Reflections on the Paris Conference

by Daniel Bodansky

[Dan Bodansky is the Foundation Professor of Law and Faculty Co-Director of the Center for Law and Global Affairs at the Sandra Day O’Connor College of Law at Arizona State University. He was in Paris last week for the climate change negotiations. This is his seventh and final post regarding the conference. Professor Bodansky has consulted for the government of Switzerland and the Center for Climate and Energy Solutions (C2ES) in relation to the Paris Summit. However, he is writing in his personal capacity and the views expressed do not necessarily represent those of the Swiss government or C2ES.]

The dust has barely settled on the Paris conference, but already it has prompted an avalanche of comment – and there will be much more written, of course, in the coming weeks and months. For a summary of the conference results, see the report by C2ES. The following are a few more personal reflections.


Invidious comparisons of the Copenhagen and Paris conferences have been a staple of commentary. Copenhagen was a disaster and Paris a triumph. That is the usual line.

But, in many ways, Paris merely formalizes and extends the bottom-up paradigm to which the 2009 Copenhagen Conference gave birth. Indeed, most of the key elements of the Paris Agreement can be found in the Copenhagen Accord:

  • The goal of holding global warming below 2° C.
  • The system of national pledges to reduce emissions (referred to in the Paris Agreement as “nationally determined contributions”).
  • The non-binding character of these contributions, and the reliance on transparency rather than legal enforcement to promote accountability and effectiveness.
  • The shift away from the binary approach to differentiation that was at the heart of the Kyoto Protocol, towards a more flexible approach that encompasses all countries.
  • The pledge to mobilize climate finance from public and private sources (with a target figure of $100 billion specified in the Paris decision text but not the agreement itself).
  • And, perhaps most importantly, the extension of the regime to address the vast majority of global emissions, rather than focusing only on the emissions of “developed” countries.

In essence, what the Paris Agreement does is tie a treaty ribbon around these key elements of the Copenhagen Accord.

To be sure, the Paris Agreement does not simply recapitulate the Copenhagen Accord; it represents a progression down the Copenhagen path. The Paris Agreement builds on Copenhagen in three important respects.

Differentiation – First, it creates a more common system for all countries than the Copenhagen Accord. Copenhagen still retained elements of the binary approach to differentiation of the Kyoto Protocol, with distinctions drawn in various provisions between Annex I and non-Annex I parties. In contrast, the Paris Agreement completely abandons the Annex I/non-Annex I bifurcation. Most of the commitments in the Paris Agreement apply to all parties, including the commitments to formulate, communicate and update NDCs, and to submit the information necessary to track progress in implementing NDCs. Instead of the binary system of MRV in Copenhagen, with international assessment and review (IAR) for Annex I parties and international consultation and analysis (ICA) for non-Annex I parties, the Paris Agreement provides for a common system of reporting and review, with built-in flexibility to take account of different national capacities and circumstances. In addition, while the financial commitments of developed countries are reaffirmed, the Paris Agreement enlarges the donor base by encouraging other countries to provide financial support.

Durability – Second, the Paris Agreement gives the Copenhagen architecture a more durable character, both by putting it in treaty form and by calling for successive NDCs every five years, to be informed by periodic global stock-takes. The Copenhagen Accord addressed only the period up until 2020, through a one-off pledging process. The Paris Agreement, in contrast, establishes an ongoing regime of indefinite duration.

A rule-based structure – Finally, the Paris Agreement has a hybrid architecture, supplementing the bottom-up system of NDCs with internationally-negotiated rules to introduce some discipline into the national pledging process, which had been something of a free-for-all in Copenhagen. For example, the agreement requires parties to provide the information necessary to ensure that their NDCs are clear and transparent and to track progress in implementation, and provides that NDCs are to become progressively more ambitious over time.

Why Did Paris Succeed? Process and Product

In every important respect, the Paris Agreement moves away from the positions of those that opposed the Copenhagen Accord (or failed to support it the final night when the conference unraveled) and towards those that supported it. The Paris Agreement is less differentiated, provides for stronger transparency, and has a stronger legal form than Copenhagen. So why was Copenhagen rejected and Paris accepted?

Some point to the deft diplomacy of the French, and this certainly played a role. But I think it is overstated.   In Copenhagen, the Danish presidency was systematically undermined by those who wanted the conference to fail. Before the conference had even begun, they leaked the Danish negotiating text; then, during the conference, they repeatedly blocked attempts to convene a smaller group to work out the agreement, arguing that such a group would be undemocratic and untransparent – even though “friends of the chair” groups have a long pedigree in the UNFCCC process and had been the primary means of resolving crunch issues at virtually every COP prior to Copenhagen.

Of course, the Danes created openings for those who wished to raise procedural objections, most importantly, by not using the text from the official U.N. negotiations as the basis for their compromise text. The French learned from this mistake. From early on, they insisted that they did not have a separate text in their back pocket and would work with the negotiating text emerging from the ADP. That helped defuse suspicions and allowed countries to feel that they had ownership of the Paris Agreement.

But if countries had wished to raise procedural objections in Paris, they still had opportunities to do so, since in many ways the Paris endgame was even less transparent than Copenhagen. Rather than convene a broadly representative “friends of the chair” group, as the Danes tried to do, the French presidency held a series of consultations with individual countries during the final day and night, and individual delegations and groups of delegations met informally with one another. Virtually no one knew who was meeting with whom, and where the text stood. The result of this fluid, ad hoc process was a text containing new provisions of unknown provenance, which most delegations saw for the first time when it was presented to them in final form on Saturday afternoon, hours before the end of the conference.

Why did countries accept this final text, rather than raise procedural objections? To say that they did so because they liked and trusted Minister Fabius and wanted to reward his excellent job chairing the conference would trivialize the issues at stake. Yes, the trust he engendered helped, but it was not the main reason why Paris succeeded. The explanation instead lies in three factors:

First, while the Copenhagen architecture had been new in 2009, countries had become familiar with it by the time the Paris conference was held. After failing to gain acceptance in Copenhagen, it had been largely adopted the following year in the Cancun Agreements, and was reflected in the decisions leading up to Paris. So the foundation of the Paris Agreement had already been laid. Countries knew at least the broad outlines of what they were getting.

Second, the positions of several key developing countries, including China and Brazil, had evolved considerably.   In Copenhagen, they still hoped for an indefinite continuation of the Kyoto Protocol; indeed two years later, in Durban, adoption of a second commitment period under Kyoto was a precondition for agreement to begin the Paris negotiations. Although no decision was ever formally made not to continue Kyoto, everyone seem to have accepted by Paris that the next phase of the climate change regime would be a single agreement applicable to all parties. The joint announcement by the United States and China in fall 2014 heralded this shift, and gave many observers confidence that Paris would succeed where Copenhagen had failed. In Paris, the Kyoto Protocol was the dog that didn’t bark, and it looks like it will now go gently into the night.

Finally, expectations for Paris were much more realistic than Copenhagen.   Leading up to Copenhagen, it was obvious to close observers that countries would not be able to reach a legal agreement. But most people expected a legal agreement and took this as the benchmark of success, so it was easy to portray Copenhagen as a failure.

For Want of a “Should” the Paris Agreement Was Almost Lost

The end game of COPs is always a process of trench warfare, in which virtually every word is fought over, and gains and losses are measured in commas and brackets. One has to be a COP-ologist, familiar with the subtle history and nuance of every provision, to follow the to and fro.

For international relations scholars who wonder whether international agreements have any effect, the obsession with words may seem peculiar. But at least sometimes they matter.

This was particularly true of the last issue to be resolved in Paris: whether the provision that developed countries undertake absolute, economy-wide emission targets (Article 4.4) would be a “shall” or a “should.”  The ability of the United States to join the Paris Agreement hinged on which word was chosen, so the issue could not have been more consequential. Arguably, if the provision said “shall” and hence represented a legal commitment, then Senate or Congressional approval would have been required for US participation, whereas if the provision was a “should,” then the Paris Agreement could be accepted by the President as a presidential-executive agreement. Although the issue first erupted publicly the final afternoon of the meeting and delayed the closing plenary by more than an hour, reportedly it had featured prominently in the negotiations for some time and had consistently been resolved in earlier drafts in favor of “should.” How the word “shall” appeared in the final text presented on Saturday afternoon is a bit of a mystery. But, in any event, the issue was finally resolved through a Secretariat statement that use of the word, “shall,” was a “technical” error, and that “shall” would be replaced by “should.” A gasp went through the room when the Secretariat read the change, but no one raised an objection, and Fabius gaveled the agreement through by acclamation.

And the Show Goes On

The Paris conference gave new hope to the UN climate change process. But much remains to be done. Countries were able to agree only to the basic structure of the new climate change regime – the cycle of NDCs, reporting, review, stocktaking, and updating. Now, they must elaborate more detailed rules for how the Paris Agreement will work in practice – rules for reporting and review, international emissions trading, and a host of other issues.

If the Paris Agreement represented a true consensus on a new paradigm, then the next phase of the negotiations might be a more technical one. But I think that many developing countries accepted the move away from binary differentiation only reluctantly. And if twenty years of following the climate change negotiations has taught me anything, it is that nothing is ever fully settled; countries constantly push to regain ground that they had previously ceded.

The Paris Agreement is a particularly important moment in the ongoing process of climate negotiations. But it is not the end. At least for the foreseeable future, the UN climate change regime will likely remain a full employment agreement for climate negotiators.

13 Responses

  1. So Dan, let me see if I have this straight. You would say that the Paris agreement is in fact a treaty, in the formal international law sense. However, because the specific provisions in the treaty do not comprise legal obligations for states, including the US, the US therefore doesn’t have to go through the senate process for consent to ratification. Is that right? Please correct where incorrect.

  2. Hi Dan, No, that’s not quite right. The first part is correct, that the Paris Agreement is a treaty within the meaning of the Vienna Convention on the Law of Treaties. But whether it requires Senate advice and consent to ratification doesn’t turn on whether it imposes legal obligations on the US. Many agreements that impose legal obligations are accepted by the US as executive agreements, either with Congressional approval (congressional executive agreements) or by the President acting alone (sole executive agreements). So the question is whether the particular kinds of obligations included in the Paris agreement fall within the President’s existing authority to accept (for example, because they are within his constitutional authority over foreign affairs or are authorized by an existing treaty), or or whether they require some type of legislative approval.

  3. Thanks, Dan. Yes I read over the Paris agreement and it definitely reads like a treaty. I read Marty Lederman’s piece over at Balkinization too. So now along with your explanation, I get the US law angle better.

    I’ve just been comparing the Paris agreement with the recent Iran deal (the JCPOA) and thinking about the similarities/differences between them.

  4. Well, from where I stand, I see it is a failure.
    There are no legally binding agreement on mitigation, there is only ‘intent’ and no ‘processes’ or ‘committed targets’. In sum developed countries have got their way and developing countries have lost out in the process.
    It may not be evident immediately, but as all other agreements will from now on be based on this treaty, in effect differentiation has been lost forever which was a key negotiating point for developing countries.

  5. I agree for the most part with Dan B.’s summary of which sorts of provisions the President can, and cannot, conclude unilaterally (without Senate or congressional approval). If, as Dan says, “particular kinds of obligations included in the Paris agreement fall within the President’s existing authority to accept” because there is an existing statute or treaty that authorizes him to accept them, then there shouldn’t be much of a constitutional problem.

    However, if there is no treaty or statute authorizing the President to commit to obligations on behalf of the United States, then his power to do so would be much more constrained. To be sure, the Court has recognized the President’s authority to conclude “sole executive agreements” in certain discrete contexts to which Congress has, in effect, long acquiesced. (Belmont, Pink.) But the Court in Medellin indicated that that category is fairly narrow. And, most importantly, to the extent Dan is suggesting that the President can unilaterally commit the U.S. to any matters that are “within his constitutional authority over foreign affairs,” I think that’s mistaken: His authority over foreign affairs extends to virtually all matters of negotiation–but, as noted above, only a very few of them may be concluded pursuant to sole executive agreements.

  6. Thanks, Marty. But doesn’t the President’s foreign affairs power include not only negotiating with foreign governments, but also communicating with them? So, to the extent the Paris Agreement involves such communication — communicating an NDC, providing information about implementation, taking part in a review process — doesn’t that fall within the President’s constitutional authority? Dan

  7. Yes, communicating with foreign states is part of the President’s constitutional authority. Indeed, I think it’s a part of his authority that is largely beyond statutory regulation.

    But that, in and of itself, is not sufficient to demonstrate that he can unilaterally commit the United States, as a matter of international law, to making such communications–after all, such a commitment binds his successors, as well.

    Having said that, I would not be surprised if such “obligations to communicate” are often established by way sole executive agreements; and if there is a long history of that practice, perhaps Congress and the courts would affirm it. But whether it is the sort of commitment that can be made without Senate or congressional involvement does not turn on whether such communications are part of the President’s constitutional authority.

Trackbacks and Pingbacks

  1. […] The dust has barely settled on the Paris conference, but already it has prompted an avalanche of comment – and there will be much more written, of course, in the coming weeks and months. For a summary of the conference results, see the report by C2ES. The following are a few more personal reflections… lire la suite […]

  2. […] Bodansky, Reflections on the Paris Conference, Opinio Juris, Dec. 17, […]

  3. […] Bodansky, Reflections on the Paris Conference, Opinio Juris, Dec. 17, […]

  4. […] in other ways. Much of the content of the Paris Agreement is not new. As Dan Bodansky has aptly observed, “in essence, what the Paris Agreement does is tie a treaty ribbon around…key elements of the […]

  5. […] wurde die Ausweiterung von Reduktionsbeiträgen auf Schwellen- und Entwicklungsländer schon in Kopenhagen […]