Setting policy targets
Targets to be achieved from public investments often seem to be determined by judgment and a fair bit of guesswork. Is there a better way?
Targets for policy investments are often set in a wishy washy way. Here is an example. Under the regionalised natural resource management (NRM) programs in Australia, the regional bodies that are responsible for planning and prioritising investments were required by the government to establish "aspirational" targets for NRM outcomes. This process injected a huge dose of unreality into the planning process – the aspirations expressed are vastly different from feasible realities in most cases.
There was also a requirement to set "Resource Condition Targets" (RCT), which were meant to be more realistic and achievable. However there was no requirement that these RCTs be determined rigorously. With few exceptions, they were not backed up by good analysis, and in truth, most of the RCTs in "accredited" regional strategies are not achievable.
The setting of targets based on wishful thinking does have its defenders. I've heard people argue that it will stretch people, and give them something to aspire to. However I think it is more likely to make people disheartened and cynical, and it makes realistic monitoring and evaluation impossible. The process seems to have encouraged monitoring and evaluation of activities, rather than outcomes. Given the current mindset around targets, it is not surprising that the regional NRM bodies and governments alike have struggled to get a good monitoring and evaluation process happening.
I think a better approach is to choose targets as the last stage of the planning process. I'd proceed as follows:
n For a range of possible NRM investments, estimate their costs and identify their likely NRM outcomes.
n Select those investments that give the best aggregate NRM outcomes for the available budget.
n Set the targets to equal the most likely NRM outcomes that were identified in the first step for those investments that have been selected for funding.
(The approach assumes that the best available investments are actually worth funding. This may not always be true.)
Note that targets do not drive the process in any way. They are an output of the process.
This approach to planning and target setting flows directly into a good monitoring and evaluation process. The targets are realistic, and failure to meet them can be understood in the context of the analysis that was done to select the targets. Monitoring and evaluation would include updating the analysis and the targets over time as new information emerges. In other words, monitoring and evaluation looks a lot like the original planning process, but adapted to account for new information.
David Pannell, The University of Western Australia, David.Pannell@uwa.edu.au
p.s. 5 Sept 2007
I read with interest your comment (Pannell Discussions No. 108; 27 August 2007) on aspirational targets. “Wishy washy” targets are certainly a waste of time and show lack of thought or understanding, or both. And the concept of “stretch” targets to challenge people may or may not have some application in performance management for individuals (I wouldn’t use them), but are certainly inappropriate in NRM. However, aspirational goals do have value – three aspects are outlined below.
Firstly, the maximum planning period that most people can relate to concerns their grandchildren. In my case, I can work out the planning period based on the possibility that my youngest child, now 18, may have a child when she is 35 that may well live to 80. Therefore, if I care about my potential grandchildren (I do), then my planning horizon is about 100 years. My goals for the environment over this timescale could only be described as aspirational. Couched in these terms, aspirational goals remind us of the investment term we should aim for, and the consequences of failure for those dear to us (most of us will be lucky enough to know some if not all of our grandchildren). I think it can fairly be argued that this type of aspirational goal is valuable – it also neatly picks up inter-generational issues.
Secondly, in planning it is vital to begin the journey with at least a very broad goal to constrain the collection of data and knowledge, and to provide guidance as to which natural resource assets are potential targets. In the first, unpublished SIF report, the planning process outlined is:
1. set broad objectives and goals
2. assess assets and risks
3. set specific goals for action
4. assess options
5. set priorities
6. take action
I had to argue hard for (1), but consider that it is essential for effective planning. On too many occasions the collection of knowledge, capacity building and other process tasks become very comfortable ends in themselves. Although (1) is effectively an aspirational goal that is likely to change with the development of a plan, it is a critical part of the planning process and provides essential direction and boundaries. Even if amended, it is a useful, documented reminder of one’s original intentions.
A particular example of the above is the way we approach planning in recovery catchments. The aspirational goal in these cases is something like “conserving/recovering the 2000 distribution and abundance of specified taxa.” Starting with this aspirational goal, we go through the subsequent planning cycle including defining the assets in more detail, undertaking a feasibility assessment, etc. In the case of the Drummond catchment the aspirational goal is achievable and becomes the operational goal. However, in the case of Buntine-Marchagee, the feasibility assessment showed that the aspirational goal is quite unachievable at this point in time. Consequently, we have developed an operational goal that involves both an amendment to the aspirational goal and related assets. However, the aspirational goal is retained, and remains the measure to which we will return if circumstances allow – it is a reminder of the next step if we attain our current operational goal, and is vital to long term planning and strategic action.
Thirdly, if one did not have an aspirational goal, then there would be little opportunity to take advantage of serendipity in the form of windfall resources, technological breakthroughs, etc. I have witnessed or experienced these on a number of occasions, and the outcomes are quantum leaps of achievement that seemed quite beyond reach. These achievements would not happen if one was always fixed on the currently achievable. Happy to provide examples.
Finally, I assume that when you talk about setting targets last, you are referring to step (5) above. However, many of your audience won’t understand that, and will confuse targets with steps (1) and/or (3). If they do this and delete steps (1) and (3), they will have no basis for establishing the criteria that determine priorities in step 5.
Thanks for this thoughtful response. I think your points are good. There is an element of semantics in this I think, mainly in your first example. I have felt "aspirational" to carry a connotation of loftiness and unattainability, rather than long-term-ness. I think one can have realistic goals over a 100-year time frame. They don't need to be aspirational in that sense of being unattainable, although of course there would inevitably be immense uncertainty associated with them.
I agree that there can be a place for a broad goal to be set to commence the process of planning. It's not a goal that you could be held accountable for, but it could help to set directions. Fine. In the SIF1 example you give, if done well, the resulting targets would be realistic. They would reflect resource availability and the actions that you'll actually take. That is good, and the earlier setting of a broad goal has helped, or at least not hindered. I guess I still worry that if the broad goal is too aspirational (too far from being feasible) then it might not provide a lot of guidance.
My piece was a reaction to the situation that is common in regional natural resource management plans in Australia, where there are broad goals that really are aspirational (unachievable) and then a set of goals that should be the ones they are held to account for (called resource condition targets), but which mostly are also aspirational because they don't account for actual resource availability, likely human responses to the program, or biophysical responses to actions. I'd be happy enough for them to keep a version of the first set of goals (probably harmless, even if they are not actually helpful), but the second set need a complete re-think in my view. (I think that's consistent with what you've said.)
I think you might be overstating the extent to which aspirational goals might help you respond to serendipitous situations. I think you could still respond without aspirational goals. You'd just adjust your realistic goals.
Yes, I'm mainly talking about your step 5.
Pannell Discussions are brief pieces on issues and ideas in economics, science, the environment, natural resource management, politics, agriculture and whatever else.
|107. Why don't farmers use futures markets more? 30 Jul 2007||109. Long jump world champion 3 Sep 2007|
URL for this page: http://dpannell.fnas.uwa.edu.au/pd/pd0108.htm