A closer examination of the fantastical numbers in Bjorn Lomborg’s new book
Watch Bob’s two minute Amazon review of False Alarm
Bjorn Lomborg’s new book False Alarm: How Climate Change Panic Costs Us Trillions Hurts the Poor, and Fails to Fix the Planet attempts to convince readers that the impacts of climate change have been exaggerated, particularly by the media, and that much of the current effort to tackle rising greenhouse gas emissions represents an over-reaction.
He has been characteristically energetic in persuading right-wing newspapers, particularly those owned by Rupert Murdoch’s News Corp, such as The Wall Street Journal, the New York Post and The Australian, to advertise his book for free in their opinion columns.
But, like his previous contributions to this issue, Dr Lomborg’s arguments are based on fantastical numbers that have little or no credibility. Overall, the numbers presented by Dr Lomborg, who has a PhD in political science, understate the potential economic impacts of climate change and exaggerate the costs of cutting greenhouse gases. And he has promoted them apparently secure in the knowledge that they will not be fact-checked by book publishers or newspaper comment editors.
This commentary details five examples of Dr Lomborg’s misuse of outdated, concocted and misinterpreted numbers that are central to his ‘lukewarmer’ argument in the following ways:
- Ignoring the costs of fossil fuel subsidies
- Illegitimately doubling the cost estimates of action by the European Union
- Misrepresenting the impact of the Paris Agreement on climate change
- Cherry-picking an outdated model estimate of the costs of climate change impacts
- Miscalculating the ‘optimal’ level of global warming.
Ignoring the costs of fossil fuel subsidies
Dr Lomborg frequently complains about subsidies provided to new clean technologies, but he often fails to mention the scale of support that fossil fuels receive. For instance, he states on page 110: “New renewable energy sources like solar and wind cost $141 billion annually in subsidies globally, and matter little in the global energy supply.” On page 173, he states: “Globally, in 2020 taxpayers will pay $141 billion to subsidize inefficient solar and wind energy.”
The book’s endnotes indicate that Dr Lomborg has estimated these figures from a graph on page 256 of the World Energy Outlook 2018, published by the International Energy Agency (IEA). The graph shows projections of total support for renewables for power generation rising from $143 billion in 2017 to a peak of about $300 billion in 2035 in its ‘New Policies Scenario’.
However, Dr Lomborg does not find any room in his book to inform readers of the much larger scale of subsidies for fossil fuels. Page 55 of the same publication by the IEA states: “We estimate that artificially low prices for fossil fuels for end-users around the world involved subsidies totalling just over $300 billion in 2017.”
Another study, published by the International Monetary Fund in 2019, concluded that all implicit and explicit subsidies for fossil fuels would have reached $5.2 trillion, equivalent to 6.5 per cent of global GDP, in 2017. Almost half of this figure is due to the fact that the prices of fossil fuels do not take account of the death and illness caused through local air pollution, an issue that Dr Lomborg overlooks.
It can be inferred from the omission of these figures from his book that Dr Lomborg seemingly has no complaints about the much larger scale of subsidies for fossil fuels!
Illegitimately doubling the cost estimates of action by the European Union
In his desperate attempt to persuade readers that cutting greenhouse gas emissions is far too expensive, Dr Lomborg throughout his book doubles cost estimates that he finds in the literature relating to policies by the EU to cut greenhouse gases.
For instance, on page 110 in Chapter 7, Dr Lomborg states: “If the European Union sticks to its climate promises for 2050, it alone could end up paying more than $2.5 trillion per year in climate costs – 10 percent of its entire GDP. This is more than all the EU’s current spending on education, health, environment, housing, defense, police, and courts. It is inconceivable that such spending will go unchallenged.”
The corresponding endnote states: “The average estimated cost of the EU 80 percent greenhouse gas emissions reduction by 2050 is a loss of 5.14 percent of GDP as estimated by seven regional models (Knopf et al. 2013). This assumes all policies are perfectly effective. More realistically, the costs will double, as they did for the EU’s 20-20-20 climate policy (Bohringer, Rutherford and Tol, 2009). That leads to a 10.3 percent cost, or €2.514, or $2.8 trillion.”
But a close reading of the papers cited by Dr Lomborg shows that he has concocted these figures through distortion of the findings of the academic studies that he references.
For instance, the 2009 paper by Professor Christoph Böhringer and co-authors presents model projections, rather than actual measurements, of the possible gross costs (i.e. without taking account of any benefits in terms of avoided climate change or local air pollution) of the EU reaching its target of cutting its annual emissions of greenhouse gases by 20 per cent by 2020 compared with 1990. It presents these costs in terms of impacts on welfare, so the GDP figures presented by Dr Lomborg are entirely his own invention.
Professor Böhringer and co-authors report that three model runs show that reaching the target through a single carbon price across all economic sectors for the whole of the EU would result in a loss of welfare of 0.45, 0.67 or 1.98 per cent in 2020. The mean of these results is 1.03 per cent, which shows that Dr Lomborg has wrongly assumed that welfare and GDP are exactly the same. The authors explicitly point out that GDP is an “incomplete” measure of welfare.
Professor Böhringer and co-authors note that the results of a 2008 study for the European Commission estimated that the “total energy system cost” of achieving the 2020 target would be equivalent to 0.50 per cent of GDP in 2020. They state: “When implemented at the lowest possible cost, climate policy is not expensive.” Predictably, Dr Lomborg does not mention this in his book.
The authors use the models to estimate the costs of reaching the 2020 target through other simplified policy combinations, including the EU Emissions Trading System (EU ETS), carbon pricing in the sectors outside the EU ETS, and targets for renewable energy deployment. They show that these other policy combinations tend to increase the overall welfare losses. In the worst case, the three models suggest that the welfare losses will increase by 67, 98 or 127 per cent.
However, Professor Böhringer and co-authors warn against misinterpretation of their results, stating: “The above results should be treated with caution. The numbers are neither accurate nor precise. They are ballpark estimates. What really matters are the insights: Climate policy need not cost a lot, but imperfect implementation implies excess costs. The excess costs are substantial relative to the costs of the first-best policy, but modest in absolute terms.”
So it is clear that Dr Lomborg both misinterpreted the results of Professor Böhringer and co-authors and used them in ways that the authors warned against. Furthermore, Dr Lomborg has ignored subsequent research and evidence that the costs of reaching the EU’s target may not have been as expensive as suggested by the 2009 paper.
For instance, a further study by Professor Böhringer and a group of other co-authors, published in 2016, used models to calculate that the combination of policies being used to achieve the EU’s target for 2020 would result in a loss of consumption of between 0.17 and 0.83 per cent compared with ‘business as usual’. Dr Lomborg ignored these far lower numbers.
The 28 EU Member States actually reached their collective 2020 target in 2014, and total emissions were 23 per cent lower in 2018 than in 1990, according to the European Environment Agency. In particular, emissions covered by the EU ETS fell by 29 per cent compared with 2005, against the original target of a 21 per cent fall by 2020.
The carbon price in the EU ETS has remained much lower than assumed in the 2009 paper by Professor Böhringer and co-authors. They expected the carbon price to be between €36 and €72 per tonne and the European Commission expected the price to be €49 per tonne in 2020. In fact, the price has never been more than €30 per tonne since the publication of the paper and was less than €10 for much of the period. This suggests that at least some of the costs of the policy were lower, not higher, than expected, a possibility that is completely overlooked by Dr Lomborg.
Dr Lomborg also misapplies his misinterpretation of the findings of Professor Böhringer and co-authors to other studies, including a 2013 paper by Dr Brigitte Knopf and co-authors. Dr Lomborg claims that that paper found that “the average estimated cost of the EU 80 percent greenhouse gas emissions reduction by 2050 is a loss of 5.14 percent of GDP as estimated by seven regional models”. This percentage figure is invented by Dr Lomborg – nowhere does it appear in the paper by Dr Knopf and co-authors. Previously, Dr Lomborg claimed that the study found that the cost by 2050 would be equivalent to 11.9 per cent of GDP, another figure that he made up without any explanation.
In fact, the paper by Dr Knopf and co-authors concluded: “Nearly all the models can achieve the long-term target of reducing [greenhouse gas] emissions by 80%, with only a moderate reduction in GDP (less than 0.7% by 2030 and below 2.3% by 2040). However, in some models, costs increase considerably after 2040, while others show costs increasing in a linear manner.” This is ignored by Dr Lomborg.
Furthermore, there is no justification for Dr Lomborg’s decision to double the costs he purportedly derived from the study by Knopf and co-authors. When I asked Dr Knopf, the lead author of the study, about this, she replied in an email message: “Concerning the doubling of costs: this is clearly not appropriate, because as I said it mixes model reality in an arbitrary way. This is similar as if I would claim: Model results show costs of x% but we know that models always underestimate progress in renewable deployment, so we can assume that is only costs half. These claims are clearly not based on solid ground and have nothing to do with the model results.”
Misrepresenting the impact of the Paris Agreement on climate change
A claim that has been made frequently by Dr Lomborg is that the Paris Agreement will have almost no impact on climate change. He has done this by wrongly suggesting that the ‘nationally determined contributions’ (NDCs) for emissions cuts by 2025 or 2030, which were submitted by countries before the United Nations climate change summit in Paris in December 2015, represent the sum total of action in relation to the Agreement. Dr Lomborg usually assumes that countries do not increase their emissions cuts after 2030 and that many allow their emissions to grow until the end of the century. Based on this assumption of 70 years of rising emissions after 2030, Dr Lomborg claims that the Paris Agreement would make very little difference to warming by 2100.
As I and others have pointed out many times to Dr Lomborg, this is highly misleading because Article 2 of the Paris Agreement commits every Party to the Paris Agreement to the objective of “[h]olding the increase in the global average temperature to well below 2°C above pre-industrial levels and pursuing efforts to limit the temperature increase to 1.5°C above pre-industrial levels, recognizing that this would significantly reduce the risks and impacts of climate change”. It is widely known that collectively the first set of NDCs are not consistent with this goal, but countries are due to submit more ambitious pledges by the end of 2020 and many, including the UK, already plan to cut emissions to net zero by 2050.
Dr Lomborg turns a blind eye to all of this to continue his attack on the Paris Agreement in False Alarm, where he uses exactly the same trick of assuming that countries do not cut their emissions any further after 2030. In Chapter 8, he writes:
“What will happen, then, if nations meet their promises under Paris? The United Nations organizers of the Paris Agreement once in 2015 (and never since) released an estimate of the total maximum impact of all carbon dioxide cuts promised by all nations. It provides the absolutely best-case scenario that we can hope for. This estimates a total reduction of 64 Gt carbon dioxide through to 2030. According to the UN’s estimate of 0.8F per 1000 Gt carbon dioxide, this translates to a reduction in temperature by the end of the century of about 0.05F. What this tells us is that even in an optimistic scenario, the Paris Agreement isn’t going to come anywhere close to solving global warming. It will have a miniscule impact on the temperature by 2100.”
In essence, this means is that Dr Lomborg assumes that countries cut their emissions by 2030 in line with their NDCs, but then in 2031 immediately increase their emissions to follow ‘business as usual’ paths until the end of the century. And his imagined 70 years of higher emissions dominates the calculation of temperature change by 2100, which would likely reach about 4˚C in Dr Lomborg’s fictional world. This is not a legitimate assessment of the intended impact of the Paris Agreement, which explicitly aims to hold global warming to well below 2˚C.
Cherry-picking an outdated model estimate of the costs of climate change impacts
Dr Lomborg’s new book draws very heavily on the work of Professor William Nordhaus, who shared the 2018 Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel for his work on climate economics.
Professor Nordhaus is a pioneer in his academic discipline and has published many influential papers during his career. He also devised the ground-breaking Dynamic Integrated model of Climate and the Economy (the DICE model), which produces estimates of the costs of action and inaction on climate change.
Recent studies have criticised the DICE model, particularly when used with its ‘default settings’, for underestimating the costs of climate change impacts. But Dr Lomborg wholeheartedly embraces the model and devotes much of his Chapter 5, ‘What is global warming going to cost us?’, to a discussion of its outputs.
Dr Lomborg writes that “a temperature increase of 7.2˚F [4˚C] around the end of the century would mean a reduction in global GDP of about 4 percent (actually 3.64 percent, but let’s call it 4 percent)”. The endnote to justify this claim about the impacts of warming of 4˚C refers to a 2018 paper by Professor Nordhaus in which he provides details of analyses using the DICE-2016R2 version of his model.
Dr Lomborg accompanies his account with a graph showing that DICE-2016R2 estimates that global warming of about 14.2˚F, or nearly 8˚C, would cost the equivalent of 15 per cent of global GDP. In fact, DICE-2016R2 in its default settings indicates a reduction in global GDP of 50 per cent when warming approaches 15˚C above pre-industrial temperature.
It is worth noting that the last time the Earth was more than 2˚C warmer than the preindustrial temperature for an extended period was about 3 million years ago during the Pliocene Epoch, when the polar ice caps were much smaller and global sea levels were 10 to 20 metres higher than today. It is difficult to be confident about the costs associated with such a transformation to a climate that modern humans have never experienced, even if it takes place over many centuries.
Such a small estimate of the economic cost of reaching 4˚C of warming by the end of this century is also at odds with the results of scientific research on the impacts of climate change. For instance, the Summary for Policymakers of the contribution of working group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC), published in 2014, stated:
“Increasing magnitudes of warming increase the likelihood of severe, pervasive, and irreversible impacts. Some risks of climate change are considerable at 1 or 2°C above preindustrial levels. Global climate change risks are high to very high with global mean temperature increase of 4°C or more above preindustrial levels in all reasons for concern, and include severe and widespread impacts on unique and threatened systems, substantial species extinction, large risks to global and regional food security, and the combination of high temperature and humidity compromising normal human activities, including growing food or working outdoors in some areas for parts of the year (high confidence). The precise levels of climate change sufficient to trigger tipping points (thresholds for abrupt and irreversible change) remain uncertain, but the risk associated with crossing multiple tipping points in the earth system or in interlinked human and natural systems increases with rising temperature (medium confidence).”
A new paper by Dr Martin Hänsel and co-authors, published in July 2020, points out that the DICE-2016R2 model makes outdated assumptions about the carbon cycle and energy balance model, describing the relationship between increasing greenhouse gas concentrations and temperature rise, and incorporates a damage function, describing the relationship between warming and economic impacts, that is not consistent with the latest findings by many researchers, such as this paper by Professor Marshall Burke and co-authors.
Miscalculating the ‘optimal’ level of global warming
Dr Lomborg goes on to use the DICE-2016R2 model to calculate the so-called ‘optimal’ level of global warming at which the combined costs of cutting greenhouse gases and residual damages from climate change are minimised. Professor Nordhaus used the model to make his own calculation in his 2018 paper that the optimal level of global warming would be 3.5˚C by 2100 and 4˚C by 2150. As noted in the previous section, this result is completely at odds with the scientific research on climate change impacts.
However, Dr Lomborg is not prepared to accept Professor Nordhaus’s calculation, particularly his assumption about the costs of cutting greenhouse gases. On page 161, Dr Lomborg states: “To adjust for the real world, we can run the numbers through William Nordhaus’s model again, but this time allow for policy costing twice as much.” He attempts to justify this instant inflation of the costs in the preceding paragraphs, stating that “the European Union has seen its climate change policy cost double from the most effective level it could have been, because politicians didn’t make the most efficient choices and couldn’t help but dish out favours to some industries over others”. Predictably, the endnote for this claim cites the 2009 paper by Professor Böhringer and co-authors.
Dr Lomborg continues: “Similarly, much climate policy in many US states, including New York, California, and Hawaii, is based on the renewable portfolio standard, a regulation that according to economists makes it at least twice as expensive as the most efficient policy would have been.” The endnote for this claim refers to a paper by Dr David Young and Dr John Bistline that was published in 2018. The abstract of the paper states: “We find that, in most cases, renewable portfolio standards are approximately twice as costly as the equivalent least-cost portfolio for achieving CO2 reductions, although the ratio can be much higher for standards with lower abatement.”
However, it seems that Dr Lomborg did not read that paper thoroughly. The authors compare the costs of attempting to achieve reductions in annual emissions of up to 39 per cent by 2030 compared with 2016. The paper finds that for low levels of emissions reductions, it is cheaper to replace coal with natural gas rather than new renewables as a source of electricity. However, the costs start to converge for higher emissions reductions that require greater deployment of renewables. The authors also point out that costs decrease significantly if a cap and trade system is assumed to operate alongside other policies, including the renewable portfolio standards.
Essentially, Young and Bistline show that the cost of emissions reductions in the United States is expected to be higher than necessary because of the lack of stringent federal policies to cut emissions, a conclusion that is ignored by Dr Lomborg. And the results of their study certainly do not seem generalisable to all other climate policies across the world, as Dr Lomborg suggests.
Nevertheless, Dr Lomborg persists with his doubling of the global costs generated by DICE-2016R2 and as a result calculates that the ‘optimal’ level of global warming by the end of this century would be even higher than Professor Nordhaus suggested in his 2018 paper. Dr Lomborg writes on page 162: “The optimal temperature rise is now 6.75˚F [3.75˚C] …because realistic climate change policies are more expensive.” Warming of 3.75˚C would be well outside the experience of modern humans and would likely create huge risks for society, as the IPCC and others have warned.
Unfortunately for Dr Lomborg, his calculation has been rendered completely obsolete by a study published in July 2020 that points out that the default settings for the DICE-2016R2 model incorporate outdated assumptions about the science and economics of climate change. The paper, by Dr Martin Hänsel and co-authors, concludes that “the UN climate targets may be optimal even in the Dynamic Integrated Climate–Economy (DICE) integrated assessment model, when appropriately updated”. The authors state: “Changes to DICE include more accurate calibration of the carbon cycle and energy balance model, and updated climate damage estimates. … When updates from climate science and economics are considered jointly, we find that around three-quarters (or one-third) of expert views on intergenerational welfare translate into economically optimal climate policy paths that are consistent with the 2 °C (or 1.5 °C) target.”
Furthermore, the DICE model does not take into consideration the potential co-benefits of climate action, particularly a reduction in the local air pollution caused by consumption of fossil fuels that leads to premature deaths and illness. As the recent study published by the International Monetary Fund concluded, nearly half of the US$5.2 trillion that fossil fuels received globally in direct and indirect subsidies in 2017 was due to their prices not taking into account the damage that they cause through local air pollution. If this economic co-benefit was taken into account it would significantly affect the calculation of the optimal level of warming, as a paper by Dr Noah Scovronik and co-authors pointed out in May 2019 – a paper that has been completely ignored by Dr Lomborg.
Hence Dr Lomborg’s central claim that the optimal level of global warming by 2100 would be 3.75˚C was completely out-of-date before his book was even published.