Gernot Wagner and Martin L . Weitzman go through the arguments in "Climate Shock," an article in the Milken Institute Review (2015, Second Quarter, pp. 55-69). The article is based on a chapter of their just-published book Climate Shock: The Economic Consequences of a Warmer Planet. Here's a sample:
We may hate the idea of countering amazing amounts of pollution with yet more pollution of a different type. But the option is simply too cheap to ignore. It’s not like anyone would literally mimic Mount Pinatubo by pumping 20 million tons of sulfur dioxide into the stratosphere. At the very least, given current technology and knowledge, the sulfur would likely be delivered in the form of sulfuric acid vapor. Sooner rather than later, we may be looking at particles specifically engineered to reflect as much solar radiation back into space as possible, maximizing the leverage.
It may only take a fleet of a few dozen planes flying 24/7 to deliver the desired amount. Some have gone as far as to calculate how many Gulfstream G650 jets it would take to haul the necessary materials. But such specifics are indeed too specific. What matters is that the total costs would apparently be low compared to both the damage carbon dioxide causes and the cost of avoiding that damage by reducing carbon emissions.
Estimates are all over the place, but most put the direct engineering costs of getting temperatures back down to pre-industrial levels on the order of $1-to-$10 billion a year. Now, $1-to-$10 billion is not nothing, but it’s well within the reach of many countries and maybe even the odd billionaire. If a ton of carbon dioxide emitted today generates $40 in damage, we are talking fractions of a penny for the sulfur to offset it. ...
Geoengineering is too cheap to dismiss as a fringe strategy developed by sinister scientists looking for attention and grant money, as some pundits would have it. If anything, it’s the most experienced climate scientists who take the issue most seriously. And not because they want to. ...Pick your favorite analogy. It’s like chemotherapy or a tracheostomy for the planet: a last-ditch effort to do what prevention failed to accomplish. ... As always, it’s a matter of trade-offs. Climate change itself will have plenty of unsavory side effects. The question, then, is not whether geoengineering alone could wreak havoc. (It could.) The question is whether climate change plus geoengineering is better or worse than unmitigated climate change.Wagner and Weitzman go on to discuss a variety of possible methods of geoengineering:
putting sulfur particles in the atmosphere; ships that spray water vapor high into the sky to generate more cloud cover; painting all roofs a more reflective white; dumping plant nutrients (like iron) into the ocean so that the resulting plants will absorb more carbon; and others.
For myself, I'm uncomfortably aware that I don't know much about the details of climate modelling. It does seems clear that a healthy majority of those who work in the area and are represented in the Intergovernmental Panel on Climate Change (IPCC) reports are concerned about the risks of climate change, so from an economic viewpoint, my usual attitude is to treat the risk as real and imporant--but focus on the economic problem of how to reduce those risks in cost-effective ways. For some earlier posts bearing on these issues, see "Climate Change Strategies (Including Mangroves)" (December 4, 2012), "Setting a Carbon Price: What's Known, What's Not" (June 25, 2013), "Short-Term Benefits of Climate Change Policy" (September 22, 2014), "Carbon Capture and Storage: An Update" (December 24, 2013), "Other Air Pollutants: Soot and Methane" (June 28, 2012), and "Should the U.S. Government Cost-Benefit Analysis Look Outside the U.S.?" (June 13, 2014).
In my reading, even though the most recent IPCC report comes out with largely the same bottom line--that climate change is a serious problem needing a substantial policy response in both the near-term and the long-term--the most recent report makes the arguments in a tone of less certainty than earlier reports. As one example, the most recent IPCC report has a highlighted discussion in a box in the first chapter acknowledging that temperature trends rise 1998 to 2012 was much less steep than the earlier trend, and less than predicted (see Box 1.1 on p. 43 of the report). The report discusses various reasons why this might have occurred, with an emphasis that additional research is needed here: for example, one possibility is that volcanoes put more sulfur into the air than expected, a form of natural geoengineering that had a cooling effect; that El Nino warmed up the globe above trend in late 1990s, making the temperature rise in the 1990s look unexpectedly rapid, and the rise since then correspondingly slower; or that oceans trapped more heat than the models had predicted.
For me, the risks of any actual efforts at geoengineering seem too high at present. But of course, this is another way of saying that I think the risks of climate change are not immediate or severe enough to be worth the risks of geoengineering. But as I noted at the start, if you believe that the risks of climate change are large and near-term--and moreover, if you have observed how difficult it seems to be for the world to take action to reduce carbon emissions substantially--then you should be looking at geoengineering very closely, even you hate the idea of needing to do so.