Thursday, April 17, 2014

Chapter 5 and the Summary for Policy Makers

Chapter 5 was one of the main chapters of the Working Group III 5th Assessment Report at the centre of the controversy this week on so-called censorship of the Summary for Policy Makers (SPM). The SPM is an executive summary of the report for the IPCC member governments. Those member governments get to dictate what points from the underlying report get included in this summary and how they are "spun". However, there is also a Technical Summary that is written entirely by the researchers responsible for the main report. The material from Chapter 5 that was in the draft SPM but eliminated in the plenary meeting in Berlin referred to emissions from specific groups of countries. This blogpost provides a quick overview of the deleted figures, some of which are still in  the Technical Summary.

The first graph breaks down emissions by broad global regions:

The developed countries are represented by the members of the OECD as it stood in 1990 (since then Mexico, Korea, Czech Republic etc. have joined). Eastern Europe and the former Soviet Union are designated "Economies in Transition" and the developing world is broken down into Asia (importantly including China and India), Latin America, and the Middle East and Africa. The left-hand panel shows emissions year by year since the Industrial Revolution and also breaks them down into energy and industrial and land use related emissions. The former continue to increase but the latter appear to have peaked. Since the 1970s, the majority of growth in energy and industrial emissions has come from developing countries and particularly Asia. In an attempt to better represent the historical responsibilities of each group of countries the right-hand panel shows the cumulative historical emissions of greenhouse gases by region.* China and particularly India have campaigned to get historical contributions to global warming better-acknowledged. But the results of our analysis show that less than half of the cumulative emissions now come from the developed countries as a whole (more when only energy and industrial emissions are considered). This, presumably, isn't the message that developing country delegates wanted to see.

The next controversial figure breaks down total and per capita greenhouse gas emissions by country income groups:

The leftmost panel shows total emissions which increased everywhere due to population growth. But they particularly increased in upper middle income countries (which includes China). The total emissions from this group are now almost equal to that from the high income countries. On a per capita basis, emissions were flat in the developed world and declining in the poorest countries (as emissions from land use declined). They rose in the middle income countries. The figure does, however, also show that in all developing country groups per capita emissions remain much below those in the developed countries.

The final deleted figure deals with emissions embodied in trade:

Looking at the emissions generated in producing imports and exports, the developed countries and economies in transition ("Annex B") import more "embodied" emissions than they export. The opposite is true of the developing countries ("Non Annex B"). Emissions that include the net emissions embodied in trade are termed "consumption emissions" in contrast to the "production emissions" that are the total emissions emitted within a country and are the usual way of calculating emissions.** These numbers are derived using input-output modelling. The results are often used to argue that developed countries have reduced their emissions by offshoring production to developing countries, which is a controversial question. But properly answering this question is more complicated than this. They are also used to claim that developed countries are responsible for their consumption emissions rather than their production emissions. But both importers and exporters gain from this trade. Because of these controversies I can understand the decision to drop the discussion and figure from the SPM.

* These do not directly correspond to the amounts of gases in the atmosphere. A large fraction of annual carbon dioxide emissions are absorbed by the ocean, vegetation etc. and methane only survives for an average of 11 years in the atmosphere before being oxidised to carbon dioxide and water. So, I am not very enthusiastic about treating cumulative emissions of carbon dioxide equivalent greenhouse gases as an indicator of historical responsibility.

** Economists would usually use the term "production emissions" to refer to emissions from production activities  and "consumption emissions" to refer to emissions by consumers. This initially caused some communication problems among researchers from different disciplines in our chapter team.


  1. This comment has been removed by a blog administrator.

  2. You are great policy maker and that's prove you in this blog.So i am definitely like..exceptional student education We provide special help for parents and child for special needs. To evaluate your child`s special needs you can take our help. Come with us for better education.

  3. Thanks for the blog post. I think it was quite bad that these graphs were removed. It is just the data!

    A couple of comments:
    * "I think these numbers are controversial as they depend on input-output modelling and its assumptions"
    Could you elaborate on that? Where is the controversy?

    * "They do not take into account the differences in emissions intensities across countries":
    That is not true. We use the GTAP database for our estimates which means we have about 130*57 separate emission intensities. They are linked by IO tables and bilateral trade data by sector. A matrix of dimension (130x57)x(130x57) links the emission intensities to consumption. The linearity assumption of IO is not so relevant here, particularly when you look at national footprints.

    * "do not show whether trade is allowing developed countries to lower their emissions"
    But that is a different research question, and one that consumption based emissions do not attempt to answer. It is an interesting question, but one needs different tools (IO may help, but more is needed).

  4. Glen - I'll look at the post again and maybe reword some things. I see this data often interpreted to answer the the question of whether offshoring of production reduces emissions. The emissions intensity point relates to that question. Anyway, I'll rewrite. All the graphs are in the technical summary and the chapter they just didn't get into the final version of the summary.

  5. I rewrote the post, what I was trying to say about the third figure wasn't well-worded in the original version.

  6. Thanks. Much better:). Yes, the data is often over interpreted...