Feb 24, 2010

Did the insulation program actually reduce fire risk?

Has the Garrett insulation scheme actually reduced the rate of installation caused fires? It’s a strange thing to say – well, it’s strange if you don’t think about it too hard.

Possum Comitatus — Editor of Pollytics

Possum Comitatus

Editor of Pollytics

Has the Garrett insulation scheme actually reduced the rate of installation caused fires? It’s a strange thing to say – well, it’s strange if you don’t think about it too hard. What we often forget is that Garrett’s insulation program dramatically increased standards in an industry where there were previously very few.

First off, some raw numbers. If we go fossicking over on the ABS site, we find some data about the numbers of buildings with insulation in Australia in 2008 – the year before the Garrett insulation program. We’ll use these numbers as our “before” data set.

61% of 5,218,000 dwellings – or 3,183,265 – had insulation in 2008 according to our favourite Bureau. According to Garrett’s department, there were also between 65,000 and 70,000 new installations done per year before the insulation program. Let’s split the difference and call it 67,500 installations.

We also know from Garrett’s department in the same letter than between 80 and 85 fires occurred every year because of insulation. That’s the numbers we need for the pre-Garrett program.

Moving on to the Garret program numbers, we know that during the program, 1.1 million installations were undertaken. So that’s our yearly installation rate and it pushes the number of houses currently with insulation up to 4.28 million homes.

We also know that 93 fires have been linked to the program.

There’s our “after numbers” we need to do the comparison.

With the 2008 figures on the numbers of fires, we don’t know how many of them were as a result of new installations and how many of them were caused by insulation that was already in the ceiling and may have been there for years. So what we need to do is differentiate between the fires caused by the insulation stock (the pre-existing insulation) and new installations.

To do this we’ll create three scenarios:

Scenario 1: 10% of fires were caused by existing insulation and 90% caused by new installations

Scenario 2: 50% of fires were caused by existing insulation and 50% caused by new installations

Senario 3: 90% of fires were caused by existing insulation and 10% caused by new installations

This gives us a range of all possible likely outcomes. If we measure the number of fires caused by new installs using these three scenarios for 2008, and for our known number of 93 for Garrett’s program, this is what we get.


From here, we can figure out how many fires there were per new installation in both 2008 (where there were 67,500 new installs) under it’s three scenarios, and with the Garrett program (where there was 1.1 million installs) – simply by dividing the number of installations undertaken by the number of fires attributable to new installations.


Under Scenario 1 where 90% of fires are attributable to new installations, 1 in 909 installs lead to fire. Under Scenario 2 it’s 1 in 1636 installs lead to fire while under Scenario 3 it’s 1 in 8182 installs.

Under the Garrett insulation program, the rate is 1 in 11,828 – a much smaller rate of fires than what existed before the program.

The Insulation Program Safety Multiple is simply the Garrett program rate divided by the 2008 rate – it shows how many times safer the Garrett program is compared to each of the three scenarios for 2008.

Interesting – that’s what the data says.


I borked a link back up there earlier. Here’s where the 65-70K installations per year for the pre-Garrett period were mentioned by the Dept head.

(Visited 5 times, 2 visits today)


Leave a comment

98 thoughts on “Did the insulation program actually reduce fire risk?

  1. New Information on the Home Insulation Scheme « Journal-ism

    […] data is the most recent, but ol’ Possum over at Crikey crunched the number in February last year and came up with similar results. He crunched the statistics and laid down some interesting ideas, […]

  2. Kevin Herbert

    Around 80% of all ceiling insulation installed in Australia since 1992 – the year of the introduction of AS3999: 1992, the bulk insulation installation standard – has been wrongly installed, thereby reducing all greenhouse gas reduction/energy savings forecasts by at least 50%…..what a joke, particularly for those consumers who’ve shelled out big dollars for ‘energy efficient homes’.

    I hear sirens..I hear Slater Gordon….I feel a a class action coming on….but who to sue? Is it the builders who flogged these white elephants, or the State Governments who endorsed the enegry efficiency schemes… or both?

  3. Insulation Fire Risk – The data is in – Pollytics

    […] What we found was that under every possible scenario, the government insulation program – far from increasing the rates of fire occurring from installing insulation – actually reduced the rate of fires and likely reduced the rate in a quite substantial manner. […]

  4. Oz Politics and Perspectives | View From the Headland

    […] actually to Garrett’s credit that the level of risk per installed household has been calculated to be about one tenth of what it had been with installations before the scheme […]

  5. John Bennetts

    Remember, not ALL of the 4 deaths were the result of electrocution. Two were heat stress. Thus, Poss, the fatality figures are halved, from one perspective.

  6. Did the Australian Home Insulation Program reduce the rate of insulation-related fires? « Compromise and Conceit

    […] I will also attempt to give some explanations for this. This builds heavily on the work of Possum at Crikey, but with the addition of a time-dependent element to the analysis, a wider range of assumptions […]

  7. The Anti-Stimulus Brigade | The Daily Bludge – Independent News

    […] 1 million homes were insulated. The anti-stimulus brigade would have you think that only stimulus related insulation carries the risk of fire, which is an absolute lie. Most people accept that the benefits of insulation outweigh the risks… Possum’s analysis again…… […]

  8. runim

    So, in effect, while there is no doubt that appropriate insulation reduces the energy requirements (and hence also the cost and GHG), of heating a fixed area, human behaviour may, over time, negate these savings.

    And thus further action needs to be taken to counter this behaviour.

    Such as rises in the price of energy – as the CPRS proposes ?

  9. fatking

    Back to Terry Williamson and the Khazzoom-Brookes postulate as it likely applies to energy use and its impact on GHG generation.

    As the breathy article in the Australian makes clear, Williamson is predicting a spike in household energy use. That is a hypothesis. Notwithstanding the rhetorical use of the present tense in the article, it is clear he bases his prediction/hypothesis on what is now almost two decades of attempts to address household heating and cooling energy consumption, primarily through improved insulation standards. Such an assertion is checkable in the public record, and is therefore empirical evidence. Since Bacon, we have more or less agreed that such evidence cannot be refuted by lofty Aristotelian rationality as attempted by Eponymous.

    Williamson’s point is simple. The evidence of the actual causality between improved insulation standards in Australia, and reduction in either household energy use, or GHG contribution by households in Australia is missing.

    Dig deep enough, such as periodic reporting by BASIX NSW, and you actually find tentative admissions of overall increases, not decreases in energy use, whether measured per household or per occupant. Those figures are always, always, masked by reference to improvements in ‘energy density’ of houses, measured as MJ/m2. The anomaly is that this ties the problem to a general increase in house sizes in Australia over the last twenty years, which both ‘explains away’ the increased energy use, but actually explains the improved ‘energy density’ as well. It is therefore inconvenient for policy makers, and quite a few ‘techies’ as well, to take on board Williamson’s emphasis on the impact of our increased expectations.

    But Williamson also introduces other contentious issues, concerning reliance on modelling, rather than empirical observations. He specifically challenges whether the thermal modelling itself is robust, not just whether it correctly deals with human behaviour. Again, if you think about it, he bemoans the lack of properly funded research. Considering the trivial cost of such research, especially in comparison to moneys spent on programs that are arguably a demonstrable policy failure, I would say he has a point.

    Disclaimer: I am the other academic referred to in the Australian article.

  10. Tom the first and best

    Did the rebate also contribute to less heating fires by meaning many people use less heating (including dangerous heating) particularly in the warmer months?

  11. Cometh the hour, cometh the FAIL – Pure Poison

    […] number of fires rose from around eighty to 120. In other words, as Crikey’s psephological blog Pollytics has demonstrated convincingly, there is no statistical evidence that the existing problem of fires […]

  12. Did the Media Get it Wrong on Batts-Gate? – The Content Makers

    […] the number of fires rose from around eighty to 120. In other words, as Crikey’s psephological blogPollytics has demonstrated convincingly, there is no statistical evidence that the existing problem of fires […]

  13. A mess? A shambles? A disaster? | Inside Story

    […] number of fires rose from around eighty to 120. In other words, as Crikey’s psephological blog Pollytics has demonstrated convincingly, there is no statistical evidence that the existing problem of fires […]

  14. Musrum

    [email protected]

    There are three things of note for me in this whole discussion:
    1. As a small community we need to find some way of broadcasting this to wider community.

    This helps: Insulation fire risk was worse before rebate

  15. Non-mainstream media acknowledges the realities of workplace fatalities « SafetyAtWorkBlog

    […] outrage against then-Environment Minister, Peter Garrett, are seriously questioned in this blog article at […]

  16. moneypenny

    Great post and discussion here. A welcome antidote to all the MSM hand-wringing. Pity KRudd doesn’t appear to read your work, Poss.

    I’d love to know how many house fires are caused less by dodgy installation of insulation and more by dodgy downlights. I think downlights should be banned – they are relatively dangerous and very energy inefficient. (If it’s good enough to ban the incandescent light-bulb…)

  17. caf

    It may (or may not) interest you to consider that 121 workers lost their lives on the Snowy Scheme.

  18. moo

    Possum, there’s one thing that you’re really not taking into account within your analysis, the time since the insulation had been installed.

    What everyone must realise is that 93 fires occurred in a very short space of time, i.e. since the insulation program started. If it weren’t rectified, that number would increase dramatically. There are still possibly around 1000 electrified roofs due to badly installed insulation, among the other issues that would raise the potential for fires and other dramas. If the program hadn’t have been stopped and been rectified, the number of fires that would have occurred due to the program would probably be over 500.

    Although there is a small rate of fires now, the number of shoddy insulation install jobs is much greater than this, which means that fires and other problems will occur at a much greater rate given time.

  19. Tri$tan

    2. It probably won’t matter anyway.
    3. The actual OHS responsabilities of this issue are interesting – but have not been explored properly anywhere I’ve seen.

  20. Tri$tan

    There are three things of note for me in this whole discussion:
    1. As a small community we need to find some way of broadcasting this to wider community.

  21. vp

    Poli ani

    Fairly forthright clientele on your blog, then?

  22. political animal

    The weekend AFR confirms (not that it needed it) Possum’s figures.

  23. Diogenes


    You are correct. The Monbiot book was where I first read about it.

    BTW What is neo-classical economic theory and what has replaced it? I know next to nothing about economics.

  24. Gaffhook

    The hypocracy of all this media and Liberal party beat up is untenable. For them to be trying to blame a Minister and be ably aided and abetted by the screaming pack of untrained hyena excuses for journalists is just apalling. How many Ministers have been held accountable for the deaths of armed forces personnell during peacetime training activities. Lots of whom died because of faulty equipment or aged equipment.
    Voyager, blackhawks, APC rollovers, Navy helicopters, fire on tanker (WA)
    It was never the Minister, the blame was always scapegoated or sheeted home to Pilot error, faulty equipment/maintainence, lack of proper equipment.

    It was usually followed by the disclaimer that there is always a risk when doing these things and sometimes things go wrong. Our heartfelt thoughts go out to the grieving families. Nothing more to see here let us move on. Next!

  25. Yaz


    The Khazzom/Brookes hypothesis is an interesting one, but of course, is true only in neo-classical economic theory, thus not all that true at all.
    I can’t be bothered looking for it now, but I think it was Monbiot in his book Heat who investigated this a bit further and found that it was true, a little, and I think found that it reduced the effectiveness of some energy efficiency measures by about 10-20% (not 100%), if memore serves. Happy to be corrected by anyone who has the book right in front of them…

  26. Possum Comitatus


    I had a borked link in the post – sorry about that folks. It was the source of the 65-75K from the opening statement to the committee by Dept Sec Kruk. I’ve fixed that up.

    Ta for those that pointed it out.

  27. Diogenes

    The Laws of Thermodynamics tell us that heat transferral will be less with insulation, except for the 30% of batts which apparently don’t make any difference.

    The Laws of Thermodynamics are silent on what humans will do when they have insulation in.

  28. Eponymous

    No evidence to show that putting in insulation reduces energy consumption?

    Apart from the 3 Laws of Thermodynamics.

  29. Diogenes

    Actually it’s not proving a negative. The hypothesis is that putting in insulation reduces energy consumption. There is no evidence to show that the hypothesis is correct.

  30. Eponymous

    Not a skerrick. But I’d rather not prove the negative.

    It seems based on the concept that people understand that they are saving energy by having insulation installed. Further, it’s based on the concept that once insulation is installed, people will suddenly find colder/warmer temperatures more comfortable. I find the assertion ludicrous in the extreme.

  31. Diogenes


    Do you have any evidence that it is incorrect? The postulate is well accepted and operates in many cases. The question is whether it applies in this case.

  32. Eponymous

    No offence intended personally Diogenes, but that is one of the worst pieces of crap I have ever heard.

  33. Diogenes


    I posted this on PB but you probably know more than about how much credibility this phenomenon has amongst economists so I’ll lob it here as well.

    Hedley Thomas has a really interesting article on insulation. I don’t think it’s an argument against the scheme but it mentions a fascinating phenomenon. The guy argues that putting in insulation actually INCREASES energy consumption, or at least doesn’t reduce it.

    [ In practical terms, comfort creep or rebound, insofar as insulation is concerned, means that people who have had it installed in their ceilings then go further, boosting their airconditioning on hot days and their heating in winter. In doing this, their energy consumption increases.]

    The guy probably doesn’t know it, but he is proposing the Khazzoom-Brookes postulate which is

    [ In the 1980s, the economists Daniel Khazzoom and Leonard Brookes independently put forward ideas about energy consumption and behavior that argue that increased energy efficiency paradoxically tends to lead to increased energy consumption. In 1992, the US economist Harry Saunders dubbed this hypothesis the Khazzoom-Brookes postulate, and showed that it was true under neo-classical growth theory over a wide range of assumptions.]

  34. Grog

    That said, I’m not suggesting she was doing any actual reporting on the issue – this was just her personal opinion.

  35. Grog

    Actually she did call me a “C-Grade Political Hack”. Which is fair enough I guess, but I did kind of hope I was at least worthy of B-Grade status.

    I’ve worked hard over the years on my hackery.

  36. Possum Comitatus

    Maybe Mark – I get paid to do this, that’s the level of basic competence that is expected of me. If not from Crikey, certainly from you lot who’d literally kick the seven shades of brown stuff out of me if I didnt 😛

  37. Mark Heydon

    Glad to hear there was a basis for the split. This is a key part of the analysis – you should have included it in the original post.

  38. Possum Comitatus

    But but but Grog! “You have to be a C-grade moron” not to understand that she’s right!


    People died, houses burned – oh the humanity!

    Where does FM radio find these people.

  39. Cuppa

    Possum, Your blood’s worth bottling. Best blogger in Australia.

  40. HellsBells

    thanks for some excellent analysis.
    The MSM jocks should hang their collective heads – and seek alternative employment.
    Abbott and his frat house colleagues should reflect on how many businesses and families now face straitened circumstances after the untimely cancellation of the program.

  41. Grog

    In the interests of accuracy her name is Bree Roberts, not Robert. btw here was her follow up:

    [no, point is the scheme failed. They didn’t have the resources to do the job properly. Four people died. Houses burned.]

    Shakes head, sighs…

  42. Grog

    Here’s an example of journalism in this country. “journalist and newsreader” for Nova FM Bree Robert tweeted Rudd “had to admit the whole thing was a disaster and just fix it”

    I tweeted her, asking if she had read the Senate Committee hearings transcript or the Minter Ellison report. This was her reply:
    [No, I’ve not read it. Why would I? It might have been a good idea if Mr Garrett read it though.]


  43. Grog

    Fuck me the media has displayed a high level of mendacity on this issue.

    I mean, even Kerry O’Brien tonight thought the Minter Ellison report meant something. Do they even know what a risk assessment is? Can they read English?

    Pretty soon, they’ll switch to the “perception is bad” line – it’s always the refuge of journos who don’t care about facts (or perhaps don’t understand them) – They’;ll start writing “Garrett may have put in place strong risk minimisation procedures, but the perception is that he failed, and in politics perception is vital”.

    Makes me want to vomit.

    The best things about Gillard putting a high focus on improving education standards is that maybe in 20 years we’ll have some more intelligent journalists than we have now.

    It wouldn’t be hard.

  44. Michael James

    Terrific work, and I’ve got to say that I suspected as much all along. You should write this up in a slightly more compact and reader-(ie. dumb Australian)-friendly and offer it to the MSM. Because unfortunately on Crikey it only confirms what the tiny thinking reader suspected anyway.

    Let us also not forget that 3 of the electrocutions happened in the state that is home to Poss. (and me): of course that state is a paradigm of strictest standards with nary a cowboy to be found. Hmmm. Turns out wiring inspections in ceilings more or less stopped decades ago when the whole thing was outsourced by government. Instead of wiring being tacked to the SIDES of beams, a lot of the time they are just loosely lying on TOP. As usual accidents came about from a conjunction of several shoddy practices–the wiring mess, the lazy installers, dumb fad for hot halogen downlights and ultimately, sadly less than observant youthful workers.

  45. Diogenes


    If you look at the national figures which show 15 work electrocutions per year in Oz, at 70K installations a year that’s 13×15= 165 electrocution deaths during the period of time it took before the program to install the 1M homes.

    I’d be staggered if a few of those 165 electrocution deaths weren’t related to ceiling activities like batts. All you would need is 1 installation electrocution death in the last 13 years to make 4 under Garrett not statistically significant.

  46. Possum Comitatus

    I didnt Diogs. Probably should have in hindsight – not that I could split the data that way, but might have been interesting.

  47. Diogenes


    Did they df between new home new installs and retrofit new installs?

  48. Possum Comitatus

    I do have a basis Mark.

    I spent yesterday morning consulting a number of industry reps, fire services people and a recently retired forensic guy who has not only worked with police and coroners to pinpoint the origins of fires, but also insurance companies. I asked them out of the fires that, in their experience, were likely caused by insulation, what proportion was from recently installed insulation.

    The figures varied – but all gave estimates substantially greater than 10% and substantially less than 90%. Hence my margins.

    I didnt know what margins should be used initially for the sensitivity – so I went and got some advice.

Leave a comment

Share this article with a friend

Just fill out the fields below and we'll send your friend a link to this article along with a message from you.

Your details

Your friend's details