- Title page
- EXECUTIVE SUMMARY
- I. Introduction
- II. Transforming the Public Sector: Putting Ideas into Practice
- III. Structure of the State Sector
- IV. Organisational Capacity
- V. Strategic Capacity
- VI. Managing Public Money
- VII. Accounting For Results
- VIII. The Spirit of Reform
- Appendices I - IV
Accountability for Outputs
Accountability is predicated on a sharp distinction between the government's role as the purchaser of outputs and the department's role as supplier. The government's responsibility is to specify the outputs it intends to purchase and to decide what it will pay; the department's responsibility is to supply the specified outputs at the set price. This distinction explains why the government maintains two sets of accounts: the Estimates detail what the government intends to spend in purchasing specified output classes; the Departmental Forecast Reports specify what each department expects to spend in producing the outputs. There is both duplication and differentiation in these documents: duplication because much of what the government buys is supplied by the departments; differentiation because the government purchases some outputs from nondepartmental sources.
Outputs are the common interest of Ministers and managers; they are the linchpin of the New Zealand accountability system. The focus on outputs is in contrast to both its previous input-based appropriations and the outcome-oriented system favoured in much of the management reform literature. The Public Finance Act defines outputs as the goods and services produced by a department or by any other public or private supplier.
New Zealand focuses on outputs because they provide a reliable basis for enforcing managerial accountability, not because they are the most important indicator of government performance. Accountability is facilitated because the supply of outputs can be directly attributed to the performance of chief executives and their departments; outcomes, by contrast, tend to be influenced by many factors, some of which are likely to be beyond the control of the relevant department. It often is difficult to trace a particular outcome to the activities of a given department, but it usually is comparatively simple to determine whether a department has supplied the specified outputs. An outcome-based system would dilute accountability by purporting to make managers responsible for results they do not control; an output-based system strengthens accountability by limiting responsibility to the results that managers do control.
The output relationship between Ministers as purchasers and managers as suppliers is intended to replicate the arms-length relationship between buyers and sellers in competitive markets and to thereby mitigate "producer capture," the proclivity of managers to impose their interests and agendas on Ministers. But this relationship opens the door to fundamental questions that relate to the ownership-purchase issue discussed earlier in this report. Should purchasers and producers be institutionally separated so as to facilitate truly independent decisions by the government on what it buys and spends, or should the two sides strive for a cooperative relationship that recognises their joint interests? Furthermore, to what extent is the gain in accountability expected from separating Ministers and managers dissipated by high transaction costs? The New Zealand reforms give a clear answer to these questions. The greater the distance between Ministers and managers, the more independent and demanding the government can be as a purchaser of outputs and enforcer of accountability.
I am not persuaded that this approach is suitable in all the purchase situations facing government. It is most appropriate when there is genuine competition among suppliers, and the government has no incentive to give preference to public entities. In these circumstances, it can bargain for the best terms by having suppliers bid against one another, and it can vigorously enforce contracts for the delivery of specified outputs. However, the rarity of Mode B net appropriations indicates that this is not the normal purchase situation.
The typical condition is one in which the government owns the departments and obtains noncontestable outputs from them through budget rather than market decisions. The relationship is not arm's-length, and pretending that it is does not make the government less vulnerable to capture. Some Ministers have been tough, involved negotiators who have sought the best deal for the government and have taken an active role in specifying the outputs to be supplied by their departments. This posture has been strengthened by the retention of purchase advisers to assist in the negotiations. But many Ministers are discomfited by this quasi-adversarial arrangement, and they have preferred instead to maintain a closer, less formal relationship with chief executives and departments. In some cases, the Minister has agreed the outputs specified by the chief executive without reviewing the details or suggesting changes. Formal accountability is maintained even when the Minister abjures an independent role, but the heavy costs of maintaining pseudo-arm's-length accountability may not be justified in these circumstances.