Cost-effectiveness is all about how you frame the question
Drive out into the rural landscape and you’ll see a variety of plantings of native vegetation. They’ll consist of different species, be of different shapes and sizes, and placed in different parts of the landscape. Indeed, about the only thing they have in common is that they all cost a lot to establish in terms of time and money.
Resources for ecological restoration are limited, regardless of whether the money is coming from the government or being privately funded. So, how do you get the best return for your limited restoration dollar? The answer depends on the type of returns you’re looking for. Appropriately framing your costs and benefits is critical to any evaluation you might undertake. To illustrate this, consider three common restoration activities commonly practiced in agricultural landscapes.
The first is a linear revegetation site or planting, commonly known as a shelter belt or tree line. The second is a block-shaped planting. Contrasting these two activities which involve planting of vegetation, our third example is a patch of remnant woodland that has been fenced to exclude grazing in order to protect and enhance its natural values. For the purposes of this ‘thought experiment’, we’ll assume each project is the same size (1.5 hectares), is being done in the same region, and has been established for 10 years. Of course, the patch of remnant vegetation is much older than ten years but in this case we assume the fence was erected 10 years ago.
The following set of numbers are all illustrative for comparison purposes (as opposed to exact numbers relating to any specific situation). In reality, there are many costs, both public and private, associated with the establishment of a restoration project. Here we focus on the basic establishment costs. Where possible, I have tried to use realistic cost estimates.
Basic costs and opportunities lost
Costs associated with the revegetation projects (linear planting and block planting) include equipment, materials and labour for direct-seeding, as well as fencing materials. The remnant patch, by comparison, only involves a fencing cost. So, if we plug in some broadly representative costs we might get something like Table 1.
If we work out the total cost of restoration actions converted to present value (compounded at 5% per annum over 10 years), and determine a per hectare cost, we can see that the remnant protection project is the cheapest. This is perhaps not surprising given the absence of planting costs with this project. Between our two revegetation sites, note the greater cost per hectare of the linear revegetation project, despite being the same size as the block-shaped planting. This is a result of the cost of fencing and the higher perimeter-to-area ratio of the linear planting. (In many situations the planting is up against an existing fence line cutting this cost in half.)
So, on basic costs, remnant protection comes out as the cheapest restoration activity. However, as these activities are taking place in
‘production’ landscapes, decision makers (which in these situations usually means land owners) need to also consider the opportunity cost of their choices; that is the value or profit from alternative use of that land. In this case, dedicating each of the sites to restoration means that they can no longer generate income from agriculture if we assume no future grazing or alternative income-generating activities.
We now calculate the total cost of our restoration factoring in opportunity cost of lost income from grazing using typical stocking rates and gross margins over the 10 year period (Table 2), with a lower opportunity cost associated with remnant protection reflecting the lower productivity of those sites for graziers. We can see that while our protected remnant remains the cheapest project per unit area, its total cost has risen substantially as a result of the incorporation of opportunity costs (as has that of the block revegetation site). Indeed, the linear planting is now roughly the same as the remnant protection. This is a result of the low opportunity costs associated with linear plantings, which are typically located along fence lines where agricultural production is marginal.It’s one of the key reasons why this shape of planting is overwhelmingly the most common found in agricultural landscapes.
These are our estimated (back of the envelope) costs of restoration. Let’s now consider the effectiveness of each project.
Assume that our restoration objective is to increase the richness of bird species (the number of species) and we conduct a series of bird surveys at each site. We discover that the greatest richness occurs at the woodland remnant, followed by the block planting and then the linear planting. The woodland represents a more established and structurally complex habitat, and we might expect greater richness in the block planting (as compared to the shelter belt) where you might find more edge-sensitive bird species that don’t use the linear planting.
To determine the cost-effectiveness of the three projects, we divide the measure of effectiveness, in this case bird species richness, by the cost to provide a benefit-cost ratio (BCR) – a measure of the conservation benefit per dollar, where the higher the number the greater the cost- effectiveness (see Decision Point #75). When we calculate BCRs we find that the fenced remnant project is most cost-effective, followed by the block revegetation and the linear revegetation sites (Table 3).
Real benefit = conservation gain
While this seems intuitive, calculating the benefit-cost ratio in this way hasn’t really told us what the return is on our restoration investment. Remnant patches often have more bird species in them. Finding that there are more birds in them after fencing them off doesn’t mean the cost of fencing has delivered us the best conservation return.
By comparing the total richness at each site we are only comparing the overall conservation value of each site, not measuring the effect that our restoration actions have had. In this case we concluded that protecting woodland remnant is the most cost-effective because it has the highest species richness and is also the cheapest (per unit area). However we are starting from a much higher base or condition, presumably with a number of species already present before restoration, when compared to a bare paddock where our plantings were undertaken.
A more appropriate measure of cost-effectiveness requires measurement of the conservation gain; that is the increase in bird species richness that has resulted from our restoration action. We can do this by measuring the change in the bird community before and after restoration, or match each restoration site with a control site that reflects the pre-restoration condition. In the case of our two planting sites this would be a cleared paddock, whereas the control for our fenced woodland remnant would be an unfenced (eg, grazed) remnant of the same vegetation type.
Let’s now imagine that we have collected data from our restoration sites as well as paired control sites, and we use the difference (?) in species richness between the pairs (ie, the conservation gain) as our measure of effectiveness (Table 4). Recalculating BCRs for the three projects reveals that our block planting is the most effective, with the greatest increase in species richness, and is also the most cost-effective, despite being the most expensive per unit area.
Of course, this is a ‘simplistic’ thought experiment. In reality, when we consider the conservation of bird species, then a combination of strategies might be best. However, it serves to illustrate how important framing our costs and benefits can be to how we decide where to invest our limited resources. If you were to only factor in upfront costs (and ignore opportunity costs) and benefits that ignore conservation gain (or loss) it’s likely your choices on what’s cost-effective would be misplaced.
More info: Dean Ansell email@example.com
Dean is a PhD student at ANU. He is studying the effectiveness and efficiency of woodland bird conservation in agricultural landscapes and looking at many block plantings, shelter belts and efforts to protect remnant vegetation.
Even in this simple ‘thought experiment’ of three treatments it’s clear that measuring the real benefit of a restoration effort can be tricky, and yet it’s critical to get right to be able to measure cost effectiveness. Here are three Decision Point resources to help you get a handle on this thing called benefit:
Where would you put your conservation dollar?
Phil Gibbons invites you to try his quick quiz in which you choose between restoration options in order to maximise the biodiversity benefit. This simple exercise helps you focus on conservation gain. “I think most people would have intuitively prioritised land that was in the best condition but when it comes to maximising your return on conservation investment the ‘wisest’ choice may not be the obvious choice.”
Decision Point #30
Measuring the effects of conservation management
Mick McCarthy asks what would have happened if it hadn’t happened? The ‘it’ here is the restoration activity. “Determining what would have happened in the absence of management will often be difficult. The best way to do this is to not only monitor the areas that are managed, but also to monitor some analogous areas that are not managed.”
Decision Point #64
When ranking environmental projects
David Pannell has spent much of his career developing systems to help managers and decision makers rank environmental projects by analysing their cost-effectiveness. Last year he distilled his wealth of experience into a twenty part blog series (simply titled Ranking Environmental Projects). You can read the key points of this series in Decision Point #75 (with a link to a single document that brings it all together).