Scholars and policymakers have long debated whether the Generalized System of Preferences (GSP) enhances development through increased trade – i.e., whether the program is effective as a form of ‘trade-as-aid’. We argue that, by itself, GSP increases poor-country exports, but that when recipients join the General Agreement on Tariffs and Trade (GATT) or its successor, the World Trade Organization (WTO), they realize fewer imports, and less gains in total trade, than GSP recipients that do not join the multilateral trading system. The logic is that GATT/WTO membership makes GSP more predictable by making it non-discriminatory, in the sense that exporters in recipient countries are less vulnerable to the program's ad hoc conditionality. This leads these exporters to lobby less against domestic protectionism, yielding higher trade barriers at home, and thus fewer imports. We test this hypothesis using a gravity model of trade, and data on all GSP programs, and find strong support for the argument that the GATT/WTO's interaction with GSP undermines trade.

The literature on escape clauses in international commerce focuses on the workings of trade remedies. The logic is that, by adhering to a strict methodology that is subject to legal review, trade remedies credibly signal that the government is only temporarily defecting from free trade. And yet, countries often turn, instead, to a measure that does not adhere to a strict methodology and is not subject to legal review: binding overhang, or the gap between a country's bound and applied tariffs. What explains a government's decision to use trade remedies or binding overhang? We argue that trade remedies are used where import surges are big enough that injury can be proven, but low enough that governments have incentive to prove it. Otherwise, binding overhang is their flexibility measure of choice. We conduct a variety of empirical analyses concerning 22 emerging economies with access to both trade remedies and binding overhang. The results strongly bear out our hypothesis, shedding new light on governments' incentives over the design of the law governing flexibility provisions.

International institutions often moderate the legal decisions they render. World Trade Organization (WTO) panels do this by exercising judicial economy. This practice, which is evident in 41 percent of all rulings, involves the decision not to rule on some of the litigants' arguments. The constraint is that it can be appealed. We argue that panels exercise judicial economy when the wider membership is ambivalent about the future consequences of a broader ruling. This is proxied by the “mixed” (that is, nonpartisan) third-party submissions, which are informative because they are costly, jeopardizing a more decisive legal victory that would benefit these governments too. We empirically test this hypothesis, and find that mixed third-party submissions increase the odds of judicial economy by upwards of 68 percent. This suggests that panels invoke judicial economy to politically appease the wider WTO membership, and not just to gain the litigants' compliance in the case at hand.

The landscape of the global economy is dotted with institutions that regulate investment and trade. In recent years, the number of bilateral investment treaties (BITs) and preferential trade agreements (PTAs), in particular, has grown at a torrid pace; practically every country is a member of at least one—if not many—of these institutions. For all the scholarly attention that these institutions have received, however, there is little research tying BITs and PTAs together. This is surprising, since both aim to increase commerce by making it more predictable. The authors seek to fill this gap in the literature. They argue that a BIT between a developed and a developing country should make it more likely that this pair of states will subsequently form a PTA. That said, the wrinkle in the story is that more is not better in this regard; the authors further argue that a developing country that has many BITs is less likely to conclude a PTA with a wealthy state. The authors test these hypotheses using annual data on pairs of developing and developed countries between 1960 and 2004 and find strong evidence in support of their argument.

Previous studies of WTO dispute settlement have sought to evaluate whether a Member's legal capacity influences its odds of bringing litigation before the multilateral trade regime. Because direct measures of legal capacity are elusive, these studies have had to use indirect proxies, such as per capita income or number of delegates in Geneva. Yet, the reliability of these measures is questionable. To gauge legal capacity more directly, we surveyed all WTO Members, posing questions concerning their professional staff, bureaucratic organization at home, bureaucratic organization in Geneva, experience handling general WTO matters, and involvement in WTO litigation. Using responses from this survey, we constructed an index of Members' WTO legal capacity that speaks more fully to the challenges of dispute settlement than do commonly used proxies, none of which are strongly correlated with our measure. We supplemented this survey with over three dozen semi-structured interviews with Members' representatives in Geneva, the results of which bear out the importance of properly measuring legal capacity.

Preferential trade agreements offer members an alternative to dispute settlement at the World Trade Organization. This gives rise to forum shopping, in that complainants can file regionally, multilaterally, or not at all. What explains this choice of forum? I argue that complainants strategically discriminate among overlapping memberships: on a given measure(s), some prefer to set a precedent that bears only on a subset of their trade relations, others a precedent that bears on all their trade relations, while still others prefer not to set a precedent. Thus, the key to forum shopping is not simply which institution is likely to come closest to the complainant's ideal ruling against the defendant, but where the resulting precedent will be more useful in the future, enabling the complainant to bring litigation against other members, rather than helping other members bring litigation against the complainant. I consider disputes over Mexican brooms and Canadian periodicals.For comments, I thank Vinod Aggarwal, Raj Bhala, Jane Bradley, Bill Davey, Rob Howse, Miles Kahler, Simon Lester, Rod Ludema, Ed Mansfield, Lisa Martin, Petros C. Mavroidis, John Odell, Joost Pauwelyn, Amy Porges, Eric Reinhardt, Peter Rosendorff, Ken Scheve, Ed Schwartz, Christina Sevilla, Michael Simon, Jay Smith, Debra Steger, Joel Trachtman, Todd Weiler, seminar participants in the Program on International Politics, Economics, and Security (PIPES) at the University of Chicago, and two anonymous referees. All shortcomings are, of course, my own. For research support, I thank the Canadian Institute for Advanced Research and the Social Science and Humanities Research Council of Canada. For research assistance, I thank Alex Muggah, Krzysztof Pelc, and Scott Winter.

Disputes filed at the World Trade Organization (WTO) are attracting a growing number of third parties. Most observers argue that their participation influences the institution's rulings. The authors argue that third parties undermine pretrial negotiations; their influence on rulings is conditioned by this selection effect. They test their hypotheses, along with the conventional wisdom, using a data set of WTO disputes initiated through 2002. Consistent with the authors' argument, they find that third-party participation lowers the prospects for early settlement. Controlling for this selection effect, the evidence also suggests that third-party support increases the chances of a legal victory at the WTO.

Does the geographic concentration of industry ‘matter’ outside the United States? Observers have long speculated that while geographically concentrated industries may be influential in American politics, this is probably not the case in countries where the electorate votes more as a national constituency. Others disagree, urging that clustered industries have an advantage regardless of how the political map is drawn. We sharpen the terms of debate and weigh in with empirical evidence from a cross-sectional analysis of intended voter turnout in eight member-states of the European Union and a multi-year study of voter turnout in the Netherlands. These tests uniformly show that, across different types of electoral systems, including those in which voters vote as a national constituency, thereby removing any effects of electoral geography per se, workers in traded industries that are physically concentrated are, in fact, substantially more likely to vote than employees in traded but geographically dispersed sectors.