Embedding wellbeing in the Public Finance Act 1989

In September 2018, the Treasury issued a discussion document seeking public feedback on the proposal to include a focus on wellbeing in the Public Finance Act 1989. This is the content of our submission.

1 Proposal to create enduring wellbeing requirements: To report on wellbeing objectives

1.1 Do you agree with the proposal to require the Government to set out how its wellbeing objectives, along with its fiscal objectives, will guide its Budget decisions? Why or why not?

We agree. The proposal will represent a significant development towards the integration of financial and non-financial information, in this case, at a more strategic level.

The Office of the Auditor-General (the Office) supports the evolution of meaningful integrated reporting, at agency, sector, and all-of-government levels. Such reporting attempts to draw links between the money and how that public money is applied to create value for New Zealand and New Zealanders. In that sense, we see the proposal as part of the development of integrated reporting of all-of-government objectives.

1.2 What do you think are the main impacts of implementing this proposal?

If public money is to be spent effectively, then it is important to specify the objectives for that spending (i.e. what the government intends to achieve with the money). We see the wellbeing element as forming part of an overall framework for specifying such objectives, with the wellbeings sitting at the higher end of the scale of outcomes objectives (i.e. what is intended to be achieved).

In that sense, a wellbeing framework has the potential to help ensure that decisions about where to direct government spending, public services, and other interventions will align with the higher level, and ultimate, objectives.

If done well, setting out wellbeing objectives should provide the public with an additional, and more specific, source of information about the economic, societal, and environmental outcomes that the government aims to focus on during its tenure. If it is well understood and well communicated by government, media, and other commentators, it can provide better information to Parliament and the public. That, in turn, has the potential to increase and improve Parliamentary and public discourse, enrich the basis for government’s accountability to Parliament and the public, and ideally improve the effectiveness of public spending.

1.3 What else does the Government need to consider when making the proposed changes, e.g. to retain sufficient flexibility for future Governments to adapt their approach to wellbeing in response to changing circumstances or new information?

We note the Discussion Document’s caveat that embedding a focus on wellbeing will take time, require careful phasing, and will continue to be developed beyond Budget 2019. We appreciate that this initiative will involve an element of experimentation. In that light, we suggest a minimalist approach to embedding the new initiative in legislation; by this, we mean carefully considering how much of the wellbeing initiatives can be introduced without needing to change legislation, proposing changes to the Act only when absolutely necessary and, in those matters for which legislative authority is needed, considering whether it would be desirable to legislate less on content than on process. Allowing a good degree of flexibility in the formative stages of this initiative would enable improvements based on experience and learning to be more easily made.

However, flexibility needs to be balanced with sufficient specificity. One of the risks is that if the information provided in the strategic Budget documents is too vague, or if the links between Budget spending and wellbeing outcomes is not apparent or is too nebulous, then the initiative might be seen as unsuccessful.

2 Proposal to create enduring wellbeing requirements: To report on wellbeing indicators

2.1 Do you agree with the proposal to require the Treasury to report on wellbeing indicators, alongside macroeconomic and fiscal indicators? Why or why not?

We agree that the proposal will represent a significant development towards the reporting of outcome aspirations and achievement at the all-of-government or system level.

The Office supports initiatives that improve how government, and public agencies, set out their performance objectives, explain their beliefs about how the spending of public money will achieve those objectives, and report their progress towards achieving those objectives.

In recent decades, this Office has strongly encouraged public agencies to improve the way they report on how well they deliver their services and how effective those services are at contributing to the outcomes they seek to influence. All wellbeing indicators are attempts to measure the conditions of the world (e.g. economic, social, and environmental) and therefore have the potential to track the achievement of outcomes over time, if well specified.

We appreciate the complexity of the interactive factors that contribute to societal wellbeing, and we acknowledge the limited progress made in outcome reporting by public agencies. We have long supported the development of outcome analysis and reporting at sector and system levels. We see the reporting on wellbeing indicators as part of the evolution in overall government performance reporting.

Augmenting the disclosure of purely economic indicators with the disclosure of specific and measurable wellbeing indicators should better inform the New Zealand public about the state of the nation; it has the potential to enrich public discussion and debate as well as strengthen Parliament and the public’s abilities to hold government to account.

We consider it important that the high-level outcomes objectives expressed in the form of wellbeing indicators are purposefully linked to sector outcomes indicators and/or outcome indicators as reported by public agencies. Acknowledging and accepting the complexity, attribution problems, and limitations in establishing linear causality, the links between all-of-government indicators and public agency indicators are important for at least three reasons:

  1. In the policy analysis and decision-making phases, there needs to be a reasonable basis for proposing how spending public money might eventuate in improving wellbeing.
  2. Public agencies need to understand how their contributions fit into the bigger picture of sector, system, and all-of-government intervention and outcome achievement. Getting a clear line of sight of why a good or service is being delivered is necessary not only to establish a sense of purpose, but also to support broader thinking and problem solving.
  3. The government needs to be able to demonstrate and explain to Parliament and the public its rationale for its spending choices and how well it has achieved what it set out to achieve.

2.2 Which of the options do you think best meets the objective to create consistent, meaningful, and enduring reporting on wellbeing indicators? Why?

The obvious tension in reporting on wellbeing indicators is reconciling the dual purposes of reporting indicators (1) that are consistent over time and independent from the preferences of the government of the day and (2) that are specifically relevant to policies, focus, and initiatives of the government of the day. Indicators under (1) are more useful for assessing the long-term progress of the nation across a wide range of ‘well-beings’ or outcomes. Indicators under (2) are more useful for assessing the performance of the incumbent government against the specific goals it is trying to achieve. (Having said that, we acknowledge that the time lag between intervention and outcome achievement can be considerable, aside from the difficulty and complexity of establishing such links to begin with.)

We consider it useful for all-of-government wellbeing reporting to accommodate both of these perspectives. We would like to see a broad, universal suite of wellbeing indicators that can track the nation’s progress over the longer term. We would also like to see a smaller but possibly richer set of indicators that reflect the government of the day’s priorities and focus, which are linked to the broader suite.

Long-term, universal set of indicators

We agree with the first option suggestion that the choice of what is measured and how it is measured should be independent of Ministers. This is essential to establish and preserve the integrity of the measures by avoiding the appearance of political bias in their selection.

The more open question is who should determine the choice of what is measured and how it is measured. Various degrees of independence and separation from Ministers could include:

  1. Parliament (i.e. the legislative branch rather than the executive branch)
  2. Treasury officials (i.e. non-political)
  3. Non-Treasury officials (i.e. separation from the agency required to report on the indicators)
  4. Non-governmental experts (i.e. separation from government altogether).

Regarding the choice of what is measured, we suggest the Treasury identify the principles that are important in making those choices and apply those principles to various options about who would determine the elements of the annual, long-term wellbeing report.

Regarding the choice, of how it is measured, we suggest this sits better with experts in the field of interest, to the extent that a consensus can be reached on what is most useful and practicable. We expect that public submissions on the draft Living Standards Dashboard would be important inputs to establishing how to measure the elements of the wellbeing report.

We comment on the important properties of performance reports, indicators, and dashboards in general in our response to Question 4.2.

While mindful of the need for consistency and stability of wellbeing indicators to ensure they are useful over time, we appreciate that there is likely to be (and probably should be) some changing and tweaking of indicators in the short term. A limited amount of short-term flexibility is desirable to allow for discussion and debate and some subsequent correction, improvement, and refining of indicators. It is unrealistic to expect the Treasury to establish the ideal set of indicators, and we think it is important for the credibility of the wellbeing reports that the public and subject area experts are able to continue to influence decisions on what is measured, and how, as the reporting is being developed.

There is a trade-off between consistency (which can improve comparability) and continuous improvement (which can improve relevance and reliability). We urge the Treasury to put effort into the continuous improvement of indicators, particularly in the short term; the earlier that a sound and generally accepted set of indicators is established, the sooner the wellbeing report will provide useful tracking data over time. Too much change once the reporting becomes established will pose a risk to the perceived integrity, and possibly the acceptance, of the reporting framework.

To help ensure the integrity of the framework while it is being developed and modified, we recommend transparent “ex-ante” and “ex-post” reporting of the indicators. This demonstrates the commitment to report against those things the Treasury determined it would report on from the outset. When changes to the reported indicators are deemed desirable, then the Treasury should sufficiently disclose what has changed and justify why the change has been made.

Because of the complexity and length of time over which outcomes are achieved, we agree the requirement to report on wellbeing needs to be enduring. To allow for this initiative to mature, and for the government, the Treasury, and other agencies involved in the initiative to learn and develop the area of wellbeing reporting, we suggest again that a minimalist approach (see our response to 1.3) to changing the Public Finance Act is preferable.

We suggest the Treasury look for ways to embed the requirement to report on wellbeing in legislation, while leaving as much as possible about the content and process to regulation or other delegated legislation. Sufficient controls will need to be stipulated to ensure the independence and integrity of the content and process of wellbeing reporting. (One risk to the acceptance of wellbeing reporting would be any perception that its content or process has been captured by special interest groups.)

Medium-term, more focused set of data

We also consider that consistent and meaningful reporting could and should be flexible enough to accommodate the focus of the government of the day.

This could include the incumbent government:

  • identifying particular indicators in the wellbeing report or dashboard indicators as higher priority or areas for special attention over the short, medium, or long term;
  • augmenting selected wellbeing report or dashboard indicators with other, related data from the Information Aotearoa NZ wellbeing database;
  • augmenting selected wellbeing report or dashboard indicators with new indicators or indicators from other sources to provide a broader, deeper, richer, more complete, or more balanced picture of a particular area of focus; or
  • adding other measures of performance, if relevant, such as indicators of changes to service delivery patterns.

We consider that allowing for a universal, long-term suite of indicators to be complemented by a customised set of indicators to reflect government priorities would manage the tension between the two potentially competing objectives discussed above.

Moreover, we consider it would be desirable to provide such flexibility because a universal, long-term set of indicators designed to provide a comprehensive view of New Zealand’s wellbeing is likely to be comprehensive in breadth but insufficiently comprehensive in depth for areas of intense government or public scrutiny.

A current example from a purely fiscal perspective is the Public Finance Act (section 26P) requirement for government to report on gross domestic product (GDP). It has been acknowledged that GDP is a useful indicator or marker of progress over the longer term. However, GDP needs to be augmented by other indicators to provide sufficiently meaningful information, for example, how the GDP was generated, GDP per capita, and how GDP is distributed across regions or demographics.

The same applies to non-fiscal measures of wellbeing. For this reason, we are supportive of a two-track model of a consistent and constant suite of indicators for long-term tracking and accommodation of a more customised suite of indicators for focusing on government priorities.

If it is considered that new indicators are desirable to better measure the wellbeing area of interest, then the question of who should determine the choice of what is measured might be influenced by Ministers. We would expect that how any newly identified outcomes are to be measured would be the product of open and informed debate among politicians, officials, and subject experts, and the public, as appropriate.

2.3 What do you think are the main impacts of implementing this proposal?

As mentioned above, the use of a wellbeing framework has the potential to help ensure that decisions on where to direct government spending, public services, and other interventions will align with the higher level, and ultimate, objectives. Focusing on achieving wellbeing or high-level outcome objectives may help better focus the government’s attention on the reasons why we have a public service rather than on the delivery of outputs as an end in itself. However, as mentioned in our response to 2.1, the strategic use of wellbeing indicators will galvanise public spending and service delivery around those indicators only if there is a clear line of sight between them.

Furthermore, if done well, the setting out of wellbeing objectives should provide the public with an additional, and more specific, source of information about the economic, societal, and environmental outcome areas that the government aims to focus on during its tenure. If it is well understood and well communicated by government, media, and other commentators, it then can provide better information to Parliament and the public. That, in turn, has the potential to increase and improve Parliamentary and public discourse, enrich the basis for government’s accountability to Parliament and the public, and ideally improve the effectiveness of public expenditure.

We should also highlight the caveat that accompanies all discussions of performance measures: the risk of perverse incentives that can arise from fixation on specific performance targets or league tables. Ideally, fixation on a performance target should encourage constructive behaviour aimed at achieving the desirable outcomes. But because the world is complex, and it is neither feasible nor desirable to create performance measures to capture all of that complexity, it is possible that one performance target when pursued in isolation could be achieved to the detriment of other important performance objectives. And not only is it important to try to improve wellbeing outcomes, it is also important how those outcomes are achieved.

Careful attention therefore needs to be given to the means by which the public sector seeks to improve wellbeing, as measured by the chosen wellbeing indicators. The Treasury will need to be aware of potential trade-offs as well as incentives to increase wellbeing indicator scores through behaviour that reduces other aspects of wellbeing, or reduces the quality of service delivery and associated customer experience.

3 Requiring consideration of wellbeing in other documents required under Part 2 of the Public Finance Act

3.1 Do you agree that the Statement on the Long-term Fiscal Position and/or the Investment Statement should be required to have a focus on wellbeing in future? Why or why not?

We consider it could be useful to explore the possibilities for expanding the Statement of Long-term Fiscal Position and the Investment Statement to incorporate wellbeing.

Given that one of the objectives for the current wellbeing initiatives is concern for the inter-generational sustainability of capital stocks, it would be worthwhile exploring the extent to which the four capitals of the Living Standards Framework could be integrated with these two statements.

One of the things that will require careful consideration is defining and distinguishing the reporting entity: When is the Treasury reporting on assets, liabilities, capital stock, and long-term projections for the Government/Crown and when is it reporting on the assets, liabilities, capital stock, and long-term projections for New Zealand (i.e. including privately owned resources and resources not owned or accounted for by anyone)?

3.2 What do you think are the main impacts of implementing this proposal?

The impact will depend on what is reported in these statements. We welcome further consideration by the Treasury and continuing discussion about how the statements could incorporate wellbeing, including stocktakes of the four capitals.

4 Do you have any other feedback?

4.1 Are there any issues not discussed in this document that you would like to bring to the Government’s attention at this stage?

It is important that, to be useful, the wellbeing indicators are credible. We acknowledge that:

  • Some things can be monetised, and some should be monetised but, for other things, there is limited value expressing a monetary value for them.
  • Some things cannot be monetised but can be quantified objectively in non-financial terms.
  • Some things cannot be quantified by direct or objective measures but can nonetheless be recorded and recognised in other ways, for example:
    • “The eye of the citizen”, i.e. citizens’ perceptions of things like happiness, security, or the mood of the nation; and
    • “The eye of the expert”, i.e. observations, assessments, and opinions of experts on specific phenomena for which specialist knowledge is needed.

We acknowledge that, in trying to create a holistic picture of New Zealand’s wellbeing, there may be aspects that fall into the unmeasurable category or a category where measurability is possible but very subjective. Subject to acknowledging the need to be aware of wellbeing aspects that are difficult to capture with hard indicators, we consider that there must be enough reliable substance in the wellbeing reports to provide credible markers of progress. This is necessary to gain the trust of Parliament and the public in the credibility of the indicators and, therefore, provide sound bases for (a) holding the government to account for those things it should reasonably be held accountable for and (b) directing attention to where improvements need to be made.

We discuss at 4.2 below some of the qualities that we consider the wellbeing reports need to have.

We also consider that the Treasury needs to give attention to where in the reporting of wellbeing, when, and by whom it needs to gain assurance over the veracity of the indicator results and to what extent that assurance can be provided from within government or from external and independent sources.

Under section 14 of the Public Audit Act 2001, the Auditor-General is the auditor of every public entity, which includes the Government reporting entity (i.e. the all-of-government accounts as reported in the Financial Statements of the Government of New Zealand (FSG)). The audits of Government departments, Crown entities, and local authorities include the auditing of non-financial performance information. For local authorities, our audits cover not only the “ex-post” financial and non-financial performance statements but also the “ex-ante” statements (long-term plans). This work commonly includes the auditing of outcome indicators, many of which measure some form of wellbeing. (The requirement for local authorities to report on wellbeing has recently been re-introduced.) However, the level of outcome reporting and audit coverage of outcome indicators in public agency annual reports is currently patchy.

Our audit work at the all-of-government level on the FSG involves auditing only financial information because there is no all-of-government performance statement. However, the current proposals relating to wellbeing indicators and reporting against the Living Standards Framework appear to be moves in that direction.

We suggest that the Government and Parliament need to consider the desirability of independent assurance on all-of-government wellbeing reporting and the role of the Auditor-General in providing such assurance. The Government should be mindful of:

  • the Auditor-General’s statutory role as the auditor of the “Government reporting entity”;
  • how an all-of-government wellbeing or Living Standards report might relate to the concept of the Government reporting entity; and
  • whether the Public Finance Act needs any modification to reflect any future role the Auditor-General might have in providing assurance over wellbeing reporting.

Notwithstanding the above, the Auditor-General has discretionary power under section 16 of the Public Audit Act 2001 to examine matters concerning the effectiveness of government activities; this could conceivably include examining the veracity of wellbeing measures reported within government or to Parliament and the public.

4.2 What submissions would you like to make on those issues?

To be credible and useful to all users of the wellbeing indicators, we consider the wellbeing reports need to have the following qualities: (1) relevance, (2) representational faithfulness, (3) understandability, (4) timeliness, (5) comparability, and (6) verifiability.

These are the qualities that are generally accepted worldwide as being necessary for credible and useful financial and non-financial performance reports, of which wellbeing (or outcomes) reporting is a part. These qualities are expounded on in various New Zealand and international professional literature and professional pronouncements and are, indeed, some of the main properties necessary for high-quality communication of information in general.

To maintain those qualities over the longer term, we would like to see formal arrangements for the review and refreshing of the wellbeing indicators to ensure their currency is maintained. As mentioned earlier, there is a trade-off here: a stable time-series is necessary to track progress (comparability over time), so changes should not be made lightly. However, some wellbeing measures may need to be adjusted over time to reflect advances in knowledge and shifts in societal values (i.e. to maintain their relevance and representational faithfulness).

From the Office’s point of view, careful attention should be given to ensuring the wellbeing reports possess those six qualities to the greatest reasonable or practicable extent, particularly if at some stage independent assurance over the reports is sought.