Thanks, "theoriginalanon99". My reference to the DSB being more of a “referee” (rather than a “court” or “enforcer”) did not imply, or call for, “real time decisions [as in] in a sports match”, but to the DSB’s core, underlying function (mind you, in the good old days of football games, the new VAR system could also take annoyingly long!). Indeed, the DSU implies an adversarial system with reasoned decision-making. And this takes time. But can the system really not do better than the 3-4 years it now takes? I think it can, and that also the DSU drafters had something in mind that would be swifter. The DSU itself is full of time-limits and even provides for disputes in situations of urgency (cutting time limits in half). Hence, time was definitely on their minds … In other systems, like ISDS, delays are made up by retroactive damages and interests, plus, in domestic courts, specific performance. The DSU, on the other hand, is a bargain whereby Members agreed to delay and cabin unilateral retaliation, in return for swift multilateral determinations on breach (hence, my pun to “refereeing”). In this sense, the DSU limits enforcement and sanctions (to avoid escalation), it is not a “court” to “beef-up enforcement”. Now, if multilateral determinations on breach are massively delayed (one side of the bargain at risk), the pressure to fall back on unilateral sanctions (i.e. suspend the other side of the bargain) mounts. The best example of this may be Art. 22.6 retaliation procedures (according to the DSU, to be completed max. 60 days after the end of the RPT!): With the best of intentions, they have become extremely complex and time- and resource-consuming (taking way beyond the 60 days limit). I appreciate the goal of “getting the numbers right”, see the book I edited years ago https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1768128 But at the same time, “getting it exactly right” is really difficult (if not impossible), and may endanger the other, arguably more important, objective of prompt decision-making (even if it may be off by some margin). As an alternative to today’s 22.6 arbitrations, I have outlined a system of swift and potentially more “correct” arbitration (that could be done in a matter of weeks), using “final-offer” arbitration as is done in … baseball (but even here we are talking about players salary arbitration, not actual baseball game refereeing!). See: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3155363

Thanks, Jessie, Christian, points well taken. Indeed, God forbid that lawyers would be “calling the shots” during a pandemic (Jessie’s point; to his “statesmen and diplomats”, I would add health experts) and, no, an international (trade) tribunal won’t be saving lives or provide ventilators directly, and certainly cannot “modify rules” (Christian’s point). But does that mean that in times of crisis, rules (and potentially adjudication) should be lifted and play no role at all? My point was mainly that when it comes to the current WTO dispute settlement system, for anything with a shorter time span than 4 or 5 years (panel, appeal, implementation period, possible compliance procedure), it may not even be worth asking the question … Yet, if you look at the WTO treaty it does cover and try to discipline at least some situations of “crisis” (e.g. balance of payments crises under GATT Art. XII, unforeseen developments under Art. XIX/safeguards, “urgency” in the DSU, “prudential measures” under GATS etc.). That said, crucially, even where a WTO panel were to say something on crisis measures (and I agree with Jessie that, if at all, this ought to be done with a “ten foot pole”), as Christian was implying, it is not as if the DSB would “tell a country what to do”, i.e., force a country to sell ventilators. That was my other main point (and many critics of the DSU overlook this): unlike the ECJ or national courts, the DSB has no power to “force countries to export ventilators”. Instead, the system is there to decide whether, for example, an export restriction on ventilators, breaches the rule book. If it does, the losing country can lift the restriction or keep it but “pay for it” (with compensation or by suffering equivalent retaliation). Also the ECJ (in 1999, Portugal v. Commission) has read this to mean that the WTO system is not a “legal system” the way we normally think of it, and hence should not be given direct effect, i.e. tie the hands of domestic lawmakers directly (plenty of "wiggle room" remains). The key underlying goal is to avoid immediate and disproportionate retaliation, escalation and (mutually harmful) trade wars (not to force a country to sell ventilators). In this sense, the DSB is merely a referee calling “foul” or “no foul”, and it is then for the parties to take it from there (comply, mutual settlement, equivalent retaliation); the DSB has no power to send a player off the pitch; its main function is to avoid an escalating brawl that would spoil or stop the game altogether. But if the system has reaction power only after 3 years, it is hard to see how it can fulfill this core mission (which, again, is different from other courts or tribunals). If the referee takes 3 years or more to make up her mind, can we really expect players not to "foul" and, in turn, when an alleged foul does occur, offended players (especially big ones!) to patiently wait on the sideline?

John -- on your/the US' "why" question: for me, the best and most productive way to read and trie to answer this question is to ask "why" what I termed earlier "appellate review 1994" turned into "appellate review 2019". Trying to answer this question must not necessarily imply that appellate review 2019 is "wrong" or a violation of the DSU (some will say it is, others not; but that is not the "why" question). Members should be able to neutrally discuss this question, starting with an attempt to objectively identify the "core features" of appellate review 2019 (e.g. high appeal and success rate on appeal; increasingly long AB proceedings and AB reports; de facto rule of precedent; low number of dissents; important role for the Secretariat etc.). Again, listing these features must not mean they are necessarily undesirable. The next step is to think of "explanatory variables", and here again there may be plenty, some may be looked at as justifiable/legitimate, others not or less so (but that again is not the main point of the exercise), starting with the text of the DSU itself, AB working procedures, lack of input/control by Members, the types of panelists/ABMs appointed by Members, the role of law firms, the AB's approach to treaty interpretation etc. And, yes, other variables may come up as well, such as what I referred to as potentially different "philosophical views" on trade remedies (ABMs and their staff v. Rules Division staff & panelists) ... Elsewhere http://tinyurl.com/y4y8sj9x, K. Pelc and I discussed why the Secretariat may have such an important role and how that, in turn, may explain certain features of appellate review 2019 (in particular, its drive for consistency). You mention "the fact that the individuals involved in handling DS cases are professionally invested in the WTO having the largest possible footprint rather than the smaller one implied by in dubio mitius". There may be some of this, but as I wrote above, in many "systemic" (SPS, TBT, GATT) cases, overall, there has been a high degree of deference (the AB in EC - Hormones even mentioned in dubio mitius!). In trade remedy cases, I would say, less so ... So increasing the WTO's "footprint" or "turf" may explain some trends, but not others ... Once all the "features" and "variables" collected, the next step is then for Members to decide what features they want to keep and which ones they want to change. They can then do so by tweaking the variables to achieve the results they want (all of this is, of course, easier said than done, especially knowing that it requires consensus!). If approached in this way (rather than a "blame game on who destroyed WTO dispute settlement"), I would agree that delving deeper into the "why" question is, indeed, an indispensable exercise to go through. Otherwise, Members will never get out of this quagmire and even if they did may soon fall back into it (if they lack an understanding of what influenced the move from AR 1994 to AR 2019, for all we know, AR 2024 may end up back again at AR 2019 by the end of the decade ...).

Thanks, Steve, Marc – I think it is most likely both. As I wrote with Rebecca Hamilton a few years ago (see: http://tinyurl.com/ydflheed) looking at “exit” (real or attempted) from international tribunals across the spectrum (WTO, ICC, ICSID/ISDS, SADC, CJEU, IACtHR, ECtHR, …) recurring “exit drivers” include both (i) tribunal-centred and (ii) state-centred ones, in particular (i) (real or perceived) judicial “activism” and/or tribunal “bias” (Marc's point), and (ii) revealed sovereignty costs (“buyer’s remorse”) and/or ex post, political change in member states (Steve's point). When it comes to US "exit" from the AB, both AB “activism” (real or perceived, I’ll leave that in the middle) and declining US political appetite for something like “appellate review 2019”, go a long way in explaining what happened … And the former (real or perceived AB “activism”) obviously fed the latter (drop in US political support) … As Steve hints at, could “appellate review 2019” have adapted so as to deal with US discontent through “voice” rather than “exit” (and this, without crossing any of the “red lines” of ITPA)? That, of course, is (especially with the benefit of hindsight) the big question … A recent statistic shared by Chad Bown (see: http://tinyurl.com/rfxwfcp) reveals a lot: There is a massive discrepancy between the types of cases filed by and against the US: 65% of cases filed against the US were trade remedy cases; only 9% of cases filed by the US against other countries were trade remedy disputes (82% of cases filed by the US were cases that Chad calls “systemic”; think of SPS, TBT, GATT etc.). Now, it is probably fair to say (but correct me if I’m wrong) that the AB has overall been rather deferential in “systemic cases” (e.g. frequently reversing SPS or GATT Art. XX panels to offer more “policy space” to regulating countries), but pretty strict in trade remedy disputes (e.g. very often reversing “Rules panels” to clamp down on AD, CVDs and safeguards). So, from the perspective of the US, the AB may have been deferential to state sovereignty, but in the "wrong" type of disputes (and this, the US would surely add, even though the AD Agreement includes an explicitly deferential standard of review (Art. 17.6), whilst SPS, TBT and GATT do not). Why this has been the case is an interesting question. If you ask me, it may at least somewhat relate to a “philosophical view” ("object and purpose"?) held by many ABMs and their support staff, skeptical of trade remedies, largely inspired by the dismissive view of economists, very critical of trade remedies, but under-estimating the “political economy” reasoning behind AD, CVD and SG, especially the "political safety valve" role they play in trade agreements.

Thanks, John. You raise a very interesting (and difficult!) question: Is "automatic adoption" by the DSB of dispute reports a necessary feature of independent third-party adjudication (ITPA)? I did mention it and, if anything, was thinking, first and foremost, of the “I” in ITPA … Indeed, if adjudicators know that their report will only have “legal force” (that is, in case breach is found, trigger an international law obligation of compliance or a formal right for claimant to retaliate) if both sides accept the report, there is the risk that adjudicators “adapt their ruling” so as to make it politically palatable, and avoid blocking by either party. That was part of the experience under the old GATT (ask Frieder Roessler …). Another fear I would have when one of the disputing parties can “block the process” (be it by vetoing the adoption of a report or appealing it “into the void”) is that “power politics” re-enters the process between a ruling and giving “legal force” to that ruling (that “power politics” re-enter the scene after a ruling gets legal force is unavoidable, and was true also under DSU 1994-2019). Larger players such as the EU, US or China will, no doubt, find it easier to push weaker opponents into, for example, dropping their veto or accepting no-appeal pacts or Art. 25 appeal arbitration. For the system to then confirm compliance or trigger an obligation to comply or allocate formal retaliation rights may then depend as much on power and power asymmetries as it depends on an independent ruling by a third-party adjudicator. This risks tainting ITPA (both its “I” and “TP”) and, over time, may undermine trust in the system itself. Think of it: If I am a relatively weak WTO Member, say, Colombia, what may then be the point in “suing” the EU or the US if I know that, when push comes to shove, the EU or US may veto adoption or appeal “into the void”?; if, in turn, I am the EU or the US complaining and the losing defendant blocks the process, the temptation for the EU or the US to retaliate anyhow (“outside of the system”) will be immense. Do not misunderstand me: It is impossible to take “power politics” out of any international dispute settlement regime (in particular the "front-end" of who can/dares to file, and the "back-end" of who ultimately gets redress), but to openly invite it into the ruling itself (or when this ruling obtains, or not, "legal force") is another thing. This raises the obvious question: Why condition panel or AB reports on adoption by the DSB? Indeed, I do not know of any ITPA system that conditions rulings on adoption by the “legislative branch” (not even FTAs do this). So, no, in this sense, automatic adoption is not a necessary feature of ITPA: In most (if not all) ITPA systems “adoption” is not called for at all. Requiring DSB adoption may, on the contrary, be an indication that the DSU was not meant to be an adjudicative system in the first place (missing the “A” in ITPA, and this irrespective of whether adoption is veto-based or automatic!). After all, pursuant to the DSU, it is not panels or the AB that have the “final word”, it is the DSB, and the DSB alone, that makes the “recommendations or rulings” (see DSU Arts. 3.4, 11, 21 etc.). But we all know that in practice Members and the DSB have not been exercising much control. This, in turn, may have given a false sense to panels and the AB that they were genuine “adjudicators”, even a “court” (to my knowledge, DSU Art. 16.2, for example, stating that “Members having objections to a panel report shall give written reasons to explain their objections for circulation at least 10 days prior to the DSB meeting at which the panel report will be considered”, has never been used; if at all, Members simply make statements at the DSB itself). One alternative would be to just let adjudicators make their rulings, without requiring adoption by anyone (as happens in all other ITPA systems I know of, including USMCA or CETA) and to then leave it to the parties to use such ruling as a basis from which to negotiate a settlement, comply or retaliate. Another alternative is, in contrast, to beef up the role of the DSB and to relegate panels/the AB to a mere advisory (rather than "adjudicatory") role (which, for the USTR, may already be the case under DSU 1994). I wonder whether something in between is possible: panels (potentially an AB), with a narrowed down mandate (see my main post above) that make ITPA rulings (for me the “A” in ITPA covers both judicial/quasi-judicial and arbitration bodies, be they standing or ad hoc), with the WTO Membership/DSB exercising genuine control (not the rubber-stamping of today), perhaps less over case-specific "win or lose" (or at least no veto for any one of the disputing parties themselves), but more systemically, and this without endangering the independence ("I") of panels/the AB … The latter may well be the "holy grail" that WTO Members should be striving for in "reform" discussions today ...

Thanks, Ulli. The key "unknown" is, indeed, whether the US remains committed to independent, third-party adjudication (ITPA) under trade agreements. And that is where the broader community can contribute: demonstrating, using hard facts and data, that "red line" ITPA is in the US' (and, for that matter, every other WTO member's) own, national interest. Looking at the dispute settlement chapters of USMCA, and USTR statements at the WTO (confirming US commitment to the 1994 DSU text), there remains hope. I wonder whether, in this context, the corona virus will help or hurt: spark more cooperation, as with the launch of the Doha Round in 2001 in the wake of 9/11, or, on the contrary, trigger the worst, nationalistic instincts and make ITPA even more politically difficult ...

In recent work https://papers.ssrn.com/abstract=3458872 Prof. Pelc and I described and (for some parts, focused on panels) empirically confirmed the important role played by the Secretariat in WTO dispute settlement. Our main point in a subsequent Washington Post op-ed referencing this work http://tinyurl.com/ydn9ap5o, was that US critiques focused on ABMs alone, neglect the broader ecosystem of WTO dispute settlement which includes WTO staff, diplomats, private lawyers and academics, all of whom played a distinct role in how the system evolved from 1995 to where things stand today. We also expressed our conviction that “an acknowledgment of, and debate about, the role of the Secretariat is a necessary step to overcoming the current crisis and can only benefit the long-term trust in, and robustness of, WTO dispute settlement”. I could not agree more with the authors of the above blog post that “personal attacks on Secretariat staff … are an unhelpful distraction” and that it is regrettable that recent debates in the press “took on a distinctly personal tone”. Any critique of the Secretariat premised on the idea that ABMs are “unable to exercise independent judgment” and that somehow staff, let alone one single individual, makes all the decisions is, indeed, untenable. The real question, to be examined with a cool head and based on the facts, is, as stated, “the balance” or appropriate division of labor, “between the Secretariat and the adjudicators”. The question is not a black or white one, of whether the adjudicators or the Secretariat “make the decisions”. Both have a role. This role may vary over time, case by case and depend on particular staff and/or adjudicators. The division of roles may be appropriate or out of balance, depending also on what benchmark is used. Yet, that ABMs sign reports, generally “make their own decisions” and “disregarding the Secretariat’s position is routine” does not negate a role also for the Secretariat. Nor does it answer the crucial question of “balance” or appropriate division of labor (or, in the hyperbole of the post’s title, necessarily “slay concerns about the WTO’s faceless bureaucrats” which, I presume, refers to WTO staff, not WTO adjudicators). Indeed, in our work, we explain how in the WTO, the aggregate role of staff generally exceeds that of support staff in other tribunals, both domestic and international. In the WTO, staff is hired and fired by the DG and staff directors, not by WTO adjudicators themselves (be it panelists or ABMs); staff is permanent, while adjudicators are appointed ad hoc or for limited terms only; staff play a role in the appointment of panelists (not ABMs), approval of the variable parts of adjudicator compensation and reimbursements, setting of timetables and working procedures, and drafting of questions to the parties; staff also attend hearings, write detailed “issues papers” with proposed solutions, and actively participate in internal deliberations and the eventual drafting of rulings. It is this aggregate role (not, for example, the example of staff drafting panel reports taken in isolation) that must be assessed. The authors of the post acknowledge this overall important role for WTO staff when arguing that any reduction in “the considerable support provided by the Secretariat” would require Members to “enhance the adjudicators’ role”. As we noted in our paper, a rebalancing or adjustment of roles may, indeed, and somewhat paradoxically in the light of current US critiques of ABMs, require “more voice for adjudicators (e.g. more experienced panelists, full-time AB members and more scope for dissents), and a reduced role for the Secretariat (e.g. minimalist “issues papers” and less involvement in internal deliberations and drafting; adjudicator rather than division director oversight of Secretariat staff)”. If the aim of the above post (as it seems from the body text) is to initiate a cool-headed debate on the role of the Secretariat, I could not agree more. If, in contrast, the goal is to outright reject the fact that the Secretariat plays a substantial (and potentially outsized) role in the first place (which the title of the post -- "Slaying concerns about the WTO's faceless bureaucrats" -- seems to indicate), it does not seem substantiated and, more importantly, forecloses rather than opens a healthy and transparent debate to improve the system.

Besides self interest, loss of reputation and risk of emulation, the threat of unilateral retaliation is, indeed, the fourth “cost” of appealing into the void that I describe in my recent JIEL editorial on WTO Dispute Settlement Post 2019, p. 306-308, see https://academic.oup.com/jiel/article/22/3/297/5609188. All four of these “costs” mean that, indeed, most likely, the blocking of panel reports (by appealing them into the void) will not become the “new normal” post 2019: "appealing an adverse panel report ‘into the void’ may actually make things worse for the losing defendant. Under a DSB adopted panel report [claimant’s] retaliation would, legally speaking, be capped by the Article 22.4 equivalence standard and monitored by a separate arbitration procedure under Article 22.6. In contrast, [claimant’s] unilateral retaliation outside the DSU may not be so capped, cannot be challenged in Article 22.6 arbitration and may hence ‘punish’ the defendant even more. In the end, power relations may, therefore, push weaker parties not to appeal ‘into the void’." And it is this last sentence that makes me doubt that, as you put it Nicolas, the “threat of unilateral retaliation” will be enough to guarantee the continued automatic adoption of panel reports in all cases (no appeals into the void): 1. Where the claimant is relatively weak (say, Colombia winning a panel case against the EU, India or China), the threat of unilateral retaliation (by Colombia) risks not being credible, so it may not stop an appeal into the void by the EU, India or China, especially if the loss before the panel (suffered by the EU, India or China) is a sensitive one (think of India’s recent loss in the export measures case, DS541). In addition, all three of these defendants can credibly claim that they appeal into the void in the hope that the AB will be resurrected soon. 2. Where both parties are equally large/strong (say, US v. China or Korea v. Japan cases), unilateral retaliation (by the US/Korea) may be credible, but it may not stop an appeal into the void (by China or Japan; again, especially if the panel loss is a sensitive one) since counter-retaliation could be threatened (by China or Japan); in other words, rather than no appeal under the threat of unilateral retaliation, what may then happen is an appeal into the void and an escalating “trade war”. So the threat of unilateral retaliation may work to ensure automatic adoption of panel reports, but probably mostly/only in cases with a strong claimant v. a relatively weak defendant (say, US v. Mexico or EU v. South Korea disputes).

Fair point, Steve. This is also why, when we tried to link the "Secretariat's increased role" to expressed "US concerns", we focused on objectively verifiable features, such as increasing length of reports and proceedings, extensive reference back to prior rulings, and low number of dissents (not e.g. the AB's interpretation on zeroing or public body). I do not think that the existence of these trends can be doubted, or are "false statements". What explains or caused them is another issue and, yes, there are most likely many factors at play. But we do venture, in the paper, an argument that the (increased) role of the Secretariat may have had an influence...

On the normative side, let me just add something on moving closer to a “clerks” system, which both of you focus on. We put this out only as an idea to debate, at the end of the paper; it is certainly not the key point of the paper. Also, it probably makes more sense for staff assisting the AB, less for staff assisting panels (as panels are purely ad hoc; ABMs are there for four years, renewable once). Moving closer to a “clerks” system could also take many different forms (e.g. individual ABMs hiring entirely new people for their term in office; assigning existing staff to individual ABMs on a case by case basis; keep staff assigned to an entire division but have them operate under the control only of ABMs, with the director of the AB Secretariat taking on a purely managerial/registrar-type role etc.). The bottom-line is that moving closer to a “clerks” system would probably do two things: (1) reduce the “collective voice” of the Secretariat, whatever that voice may be inclined to do (I do think that it can systemically be linked to a preference for consistency, sticking to precedent, upholding of past practices etc.), and (2) enhance the voice of individual (Members-selected) AB members, as they would be the ones directly controlling their "clerk(s)", and also because individual ABMs would then have “their own assistant” to help explain and defend their views and positions in a given case (now, if you are not a native English speaker or are completely new to the system or have more of a diplomatic rather than a legal technical background, it may be very hard for individual ABMs to enter the internal debates, be able to phrase their position in the legal/technical/precedent-laden discourse that currently prevails etc.). In other words, re-arranging staff support closer to a “clerks” system (which, by the way, can be done without changing the DSU, no “can of worms” opened there either; one just needs to change internal, WTO Secretariat policy) may be one element in re-calibrating the playing field somewhat between adjudicators and staff. On Sungjoon’s point that this would create all kinds of concerns related to nationality, who the clerks would be etc., the same concerns could already now be raised in respect of current Secretariat hires. Indeed, as pointed out above, staff influence under the current system is more important than what it would likely be under something closer to a “clerks” system. Obviously, either way, existing Rules of Conduct (which do regulate potential conflicts of interest) would continue to apply.

Thanks, Simon, and Sungjoon, for your comments. Interesting how both of you jump to the normative parts of the paper, i.e. the impact of the increased role of the Secretariat and what to correct/change, if anything. In my view, the main contribution of the paper is rather descriptive: (i) offering empirical proof of the Secretariat’s involvement in the drafting of panel reports (the data-driven methods we use are quite powerful); and (ii) describing -- and offering reasons for -- the rise, over time, of the role of the Secretariat across a number of functions (issues paper, questions to parties, internal deliberations, drafting of reports etc.), both in panel and, importantly, also in AB proceedings. To be clear, and in support of what Sungjoon writes, there is, and remains, an absolute need to have the WTO Secretariat support both panels and the AB. As I pointed out after involvement in a CAFTA state-to-state trade dispute, the fact that PTA dispute mechanisms do not have a (permanent) Secretariat is a serious drawback. See http://tinyurl.com/jofkky4 Yet, when it comes to the WTO, the question is whether today’s division of labor between (part-time) adjudicators and (permanent) staff is out of kilter and may even partly -- I repeat, partly -- explain some of the outcomes the US is so upset with. If so, this is not a question of re-designing the system to improve it (and thereby unnecessarily “opening a can of worms”, as Simon puts it), but rather an existential exercise to save the system …

"Farming out" drafting in itself is, indeed, not necessarily a problem (although in a fine-grained precedential system, like today's WTO, it is worth knowing about). The crux of our paper is, rather, that the Secretariat's role in drafting must be considered in the context of its other roles: panel selection, writing of issues papers, questions to parties, active participation in internal deliberations etc. And this, crucially, realizing also the dynamic of increasing asymmetries between (permanent) staff and (part-time) adjudicators. This, undoubtedly, raises questions of division of labor, who does what (and varies case by case). In terms of "efficiency", that is exactly the other point we make in the paper: the (increasing) role of the Secretariat, and how it gets involved, may, over time, actually make things more complicated, and longer, so less efficient. On top of that, it may partly explain some of the outcomes that at least one large WTO Member currently objects to (e.g. de facto rule of precedent, expansive rulings), to the point that it risks blocking the entire system. Hence, simply making sure that the adjudicator reads and agrees with what the Secretariat drafted (which, I have no doubt, is already generally the case today) does not ensure "we are all good" or "fixed" the problem.

Yes, just email: [email protected]

Only one caveat, Sungjoon: 90% compliance rate is with WTO rulings; not with WTO rules ... We do not know exactly how many WTO violations are out there & not challenged. Rule of law is not merely court compliance, but primary norm compliance, and my fear is that the WTO does much better on former (compliance with rulings) than on latter (compliance with rules). Based on WTO disputes actually filed (these days predominantly in trade remedies), there are entire fields of WTO law that are becoming "black holes" (e.g. GATT Art. XXIV, major parts of GATS & TRIPS etc.).

Thanks, Sungjoon, you are of course right. But through your lens, I am not even sure that WTO suspension, as calibrated in the case law, falls under the pole of "enforcement/retaliation" (and, yes, it is certainly not "remedy/compensation"): it is merely reciprocal suspension, tit-for-tat reduction in trade flows (along the lines of Art. 60 VCLT, a kind of exceptio non adimpleti contractus), that as such is not really "instrumental" to achieve the purpose of repealing the violation (although factors other than formal WTO suspension are at work to achieve the impressive 90% compliance rate with WTO rulings). And of all things, I have a hard time seeing how WTO suspension, which has recently led to all kinds of settlements and compensations keeping the violation in place (Cotton, Clove, Hormones etc.) sustains a "trade rule of law". Again, it is much more a system of balance, than one of strict rule of law. And, mind you, this is not necessarily a bad thing!

Great points, Jonathan. But in Chorzow / general international law the quest is to define and compensate all "injury" caused by the breach. And, yes, in doing so, you do not want to compensate for speculative losses, so "full compensation" may not be achieved. Still, it makes sense to require proof/some proximity before making a violator pay. In the WTO, in contrast, identifying "injury" is not even the starting point/goal. The WTO follows the very mercantilistic idea that lifting a trade restriction is a "concession" (will "cost you") and restricting trade (or suspending a trade concession in retaliation) is a privilege (will "help you"). So what it asks following breach is: how much trade did the violator keep out; in return, the victim can do the same. Now imagine that, instead, we trie to identify the "injury" caused by a WTO breach. You would then indeed have to decide, first, injury to whom? To the entire country ("net loss") or only the harm caused to affected traders? If one considers net loss (e.g. impact on Canada as a whole of the US COOL measure), one could indeed en up with surprises: being the victim of a WTO breach may actually increase a country's overall welfare (think of export subsidies!); conversely, imposing a WTO inconsistent trade restriction will, in many cases, harm (not help) the violating country as a whole (steel tariffs may help steel producers but harm the overall economy). Still, if general welfare impact were the standard, economists can calculate this (see chapters by Schropp and Breuss in this book I edited a while ago: http://tinyurl.com/haq22tf). But it would mean a massive shift in approach for the WTO. If, instead, one considers injury only to affected traders (e.g. Canadian cattle/meat traders in the COOL case), the way one considers damages to a specific investor in ISDS (without offsetting benefits that may flow elsewhere in the home state), then what Canada/Mexico asked for in the COOL arbitration (counting also domestic losses) makes sense. Indeed, also in the Cotton retaliation for US actionable subsidies, the loss to Brazilian cotton farmers was calculated (without offsets for lower cotton prices in Brazil benefitting Brazilian consumers). Either way, if one adopts an injury approach (be it net loss or affected traders loss), one must then do the same when calculating the impact of the retaliating measure, so as to ensure "equivalence" between the impact of the violation and that of the retaliation (in COOL, this would mean that one would then have to account for the net loss on the US economy as a whole, or all losses suffered by affected US traders, as a result of Canada's retaliation for it to be equivalent to the injury caused by US COOL). Again, this all sounds complicated, but it can be done (approximately) by economists. I am not favouring any of these three approaches (trade losses, national net loss, economic impact on traders); only pointing out the systemic/distributional differences for the broader system as a whole (for one thing, if you are a large country / price maker, like the US, a trade restriction by a small country is unlikely to affect US domestic prices so for the US not counting for domestic losses in retaliation is less of an issue).

Yes, and this could also deal with some of the WTO-FTA issues we discussed earlier ...

Indeed, not all provisions can be contracted out from; contracting out that would undermine the treaty as a whole (whatever that means!) is prohibited. But the fact that an inter se agreement changes a provision in a multilateral treaty (here, right to a panel or prohibition of certain tariffs under Art. 4.2 AoA for specific Peruvian measure) is not, as such, enough to be prohibited by 41.1(b)(ii); you need more than that, a special case where the derogation destroys the treaty as a whole (if it does so one would think it also affects third party rights, so the two limitations in 41 at least partly overlap). AB already allowed for contracting out of right to a panel under DSU settlement so I do not see why doing the same under an FTA would undermine the effective execution of the WTO treaty any more than that ...

The WTO never publishes party submissions. Some members (such as EU, US) make their own submissions public on their own websites. I could not find anything either on Peru's external trade ministry website. But Peru's core arguments are summarised in the Panel report.

On (i), absolutely, Art. 11.1(b) of the Safeguards Agreement is one of few examples where the WTO explicitly prohibits certain inter se agreements (to be distinguished from the many instances where the WTO prohibits individual state behaviour, in regard of which WTO members can waive/consent inter se, as long as they do not affect third party rights) and, pursuant to VCLT Art. 41.1(b), in that case (Art. 11.1(b) SGA), contracting out is not "legal". On (ii), very much like for the VCLT, it depends on the specific ASR article: some are now commonly accepted as custom/general principles, others not (yet?).

Thanks, Julia. On (i), yes, international law allows states to consent to what would otherwise be a violation ("agree not to comply"), and they can do so unilaterally or by treaty (Art. 20 ASR); treaty law also allows for inter se modification of a multilateral treaty as long as, essentially, the WTO treaty does not explicitly prohibit this and third party rights are not affected (VCLT Art. 41 and Art. 30/31.3(a)). Those are two important conditions and the WTO does not generally prohibit inter se modifications or inter se contracting out (on the contrary, it explicitly allows for FTAs etc.). Also, the fact that one Member consented/agreed/waived, cannot take away the rights of other WTO members. A good example is when the EU and US were allowed to "contract out" of DSU Art. 17.10 on confidential AB proceedings. The DSU prohibits open hearings, so to hold open hearings would violate the DSU. But that does not stop 2 WTO members from waiving their rights by agreement (nothing in the DSU prohibits such agreement). And the AB permitted this for as long as the confidentiality rights of other members were not affected. On (ii) no real difference in status between VCLT and ASR if and to the extent both represent customary international law. But ASR is somewhat easier to digest since it waives/consents to case/claim/measure specific issues. VCLT, in contrast, is more drastic and relates to legal modification of treaty between parties in general; or interpretation of treaty between parties generally, also beyond the specific case/measure.

The Spanish original states, in full: "Perú podrá mantener su Sistema de Franja de Precios, establecido en el D.S. No 1152001EF y sus modificaciones, respecto a los productos sujetos a la aplicación del sistema indicados con un asterisco (*) en la columna 4 en la Lista de Perú establecida en este Anexo". And keep in mind also, as I pointed out in my main post, that there still is the issue of the FTA not being in force (yet?): a major hurdle, but not insurmountable, for Peru, depending on the broader facts in the case. Press statements (ICJ Nuclear Tests cases) and promises during panel proceedings (US - Section 301) have been found to unilaterally commit states. But here, if there is nothing beyond the FTA as such, Guatemala could argue that its consent re the PRS was dependent on Peru's reciprocal ratification of the FTA. So as long as Peru does not ratify the FTA, arguably, the consent is in suspense. So best would, no doubt, be for Peru to ratify the FTA; in that case, difficult for Guatemala to deny there is consent in the sense of Art. 20.