Departmentalising the State Sector

New Zealand has almost twice as many departments as Ministers, making it common for Ministers to have multiple departmental responsibilities. The thirty-nine departments vary considerably in size; the two largest (Social Welfare and Inland Revenue) account for about 35 percent of State sector employment.

New Zealand departments represent much more than entries on an organisational chart. Departments are the structure in which managerial discretion is exercised and accountability is enforced, whereas the Votes they administer provide the basis for structuring the budget and managing funds. Each department is responsible for a vast stream of reports, financial statements, and other documents prescribed by law, Ministers, or the central agencies. No matter how small it is, each department is headed by a chief executive, negotiates and implements annual performance and purchase agreements, reports monthly on financial operations, quarterly against purchase agreements and half-yearly on chief executive performance, and produces a departmental forecast report and an annual report. Each must handle the flow of paperwork from Cabinet and the informational demands of central agencies and Parliamentary committees, and each must interact with outside clients and other interested persons. A department must vie for its Minister's time, especially if the Minister is responsible for more than one, as typically is the case with small departments. Thus, while all departments have good reasons for being established, they collectively impose substantial running costs on government. These costs account for a large portion of the operating budgets of small departments.

A large department is likely to have more scope in its operating budget to pay for complying with accountability demands than might a small one. Not surprisingly, managers in small units have griped about these demands. Small departments may have another shortcoming. They may not offer ample career advancement opportunities for staff pigeonholed in small organisational enclaves.

Having numerous small departments generates diseconomies of scale. In addition to accountability and reporting burdens, each department must allocate staff resources to internal management, such as running its own budget and dealing with personnel matters. After these and other responsibilities - which may include monitoring Crown entities - have been spoken for, a small department may have meagre resources left over for policy advice. Perhaps this is the reason why some small departments appear to be listless.

The government may wish to explore means of consolidating some departments or converting them into nondepartmental bodies. (The place of Crown entities in New Zealand's organisational structure is considered in a later section of this chapter.) If the number of departments were reduced, some elements may be recast into Crown entities. Alternatively, New Zealand might explore the feasibility of establishing a new form of organisation that is a hybrid between the core departments and Crown entities. One possibility would be to create executive agencies within departments along the lines of the United Kingdom's Next Steps initiative. These types of agencies would have substantial overall freedom but would be accountable to the Minister and chief executive. They would not have independent governing boards and their budgets would be incorporated in departmental output classes.

It would not be helpful to prescribe a maximum number of departments, but a review of the number maintained in other countries suggests that it may be appropriate to retain somewhat more than half the number of departments currently in place.

Regardless of their number, departments still will constitute the core of the State sector. There still will be great differences in their size, as measured by staff, budget resources, and other indicators. Some are small because service provision has been split off from policy advice; others remain large because policy and operations still are joined together. In New Zealand, the case for decoupling has been based on concern that opportunistic service providers may capture policymakers (including the Minister) by selectively feeding them information. Other countries have acted on different premises in separating the service and policy functions. Sweden long ago separated operating agencies from policy ministries to promote professionalism and political neutrality in the Public Service; the United Kingdom devised Next Steps agencies as a means of freeing service providers from the stranglehold of Whitehall. It wanted operating decisions to be taken by those responsible for providing the services, not by central controllers ensconced in headquarters fortresses. If New Zealand's fear was that Ministers and senior managers would be captured by subordinates (the agent-principal problem), the United Kingdom's concern was that operating managers would be impeded by the top.

In New Zealand, the form of organisation has been decided case by case. For example, the present chief executive of the Department of Social Welfare heads a "coupled" organisation; policy advice and operations are housed in the same department. In her previous post, she divested the Ministry of Transport of its operating units and thereby reduced its staffing from 4000 in 1987 to about 50 now. Justice was a coupled department until it was split into several entities by a 1995 reorganisation; the Department of Labour remains coupled, as are Customs and a few other large departments.

When operations are split off, the surviving department is likely to be involved in much more than advising the Minister on policy. It has ongoing responsibility for internal management and accountability reports and (in some cases) also manages contractual relations between Crown entities and their Minister. This role entails negotiating contracts and monitoring performance, two specialised, labour-intensive tasks.

The reformed State sector relies on written contracts to define organisational responsibilities, allocate resources, programme the work to be done and the priorities to be emphasized, and maintain accountability. Managing these contracts may require a somewhat different mix of personnel skills than has traditionally been found in the Public Service. In some small and medium-sized departments, upwards of one-quarter of the staff is allocated to these tasks and to related activities. The ongoing pressure to retrench running costs gives these departments less room for shifting resources and investing in training than might be available in larger ones. I have been told by managers of some small departments that their operating budgets have been stretched so thin that no money is available for training in newly required skills. Even when per employee operating funds are equal to those of larger entities, small departments may be deterred from making substantial investments in training by the lumpiness of these expenditures.

The large, coupled departments generally appear to be well managed. Some have been reorganised along business lines, with a general manager in charge of each business unit. Each unit has its own performance targets and operates independently of any other units in the same department. The headquarters of these departments have been transformed into corporate offices whose principal responsibilities are to provide policy guidance, monitor the performance of the business units, and manage external relations with the Minister, central agencies, Parliament, and the public. These reorganised departments have achieved many of the benefits of decoupled organisations.

A strong argument can be made either for splitting up a department that is responsible for an array of loosely related activities, or for creating semi-autonomous agencies within it. In these cases, decoupling would give operating units a clearer sense of what their managerial responsibilities are, and they would be able to operate with less interference from headquarters. The old Department of Justice, for example, was a conglomerate in which various activities had little connection with one another. There was no synergy from placing them all in the same department, and some responsibilities lacked adequate attention from policymakers. Transport was another conglomerate which benefitted from hiving off responsibility for various modes of transportation - such as air, water, and roads - into separate entities. When conglomerates are split up, management can be delayered, the lines of communication and control shortened, and responsibilities more precisely defined. The case for divestiture is even stronger when conglomerates mix together regulatory and operating functions or commercial and non-market activities.

But when a department has homogeneous or closely related operational functions and does not both regulate and deliver services, decoupling may be counterproductive. A number of respected observers have argued that separating the policy functions of the Ministry of Defence from the operational responsibility of the New Zealand Defence Forces has increased overhead and transaction costs. The two units have been compelled to build new organisational linkages, so as to provide policymakers with advice from those who command defence forces. Although I do not take a position on defence or any other organisation, in the future the government should decouple departments only after it has considered the feasibility of clarifying objectives and improving performance through internal reorganisation.

Decoupling does not necessarily separate policy and operations. Service providing units tend to "grow" their own policy capacity after they have been split off from the department. It should not be surprising, for example, that a roads authority would deem it appropriate to advise policymakers on the measures that might be taken to reduce highway fatalities, or that defence forces would want to influence decisions on New Zealand's military posture. When this occurs, splitting service provision into a separate entity may have the effect of expanding the range of advice available to Ministers. But even before the reforms were introduced, Ministers had external sources of advice and were not as captured by providers as some feared. The "Yes, Minister" syndrome pertains as much to the disregard of the Minister's preferences in carrying out decisions as in taking those decisions in the first instance. Adding channels for policy advice in operating entities may be beneficial, but the government should be mindful of the elevated running and transaction costs.

Regardless of whether it is integrated or decoupled, a department must have the capacity to provide strategic guidance to operating units. One cannot expect significant strategic thinking to bubble up from below, nor can one expect a department to thoroughly examine its direction if the top is not engaged in this activity. In fact, questioning programmes and priorities, reviewing the allocation of resources to ensure that they promote achievement of key objectives, and evaluating outcomes to determine whether progress is satisfactory and on schedule are among the key responsibilities of top departmental managers. They must assign time and other organisational resources to this work and not allow ongoing operational tasks to crowd out attention to big questions concerning the department's purpose and effectiveness. This matter is treated more fully in chapter 5.

Last modified: