Gabriel J. Michael / gmichael at gwu dot edu
This post is licensed CC-BY SA 3.0.
I’m back from South Africa, and today I’ll be attending “stakeholder presentations” on the Transatlantic Trade and Investment Partnership (TTIP)/Transatlantic Free Trade Area (TAFTA), a proposed bilateral agreement between the United States and the European Union. I’ll be live-tweeting from DC, so if you’re interested in what’s going on, please head over to Twitter. Right now, however, I wanted to update my sneak peek from last week, and offer some answers to the question: Which TPP proposals and chapters are most contentious?
As before, this analysis is based on the Wikileaks data reporting country positions on various proposals. Unlike before, rather than plotting differences between countries, in this post I plot differences between chapters and proposals. Thus, the farther apart two proposals/chapters are on a plot, the more disagreement exists among the negotiating parties with respect to that proposal/chapter.
First, we’ll take a look at all chapters combined. I struggled with the best way to approach this, but eventually settled on simply using column sums or column means of chapters to produce the distance matrix. Column sums are non-normalized, i.e., distance may be affected by the total number of proposals in a chapter. Column means are normalized. There’s no way to say which approach is better: column sums might exaggerate differences between chapters simply because we have more data on some chapters than others; on the other hand, if there really are fewer proposals in some chapters, column means might exaggerate the similarity between chapters. Given this, I’m including both charts.
The column sums approach will likely make more intuitive sense, given that most of the discussion has focused on chapters for which we have the most data. The intellectual property chapter is a huge outlier, denoting serious contention. Other contentious areas include the environment, market access, and investment chapters.
As noted above, the column means approach normalizes the number of proposals in each chapter. This almost certainly overstates the level of contention for chapters with only one leaked proposal (competition and customs). In spite of this, although the customs chapter now appears to be the most contentious, the intellectual property chapter is still quite contentious.
Before moving on to proposals in each individual chapter, I should emphasize that lack of contention does not mean acceptance. That is, proposals falling close to the origin may either be acceptable to most parties, or unacceptable to most parties.
As I noted in last week’s sneak peek, looking at proposals in the intellectual property chapter, the most contention surrounds the issues of the TRIPS/WIPO exclusions to national treatment, the Chilean ISP copyright proposal, copyright technical protection mechanisms, and scent trademarks.
In the environment chapter, there is significant disagreement on most proposals, although “subnational coverage” (presumably whether provisions apply to localities like states or provinces) and a biodiversity proposals appear uniquely contentious.
From what I can gather, the “DL600 Annex” is a reference to investor-state dispute settlement (ISDS) provisions, one of the most troubling aspects of the TPP. Such provisions allow businesses to sue states for actions perceived to harm investment. One of the most high profile ISDS cases currently involves Philip Morris suing the Australian government over Australia’s plain packaging tobacco regulations.
The legal chapter contains the controversial “medicines annex”, as well as a joint U.S.-Malaysia proposal on a “tobacco exception”. Earlier reports had indicated that the U.S. would not seek to exempt tobacco from tariff reductions, causing some public health groups to criticize the U.S. position. The “medicines annex” has been reported to be an attempt by the U.S. to undermine cost reduction measures for drug purchases in other countries.
As discussed above, it’s important to remember that less contention does not mean acceptance. The “medicines annex” in this chapter is a good example: while it is relatively close to the origin, this is because 9 of the 12 parties are rejecting the proposal.
The following graphs report issue distances/contention for the market access, rules of origin, services, and technical barriers to trade chapters.
The competition, customs, and government procurement chapters have too few proposals to perform multidimensional scaling, so they are not included. The sanitary/phytosanitary chapter is not included because one of the proposals (technical consultations) consists almost entirely of reserved positions.
Unlike my graphs of distances between country negotiating positions, in this post I’ve eliminated the assumption that a “reserved position” falls equidistant between rejection and acceptance of a proposal. Instead, reserved positions are treated as missing data.
Strictly speaking, it’s not clear that these divisions represent “chapters” or if they are just convenient divisions from the leaked document. While we know that there is an entire intellectual property chapter, other reporting has suggested that the “medicines annex” referred to in the “legal” chapter above is in fact part of a “transparency” chapter, for which we have little other information.
I am happy to share the code and data used in this post, although it is a mess and I am swamped with work right now, so it may be a while before I am able to make it presentable and respond to requests.