Response to Bob Ward

Activism Dressed Up as Science

Bob Ward of the Grantham Institute on Climate Change at the LSE is aggressively promoting a short text that he claims to document a “fundamental methodological flaw” in my research paper, “Impact of Current Climate Proposals”.

In a press release, a letter to the Financial Times, and on social media, Ward suggests that his musings justify the withdrawal of my peer-reviewed research paper.

Communications professional Ward has every right to posit his personal view on how climate impact should be measured, even if in doing so he demonstrates a misunderstanding of methodology, modeling and my paper.

However, Ward is going much further than that. He is trying to smear my research, the journal that published it, and revoke my right to publish, simply because he thinks I should have approached the paper differently.

That is deeply troubling for any researcher, regardless of his or her personal beliefs.

What my research paper does

In my paper, I analyze the temperature impacts of the promised CO₂ reductions made by governments ahead of the Paris climate summit:

I identify very similar reductions by 2030 as the UNFCCC, the UNEP, the Dutch EPA, and researchers from LSE (including Ward himself).
I use the standard IPCC climate model, funded partly by the US EPA.
I estimate the effect assuming that all nations keep their promises until 2030 (very optimistic, compared to history.).
I estimate the impact of both pledges fading after 2030 (pessimistic) and remaining throughout the century (optimistic).
I find that globally, the impact of just keeping the promises from Paris until 2030 results in a temperature reduction by 2100 of 0.05°C. Keeping the Paris promises all the way through the century reduces temperatures by 0.17°C.

It is evident that Ward doesn’t like my conclusion, but it is hard not to see his furious objections as mostly a misplaced belief that Paris should make much more difference to global warming.

Ward’s ‘fatal flaw’ is included in other analyses

Ward’s central claim is that my paper suffers from a “fundamental methodological flaw” because it does not take into account the vague undertakings made by politicians after 2030.

There are five analyses of the impacts of Paris: Climate Interactive, Climate Action Tracker, MIT, IEA and Lomborg.

Only one (Lomborg) is peer-reviewed and three of the five use the same methodology – namely, estimating the impact just of the INDCs from 2016-2030, and extending them to 2100.

In effect, Ward claims that three out of these five analyses are wrong. My paper, the only peer-reviewed one, is the only one that he has called to be withdrawn.

Difference in perspective does not equal a ‘methodological flaw’

It is telling that Ward believes that he has looked into a crystal ball and identified what will happen to emissions after 2030. This unobtainable knowledge is the only basis upon he rests his case that my approach is ‘wrong’.

Ward believes in this point so much that in his short, curious document, he repeats the argument three times.

He insists that my analysis for the EU is wrong, because of the two scenarios I establish for the EU after 2030 “neither of these scenarios corresponds to expected policies beyond 2030.”

Unfortunately, Ward does nothing to explain how he would know what the “expected” EU policies are. This, of course, is because what he means to say is that “neither of these scenarios corresponds to what I expect from EU policies beyond 2030.”

When Ward states that my paper contains a “fundamental methodological flaw”, it appears that he does not understand what methodology means.

I clearly set out my methodology in the introduction of my paper, following the landmark (Wigley 1998).

I explain what I will examine in my paper. I am looking at the impact of the INDCs on emissions. Therefore, I define the cutoff for policy impacts from the INDCs for 2030:

I will investigate policies that have practical political implications soon and have a verifiable outcome by 2030, but not policies that promise actions only or mostly starting after 2030. Of course, policies that can be evaluated by 2030 will still impact emissions long after 2030, and hence affect the temperature trajectory all the way to the end of the century.

When Ward disagrees with this approach, he makes a category mistake calling it a “methodological flaw.” He simply believes that he knows the future better.

Misrepresentations of my paper

Ward, who has an undergraduate degree in geology and an unfinished PhD thesis on paleopiezometry, appears to have read my article only cursorily, and to misunderstand the approach. He certainly drastically misrepresents it. Understanding a research paper, in my opinion, should be a first step before calling for its withdrawal.

Ward’s response to my paper is just three pages long, but contains five serious examples where he apparently misunderstands or misses fundamental points:

First, Ward claims that I exclude feel-good but meaningless promises from the EU on emission reductions in 2050 “without explanation”.

This is simply incorrect. My Methodology states (repeated above) that I will “investigate policies that have practical political implications soon and have a verifiable outcome by 2030, but not policies that promise actions only or mostly starting after 2030.”

Ward may disagree with that methodological decision, but he cannot truthfully say I don’t explain my decision. The same is true for both China and the USA.

Second, Ward apparently misses or misunderstands two out of the four scenarios I look at in my paper. He writes: “In each of these cases, annual emissions are assumed not to reduce any further.”

He somehow misses the fact that I have also investigated two scenarios with much higher reductions.

Third, Ward writes that “It is clear that the post-2030 assumptions largely obliterate the benefits of the emissions cuts up to 2030.” This is factually incorrect and betrays a lack of understanding of marginal analysis. There is no cancelling out of reductions before 2030 with extra emissions after.

The paper is careful in assuming the same emissions except with reductions that come from the emission reduction promises from 2016-2030 and the emission reductions that are extended from that period, 2031-2100.

Fourth, Ward claims that because the temperature rises in my analysis are higher than in the CAT, that “shows just how extreme the assumptions about post-2030 emissions made by Lomborg (2015) really are.” Again, this shows he has fundamentally misunderstood the marginal analysis of my paper.

I similarly do an analysis with RCP6, which consistently shows lower temperatures than CAT. But all I show is the marginal change between running RCP8.5 with/without CO₂ reductions and similarly with RCP6 with/without CO₂ reductions.

Fifth, Ward states that my scenarios “suggest that the INDCs would lead to an increase in warming compared with scenarios without the INDCs”. This is particularly silly. It demonstrates that Ward either doesn’t understand the science, or has no interest in accurately representing what I did.

He could easily have avoided the error if he had glanced even cursorily at Figure 11. This misunderstanding – for want of a better word – leads Ward to suggest conspiratorially (and completely wrongly) that I somehow “hide this fact” by using RCP8.5.

Ward questions why the journal Global Policy published my paper and ‘how it passed peer review’. The more salient question is how he produced – and is publishing – a response that so fundamentally misunderstands the paper it responds to.

Kyoto Analogy

Stripping away the intemperate rhetoric and over-the-top claims in the response, fundamentally Ward has a personal opinion on how to measure climate impact which differs from my methodological approach.

To better understand the difference between us, it is illustrative to imagine if my paper had been written in 1997, and had been estimating the Kyoto promises.

My perspective is that it would have been appropriate to examine the impact of the promises on the table in Kyoto. This would show that, even if extended throughout the century, they would do very little (as the predecessor of my paper, Wigley (1998) did – surprisingly Ward hasn’t called for that paper to be withdrawn yet).

I would have used a pessimistic scenario that showed the Kyoto promises crumble after 2010, and an optimistic scenario with the Kyoto promises being extended throughout the century.

By Ward’s logic, I should have assumed that not only would this treaty be implemented, but that stronger and ever-increasing cuts would consistently be made as a result of policy (and not economic downturns) for decades. History shows that this would have been utterly wrong.

Should such an analysis of Kyoto have included President Bill Clinton's 1993 announcement that the US would reduce its emissions by 2000? That promise was never fulfilled. According to the Washington Post, the US administration's excuse was that the "goal is no longer possible because the economy has grown more rapidly than expected." The commitment failed even though it was for just seven years later, was to be implemented right away, and under the same president who made it.

Every industrialized nation actually promised in 1992 to return their emissions in 2000 to 1990-levels, - and almost every single one outside the former Warsaw pact missed that target.

Even the commitments made in the Kyoto Protocol itself ended up meaning nothing. The treaty was abandoned by the USA, and eventually by Russia, Japan and Canada.

We would clearly not have known this if we were conducting analysis in 1997 - but the examples show why it is folly to assume that we can realistically believe targets much further ahead to be right.

Of course others are welcome to look at hyper-optimistic scenarios; many do. I simply chose to focus on more realistic ones. Suggesting that this means my paper should be withdrawn for doing so appears to be much more of a political statement than a scientifically, well-informed one.

Why evaluating Paris after 2030 makes little sense

In my earlier response to climate blogger Joe Romm, I explained why I excluded carbon cutting promises that do not have a verifiable outcome by 2030. The comments broadly apply to Ward, who repeats the same criticism even though he was provided with this commentary.

I reproduce that commentary here:

First: It is difficult to defend the inclusion of targets with a very low likelihood of implementation.

In my article, I only include policies that have practical political implications soon and have a verifiable outcome by 2030.

It is undeniable that political targets further away are less likely to be implemented. Recent history clearly indicates that climate promises even 10-15 years ahead will be routinely flouted.

When China commits to reduce its carbon intensity of GDP by 60% to 65% below 2005 levels by 2030, we can analyze the progression towards that goal very clearly over the next 15 years and clearly determine if it is met by 2030 - so this is included in my analysis.

However, the promise to "achieve the peaking of carbon dioxide emissions around 2030 and making best efforts to peak early" (often curiously misquoted, as for instance in "peak CO₂ emissions by 2030 at the latest") is something that will only have an effect after around 2030, and it is something that will first be verifiable around 2035 or later.

This is especially true given that Chinese energy statistics are notoriously opaque. Just in the last few weeks it became clear that China burned perhaps 17% more coal per year in recent years than was previously understood.

China's 'peaking' promise is very unlikely to be achieved based on economic reality alone. The cost can be identified from the Asia Modeling Exercise which indicates that the lowest GDP loss would be about $400bn or about 1.7% of GDP, and likely twice that. It strains credibility to expect China to commit such economic self-harm.

(It is worth noting in passing that China also promises in its INDC to be "democratic" in 2050. The one-party state's vow should probably be treated rather similarly to the suggestion that it will rein in economic growth so dramatically).

Second: my approach is methodologically clear. The alternative is unable to avoid a slippery slope that would include every target, vow, promise, or vague political undertaking.

In my analysis, I was consistent in ruling out longer-term promises that were further off and economically implausible.

I also left out the US promise of "deep, economy-wide emission reductions of 80% or more by 2050." Data from the Stanford Energy Modeling Forum for the US shows an average GDP loss at more than $1 trillion annually, if done efficiently. If not, which seems to be the only constant in climate policy, the cost will likely double to almost $2.5 trillion or 7.5% of US GDP in 2050.

And I left out the EU promise "to reduce its emissions by 80-95% by 2050 compared to 1990." Data from the Stanford Energy Modeling Forum shows the average GDP loss at almost €3 trillion annually, if done efficiently. If not, the cost will likely double to almost €6 trillion or 25% of EU GDP in 2050.

If we were to include the Chinese 'peaking' promise, why not also include the US promise to cut 80% by 2050 and the EU promise to cut 80-95% by 2050, both of which are mentioned in their INDCs?

Including these promises would make a mockery of any real analysis of what the Paris treaty can achieve.

Indeed, since almost every nation has signed up to reduce temperature rises to 2°C, and about 80-90 nations including the EU and the US 'endorse' this target in their INDCs, where should we draw the line?

Third: the commitment period of 2016-2030 is by far the most common understanding of what Paris constitutes.

This is true whether we pay attention to the United Nations or at the official material from nations themselves:

The UNFCCC in its "Synthesis report on the aggregate effect of the intended nationally determined contributions" describes the central results as emission reductions achieved in 2025 and 2030, not further. It specifically labels possible emission reductions after 2030 as actions taken by nations "beyond the time frames stated in their INDCs (e.g. beyond 2025 and 2030)."
The US clearly states that its understanding of its INDC is for 2025 and not further: "The U.S. target is for a single year: 2025."
The EU sets its targets for 2030 and not any further.
In its own INDC, China clearly writes what it expects from the Paris agreement, namely to "formulate and implement programs and measures to reduce or limit greenhouse gas emissions for the period 2020-2030." So even China itself is unequivocal that the Paris deal is not about promises after 2030, but up until 2030.