Share:

The performance measurement advice that government gives its entities causes the biggest measurement problem in the public sector: trivial activity measures.

Everyone wants government organisations to perform well. Unlike the private sector, they can’t be judged by profit. But we all want to know that the tax we pay is being put to good use.

Notoriously, public sector organisations don’t measure their performance well. From my 2017 study, ‘How many organisations have meaningful KPIs‘, I found that only 6% have measures that provide direct evidence of their strategic goals.

In part it’s because public accountability is tough. To avoid the distraction of dealing with reactive public scrutiny, government leaders can resort to vanity metrics and measures of how much activity they’re doing. They will avoid measurement of the real results their organisation exists to create for the community.

But another reason why public sector organisations don’t measure performance well is likely due to the advice they get from their regulators about how and what to measure. This advice unwittingly sends them in the wrong direction:

1. To believe that strategic objectives are too high-level to measure, rather than making strategic objectives measurable.
2. To focus on what is within the organisation’s control, rather than the social results it exists to influence.
3. To measure activity, rather than measuring results.

We will see this problem clearly in the advice given in the Australian Government’s Resource Management Guide 131 (RMG-131) for developing good performance information. It exists to support the PGPA Act for Australian Government entities. But the same goes for many other countries’ government advice for performance measurement.

If public sector organisations are going to truly improve and demonstrate the value they generate with taxpayer’s money, a few things need to be fixed in the advice they follow for developing performance measures.

Fix 1: Strategic objectives can be high-level and measurable.

The first fix to the advice government gives itself on how to measure performance is to provide guidance on how to measure strategic objectives. Any change we intend to make should be observable, and therefore should be measurable. Our strategic objectives describe intended changes, and can be written measurably with a little deliberate effort.

Item 15 in RMG-131 says that because an entity’s purposes (which they define in item 16 as strategic objectives) are naturally high-level, performance measurements should be based on lower-level objectives derived from it’s purposes.

The problem with this advise is that it essentially advises government organisations to not bother looking for ways to directly measure their strategic objectives, but to focus on measuring their activities.

And this leads us to the next problem: measuring only what they can control, not what they should influence.

Fix 2: Public sector organisations exist to influence, not control.

The harsh truth is that no entity can have control over everything it wants to change. Control assumes things are predictable. But our world is somewhere between predictable and unpredictable. And that’s the realm of influence, not control. The most meaningful performance measures for improvement of effectiveness and efficiency (government loves these words) are measures of influence, not control.

Item 20 in RMG-131 suggests government entities use the Logic Model to more easily measure the things they have direct control over. They allude to the notion that they have too little influence over the outcomes they contribute to, to measure their impact on those outcomes.

The problem with this advice is that government organisations will resort to measuring only what they have direct control over. And what they have direct control over is what they choose to do; how they spend their allocated budget. They assume that what they have chosen to do will, indeed, positively impact the outcomes they contribute to.

This, again, leads to another problem: measuring the doing of activity, not the achievements from activity.

Fix 3: Measuring activity means measuring the results of that activity.

In PuMP, we do measure activity. But it’s the results of activities that we design performance measures to evidence. Activities are just smaller parts of the larger organisational system of processes, functions, goals and purpose (as described in the Results Map, activities are in the outer layer).

Item 17 of RMG-131 does suggest that performance measures should focus on what the activity is intended to achieve. But the measure examples given include many actions or measures of how much action, like these:

• “Effective administration of investment in road infrastructure”
  (what should be quantified to know how effective?)
• “Coordinate the National Road Safety Action Plan 2018-2020 through the Council of Australian Governments’ Transport and Infrastructure Council”
  (what is the result of this that should be quantified?)
• “Number of site visits undertaken nationally”
  (what is the result of a site visit that should be quantified?)

The problem with the advice of item 17 is that it’s not reinforced with consistent and proper examples of measures of activity results. If you take just 10 minutes to look at your own strategic plan, or that of any public sector organisation you immediately think of, chances are you’ll find too many action-oriented measures and too few results-oriented measures.

***

Is the return on our taxes quantified by how much activity government does with those taxes? Where is the evidence of how it’s improved the quality of life for everyone? Isn’t that the evidence we most want them to use to spend taxes ever more wisely?

Do government organisations measure their performance with activity measures, because they focus too much on what they can control, and not on what they should influence? [tweet this]