I have repeatedly pointed out on these pages an interesting pattern of the debating style of the loudest activists clamoring for massive government intervention to fight climate change. First, they beat their opponents to a pulp, chanting “the science is settled” and pointing everyone towards the “consensus” as epitomized in the “summary for policymakers” that accompanies the periodic reports from the Intergovernmental Panel on Climate Change (IPCC). (Here’s just one example of Joe Romm linking the IPCC reports to the supposed scientific certainty involved.) Then, when the recipients of this public lashing actually read the peer-reviewed science and realize the case for government action is very weak, the activists change their attack completely, and now all of a sudden the IPCC reports are woefully inadequate.

In the present post, I’ll walk through yet another example of this phenomenon, in this case a recent ThinkProgress article that complains that GDP (Gross Domestic Product) isn’t a good metric when it comes to the debate over climate change. As we’ll see, when confronted with very compelling arguments that the IPCC reports and leading computer models do not justify the aggressive government intervention that the people at ThinkProgress seek,[1] they don’t dispute the point. Instead, they rattle off all sorts of reasons that the IPCC is essentially wrong, because the computer models used in the IPCC reports leave out important details, and because the standard cost/benefit approach to judging policy recommendations doesn’t work when it comes to climate change.

All of this should make innocent onlookers very suspicious. For years, advocates of heavy restrictions on energy use and individual liberty have cited the IPCC reports in their proclamations that “the science is settled” and that only “deniers” could possibly dispute the need for immediate and strong government actions. Now all of a sudden, the leading advocates are changing their case mid-stream, implicitly admitting that the weight they originally put forth on the IPCC reports will no longer give them the conclusion they want.

The Strong Case Against Government Intervention

The opening paragraphs of the ThinkProgress article inadvertently showcase just how strong a case the critics of aggressive government intervention have made, ironically relying on the IPCC reports themselves:

When it comes to obstacles to climate action, the climate change deniers in American government are pretty well known. But there’s also a subgroup of “reasonable” critics who concede the science of climate change, then deny that anything particularly dramatic needs to be done about it policy-wise.

Their argument revolves around something economists call GDP, and they use to wriggle out of supporting major climate action by effectively saying, “Yeah, even if climate change is real, it’s not going to be a big deal.”

For the uninitiated, “gross domestic product” (GDP) is the total value, in dollars, of all the goods and services produced by the American economy — or whatever economy is being measured — in a given year. It’s become the go-to metric of our society’s material standard of living and even its general well-being. The U.N.’s Intergovernmental Panel on Climate Change (IPCC) generally projects that losses to global GDP from climate change will be between one and five percent per year by century’s end…One to five percent is a seemingly small number. So the “reasonable” skeptics then argue an all-hands-on-deck effort to cut greenhouse gas emissions is unnecessary, and that it would likely reduce GDP even more than doing nothing.

But to hear Kate Gordon [Vice President at Next Generation, the nonprofit founded by hedge fund manager turned environmental activist Tom Steyer] tell it, this is a terrible way to frame the debate.

The rest of the ThinkProgress article then goes on to enumerate the various reasons that policymakers should not try to evaluate climate change policies in conventional terms, to see if the aggregate benefits of the interventions outweigh the aggregate costs. (Note that the article relies on the analysis provided by Kate Gordon, a lawyer and city planner who worked at CAP with John Podesta before she went to work for billionaire Tom Steyer.) The ThinkProgress article does not circle back and say that the “reasonable skeptics” are wrong insofar as their arguments go, but merely that these “reasonable skeptics” are relying on IPCC computer models that are too simplistic.

Uh Oh, ThinkProgress Starting to Sound Like “Deniers”

Yes, you read that right: It turns out that all of the much-ballyhooed “consensus signed on by all major governments and scientific organizations around the world” actually rests on quicksand. (To repeat, here’s an example of Joe Romm using this type of rhetoric to enshrine the IPCC as the epitome of scientific “consensus.”) Here’s how the new ThinkProgress article describes the IPCC reports and computer models on which they base their assessments of the costs and benefits of climate change policies:

[W]e’ve been tallying up GDP for most of this century, but projecting climate change’s impact on the economy is a whole other ballgame. On their own, the climate and the economy are enormously complex systems for computer models to even crudely replicate. Accounting for the effects of the first system on the second simply compounds the problem. So complexities, hidden factors, and feedback loops that could have profound ripple effects are all simplified away, because we just don’t have the information to know how to appropriately model those changes.

Other models can also descend into ad absurdum results pretty quickly. For example, most scientists agree an 18°C rise in global temperatures would literally render the Earth uninhabitable. But the standard computer model used by the IPCC projects that rise would only cut global GDP by half.

The point isn’t so much that one approach is better than the other. It’s that GDP projections are flung all over the map even by small changes to the input information or underlying assumptions of the models. It’s just an inherently bad metric for understanding the damage climate change will do.

Wait a second! The ThinkProgress writer is now telling us that it’s a really hard problem to model the global climate and economy? That the computer models used by the IPCC spit out nonsense results? That we can get results “all over the map” by tweaking the inputs? These guys are sure starting to sound like “deniers,” aren’t they? Is the above quote coming from a ThinkProgress post or an IER one?

So now we see their whole (original) case unraveling. We at IER have been making these points all along. For example, we reported that the computer climate/economic models were grossly simplistic and useless for policy analysis. Way back in 2009, we showed on these pages that using the analysis put out by the Congressional Budget Office (CBO), the costs of the Waxman-Markey cap and trade bill were far higher than the benefits that would accrue to Americans, and that they would plausibly be higher than the benefits accruing to the entire world—again, using the CBO’s own numbers.

The people at ThinkProgress apparently realized that they’re fighting a losing battle. Of the links to the “reasonable critics” they cite, the analysis by Jim Manzi is particularly insightful. He carefully walks through the analysis step by step, but here’s a good summary of Manzi’s thesis, written in 2008 (so the numbers may have slightly changed since then):

The current IPCC consensus forecast is that, under fairly reasonable assumptions for world population and economic growth, global temperatures will rise by about 3°C by the year 2100. Also according to the IPCC, a 4°C increase in temperatures would cause total estimated economic losses of 1–5 percent of global GDP…

This is the central problem for advocates of rapid, aggressive emissions reductions. Despite the rhetoric, the best available estimate of the damage we face from unconstrained global warming is not “global destruction,” but is instead costs on the order of 3 percent of global GDP in a much wealthier world well over a hundred years from now.

It should not, therefore, be surprising that formal efforts to weigh the near-term costs of emissions abatement against the long-term benefits from avoided global warming show few net benefits, even in theory.

After demonstrating that the standard IPCC reports have never justified the alarmist rhetoric characterizing the climate change debate, Manzi then begins to list the various reasons that even the alleged theoretically possible yet small “net benefits” will be elusive in the real world; for example, the models assume a globally coordinated regime implemented by all major emitters. In light of these difficulties, Manzi concludes that there is no case for aggressive government intervention to reduce emissions.

The ThinkProgress folks seem to realize that they can’t beat someone like Manzi (or us at IER, for that matter) on the battlefield of standard policy analysis—again, using the physical science “consensus” as stipulated by the peer-reviewed climate scientists. That’s why ThinkProgress moves the goalposts. Now all of a sudden, we can’t trust those simplistic computer models showcased in the IPCC reports, and we can’t rely on macro estimates of costs and benefits. Instead the ThinkProgress writer points us to regional (not global) impacts and brings up an insurance analogy.

Go Ahead and Use Regional Analysis—But Be Consistent

To repeat, the ThinkProgress folks realize that they can’t justify their desired government policies using conventional tools of macro cost/benefit analysis. That’s why they switch to a regional analysis:

Another problem with measuring climate change with one single GDP number is that it requires ignoring vast differences between states, regions, communities, and socio-economic strata. That can hide tremendous amounts of real human suffering that’s hard to put a price tag on. “You have to aggregate all of the impacts up to one number,” Gordon said. “Which is insane.”

She offered agriculture as an example: “We have a gigantic country with extremely different climate zones. If you want to move little corn symbols around on a map from place to place, you can move them all up north and then they’re still there,” Gordon said. “So that’s not that big a hit to GDP overall. But it’s a huge hit to the Southern Midwest and the Southeast.”

But hold on a second. The point of aggregating the damages of climate change into a single number is to be able to compare it to the damages from government restrictions on the economy. That’s the standard way that policymakers decide, “Does this proposed policy—such as a carbon tax—actually help on net or does it make Americans worse off?” The costs and benefits of a policy may be different, region by region, but in order to decide whether in the aggregate it’s a wise or foolish policy, you need to somehow aggregate those regional impacts into a grand total figure.

The ThinkProgress writer is correct that such aggregation ignores many important details. But if we’re going to quibble about it on the side of climate change impacts, then we have to be consistent and do it with the harms from government intervention.

For example, progressive writers often pooh-pooh the obvious harm that a carbon tax would pose to the coal industry, and all the jobs it would destroy. The progressives are quick to point out that a carbon tax would “create jobs” in the renewable sector, hence offsetting the losses in coal. Notice that this is exactly the kind of “aggregation” and “ignoring of specific impacts” that the ThinkProgress writer was just lamenting.

Later in the article, the ThinkProgress piece explains all of the hard-to-quantify damage that unrestricted climate change would pose to poor foreigners. Okay, but again, if that’s the route we are going to take, we need to be consistent. For humans to have any appreciable impact on climate change, it will take more than U.S. action. If the “solution” involves a slowing of the electrification of Africa, it will mean deaths. For example, a Forbes article in June reported:

Since 1990, 650 million Chinese have been lifted out of poverty, infant mortality has been reduced by 70%, and life expectancy has increased six years – a historic evolution powered by coal, and what the IEA has referred to as an “example” for other developing nations…There is only enough electricity generated in Sub-Sahara to power one light bulb per person for three hours a day. Over 65% of the population lives without any electricity at all.

Right now, fossil fuels (especially coal) are a very economical way to quickly bring electricity to hundreds of millions of desperately impoverished people around the globe. Part of the hard-to-quantify impact of restricting the growth in global CO2 emissions thus includes the lower standard of living, and shorter lifespans, of these desperately poor foreigners.

We do not deny that it is ultimately arbitrary to add up human lives and other important social goals on a scale of dollars and cents. But the ThinkProgress crowd is wrong to think that this consideration tilts the deck in their favor. If they can’t justify their policies using standard tools of analysis, it’s not enough to bring up all of the limitations of that original analysis—they still need to explain why their desired policies do more good than harm.

The Insurance Analogy

The ThinkProgress article then goes on to liken climate change policies to the purchase of insurance:

To their credit, most actual economists realize GDP was never meant to measure something as sweeping as the well-being of a society. It’s politicians and the figures and writers encased in the macroeconomic debate in Washington, DC that have focused on it to the exclusion of all else. Gordon argues against using GDP at all, and concentrating instead on the risks of what climate change could actually do to people and communities.

On the question of how much we should spend to ward off climate change, Gordon uses the common experience of buying insurance as an analogy. When buying a health care plan, everyone instinctively considers factors like their own risky behaviors (like smoking), their own family history (say, heart disease), their future economic prospects and their future lifestyle, among other things.

This rhetorical move to discount the value of GDP as a metric—which has been afoot for some time now—shows once again that the alarmist crowd realizes that they have a very weak case on conventional grounds. Rather than claiming that it’s a no-brainer that their recommended policies will deliver more benefits than costs, now they are merely claiming these policies make sense by eliminating threats that might occur if no action is taken. (That’s what they are ultimately arguing, by switching to an insurance analogy.) To repeat, this is a definite change in rhetoric; that wasn’t the standard case that was being made ten or even five years ago.

In any event, just by labeling something “insurance” doesn’t mean it’s automatically a good deal. You still have to make a case that the benefits (broadly construed) outweigh the costs of the insurance policy. After all, not everyone buys the most expensive insurance plan available.

When it comes to climate change, in a previous post I walked through the numbers and showed that no one in his right mind would buy an insurance policy that had the same characteristics as the climate change issue. After reviewing the data from the latest IPCC report to get a ballpark estimate of the numbers involved, I wrote:

Suppose someone from an insurance company came to you in the year 2050 and said, “We’ve run computer models many thousands of times using all kinds of different assumptions. In the worst-case scenario, a very small fraction of the computer runs—about 1 in 500—has you losing 20% of your income in the year 2100. In order to insure you against this extremely unlikely outcome that will occur in half a century, we want to charge you 3.4% of your income this year.”

Would you want to take that deal? Of course not. The premium is way too high in light of the very low probability and the relative modesty of the “catastrophe.”

Once again, we see that the ThinkProgress crowd doesn’t actually take their own analysis seriously. Sure, use an insurance analogy if you want. But when you plug in the actual numbers, you see that their recommended government interventions would be an outrageously expensive “insurance policy” relative to the benefits it delivers. To be sure, the ThinkProgress people can come back and complain that my analysis relies on the IPCC numbers and that they aren’t really accurate, but it’s not my fault the interventionists have been lecturing us for years that we need to trust the “consensus” as codified in the IPCC and other official documents.


For more than a decade the advocates of aggressive government intervention in the name of fighting first “global warming” and now “climate change” have had a field day labeling their critics as “deniers.” Yet as their critics began reading the actual analyses put out by the IPCC, Congressional Budget Office, and other allegedly neutral parties, a funny thing happened: The critics saw that the aggressive government policies could not be justified using standard metrics.

Realizing this, the proponents of aggressive government measures have begun shifting their rhetoric. Many of them now admit that if we make middle-of-the-road assumptions on emissions growth and the climate’s sensitivity, their recommended policies turn out to be as expensive as the alleged climate damages they seek to prevent. That’s why the alarmists now focus on possible (if unlikely) threats, regional impacts, and insurance analogies.

We can bring the argument to this new battleground; the case for intervention is still weak. But it’s worth noting that the aggressive interventionists have moved the goalposts. By doing so, they implicitly admit that analysts like Jim Manzi, and your humble IER team, have been right all along: Using the government’s own preferred data sources and computer models, it is very difficult to justify policies to restrict carbon dioxide emissions with the metrics that are used for all other government policy debates. The alarmists who have been yelling, “Case closed!” for years were simply bluffing; now they’re trying to reopen the case they realize they’ve lost on their initial terms.


[1] To convey the full context, I note that ThinkProgress is a product of the Center for American Progress (CAP), and the former President and Founder of CAP is John Podesta—President Obama’s climate change advisor.

Print Friendly, PDF & Email