Outcome Budgeting

Posted on May 16, 2019

Outcome Budgeting (OB) was announced by the NSW Government in the 2017-18 budget and is part of an international trend to reform public sector budgeting systems and decision making. This explainer will help you understand what OB is, how it’s applied in NSW and what problems it may cause in the future.

What is Outcome Budgeting?

At its most basic level, OB is a budgetary system that focuses on the outcomes of funded activities instead of the quantity of activities. OB was first popularised in the early 1990s by David Osborn and Ted Gaebler (1992) in their book Reinventing Government.

Outcome Budgeting in NSW

The NSW Government is currently implementing OB into the budget process. Two key factors are driving this change.

It increases accountability by providing a common framework for ministers, government agencies, non-government providers and the community to gauge the effectiveness of public services.

It helps the Government use performance-based and outcomes-focused measures to gauge the most effective way to achieve outcomes for the people of NSW.

The NSW Government has put forward 46 Cabinet-endorsed state outcomes, which cover the total budget spend. By doing this, it has shown that it intends to integrate program performance and its impact on state outcomes into budget decision making.

These state outcomes can be explored using an interactive graphic prepared by NSW Treasury. It lists the outcomes by sector and details what expenditure has been allocated to each of them.

The approach being used by the NSW Government requires cross-agency collaboration, to ensure outcomes are not owned by any one agency. The following figure demonstrates the practical features of this approach to OB:

For more details on how these four elements of the OB framework will be implemented, check out the NSW Treasury Policy Paper on OB.

Outcome Budgeting challenges

The process of incorporating OB into government decision making is complex, partly because of a number of challenges with implementation and maintenance. These include:

  • the identification of outcomes and the attribution of cause-effect links
  • the creation of perverse incentives
  • its use in decision making
  • data systems and capability.
The identification of outcomes and the attribution of cause-effect links

Shifting the emphasis in budgeting from activity to outcomes is laudable. However, it is often easier said than done.

Firstly, specifying the outcomes of government agencies is not always simple. And finding a causal link between government activity and those outcomes is almost impossible. For example, what is the effect of the hospital system on the health of the population? Or the Environmental Protection Agency on the quality of the environment? Or the police force on the incidence of crime?

For these outcomes (and indeed many of the 46 outcomes listed in the NSW framework), no one agency is solely responsible for achieving them. For example, improved health is a function of housing standards, level of education, social class and a range of other factors beyond the remit of the health system.

For this reason, it is often difficult, if not impossible, to attribute changes in these outcomes to individual agencies. Hence the grand claims that OB will lead to a fundamental reallocation of resources to the most effective government actions are often overstated.

The difficulties in identifying outcomes and causal attribution means that OB often lapses into OUTPUT budgeting because outputs (the ‘products’ of government activities) are easier to measure. For example, the number of graduates produced by a university.

It’s interesting to note that one of the 46 outcomes in the NSW framework is a reduction in waiting times in emergency departments. This is of course a measure of activity, not a health outcome!

Perverse incentives

This brings us to the second challenge: the risk of perverse incentives. A recent book by Jerry Muller (2018) called The Tyranny of Metrics provides a number of examples across government where agencies try to game the system to attract or retain funding.

Reclassifying types of crime, teaching to the test for standardised testing of school students, or not performing surgery on patients with complex medical problems are all examples of perverse incentives. This scene from The Wire speaks to these perverse incentives quite well.

Decision making

Another challenge with OB relates to its use in decision making. There are often grand claims about how OB will help governments redirect resources from activities that are producing poor outcomes, to those that are more effective. While this is certainly attractive (who could argue against reducing funding for government activities that don’t work?) it is harder in practice and may not even be desirable.

At its worst, it conjures up the image of a bunch of technocrats twiddling dials, suggesting that government decision making and the allocation of budgets is a purely technical exercise, when in fact it is a highly political, contested and messy business.

We also need to consider whether it’s fair that a community doesn’t get access to particular resources, because the previous allocation wasn’t producing expected results thanks to poor design and implementation by government.

Data systems and capability

A final challenge relates to data systems and capability. Putting aside the above concerns, there are real doubts as to whether governments have the data systems and analytical capabilities to implement OB effectively.

For example, in 2018 the NSW Department of Finance, Services and Innovation conducted a data skills survey involving 1,156 NSW public sector employees. While the survey found that 94% of staff would benefit from using data in their work, these results are contrasted by the significant struggles they face in carrying out data-driven work.

Out of 12 possible answers, ‘accessing the data’ was nominated as the greatest data challenge faced by NSW public sector employees. This was followed closely by ‘using data’ and ‘quality of available data’. In another question, only 19% of survey participants chose to self-identify as a data professional.

Nexus has extensive experience, at whole of government and organisational levels, in adopting outcome-oriented budgeting systems to facilitate better policy making and resource allocation. In doing so, we are to our attuned the inherent challenges and risks involved.


  • Osbourne & Gaebler (1992). Reinventing government: how the entrepreneurial spirit is transforming the public sector. Plume, New York.
  • Muller, J 2018. The Tyranny of Metrics. Princeton University Press.

Useful Resources