On 16th June this year, the Guardian ran an article with what seemed to me to be a remarkable headline, namely “How millions of lives can be saved if the US acts now on climate” with a secondary headline that is possibly even more remarkable: “Researchers have now calculated how many people could be saved from heat-related death if the US takes meaningful action”. And they do mean how many people – exactly how many people, and by US, state and even city.

I followed the helpful link to the Climate Impact Lab’s lives saved calculator with considerable curiosity, and was amazed to find that the people behind it claim to have calculated precisely how much adaptation cost and how many lives can be saved if the country, state or city goes “clean electric” or net zero by 2050.

The calculator is fun (believe it or not, given the gloomy nature of the subject). Do you live in Oakville, California and want to know what can be saved if your city achieves 100% “clean” electricity by 2050? Then you’re in luck, as the “lives saved calculator” will tell you: $62,300 in adaptation costs globally by 2100, apparently, but regrettably, no lives would be saved. Hang on a minute. Nil desperandum, residents of Oakville. Think big. Don’t just achieve 100% “clean” electricity by 2050, but instead achieve net-zero emissions. What then? $449,000 adaptation costs saved by 2100 and, what’s truly amazing, one life saved too!

Not enough? Never mind, let’s think bigger. What about California as a whole? That’s more like it – a saving globally by 2100 of $33.8 billion in adaptation costs and 73,800 lives saved. Or net zero; then 508,100 lives and $243.4 billion by 2100. How about the USA? Well, apparently, if the United States achieves 100% clean electricity by 2050, then 1.6 million lives and $742.2 billion in adaption costs will be saved by 2100. Better still, if the USA achieves net zero by 2050, then those figures become 7.3 million lives and $3.7 trillion in saved adaptation costs. Obviously the costs of going net zero will then be worth it. It’s obvious. Isn’t it?

Methodology

How is this all established with such confidence? Well, the website lets us into the secret of their approach. And we should be impressed and have confidence in it:

The Climate Impact Lab’s team of economists, climate scientists, data engineers, and risk analysts are building the world’s most comprehensive body of research quantifying the impacts of climate change, sector-by-sector and community-by-community around the world.

OK, what does that involve?

The Climate Impact Lab is…leveraging a first-of-its-kind, evidence-based, data-driven approach to quantify the impacts and costs of climate change, sector-by-sector and community-by-community around the world. This research will allow decision-makers in the public and private sectors to understand the risks climate change presents and mitigate those risks through smarter investments and public policy. The research will also produce the world’s first empirically-derived estimate of the social cost of carbon — the cost to society from each ton of carbon dioxide emitted. This figure can serve as the basis for energy and climate policies.

Well, it sounds good, I’ll grant you that, but so far it’s all just words. And if you read on, the website will offer you lots more such words, but nowhere does it set out the precise methodology that offers up such specific numbers, both human and financial. So, to try to work out what’s going on, I had to keep digging, and if I’ve understood things correctly, the important piece of the jigsaw puzzle is an article published on 21st April 2022 and titled “Valuing the Global Mortality Consequences of Climate Change Accounting for Adaptation Costs and Benefits”. In many ways, it’s a hugely impressive piece of work, and I urge you to read it. The problem I have with it, however, is that while the Guardian article and (to a considerable extent) the Climate Impact Lab’s website both deal in certainty, the study on which such apparent certainty is based very properly is in reality full of uncertainty. The abstract alone makes it clear that, for all the talk of empiricism and use of impressive levels of global data, the conclusions remain founded on estimates:

This paper develops the first globally comprehensive and empirically grounded estimates of mortality risk due to future temperature increases caused by climate change…we estimate age-specific mortality-temperature relationships that enable both extrapolation to countries without data and projection into future years while accounting for adaptation….Under a high emissions scenario, we estimate the mean increase in mortality risk is valued at roughly 3.2% of global GDP in 2100 …Finally, we estimate that the release of an additional ton of CO2 today will cause mean [interquartile range] damages of $36.6 [-$7.8, $73.0] under a high emissions scenario and $17.1 [-$24.7, $53.6] under a moderate scenario, using a 2% discount rate that is justified by US Treasury rates over the last two decades. Globally, these empirically grounded estimates substantially exceed the previous literature’s estimates that lacked similar empirical grounding, suggesting that revision of the estimated economic damage from climate change is warranted. [My emphasis].

I take no issue with the authors’ work. Rather my issue is with the relative certainty to which their properly caveated work is put. The “calculator” makes no mentions of estimates. Instead it invites us to “See how reducing emissions in your community improves public health globally.” And when concerned citizens of the United States look to see what impact actions taken in their local community, state or country as a whole might make, they are not offered any caveats; rather, they are told bluntly and simply that achieving “clean electricity” or “net zero” would save x lives and avoid $x million (or billion or trillion) in adaptation costs globally.

The other obvious shortcoming of this approach (or so it seems to me) is that I can’t see that it considers the inevitably huge uncertainties as to what the rest of the world, outside the United States, might do vis-a-vis climate change and emissions policies by 2050. If China, India and Russia “go rogue” (which I venture to suggest isn’t beyond the realms of possibility, given emissions policies to date), then surely all bets are off? Alternatively, in the unlikely event that COP27 (or some subsequent COP) actually achieves its objectives rather than continues as all the talking shops to date, then again I would assume a reassessment of some of the key assumptions might be required.

And, colour me cynical if you will, but I simply fail to accept that it can be stated with such categorical certainty that if (say) Tacoma in Washington State achieves net-zero emissions by 2050 then that alone, without qualification or caveat, will save 3,400 lives and avoid $1.6 billion in adaptation costs by 2100.

Conclusion

The Guardian tells us that this research and the calculator mean that:

A total of 7.4 million lives around the world will be saved over this century if the US manages to cut its emissions to net zero by 2050, according to the analysis.

The financial savings would be enormous, too, with a net zero America able to save the world $3.7tn in costs to adapt to the rising heat. As the world’s second largest polluter of greenhouse gases, the US and its political vagaries will in large part decide how many people in faraway countries will be subjected to deadly heat, as well as endure punishing storms, floods, drought and other consequences of the climate emergency.

Well, maybe, maybe not. Of course, if one accepts that greenhouse gas emissions inevitably mean a warming planet, and if one accepts that this inevitably means that things will get worse rather than better, then no doubt a real reduction in emissions by (rather than merely exporting them from) the USA might save lives and might save money. But it’s a big and complex world, both as regards those assumptions and the geopolitical realities that the USA (and the rest of us) will face over the next 28 years up to that magic 2050 date. Realistically, it can all only be guesswork, and offering this veneer of near-certainty strikes me as massively over-egging the pudding.

Perhaps I’m not alone in my thoughts. So far as I’m aware, in the UK at least only the Guardian seems to have made anything of this story. Even the BBC, which so often reports on studies such as these in tandem with the Guardian, seems to have decided not to run with this one. For once they may be right.

25 Comments

  1. Excuse me whilst the empirical sceptic in me goes to throw up in the bathroom.

    Like

  2. …”and then we all die”. Or perhaps that decimal point we argued about last week really was a poppy seed off your breakfast bun after all.

    Like

  3. Mark,

    I may have much less regard for the Climate Impact Lab’s work than you have. As I see it, the whole thing is junk science. Here are four perceived problems to be going on with:

    Problem 1:

    When it comes to mortality causation, they have no causal model upon which to base their speculations regarding how many lives could be saved by preventing an increase in temperature. What they have instead is a massive database of correlations between temperature extremes and mortality rates and other factors such as wealth and average temperature. Medical factors are assumed to be mediators but these are not modelled. Consequently, they do not have the information they need to make the speculations they are making. For example, if stroke is a major medical factor, and incidence of stroke goes up with temperature, then spending money on preventing heat increases may have a beneficial effect that is offset by the downside of diverting money from dealing with other causes of stroke. How this will pan out depends upon the form of the causal model used. But, as I said, they don’t have a causal model.

    This is a classic example of how people mistakenly believe that Big Data analytics can uncover causation without the need for a causal model.

    Problem 2:

    The range of uncertainty is partly determined by the possible RCPs. They capture this uncertainty by using RCP4.5 to represent the lower risk band and RCP8.5 to represent the upper. Actually, RCP4.5 is a mid-range RCP and RCP8.5 is a fanciful upper range that has no credibility. As a result, the uncertainties are grossly skewed towards the alarmist end of the scale.

    Problem 3:

    Another uncertainty is the range of projections made by the various climate models used. This is handled by averaging their outputs using aleatoric methodology. This is inappropriate since the uncertainties are epistemic. It is a habit that the climate science community seems to be incapable of breaking. Those that make this mistake are either unaware of their error, or they don’t care because they reside in a comfortable consensus.

    Problem 4:

    Much is made of the cost of adaptation and how this should be converted into additional lives lost. Nothing is said regarding the costs of preventing temperature increases. They appear to be assuming that these costs will have no mortality impact. How the grim reaper knows where the money is being spent is not explained.

    Liked by 1 person

  4. Just to elaborate on problems 1 and 4:

    Because they have no causal model by which they can investigate how various interventions may pan out, they have to resort to a very crude measure when it comes to evaluating the morbidity associated with adaptive costs:

    “These projections account for the beneficial effects of adaptation to climate change, but not their costs. Using a commonly accepted statistical metric for valuing death risk, the authors convert these adaptation costs, representing about 13 percent of the total impact, to equivalent changes in the death rate.”

    Suddenly, their ‘empirical grounding’ is nowhere to be seen! They just apply a ‘commonly accepted statistical metric’. Worse still, now they have let the accepted metric genie out of the bottle, they have to apply it when discussing costs of preventing temperature increases. However, all they say is:

    “When mortality costs around the globe are totaled, the researchers find that the present-day value of emitting an additional ton of CO2 is $36.6 per ton under a scenario of continued high emissions…Stated another way, we should be willing to pay $36.6 per ton today to avoid the future mortality consequences of climate change.”

    There is no statement regarding how the ‘commonly accepted statistical metric for valuating death risk’ converts this $36.6 per ton back into deaths. It’s called killing Peter to save Paul.

    Also, remember that this $36.6 per ton is based upon RCP8.5. The figure falls to $17.1 for RCP4.5. What it would be for the lower bound RCPs is not given, because they didn’t bother going there. So much for a team of ‘risk analysts’.

    Liked by 1 person

  5. John, thank you for a dose of expertise combined with common sense.

    It was obvious to me that criticisms aplenty could be made of the calculator and the study on which it is based, but I didn’t feel qualified to do a good enough job. My hope was that by drawing attention to yet more bizarre propaganda someone better qualified than me would comment appropriately, and you haven’t disappointed me.

    Like

  6. Mark,

    The reality is that I am one of those fake experts that John Cook warned you about. Still, you don’t have to be a vet to know the smell of bullshit.

    Richard,

    The vomit started out as theoretical and very nearly turned empirical, but I think it may have been a bout of lactose intolerance that just happened to coincide with some ‘utter junk’ intolerance.

    At first blush, one might expect the empirical sceptic to warm to the ‘empirically grounded’ approach taken by the CIL. After all, they say this in one of their papers:

    “Researchers have struggled to provide a credible understanding of the damages inflicted by climate change at a local and global scale. Available data was limited to wealthy parts of the globe, resulting in insufficiently complete analysis that lacked the scale and scope to credibly describe a global response to warming. Further, there has not been a clear method to quantify the costs and benefits of adaptation. Because of these difficulties, theoretical models have been widely used to capture the global nature of the problem and inform understanding of the global costs of climate change. However, the many untested assumptions implicit in these models weaken the authority of these answers. Thus, to date, there have been no globally comprehensive, spatially resolved, data-driven studies of climate impacts that account for adaptation.”

    All that talk of untested assumptions in models should be making me smile with approval – and it is. However, whilst the CIL is following the data and claiming to resist theories containing untested assumptions, they are still not being empirically sceptical. Firstly, their work is still grounded in the climate models. Secondly, they seem to be fully prepared to accept uncritically the theorising that lies behind RCP8.5. Thirdly, using a ‘commonly accepted statistical metric’ is not being empirical. Finally, not bothering with a causal model but replacing it with a model based upon empirically derived correlations is not my idea of keeping an open mind towards the data. It may give you some insights, but causation won’t be one of them.

    Liked by 1 person

  7. We all love cartoons I’m sure – thanks Joe Public for pointing to one that played such a significant role on Climate Audit in the unveiling of the first tranche of Climategate emails on 17th November 2009. Miracles indeed.

    But what about this cartoon?

    Would I do a main post on Cliscep saying what a good effort it was?

    Er, no. Because it presupposes that there is no cost at all to decarbonisation.

    That isn’t the only reason I struggled with Mark’s main post here but it’s a very big one.

    The others have to do with understanding deeply (I hope) what computer models can and can’t deliver.

    The ‘smell of bullshit’ as John puts it was overpowering in this case.

    This is the Big Lie (as Goebbels used to put it) of climate alarmism. No cost to ‘mitigation’ – yet another misnomer among so many. And that’s now empirical, folks.

    Liked by 2 people

  8. This was my comment when someone used the above cartoon as some kind of riposte to Ben Pile in February 2020.

    The dumbness is numbing yet it’s very dangerous dumbness. This latest paper and software belongs in the same pile of junk.

    Liked by 2 people

  9. Richard,

    To you and I the CIL work may be numbingly dumb, but it isn’t meant to be seen as such. It’s supposed to go like this:

    “You deniers love to harp on about models with their untested assumptions, so we developed a ground-breaking and empirically grounded approach that circumvented their use. And what do you know? The results were even more alarming. You guys just never tire of being proven wrong”.

    It’s like Lewandowsky’s attempt to make our concerns regarding uncertainty backfire on us. It might have worked had he not got into bed with economists who are out of their depth. And here they are again…

    Liked by 1 person

  10. Yep, that’s how it’s meant to be seen. And it isn’t.

    Not seen that way by us. And not actually that way.

    This is what one gets when there are trillions devoted to a false cause.

    It will turn into full-scale corruption of academia if reality doesn’t step in. But it still might.

    Like

  11. I should add that Goebbels accused the Brits of using the Big Lie but never admitted to stooping so low as to use it himself. For more recent commentary on the same see Robert Malone is Fooled from March. I still like Malone but he probably did commit some technical misattribution here. And we all I’m sure care about the posthumous reputation of Herr Goebbels. (Actually, we care about getting all details right, no matter what.)

    Like

  12. Richard, what do you mean by “it will turn into full-scale corruption of academia if reality doesn’t step in”? Whole segments of academia have already been totally corrupted for more than a decade and are becoming more and more strident about it. Even more worrying are the first indications that sceptical views are not to be tolerated in academia and without, and will be actively legislated against in the not too distant future. Our reality may be doomed for some considerable time. A new dark age?

    Liked by 1 person

  13. Alan: I see remaining pockets of resistance plus the reality-check of the disastrous impact of the policies flowing from a corrupted academy, leading to alternative centres of learning springing up. But these may not be enough, I grant you that. A new dark age is a real possibility.

    Like

  14. Eg

    Universities devoted to the unfettered pursuit of truth are the cornerstone of a free and flourishing democratic society.

    For universities to serve their purpose, they must be fully committed to freedom of inquiry, freedom of conscience, and civil discourse.

    In order to maintain these principles, UATX will be fiercely independent—financially, intellectually, and politically.

    from the University of Austin in Texas. Commentary from Spiked in The University of Austin puts the rest of academia to shame and from one of the founders, Niall Ferguson, in the first hour of a long podcast with bright spark and robotics coder Lex Fridman. Both over six months back. I don’t know how they’re getting on.

    Like

  15. Richard plans for the U of T at Austin look promising. I was once invited for a interview there for a position in the School of Geology. At the time this must have been one of the best geoscience schools in the world. I had been recommended by a retiring eminence, but when I arrived I was told by a University Vice Chancellor that they only had funds for the most junior of positions which ruled me out. Set in one of the most pleasant parts of Texas (landscape and climate) a good site for an academically -free institution. But who judges what should be taught?

    Liked by 1 person

  16. Joe, I never saw that from Josh. He was one of those who retweeted my tweet above, presumably before penning the cartoon. That is brilliant, the way it makes the link between dumb climate obsession and pandemic unpreparedness. Thank you.

    Liked by 1 person

  17. Mike, Thanks for the Stanford Uni link

    quote – “Stanford University has been dealing with a power outage for three days now. The power first went out on Tuesday afternoon, after a fire damaged a transmission line.

    According to the San Francisco Chronicle, one reason it’s taking so long to restore power is that the university’s power is now completely renewable, mostly solar, and most of the solar power comes from plants in Kern and Kings counties.

    The university has some power, from generators, but does not have enough to fully turn on.”

    great Advert for the Net Zero future !!!

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.