Climate Change and the $25 Million Free Lunch of the Dwarves

There is a disturbing passage towards the end of the final book in C.S. Lewis’ Narnia septet:

Aslan raised his head and shook his mane. Instantly a glorious feast appeared on the Dwarfs’ knees: pies and tongues and pigeons and trifles and ices, and each Dwarf had a goblet of good wine in his right hand. But it wasn’t much use. They began eating and drinking greedily enough, but it was clear that they couldn’t taste it properly. They thought they were eating and drinking only the sort of things you might find in a Stable. One said he was trying to eat hay and another said he had got a bit of an old turnip and a third said he’d found a raw cabbage leaf. And they raised golden goblets of rich red wine to their lips and said, ‘Ugh! Fancy drinking dirty water out of a trough that a donkey’s been at! Never thought we’d come to this.’ But very soon every Dwarf began suspecting that every other Dwarf had found something nicer than he had, and they started grabbing and snatching, and went on to quarreling, till in a few minutes there was a free fight and all the good food was smeared on their faces and clothes or trodden under foot. But when at last they sat down to nurse their black eyes and their bleeding noses, they all said: ‘Well, at any rate, there’s no Humbug here. We haven’t let anyone take us in. The Dwarfs are for the Dwarfs!’

‘You see,’ said Aslan. ‘They will not let us help them. They have chosen cunning instead of belief. Their prison is only in their own minds, yet they are in that prison; and so afraid of being taken in that they can not be taken out.’

Never mind Lewis’ peccably orthodox theology of hell, rejecting the universalism of his inspiration George D. Macdonald and of the Cappadocian Fathers. The Dwarfs’ banquet is a pretty good allegory of the mess we are in over climate change.

Paolo Veronese, The Wedding at Cana

Paolo Veronese, The Wedding at Cana

High-res image link. You will need a big cinema screen to see it life-size, the original is 6.7m x 9.9m.

Brad Plumer, in a solid and much-cited Vox post, sheds gloom on the premature celebrations of success in Paris. The “intended national contributions” (INDCs) add up, says Climate Tracker, to cutting global warming from 3.6 deg C by 2100 to 2.7 deg C. That is far too high, seeing that the 2 degrees limit agreed at Copenhagen is reckless: it has a 50% chance of being just acceptable. Normal prudence argues for 1.5 degrees. He’s dead right there. The INDCs are agreed to be just a down-payment, to be ratcheted up later – this will be in the agreement. Plumer underlines that the ratcheting up isn’t a marginal tweak, but a huge effort, more or less doubling what’s already on the table.

The knowledgeable and acute Plumer is still a paid-up member of the Beltway political class, and shares two of its biases. One is paying too much attention to what politicians say as opposed to what they do. The other is a neglect of the impersonal forces of economics and technology, combined with the small decisions of billions of ordinary people.

The first bias is at work in his taking the INDCs at face value. These present the emissions of the USA and the EU on a sedately declining path. Those of China will peak close to 2030; India has not accepted any absolute cap at all, and plans a massive expansion of coal-burning along with its dramatic push for wind and solar power (160 GW by 2022). A few other middle-income countries like Turkey offer similar scenarios, but those are the two that count. It’s bad news, but let’s take a closer look.


Last November’s agreement between Obama and Xi was hugely important politically. Both countries accepted hard emissions targets, and so opened the path to a successful agreement in Paris.

The content of the undertakings was far less impressive. The US promise of a 26-28% decrease in emissions from 2005 to 2025, implemented mainly through the Clean Power Plan regulations on coal fossil power generation, only represented a continuation of the trend decline. On the Chinese side, the undertaking was only to peak emissions “around 2030 and to make best efforts to peak early”. The pessimism of the scenario is extreme. The recent absolute decline in coal output suggests to me that China’s emissions will peak before 2020 and may have already done so. endnote. Optimism is reinforced by the brutal effectiveness of the government’s coercion of local officials: some district heating systems were shut down in 2010 to meet energy and pollution targets (pdf, page 5).

I am not the only observer concluding that China’s carbon emissions will peak soon. Deutsche Bank think 2016, Bernstein Research think this year (ibid, page 7). Xi could safely have promised a much earlier emissions cap. I will take a small bet that he will do so, either at Paris or – better theatre – at or just before the top-level signing junket the diplomats are planning for the spring of 2016. Will they try for U2 or the Pope or both? No bets on that one.

The other political effect was that both leaders became more committed domestically. Policy announcements have speeded up. Obama has not only made good on the CPP, but killed off oil exploration in the Alaskan Arctic. Xi has raised China’s 2015 solar target to a staggering 23 GW, equal to all the panels installed in the world in the first 60 years of the technology (1954-2009).  This determination affects the perceptions of businessmen and the public. American electric utilities have not joined the reflex Republican lawsuits against the CPP, which they can live with.


India still claims it will double coal-burning to meet its growing electricity demand. It also stands by its anti-colonialist rhetoric that it’s now the turn of the poor former colonies to trash the playground. Closer up, the coal plan looks more than shaky. Two large Indian resource companies, Adani and Reliance, have walked away from large coal projects they were developing in Australia and Indonesia, and in Reliance’s case, also from a huge Indian power station. The last time India tried for a massive boost in coal-fed electricity was in 2005, with the plan for 16 “ultra mega” power plants. Today, two are in partial operation and no others are under construction. Ten others are paper only, marking time in the face of a sceptical business community. The dominant coal supplier, the state-owned behemoth Coal India, is no better than British Rail at keeping to timetables. (Greenpeace pdf, page 15; EIA chart update).   Geological common sense and rising wages suggest that the days of cheap coal in India are numbered. State politicians get much more out of solar and wind, which give quick results and put money into the dhotis and pockets of in-state voters and businesses.

The coal policy looks incoherent and incompetent. But Narendra Modi is one of those rare politicians for whom the conspiracy theory is a better prior than cockup, as exemplified by Trey Gowdy. He is both very unscrupulous and very intelligent: Lloyd George more than Mussolini. I’d be surprised if he expects the coal plan to succeed, or the solar one to fail. In that case India’s emissions will grow much less than its INDC. In five years, India will have a large and low-cost domestic solar and wind supply chain – the policy to achieve this is careful and already working. Modi can then follow Xi into nobly announcing his emissions cap.

The great free lunch

I’ve been pointing out for a while that following the steep decline in the cost of wind and solar energy, an energy transition is a free lunch.  It’s a simple calculation.

Citibank are just the latest team of experts to confirm the “no or negligible net cost” conclusion. (Pdf report page 23; press coverage.) They make the total world outlays on energy to 2040 in a fossil-heavy BAU at $192 trillion, those in an energy transition at $190.2 trillion, net saving $1.8 trillion. The last number is well within the error range, and should not be taken too seriously. The less precise “next to nothing” conclusion is solid. I stress that is the consensus expert position: the IPCC WG3 (pdf, page 15) say the same thing, as do Fraunhofer IWES in Kassel for Germany. Can any reader cite a credible study suggesting the contrary? In case you are worried, these professional scenarios do include the costs of grid integration for variable wind and solar energy, it’s not back to candles.

  • Net cash cost of energy transition to 2040 ≈ $0.

Fossil fuels are also responsible for huge amounts of local and regional air pollution. This contributes to a steadily lengthening list of health problems and premature deaths. UNEP gives the world total of premature deaths from all air pollution at 3.5 million a year.  The OECD has estimated the annual health costs (basis 2010) attributable to outdoor air pollution in its member states plus India and China at $3.5 trillion a year (pdf, page 2). Half of this is due to road transport. This estimate (endnote) excludes a good number of countries including Brazil and Indonesia, plus the effects of indoor air pollution from wood fires and kerosene stoves. It is almost certainly a major underestimate. A full energy transition eliminates all of these costs in time. There will be a few deaths from solar and wind installers falling from heights, and electrocutions from electric vehicles. More significantly, air pollution has improved in OECD countries since 2010, but not enough to change the overall picture.

A complete energy transition would eliminate these health costs gradually, and in the end completely. Using a straight-line reduction from $3.5 trn a year in 2015 to zero in 2060, we get a total saving of $25.3 trillion.

  • Health saving to 2040 from energy transition ≈ $25 trillion.

So the net undiscounted cost to 2040 of the energy transition (cash for energy plus health) is minus $25 trillion. That’s a $3,500 bonus for every man, woman and child alive today.

I have not included anything for avoided GDP costs from extreme weather, hotter summers, and sea-level rise. Citi guesstimate these at $44 trn to 2060 (pdf, page 8).  Here’s a higher estimate. On top of that, there are avoided burdens in loss of biodiversity and the risk of catastrophic disaster. These impacts are all real but methodologically difficult.  It’s not downplaying these to leave them aside as superfluous to the proof. Discounting (what rate?) would also add a layer of technical difficulties and does not affect the conclusion.

We are faced with the greatest free lunch in the history of humankind. Cue Veronese’s knockout Renaissance banquet.

Turning it down

One objection is that if it’s so, why haven’t we tucked in already? Replies:

  • Cultural, economic and political inertia

The world’s energy system is enormous. The trillions in physical and financial assets and the millions of workers are supported by a whole ideosphere of professionals who have spent their lives thinking and writing about fossil fuels. One example among many: the energy statisticians at the EIA do a good job on fossil fuels in the USA, but their forecasts on wind and solar have been laughable. The IEA in Paris have been almost as bad. There is something wrong when the best forecasters on global solar pv installation have been at Greenpeace.

  • Vested interests in fossil fuels

The Kochs and others have been fighting with desperation and complete lack of scruple for their businesses to survive. It’s already too late for several large coal companies. Curiously, the industry has not yet mobilised to stop the electric car threat. Possibly because Big Oil is still living in a dreamworld where oil and gas continue to grow to 2035 and carbon emissions don’t fall at all. If they ever wake up, they will be up against the electric utilities and a good part of the car industry.

But don’t underestimate their past work. The Cloud of Unknowing created in the English-speaking world by Exxon and the Kochs on climate change must count as one of the great achievements of agitprop. The denialists still show up in dozens on Joe Romm’s blog and others, and it’s not likely they are all paid: many are Willi Münzenberg’s “useful idiots”. The denialist control of the Republicans in Congress is partly venal, but partly also convinced. It’s not like the political protection of Florida sugar barons from cheap imports, which everybody accepts as purely mercenary, and has no ideologues.

There are other losers more deserving of our sympathy, like coal-miners and roustabouts. The benefits of the transition would allow generous compensation, retraining and community regeneration policies, but on past experience these are sadly unlikely.

  • Front-loading of the spending

In the early years of the transition, spending is higher than under BAU. Renewable energies and electric vehicles are nearly all upfront capital expenditure, with very low operating costs. In their infancy, they also need subsidies to get them along the learning curve to cost parity. Fraunhofer’s scenario of the costs of the Energiewende in Germany shows the problem.

Outlays and savings of the energy transition in Germany

Outlays and savings of the energy transition in Germany

(The green bars above are the saved fossil fuel costs, the bars below the outlays. The red line is the net cash balance.)

The problem should not be exaggerated. The infant subsidies to wind and solar are largely past history, or embedded legacies like FIT guarantees that don’t affect future investment decisions. Brazil and Chile are installing wind and solar without subsidy. Electric vehicles are still dependent on subsidies, but the learning curve on batteries is so steep that the subsidies may go in five years. In any case, OECD countries are operating well below full employment, so the opportunity cost of additional public and private investment will be negligible for some time to come.

  • Small government ideology

As Mike O’Hare has stressed, climate change is the problem from hell because it’s all about externalities: the harm to other people and the biosphere caused by actions that look beneficial to individuals and organisations like companies. It is therefore an issue of collective action, and the solution goes through government. The dominant ideology of today, free-market liberalism, is instinctively opposed to such solutions and therefore minimizes the problems. Combined with the front-loading, we are facing some combination of carbon taxes, costly emissions regulations, subsidies to clean technologies, and direct public spending say on railways and buses. Merkel and Cameron, fearful of the impact of household electricity surcharges, have already backtracked on solar subsidies.

The world may get lucky on some of this. The subsidies to wind and solar energy are largely past (Danke, Tak, ArigatÅ). Those to electric cars – up to €10,000 a pop in France  – won’t need to last long. Electric buses are competitive on a lifelong cost of ownership basis. Green technologies may win in power generation, transport and heating/cooling without any externality pricing. I don’t though see a transition happening in aviation, shipping, steel and cement, or large-scale reforestation, without the heavy hand of government. Large scale carbon sequestration, if we find it necessary (and James Hansen is usually right), cannot be a commercial proposition without carbon pricing or equivalent regulation.

  • The imaginary free rider problem

The international free rider problem was a major obstacle to a global agreement when it was assumed that the net costs of climate mitigation were large. The benefits of action in one country would mainly accrue to foreigners, so you needed an impossible global deal before anybody got started. With the falling costs of renewable energies, and the realization of the local costs of air pollution, the problem has disappeared in fact. A rational policymaker in most countries will find that national climate action has net benefits in isolation, though these are much higher if everybody else joins in. This has made possible the turnaround in the prospects of climate negotiations, seized on by Christiana Figueres, and then by Barack Obama and Xi Jinping. However, not all policymakers are rational, and the free rider fallacy still grips many in the US Republican Party and elsewhere.

The Dwarfs’ banquet?

The free lunch explains why politicians as different as Barack Obama, Xi Jinping, Narendra Modi, Angela Merkel, Dilma Rousseff and David Cameron are all now backing the energy transition. It doesn’t cost an arm and a leg, and on balance their voters or subjects will not punish them for the policy.  They and other nervous leaders are reassured by the examples of California, Denmark and Germany, whose aggressive climate policies are plainly no obstacle to prosperity. After decades of flaffing around, a decent climate agreement in Paris next month is now odds-on.

The obstacles to taking up the offer explain the hesitations, delays, ambiguities, and backtracks of these same leaders. The five reasons for inaction are incoherent in logic, but psychologically and politically they add up to a formidable constellation of nay-saying forces. Only a handful of countries have adopted a hard zero-carbon target, and they don’t include Germany. What politician has had the courage to tell the coal and oil industries that they must go out of business well before 2050?

The banquet is not quite like Aslan’s. The dwarfs are behaving more like terrified and hungry savages, darting in to seize a pie from the table, then running off to consume it in safety. There must be a catch. We won’t let go of our fears of joy and plenty.

Humankind may still fail. The rejection of the $25 trillion free lunch, if it continues, will be the greatest ever failure of public policy, worse than August 1914: since this time, all our grandchildren will be in the trenches at at Verdun.


Endnote 1: coal in China update

China’s coal output has been in absolute decline for 18 months, as I first reported here  and here. Consumption in 2013 was revised upwards, so there may be an element of data massaging, but the short-term trend is indubitable. This slowdown allowed global carbon emissions from energy to go flat in 2014, according to the IEA.

There is some element of cyclical downturn in this, but the shift away from heavy industry  in the makeup of GDP is both rapid and longstanding. In 2014 China added 21 GW of hydro, 23 GW of wind capacity and 11GW of solar, and will add more this year. This new renewable capacity is not only meeting the bulk of incremental demand, itself slowing from efficiency gains as well as tertiarisation. It has started to cut into coal burning: “utilisation rates at thermal power plants – nearly all coal-fired – have dropped to 52.2 percent in the first two months of this year” (Reuters).

You can construct scenarios in which these trends are reversed: a badly designed Keynesian stimulus programme to offset the current slowdown; a large increase on gasoline-fuelled cars; or an expansion of coal-to-gas or coal-to-gasoline plants. Fine, if you think Xi and the rest of the Chinese Politburo Standing Committee are incompetent technocrats unaware of the political danger of urban air pollution. Do they look like that to you?

Endnote 2: methodology of health costs

The OECD estimate of the costs of deaths is based on “the standard method for calculating the cost of mortality – the Value of Statistical Life (VSL) as derived from individuals’ valuation of their willingness to pay to reduce the risk of dying.” This is ethically problematic, as it values lives according to income per head. It is politically realistic in that the VSL should be reflected in the respective amounts that the citizens of Norway and Chad will be willing to pay in taxes or regulatory costs to mitigate the risk.

The OECD people are winging it on morbidity. “Indicative estimates suggest that morbidity would add 10% to the mortality cost figures, but work is needed …” This is strange; you would think that morbidity costs are fairly objective – health care plus lost work plus a malus for suffering – and putting a price on death problematic, but apparently it’s the other way round. For every death, there should be many more people made seriously ill. In road accidents, the ratio in Europe is 12 serious injuries to every death.  The QALY loss in life expectancy from illness must be very large, and people would pay a lot to avoid it as well as death. In some cases, like lifelong care of a baby brain-damaged in pregnancy, medical costs may be as much as a VSL life. My intuition is that the cost of illness from air pollution is in the same ballpark as that of death.

[Cross-posted at The Reality-Based Community]