Deciding how to govern Shared Activities

The diagram below shows the major elements involved in deciding how to govern shared activities.

A. Defining the problem

In each instance where governance arrangements are required, three basic questions should be asked:

- Why is a shared activity being developed/used (e.g. achieve economies of scale, spread capital requirements, mitigate risk, support a shared output, increase capability etc)?

- What is to be governed, and what activities will be required under a governance regime (e.g. ownership, contract management, standards setting, pricing and cost recovery, negotiation with suppliers, dispute resolution etc)?

- Who is involved in the activity, or will depend on it somehow? (e.g. agencies, Ministers, customers, citizens, other stakeholders etc)?

Thought should be given to whether, when, and how all the stakeholders in a shared activity should be involved in deciding how it is to be governed.

B. Finding the right solution

Options

There cannot be a 'one-size-fits-all' approach to governing shared activities. What matters is that some reasonably robust process has been followed in choosing a governance model. The two options considered under this approach are:

• model: agencies participating in the activity are part of a collective governance environment.
• model: a single decision-maker governs the activity on behalf of its participants.

Under the unitary model, there are two options for choice of decision-maker - a central agency, or some other lead agency. In general, a central agency would be an appropriate choice where a shared activity will have very wide usage (e.g. the government portal) or is very generic (e.g. the New Zealand Government Locator Service metadata standard and the supporting Metalogue™ system for creating and storing metadata records). It is conceivable that this could mean establishment of a new entity to govern an activity.

Hybrid arrangements are possible. For example, under the club model the club may decide to allocate decision-making rights to a lead agency.

Club vs. Unitary model - how to decide?

In practice, the two options are likely to be very close to one another. For example, it is probable that a unitary decision-maker would establish some form of consultative body representing the interests of the participants in the activity in question. It is also possible that a club will appoint a lead agency for a variety of reasons.

Where they are distinct is in the area of decision-making rights. Under the unitary model, rights and responsibilities would reside with an individual agency, even if exercised by another party (e.g. the e-GIF Management Committee [The State Services Commissioner is the Steward of the e-GIF, and the Committee acts on his behalf.] ). Under the club model, even if delegated to a lead agency, these rights and responsibilities would still ultimately reside with the club.

A range of factors influence which option is the best choice in any particular situation. One obvious requirement is that agencies' individual accountabilities should either not be compromised, or be open to being modified to reflect use of a shared activity. Other questions include asking who is truly accountable if there is a failure, who is paying for the activity, and how cost-effective different governance arrangements will be.

A Treasury paper [Treasury, Draft Paper,Towards a Centralisation/Decentralisation Framework, May 2002.] suggests that the following a set of issues might also be considered when allocating decision-rights:

• information asymmetry
• economies/diseconomies of scale and scope;
• incentives;
• risk management;
• agency participation;
• coordination and consistency; and
• constitutional issues.

The tables below set out a series of questions that can be used to look at each of these issues when deciding how a shared activity should be governed. It is not an exhaustive set of questions - what is presented is indicative, not prescriptive. Not all questions need be used, and others relevant to a particular issue could supplement or replace them.

What is most important is that in reaching a decision, some questions of this type are asked, and the results are documented so that stakeholders can see for themselves what concerns and logic has informed that judgment.

Applying the questions

Most of the questions are not closed. It is possible that answering these questions will lead to an ambiguous answer. They should, however, enable a general preference for a unitary or a club approach to be formed. There may also be a preference for a hybrid arrangement (e.g. a unitary decision-maker working through a club, or vice versa). Making a final choice will be a matter of judgment for those who are party to the decision-making process.

Guiding principles and constraints

Once there is a preferred governance arrangement more questions need to be asked. First, the proposed solution should be evaluated against a set of governance principles. This can help test the broad suitability of a governance proposal, and help in the design of detailed governance arrangements (e.g. design of terms of reference for governing bodies, processes for dispute resolution, management of accountabilities, etc). A final test of any proposed governance arrangement is that it not be in conflict with any major constraints.

As with the questions provided above, the principles and constraints suggested below are indicative. Agencies using this approach should test them for their relevance, and adjust them as appropriate.

C. Implementing the solution

Once a preferred option has been identified, the key to its success will lie in details of its implementation.

Implementation analysis will be the stage at which attention is given to matters such as:

• giving effect to governance principles;
• addressing constraints;
• ensuring that the proposed governance option meets requirements;
• ensuring that incentives are in place or are addressed, and
• engaging stakeholders in the design process.

There will also be a need to design the ways in which the preferred option will be put into action. This will involve developing things such as:

• terms of reference for governance bodies and/or decision-making processes;
• advisory/user/stakeholder groups;
• strategies and/or business plans;

- contractual or quasi-contractual instruments (e.g. service level agreements, charters, memoranda of understanding, interoperability agreements);

- agreements over allocation of roles and responsibilities (e.g. data steward/custodian);

- disputes resolution processes;

- funding arrangements (e.g. central vs. departmental, membership fees vs. user pays, capital arrangements/balance sheet issues etc); and

- design of formal Cabinet mandates (if required).

Once this design work has been done, it is assumed that the whole package of proposals for governance of the activity in question will pass through any formal process of consultation and ratification that may be required. It is also assumed that the final arrangement will be periodically re-evaluated to ensure it remains fit-for-purpose.

The diagram below shows all the steps in the decision-making approach.

Conclusion

The key assumptions informing this approach are that:

- E-government will enable/require agencies to share or integrate a variety of information and technology activities;

- sharing activities requires good interagency governance arrangements;

- the essence of governance is the allocation and exercise of decision-rights;

- there are two broad options for making decisions about shared activities - the club model and the unitary model;

- the process for deciding which model suits a particular situations is as important as the outcome.