flows.

Very real problems can be created for things such as state budget processes when the timing and scale of payments from the Commonwealth is uncertain. Further, coordinating agencies are concerned about ancillary costs that can arise in implementation that are not covered and which can affect other parts of state budgets.

If we now look at the Northern Territory report into indigenous housing, we find that program delivery there suffered from, among other things, from a multiplicity of sometimes conflicting and unranked objectives with over-complicated decision structures including no less than six decision layers.

Turning to the school building program, the ANAO summarised some of its criticisms this way:

The program design and funding variation process, which is focussed at project/school and ties funding to individual projects within schools, has not allowed the states to manage BER at a program level. The complex and constantly amended funding variation process has not enabled states to readily transfer funds between schools in order to achieve the agreed outcomes. The variation process and adherence to a narrow definition of approved project is considered increased input control, which is at odds with the principles of the IGA (Inter Governmental Agreement) to focus on outcomes.

Without going into the full detail in the report, some of the problems that arose can be summarised this way:

1. The use of bands based on student numbers to determine project size created inequities between schools at the margin (a shift of a few pupils could have dramatic funding impacts), as well as an incentive to manipulate student numbers.
2. The way that funding was announced meant that was meant to be a maximum spend per school actually became a target spend.
3. The requirement that the Commonwealth approve spend on a school-by-school, project-by-project basis, together with over-complicated objectives and reporting requirements, increased administrative load and reduced flexibility. This was further complicated by a ruling that spend below maximum on particular school projects could not be transferred to other projects in the same jurisdiction but must be returned to the Commonwealth. School systems had no flexibility in adjusting spend between schools to achieve maximum value.
4. Conflict arose between the Commonwealth's approach and the centralised approaches to public education in the various states. This was complicated by Commonwealth rules that created tensions between principals and their state employers.
5. Budget confusions arose. In preparing costings, the Commonwealth Finance Department worked on the basis that spend would equal 90% of the maximum per school. This was interpreted by the Department as 90% of schools participating. Since the Department's aim was 100% participation, while the maximum spend had actually become target, it quickly became clear that spend would be greater than budget. This led to financial adjustments, including transfer of funds from social housing, to try to keep spend within aggregate budget approvals.
6. Given the size of the program, variations in Commonwealth payments created cash flow problems in the states. While these were resolved, it cannot have helped state budgeting.
7. In addition to difficulties associated with transfers of money between school projects, initial Commonwealth inflexibility created on-ground difficulties because of varying on-ground conditions. The intent was that all schools and areas should benefit from stimulus spend and in broadly the same time horizon. However, the actual impact of the downturn was quite variable across Australia, while there were also considerable variations in the available supply of skilled labour.
8. Not only were reporting requirements overly complicated, but the data provided inevitable contained so many variations and was based on so many assumptions as to be unusable in measuring progress. 

I said at the outset that the report was unlikely to provide the type of smoking gun desired by the opposition. Despite building delays of the type I forecast at the time, the report concludes that the program did broadly meet the stimulus objectives. Further, in criticising both the Department and other Commonwealth bodies involved, the ANOA explicitly recognised the size of the program and the difficulties involved. The analysis makes it clear that the Department did try within its limits to be flexible and responsive.

  To my mind, and putting aside the inevitable problems always associated with the delivery of such a huge program, the problems that arose were due to the combination of failures in the way the way the program was specified in combination with the broader systemic features such as inflexibility, over-complication, over-specification and over-control that mark the general Rudd Government approach to public policy.

You can see why some of us are so cautious about things such as the Health proposals. Regardless of the general in-principle arguments involved, there has to be a question mark over the likely effectiveness of the proposals in the absence of a change in the Commonwealth Government's approach to public policy and administration.