333. Reducing bushfire risks

All the world has seen media coverage of the extraordinary fires that Australia has endured this summer. They sparked an intense discussion and debate about what Australia should do to reduce bushfire risks going forward. What does our economic modelling of fire management say about that?

I had a direct stake in the fires. We have a small cabin in a coastal town an hour north of Perth, and before Christmas we had about a week of temperatures above 40°C (104°F), and a big fire threatened the town. It got to the point where they evacuated everybody, but thankfully the weather improved just enough and just in time for the firefighters to get it under control.

Over the past 8 years, I’ve been involved in a number of projects looking at the economics of bushfire risk mitigation in different contexts. We’ve done case studies in four Australian states (Western Australia, South Australia, Victoria, New South Wales) and in New Zealand.

The different studies analysed different combinations of fire management options. The one common element in all the studies was prescribed burning to reduce fuel loads, but in particular cases we also looked at land-use planning to exclude assets from high-risk areas, retrofitting of houses to make them more fire-resistant, fire breaks, and community education.

It was a mix of research and consulting projects, and some of it still needs to be published.

We found that the economics of bushfires are really complex and data-intensive, and often there is a lack of data to do a bullet-proof analysis. In some cases, we met some of the data needs using sophisticated fire simulation models, but in others, we relied on data synthesis and expert judgments.  Despite the variations in context and methods, we did learn some pretty clear lessons across the various analyses.

The results for prescribed burning were somewhat variable in different contexts, but overall, we found that the long-run benefits of prescribed burning tended to outweigh the costs in most cases.

Interestingly, though, we found that prescribed burning leads to reductions in average losses that are modest relative to the overall losses due to fires. These modest reductions in losses are worth pursuing – they exceed the costs of prescribed burning – but it means that we need to have realistic expectations about what prescribed burning can do for us. Especially in extremely bad (“catastrophic”) fire conditions, losses can still be large, even with the best possible prescribed burning strategy in place.

One of the questions we looked at was whether prescribed burning should be done close to properties or more distantly. Closer to properties has higher benefits, because it increases the chance that a recent prescribed burn will block the passage of a bushfire. But closer to properties also has higher costs, because of the need for additional fire-fighting crews and equipment to reduce the risk of prescribed burns escaping and doing more damage than they prevent. In a detailed analysis of this (Florec et al. 2019), we found that the costs of burning close to houses outweighed the benefits.

In the analysis we did for the Perth Hills, we found that strong land-use planning was the most cost-effective strategy. While it is easy to see the sense in that, it comes up against the reality that some people especially like to live in locations where the fire risk is especially high. We didn’t factor in the likely transaction costs to government in trying to impose a policy that a group of people is opposed to.

Another interesting finding was that a broad policy of retrofitting houses to reduce their likelihood of burning was not economically efficient. Such a policy imposes substantial costs on a large number of houses to avoid the loss of a much smaller number of houses, and we found that the numbers just don’t stack up. Interestingly, new houses in bushfire-prone areas of Western Australia are now required to meet high standards of fire resistance.

Finally, in one case where we had access to a fire simulation model, we looked at possible impacts of climate change on future losses from fires (in 2030 and 2090). In brief, the additional losses due to climate change were large, potentially very large. In the media discussions during and following the recent fires, some politicians were suggesting that bushfire risk is not a reason to pursue stronger policies to mitigate climate change; all we need to do is a better job of prescribed burning. While there certainly can be benefits from prescribed burning, our analysis shows that there is no way increased prescribed burning could come close to offsetting the worsening fire risk from even modest climate change. Indeed, increasing prescribed burning may not even be feasible following climate change, as climate change narrows the window of time within which prescribed burning can be done without excessive risk.

From a bushfire perspective, Australia would have a lot to gain from effective international action to mitigate climate change. This suggests that we should be playing a stronger role in the global climate policy process.

Further reading

Florec, V., Burton, M., Pannell, D., Kelso, J. and Milne, G. (2019). Where to prescribe burn: the costs and benefits of prescribed burning close to houses, International Journal of Wildland Fire https://doi.org/10.1071/WF18192


  • 17 March, 2020 - 6:11 am | link

    Great article and simply explained!

  • Mike Wouters
    17 March, 2020 - 6:35 am | link

    Thanks David, you have summarised the complex situation around the economics of bushfire mitigation very well.

  • David Godden
    17 March, 2020 - 2:53 pm | link

    A couple of comments. The following, albeit anecdotal, report is from someone who experienced a NSW fire early in mid-Sept 2019 (statutory NSW Bush Fire Danger Period normally starts on 1 October and continues through the following 31 March): https://www.gleninnesexaminer.com.au/story/6492701/opinion-we-did-burnoffs-badja-sparks-hits-back/ see also https://www.theguardian.com/commentisfree/2019/nov/11/weve-been-in-bushfire-hell-in-glen-innes-and-the-scientists-knew-it-was-coming
    Note that the second fire burned over previously bushfire-burned ground – this is a comment on what happens in serious fires. I remember Phil Koperberg saying something similar probably 20 years ago. Prescribed (in NSW, hazard reduction) burning is only of limited use in bad, not just catastrophic, fire seasons.
    With regard to possible impacts of climate change. The “window” for hazard reduction burning in NSW in 2019 (and recent years except 2016) was probably just June-July. In NSW at least, climate change is *now*, not something in the future.

  • Jeff Thomas
    17 March, 2020 - 4:47 pm | link

    Very interesting conclusion ‘re close/ distant to houses, David have you done any work on the economics of ecological impacts of various fire regimes on the conservation objectives of the area under question?

    • 18 March, 2020 - 10:12 am | link

      Note that it is still very worthwhile reducing fuel loads close to houses, but using methods other than prescribed burning.

      We have included ecological impacts in some of the studies, but it’s been rather crude and simplistic. I wouldn’t like to draw any definitive conclusions from what we’ve done (which is why I don’t mention it in this post). It’s an area where the experts don’t all agree, which adds to the challenge of producing results we can be confident about.

  • Leave a Reply

    Your email address will not be published. Required fields are marked *

    Please solve this to show you\'re a real person *