*© Polity Press 2021. All rights reserved.
The idea of solar geoengineering — managing climate change by blocking some solar radiation — has been kicking around for decades. But only in the last few years, when it finally dawned on the globe’s chattering classes that climate change is a clear and present danger and that containing greenhouse gas emissions is a truly formidable task, has geoengineering seeped into the public discourse. Gernot Wagner has taken on the role of fair-minded interlocutor on geoengineering’s promise and perils. And nobody, it is safe to say, is better positioned for the job. He’s not only an economist at New York University (currently on leave at Columbia Business School) and the former executive director of Harvard’s Solar Geoengineering Research Program, but a Bloomberg Green columnist who offers wise, accessible analyses on matters environmental (and more). Here, we excerpt a chapter from his latest book, Geoengineering: The Gamble,* in which he tackles the question of whether path-of-least-resistance technofixes for climate change could sap our will to confront emissions containment head-on.
— Peter Passell
Illustrations by aad goudappel
Published April 25, 2022
The first thing to know about “moral hazard” is that, more often than not, the term is woefully misused. The dictionary definition is clear:
Moral hazard: Lack of incentive to guard against risk where one is protected from its consequences, e.g., by insurance.
There are, indeed, plenty of examples where the term does apply directly. Think health insurance, where the knowledge that a doctor will attempt to fix things later might lead to riskier behavior now. Or think about a direct technological intervention, like condoms or seat belts. Using either might lead to some riskier behavior.
Part of that reaction might be perfectly rational; driving faster, after all, has its advantages. If seatbelts make driving faster safer, then doing so might, indeed, make sense for the individual driver, and the introduction of seatbelts as a matter of policy might be a perfectly sensible step for society.
All too often, though, the concept of moral hazard is misapplied in two ways. One is that not every adverse reaction can be ascribed to moral hazard. There are, for example, plenty of other horsemen of the climate policy apocalypse. The energy-efficiency rebound effect comes to mind, which implies that energy efficiency standards, say, making cars more fuel-efficient will lead to more driving and, thus, more energy use. Others are spatial and temporal emissions leakages, whereby policies introduced in one particular place and time might lead to emissions increases elsewhere. Some of those emissions increases may, indeed, at least indirectly be linked to moral hazard-style shifts in behavior. Much of it is not.
Moral hazard in the context of geoengineering is no different. There, the concern is that the potential use of either carbon removal or, perhaps more so, solar geoengineering, might lead to increased emissions. The concern as such is both broader than moral hazard — any form of “crowding out” one action by another — and more specific. Instead of moral hazard, it is really “mitigation deterrence” or outright “mitigation obstruction.”
With geoengineering, it is the environmental Left that worries about moral hazards. And this does not only apply to geoengineering. It extends to most anything where a mere technofix does not go far enough and might preempt more fundamental, complex societal changes.
That possibility might extend to the mere mention of either technology: Think that a relatively cheap technofix might be available one day? Ease off the necessary emissions reductions now. This concern is well founded. Newt Gingrich, the former Republican Speaker of the House, penned an op-ed saying as much in 2008 during the ultimately futile fight in U.S. Congress to pass comprehensive climate legislation. Then again, that’s exactly what Gingrich would say.
Moral Hazards = Politics
The second thing to know about moral hazards is that, more often than not, invoking them is intensely political. But politics, here, comes with a twist.
Typically, the political Right invokes moral hazards by pointing out how policy might shield individuals from their moral failings and encourage amoral behavior. That goes for health insurance in general as well as for more specific policies and technologies. Moral hazards, of course, are typically only applied to disliked policies. Universal health care? Moral hazard! Federal bailouts of favored political constituents? Silence.
Birth control interventions, including abortions, might be the most controversial — and, thus, political — example. Their mere availability, the argument goes, makes unprotected sex less risky and, thus, encourages more of it. (Full disclosure, it is also an example that hits home rather directly, with my wife working as a gynecologist at NYU Langone Health and as the division director of reproductive choice at Bellevue Hospital in New York.) One constant moral and political question, then, is whether to advertise and promote or whether to suppress any particular technology. Many on the Right would rather suppress abortions and certain types of birth control, declaring their very use immoral.
With geoengineering, it is the environmental Left that worries about moral hazards. And this does not only apply to geoengineering. It extends to most anything where a mere technofix does not go far enough and might preempt more fundamental, complex societal changes:
Green moral hazard: Lack of incentive to fix deeply seated, complex environmental problems because of the possibility of a quick technofix, e.g., geoengineering.
Such worries may not stop with technologies alone, nor with geoengineering. Even comprehensive carbon pricing might fit the bill as an intervention that “merely” leads to lower CO2 emissions, while not addressing all sorts of other environmental and societal problems. Yes, a sufficiently high price and other climate policies might “stick it to CO2,” but they do not “stick it to the Man” at the same time — at least not as much as some on the Left might hope.
There are, indeed, real arguments to be had about the best environmental policies and technologies. By now, for example, there is broad understanding that carbon pricing alone would not adequately address climate change. Comprehensive climate policy also ought to include investments in low-carbon technologies in the form of direct subsidies.
Instead of crediting Rockefeller and fossil fuels with serving as a backstop technology of sorts, it might be more accurate to credit Edison and the electric light for doing so.
There are similarly real arguments to be had about how to pass the most ambitious set of climate policies. One strain of this argument is whether to focus on climate policy in isolation, or to see it as an issue intimately linked to other, broader societal concerns.
Policy sequencing matters, too. Hardly any country or jurisdiction has passed carbon pricing in isolation. Doing so typically goes hand-in-hand with (or follows) policies focused on subsidizing low-carbon energy or infrastructure more broadly. All these are important active policy debates.
Now add geoengineering into the mix. It’s easy to see how some environmentalists would be concerned that any focus on carbon removal (especially solar geoengineering) might detract from the need to cut CO2 emissions in the first place: Finally, we have some momentum on [much-needed policy X]! Why detract from it by talking about geoengineering?
Moral Hazards Throughout Environmental History
Geoengineering is hardly alone in invoking feelings of strong moral hazard. It is not even unique within recent climate policy discussions. Carbon removal is now roughly where talk of adaptation was in the mid-1990s, solar geoengineering not yet even that far. Back then, mere talk of the need to adapt to climate change already in store was often dismissed as a distraction from carbon mitigation. Even Vice President Al Gore was on record saying as much.
Today, adaptation has a well-established place in comprehensive climate policy. Now, though, solar geoengineering and to a lesser extent carbon removal are often the odd ones out. The history of moral hazard in environmental thought, meanwhile, goes back much further. By some accounts, modern environmentalism’s very existence is intertwined with green moral hazard logic. That logic, in turn, comes in two forms. One is a principled rejection of any and all technofixes to environmental problems. The second, intertwined with the first, is a rejection of any policy that falls short of a broader social revolution with widely held notions of justice at its core.
To be clear, I subscribe to many of these ideals. Rampant inequality, systemic racism and all sorts of other societal ills clearly demand much broader, fundamental reforms. But there is a rather big step between arguing how a simple technofix can’t solve it all — it usually can’t — and how said technofix should never be touched because it doesn’t solve it all. Most interventions involve trade-offs, technological or otherwise.
Tales of Two Technofixes
The two stories most often told in this context are of whale oil and horse manure. Both stories played out in the mid- to late-19th and the early 20th centuries. Both were, at first, seen as clear-cut success stories of technological innovation and entrepreneurial spirit. Both also involve fossil fuels saving the day.
Not least because of that, both stories have since become highly politicized.
Looking at the whale-oil story, for example, Matt Ridley, the self-described Rational Optimist, likes to reference Warren Meyer, a self-declared climate skeptic, by saying, “Greenpeace should have a picture of John D. Rockefeller on the wall of every office.” That quotation serves several purposes, not least it appears to turn environmentalists’ morality against themselves. See, if it hadn’t been for the discovery of rock oil, the argument goes, the world would have continued to hunt right whales for their blubber (aka whale oil). The truth, it turns out, is quite a bit messier, and it took Italian chemist Ugo Bardi looking through a wealth of whale-oil extraction and pricing data to piece it together.
North Atlantic whale-oil production peaked before 1850, well before rock oil became available in 1859. By the time kerosene became commercially available, whale-oil extraction was less than half its pre-1850 peak, leading Bardi to conclude, “the peaking and the initial phase of decline of whale-oil production were not caused by the availability of a better technology, but by the physical depletion of the resource.”
The U.S. Civil War played a role here, too, as it partially grounded the whaling fleet during the early 1860s. Moreover, kerosene lamps, revolutionary as they were in their own right, would soon be made obsolete by another technical leap: Thomas Edison invented the incandescent lightbulb in the late 1870s, a technology that would rapidly replace oil lamps and democratize access to artificial light over the following decades.
So, first, it wasn’t rock oil that led to the initial 50 percent drop in whale-oil supply. That was due to woeful overexploitation of right whales. And second, instead of crediting Rockefeller and fossil fuels with serving as a backstop technology of sorts, it might be more accurate to credit Edison and the electric light for doing so.
Brand offered a techno-optimistic vision of environmentalism that is pro-nuclear power, pro-genetically modified organisms, and pro-geoengineering — not blindly so, but clearly as part of the overall solutions package.
All that doesn’t just put a jinx on stories of fossil fuels as an environmental savior, it also complicates how this story fits within the green moral hazard framework. New technologies often come with pluses and minuses. Sometimes the overall picture is indeed mixed or dominated by negatives. Sometimes new technologies might do a lot of good. The result often depends on broader societal (political) choices.
The horse manure story paints a similar, and similarly complex, picture. The setup is one of fossil-fueled technological salvation. While whales were declining, horses multiplied — and not out in the wide-open plains, but right in the heart of major cities. Horses had been the most important mode of transportation for thousands of years. It wasn’t until the late 1800s, however, that they really took off.
New Yorkers made over 30 million trips on horse-drawn carriages in 1860. By the end of the decade, that figure topped 100 million and there was no end in sight. In the 1890s, 200,000 horses produced 2,500 tons of manure a day. It took thousands of horses just to haul the manure away. Much of it was simply dumped into empty lots. One doomsday scribe infamously predicted that by 1930 horse manure would blanket Manhattan three stories deep.
In 1898, the world’s first urban planning conference broke up in disarray after only three of 10 scheduled days. Delegates could not see a way out of the gridlock and stink. Mobility was as essential and addictive back then as it is today. It was a transportation, environmental and public health nightmare rolled into one. The model of cities looked wholly unsustainable.
We know what came next. John D. Rockefeller, Henry Ford and a host of other inventors and entrepreneurs saved the day with oil-powered cars that pushed horse-drawn carriages out of New York. In 1900, cars were still a luxury turning heads. That year, 4,192 cars sold in the entire United States. But by 1912, sales topped 350,000 and for the first time, cars outnumbered horse-drawn carriages in New York City.
The story isn't about a triumph of technological innovation and entrepreneurs, of capitalism over environmental despair. It is about the importance of government policy and public investment — from interstate highways to Navy ships patrolling the Persian Gulf.
Five years later, the last horse-drawn streetcar was retired for good. By the 1920s, horses were all but gone from city streets save for horse carriages shuttling tourists through Central Park and the occasional mounted cops, who to this day roam through Times Square and Central Park.
Yet, once again, the story isn’t quite as simple. Yes, cars provided a superior technology compared to horse-drawn carriages. They were faster, safer and cleaner — at least to the extent that the tons of pollutants coming out of their tailpipes tended to dissipate instead of accumulating like horse manure. But it wasn’t nearly as simple as most versions of the story seem to suggest.
It wasn’t a triumph of technological innovation and entrepreneurs, of capitalism over environmental despair. Far from it, though the lesson here is different from that of the whale-oil story. It is about the importance of government policy and public investment — from interstate highways to Navy ships patrolling the Persian Gulf.
When General Motors bought up streetcars and converted systems to buses through a wholly owned subsidiary, it effectively destroyed mass transit in over 40 cities throughout the United States between the 1930s and ’50s. New technologies don’t arise in a societal or political vacuum. Some might appear to grow into a void, but even the internet famously depended on heavy public investment.
These decidedly mixed histories make the skeptical reactions of many environmentalists to geoengineering technologies more than understandable. An even clearer example is nuclear technology, pitting Promethean promises directly against nature.
The advent of the nuclear age also coincided with the birth of modern environmentalism. Rachel Carson, Barry Commoner and others central to the founding of the modern environmental movement pointed to nuclear fallout as the clearest example yet of how, in Commoner’s words, “we tend to use modern large-scale technology before we fully understand its consequences.” Carson pointed to nuclear fallout to explain the dangers of DDT.
It was against this backdrop that solar geoengineering was introduced, and it was at first quite deliberately packaged as a technofix. The 1965 report to President Johnson, for example presented solar geoengineering as the sole response to climate change, very much reflecting this technofix mentality. This all but guaranteed that geoengineering would be broadly understood as a moral hazard.
Even before the first Earth Day in 1970, Stewart Brand declared in his Whole Earth Catalog’s statement of purpose, “We are as gods and might as well get good at it.” Brand offered a techno-optimistic vision of environmentalism, one that is pro-nuclear power, pro-genetically modified organisms, and progeoengineering — not blindly so, but clearly as part of the overall solutions package. In fact, Brand participates in these debates to this day.
I can see the appeal of this technooptimistic brand of environmentalism. But despite Brand’s prominence, it was clearly a minority view then and it remains one to this day. It has often also been co-opted by selfdeclared “rational” or “pragmatic” environmentalists who seem to take pride in controversy and in being seen as either having left the environmental movement or in never truly having been a part of it.
Solar Geoengineering and Moral Hazard
Both carbon removal and solar geoengineering occupy a rather complex role in the climate change policy debate. Neither technology makes sense even to contemplate without a clear understanding that climate change is sufficiently bad to potentially warrant the intervention. That still raises plenty of scientific and ethical questions. One way to encapsulate that debate is to ask the deceptively simple question of whether solar geoengineering should be seen as a first or last resort. The correct answer may well be neither. In fact, I would emphatically argue that it is neither. But looking at either extreme position is still instructive.
To those few labeling either carbon removal or (especially) solar geoengineering a first resort, it might seem unimaginable to wean the world off fossil fuels. The 1965 U.S. presidential report (implicitly) took that position. Newt Gingrich’s 2008 op-ed did the same. Why pass climate policies aimed at decarbonizing the economy if there is a simple technofix? That thinking, once again, directly invokes solar geoengineering as a green moral hazard. It seemingly avoids any of the harder, more expensive steps necessary to address climate change. I would describe any such position as wishful thinking, willful blindness — or worse.
What if it doesn’t work as advertised? And even if it does, solar geoengineering is no solution. Carbon removal at least addresses the proximate root cause. Solar geoengineering doesn’t even do that. And I’m saying “proximate” here because carbon removal does not, in fact, address the causes those concerned about moral hazard might point to. Solar geoengineering would, indeed, show faster results than cutting CO2 emissions or removing it once the carbon is in the atmosphere. But fast is not the only objective.
The way the moral hazard question is asked matters. That finding alone doesn’t invalidate any and all prior surveys pointing to moral hazard, but it should certainly add a hefty dose of skepticism.
The only sensible way to approach either carbon removal or solar geoengineering, then, is not as an either/or but as a yes/and. Neither should stand on its own but must instead be seen as part of a much broader climate policy portfolio that includes cutting CO2 emissions in the first place, as well as plenty of adaptation to what’s already in store. That also makes any geoengineering deployment scenario neither a first nor a last resort.
Of course, saying so will not make it so. While there are, indeed, some small rational trade-offs among all four interventions — mitigation, adaptation, carbon removal, and solar geoengineering — moral hazard is not about what would happen in an idealized, rational scenario. Moral-hazard worries around solar geoengineering are all about what would happen in the irrational, fickle world we inhabit. That holds true for carbon removal, but even more so for geoengineering.
Given the “free-driver” properties of solar geoengineering leading to the “not if, but when” conclusion around its eventual deployment, geoengineering governance ought to focus on channeling these fundamental forces into a productive direction. To a first approximation, this implies simply suppressing them — and globally. Given how the moral hazard of solar geoengineering cannot be wished away, it also means channeling the moral hazard forces themselves in the right direction.
Moral Hazard and Its Inverse?
It’s clear that solar geoengineering as a technofix implies a trade-off with the desire to cut CO2 emissions in the first place. Or does it? The empirical evidence is decidedly mixed. When we scoured the academic literature about five years ago, we found some 30 surveys attempting to elicit public opinion on solar geoengineering’s moral hazard. Most of these surveys concluded that there is some kind of moral hazard. The mere availability of solar geoengineering, it seems, would encourage survey respondents to want to ease off CO2 emissions cuts. But, alas, these surveys also often raise more questions than they answer.
One concern is that very few respondents know what solar geoengineering is before taking the survey. That’s not an insurmountable challenge per se. Surveys often try to elicit basic familiarity with some obscure (or not so obscure) topic, scientific or otherwise: Have you heard about the SARS-CoV-2 virus? … Which branch of the U.S. government makes laws? … The fraction of respondents showing some basic familiarity with the subject might itself be revealing, one way or another. The likes of Jay Leno and Jimmy Kimmel have for years been milking admittedly biased versions of such tests for laughs.
Real framing choices are much subtler. One that was apparently ignored by prior surveys is what’s typically called “acquiescence bias.” It turns out that when survey respondents don’t have strong feelings either way, they have a tendency to agree with the way the question is phrased.
The public’s unfamiliarity with solar geoengineering is revealing in itself. But surveys typically don’t stop with this finding. They then go on to elicit a number of other responses, including responses around moral hazard. A lot then depends on how the information is presented. Describing it as a riskless, costless, innocuous intervention would elicit very different reactions than a description invoking images of madman scientists taking over the world’s climates on behalf of fossilfuel interests.
Real framing choices, of course, are much subtler. One that’s rather obvious in hindsight but was apparently ignored by all such prior surveys is what’s typically called “acquiescence bias.” It turns out that when survey respondents don’t have strong feelings either way, they have a tendency to agree with the way the question is phrased.
Ask whether people think solar geoengineering might detract from the need to cut CO2 emissions, and to someone who had not previously thought about the question, it appears like a reasonable possibility. Ask the same person whether they think the availability of solar geoengineering might encourage more CO2 emissions cuts, and they consider that a reasonable possibility, too.
The way the moral hazard question is asked matters. That finding alone doesn’t invalidate any and all prior surveys pointing to moral hazard, but it should certainly add a hefty dose of skepticism.
Another dose of skepticism is in order since these surveys just elicit answers to hypothetical questions. There are better and worse ways of doing that, but there’s no escaping the fact that, in the end, it’s “just” a survey. Ask a silly or biased question, get a silly or biased answer.
Potentially one better: observe how people actually behave. Social scientist Christine Merk did just that with two of her colleagues in the first-ever revealed-preference survey of around 600 test subjects. Instead of asking them to answer hypothetical questions around moral hazard, she let them vote with their feet — or rather, with their euros.
What about those who end up making the decisions? Structural biases and deeply seated, vested fossil-fuel interests surely matter more. This is, indeed, just that type of moral hazard that should worry us.
The experiment, in rough terms: split 600 participants into three groups. The first two groups are told about climate change in varying detail. The third is told about climate change and the possibility of solar geoengineering in the form of stratospheric aerosols.
All three groups were paid for participating in the experiment, but with one twist: they could choose how much of their survey payments they’d like to use to pay for carbon offsets. (All that raises further questions about the effectiveness of carbon offsets as a response to climate change, but let’s leave that for another day. It should not bias the results here one way or another.) The resulting paper’s title summarizes the findings rather modestly: “Knowledge about aerosol injection does not reduce individual mitigation efforts.”
I’d go one step further: Merk and her colleagues’ experiment shows the very real possibility that telling the uninitiated about solar geoengineering might indeed increase their desire to cut CO2 emissions. The direct comparison of the effect of telling 200 subjects about solar geoengineering versus just telling them more about climate change itself shows just that. It’s solar geoengineering that has the additional motivating effect, not reading more about climate change in general.
Why would that be? One hypothesis for this inverse moral hazard finding is that solar geoengineering sounds so scary that it makes climate change itself look like a more serious problem. As in, wait, serious scientists are talking about doing what? Perhaps there’s something to this whole climate change thing after all.
To be clear, that’s just one hypothesis, and Merk’s study does not offer conclusive evidence for or against this hypothesis. (Full disclosure: Merk has since become a co-author of two essays on moral hazard, and we have been running extensive online experiments to try to get to the bottom of this very question.)
Moral hazard is a clear possibility and so is its inverse. It all seems to depend on how things are presented, who is asked and, as usual, on the broader context. That raises the all-important question of whether any of this actually matters. The public might think one way or another. But what about those who end up making the decisions? Structural biases and deeply seated, vested fossil-fuel interests surely matter more.
It is, indeed, just that type of moral hazard that should worry us. Fossil-fuel interests advocating for either carbon removal or solar geoengineering solely as a way to delay cutting CO2 emissions would likely be the most consequential type of intervention.
That kind of fossil-fuel advocacy might take several forms. One might be classic lobbying. Another might well involve attempts at misleading the public by introducing either carbon removal or solar geoengineering as the easy way out of our climate policy dilemma. Don’t listen to those environmentalists telling you that you can’t drive your gas guzzler… .
The real fossil industry actions, of course, would be much subtler and, thus, dangerous. For example, given the general public’s low level of familiarity with carbon removal and especially with solar geoengineering, part of vested interests’ advocacy may well be focused on trying to shape a positive view of either set of technologies, downplaying any risks.
Education, Education, Education
There’s no silver bullet to addressing any of these concerns. At the very least, it is important for everyone involved in geoengineering — and climate conversations more broadly — to be aware of them. That also means having more inclusive conversations in the first place. Education is key. The goal is nothing short of elevating the climate conversation by including carbon removal and solar geoengineering in their proper place.
I realize I’m probably preaching to the choir here. The overall goal, of course, must be much broader. One part of this equation, for example, is having balanced takes on geoengineering in climate books that aren’t focused on geoengineering itself but paint a more comprehensive picture of climate change and climate policy.
I tried to do that with my co-author Marty Weitzman in Climate Shock, a book that is primarily about low-probability, high-consequence climate impacts.
I am, of course, far from alone. The most prominent recent example is Eric Holthaus’s The Future Earth. The subtitle alone says a lot about Holthaus’s general outlook: “A Radical Vision for What’s Possible in the Age of Warming.” Holthaus goes decade by decade through a “radical” yet hopeful vision of the world.
That vision, for example, includes an entirely decarbonized U.S. economy by midcentury. Despite all that, Holthaus’s vision embraces a preview of solar geoengineering — and certainly not because he likes the idea. Nobody should. Despite conceiving of a world that includes plenty of seemingly radical decarbonization technologies and policies, he envisions solar geoengineering playing a specific, limited role.
Many climate books portray geoengineering in anything but a balanced fashion. Some are naively optimistic. Many more appear to be openly hostile. And broadening the conversation doesn’t end with books or erudite writings.
For every Future Earth, of course, there are many climate books that portray geoengineering in anything but a balanced fashion. Some are naively optimistic about carbon removal and solar geoengineering. Many more appear to be openly hostile. And broadening the conversation, of course, doesn’t end with books or erudite writings.
Hollywood, as usual, plays an outsized role. Sadly, “education” is hard to find there. Solar geoengineering is simply too tempting to portray through some kind of gee-whiz lens. The movie “Geostorm” might be the most prominent such example, with a globe-spanning satellite network controlling the world’s weather. (No, that is not a sensible solar geoengineering technology.)
I’m also ignoring “Snowpiercer” here, which has a failed solar geoengineering attempt that instead created Snowball Earth as backdrop, but really plays out aboard a train circling the globe, cannibalism and all. (The movie makes for some morbid entertainment, but it is not a good introduction to solar geoengineering, either.)
Meanwhile, the first sci-fi novels have come out portraying solar geoengineering in relatively realistic terms, while helping their readers think through some of the more profound social and political implications. Eliot Peper’s Veil might be the best example in this genre, Kim Stanley Robinson’s Ministry of the Future the most prominent. Peper’s story is replete with a solar geoengineering deployment scenario largely grounded in reality, a sinister oil conglomerate SaudExxon selling the Earth for profit, and still plenty of humanity to contemplate how solar geoengineering might play out on a planet struggling to bring global warming under control.
None of this guarantees an elevated geoengineering conversation. Nor does it reduce the need for deliberate, public engagement processes. That is especially true in light of much more deeply seated problems. Technological choices and technology policies are not typically egalitarian, democratic or pluralist, especially when viewed through a global lens. Yet it is precisely an imaginative portrayal like in Peper’s Veil that might lead toward the kind of rapid, deliberate education needed to have sensible discussions around geoengineering and its role within a wider response to climate change — green moral hazards and all.