Joe Romm inadvertently exposes why Paris Climate Agreement aims are unachievable


Joe Romm promotes a myth that the Paris Climate Agreement will make a huge difference to future greenhouse gas emissions. Below I show how the modelled impact of think tank Climate Interactive conclusion of a large difference is based on emissions forecasts of implausible large emissions growth in policy countries, and low emissions growth in the non-policy developing countries.


In the previous post I looked at how blogger Joe Romm falsely rebutted a claim that President Donald Trump had made that the Paris climate deal would only reduce only reduce future warming in 2100 by a mere 0.2°C. Romm was wrong on two fronts. He first did not check the data behind his assertions and second,  in comparing two papers by the same organisation he did not actually read the explanation in the later paper on how it differed from the first. In this post I look at how he has swallowed whole the fiction of bogus forecasts, that means the mere act of world leaders signing a bit of paper leads to huge changes in forecast emissions.

In his post  Trump falsely claims Paris deal has a minimal impact on warming, Romm states

In a speech from the White House Rose Garden filled with thorny lies and misleading statements, one pricks the most: Trump claimed that the Paris climate deal would only reduce future warming in 2100 by a mere 0.2°C. White House talking points further assert that “according to researchers at MIT, if all member nations met their obligations, the impact on the climate would be negligible… less than .2 degrees Celsius in 2100.”

The Director of MIT’s System Dynamics Group, John Sterman, and his partner at Climate Interactive, Andrew Jones, quickly emailed ThinkProgress to explain, “We are not these researchers and this is not our finding.”

They point out that “our business as usual, reference scenario leads to expected warming by 2100 of 4.2°C. Full implementation of current Paris pledges plus all announced mid-century strategies would reduce expected warming by 2100 to 3.3°C, a difference of 0.9°C [1.6°F].”

The reference scenario is RCP8.5, used in the IPCC AR5 report published in 2013 and 2014. This is essentially a baseline non-policy forecast against which the impact of climate mitigation policies can be judged. The actual RCP website produces emissions estimates by type of greenhouse gas, of which breaks around three-quarters is CO2. The IPCC and Climate Interactive add these different gases together with an estimate of global emissions in 2100. Climate Interactive current estimate as of April 2017 is 137.58 GtCO2e for the the reference scenario and the National Plans will produce 85.66 GTCO2e. These National would allegedly make global emissions 37.7% than they would have been without them, assuming they are extended beyond 2030. Climate Interactive have summarized this in a graph.

To anyone who actually reads the things, this does not make sense. The submissions made prior to the December 2015 COP21 in Paris were mostly political exercises, with very little of real substance from all but a very few countries, such as the United Kingdom. Why it does not make sense becomes clear from the earlier data that I extracted from Climate Interactives’ C-ROADS Climate Simulator version v4.026v.071 around November 2015.  This put the RCP8.5 global GHG emissions estimate in 2100 at the equivalent of 139.3 GtCO2e. But policy is decided and implemented at country level. To determine the impact of policy proposal there must be some sort of breakdown of emissions. C-ROADS does not provide a breakdown by all countries, but does to divide the world into up to 15 countries and regions. One derived break-down is into 7 countries or regions. That is the countries of USA, Russia, China and India, along with the country groups of EU27, Other Developed Countries and Other Developing Countries. Also available are population and GDP historical data and forecasts. Using this RCP8.5 and built-in population forecasts I derived the following GHG emissions per capita for the historical period 1970 to 2012 and the forecast period 2013 to 2100.

Like when I looked at Climate Interactives’ per capita CO2 emissions from fossil fuels estimates at the end of 2015, these forecasts did not make much sense. Given that these emissions are the vast majority of total GHG emissions it is not surprising that the same picture emerges.

In the USA and the EU I can think of no apparent reason for the forecast of per capita emissions to rise when they have been falling since 1973 and 1980 respectively. It would require for energy prices to keep falling, and for all sectors to be needlessly wasteful. The same goes for other developed countries, which along with Canada and Australia, includes the lesser developed countries of Turkey and Mexico. Indeed why would these countries go from per capita emissions similar to the EU27 now to those of the USA in 2100?

In Russia, emissions have risen since the economy bottomed out in the late 1990s following the collapse of communism. It might end up with higher emissions than the USA in 1973 due to the much harsher and extreme climate. But technology has vastly improved in the last half century and it should be the default assumption that it will continue to improve through the century. It looks like someone, or a number of people, have failed to reconcile the country estimate with the forecast decline in population from 143 million in 2010 to 117 million. But more than this, there is something seriously wrong with emission estimates that would imply that the Russian people become evermore inefficient and wasteful in their energy use.

In China there are similar issues. Emissions have increased massively in the last few decades on the back of even more phenomenal growth, that surpasses the growth of any large economy in history. But emissions per capita will likely peak due to economic reasons in the next couple of decades, and probably at a much lower level than the USA in 1973. But like Russia, population is also forecast to be much lower than currently. From a population of 1340 million in 2010, Climate Interactive forecasts population to peak at  1420 million in 2030 (from 2020 to 2030 growth slows to 2 million a year) to 1000 million in 2100. From 2080 (forecast population 1120) to 2100 population is forecast to decline by 6 million a year.

The emissions per capita for India I would suggest are far too low. When made, the high levels of economic growth were predicted to collapse post 2012. When I wrote the previous post on 30th December 2015, to meet the growth forecast for 2010-2015, India’s GDP would have needed to drop by 20% in the next 24 hours. It did not happen, and in the 18 months since actual growth has further widened the gap with forecast. Similarly forecast growth in GHG emissions are far too low. The impact of 1.25 billion people today (and 1.66 billion in 2100) is made largely irrelevant, nicely side-lining a country who has declared economic growth is a priority.

As with the emissions forecast for India, the emissions forecast for other developing countries is far too pessimistic, based again on too pessimistic growth forecasts. This mixed group of countries include the 50+ African nations, plus nearly all of South America. Other countries in the group include Pakistan, Bangladesh, Myanmar, Thailand, Indonesia, Vietnam, Haiti, Trinidad, Iraq, Iran and Kiribati. There are at least a score more I have omitted, in total making up around 42% of the current global population and 62% of the forecast population in 2100. That is 3 billion people today against 7 billion in 2100. A graph summarizing of Climate Interactive’s population figures is below.

This can be compared with total GHG emissions.

For the USA, the EU27, the other Developed countries and China, I have made more reasonable emissions per capita estimates for 2100.

These more reasonable estimates (assuming there is no technological breakthrough that makes zero carbon energy much cheaper than any carbon technology) produce a reduction in projected emissions of the same order of magnitude as the supposed reduction resulting from implementation of the National Plans. However, global emissions will not be this level, as non-policy developing nations are likely to have much higher emissions. Adjusting for this gives my rough estimate for global emissions in 2100.

The overall emissions forecast is not very dissimilar to that of RCP8.5. Only this time the emissions growth has shift dramatically from the policy countries to the non-policy countries. This is consistent with the data from 1990 to 2012, where I found that the net emissions growth was accounted for by the increase in emissions from developing countries who were not signatories to reducing emissions under the 1992 Rio Declaration. As a final chart I have put the revised emission estimates for India and Other Developing Countries to scale alongside Climate Interactives’ Scoreboard graphic at the top of the page.

This clearly shows that the emissions pathway consistent the constraining warming to  2°C will only be attained if the developing world collectively start reducing their emissions in a very few years from now. In reality, the priority of many is continued economic growth, which will see emissions rise for decades.

Concluding Comments

This is a long post, covering a lot of ground. In summary though it shows environmental activist has Joe Romm has failed to check the claims he is promoting. An examination of Climate Interactive (CI) data underlying the claims that current policies will reduce global temperature by 0.9°C through reducing GHG global emissions does not stand up to scrutiny. That 0.9°C claim is based on global emissions being around 35-40% lower than they would have been without policy. Breaking the CI data down into 7 countries and regions reveals that

  • the emissions per capita forecasts for China and Russia show implausibly high levels of emissions growth, when they show peak in a few years.
  • the emissions per capita forecasts for USA and EU27 show emissions increasing after being static or falling for a number of decades.
  • the emissions per capita forecasts for India and Other Developing Countries show emissions increasing as at implausibly lower rates than in recent decades.

The consequence is that by the mere act of signing an agreement makes apparent huge differences to projected future emissions. In reality it is folks playing around with numbers and not achieving anything at all, except higher energy prices and job-destroying regulations. However, it does save the believers in the climate cult from having to recognize the real world. Given the massed hordes of academics and political activists, that is a very big deal indeed.

Kevin Marshall 

Joe Romm falsely accuses President Trump understating Impact of Paris Deal on Global Warming

Joe Romm of Climate Progress had a post two weeks ago Trump falsely claims Paris deal has a minimal impact on warming

Romm states

In a speech from the White House Rose Garden filled with thorny lies and misleading statements, one pricks the most: Trump claimed that the Paris climate deal would only reduce future warming in 2100 by a mere 0.2°C. White House talking points further assert that “according to researchers at MIT, if all member nations met their obligations, the impact on the climate would be negligible… less than .2 degrees Celsius in 2100.”

The deeply prejudiced wording, written for an extremely partisan readership, encourages readers to accept the next part without question.

The 0.2°C estimate used by Trump may be from another MIT group; the Joint Program on the Science and Policy of Global Change did have such an estimate in early 2015, before all of the Paris pledges were in. But, their post-Paris 2016 analysis also concluded the impact of the full pledges was closer to 1°C.


This states

New in this edition of the Outlook are estimates of the impacts of post-2020 proposals from major countries that were submitted by mid-August 2015 for the UN Conference of Parties (COP21) meeting in Paris in December 2015.

So what INDC submissions were in by Mid-August? From the submissions page (and with the size of total 2010 GHG Emissions from the Country Briefs) we get the following major countries.

In box 4 of the outlook, it is only Korea that is not included in the 0.2°C impact estimate. That is just over half the global emissions are covered in the MIT analysis. But there were more countries who submitted after mid-August.

The major countries include

My table is not fully representative, as the UNFCCC did not include country briefs for Nigeria, Egypt, Saudi Arabia, Iran, Iraq, Kuwait and UAE. All these countries made INDC submissions along with a lot of more minor GHG emitters. I would suggest that by mid-August all the major countries that wanted to proclaim how virtuous they are in combating climate change were the early producers of the INDC submissions. Countries like the Gulf States, India and Indonesia tended to slip their documents in somewhat later with a lot of measly words to make it appear that they were proposing far more than token gestures and pleas for subsidies. Therefore, the 0.2°C estimate likely included two-thirds to three-quarters of all the real emission constraint proposals. So how does an analysis a few months later produce almost five times the impact on emissions?

The second paragraph of the page the later article Joe Romm links to clearly states difference in methodology between the two estimates.


A useful way to assess that impact is to simulate the effects of policies that extend the Agreement’s 188 pledges (known as Nationally Determined Contributions, or NDCs) to the end of the century. In a new study that takes this approach, a team of climate scientists and economists from the MIT Joint Program on the Science and Policy of Global Change led by research scientist Andrei Sokolov finds that by 2100, the Paris Agreement reduces the SAT considerably, but still exceeds the 2 C goal by about 1 C.

The primary difference is that the earlier study tries to measure the actual, real world, impacts of existing policy, and policy pledges, if those policies are fully enacted. In the USA, those pledges would need Congressional approval to be enacted. The later study takes these submissions, (which were only through to 2030) and tries to estimate the impact if they were extended until 2100.  That is renewables subsidies that push up domestic and business energy costs would be applied for 85 years rather than 15. It is not surprising that if you assume policy proposals are extended for over five times their original period, that they will produce almost five times the original impact. To understand this all that is required is to actually read and comprehend what is written. But Joe Romm is so full of bile for his President and so mad-crazy to save the planet from the evils of Climate Change and (mostly US) big business that he is blinded to that simple reality-check.

The fuller story is that even if all policies were fully enacted and extended to 2100, the impact on emissions would be far smaller than Joe Romm claims. That will be the subject of the next post.

Kevin Marshall