The Empirical Turn in International Economic Law
The Empirical Turn in International Economic Law*
In November 2010, the American Society of International Law’s International Economic Law Interest Group (ASIL IELIG) convened a broad cross-section of scholars, practitioners, and students of international economic law. The focus of this conference was International Economic Law in a Time of Change: Reassessing Legal Theory, Doctrine, Methodology and Policy Prescriptions. Surveying the field, we became aware of certain swings in attitudes—from skepticism to euphoria and back to skepticism again—toward the empirical work that, of late, has been seeping into the curriculum and research of legal academies. If we want to know how the world is changing—and how our legal rules shape and should respond to that change—empirical studies are simply unavoidable. At the same time, the real benefit of empirical studies is always a function of how intelligently such studies are conceived and executed. The empirical turn in international economic law is inevitable, but the meaning of that turn is not. Thus, as this paper will discuss, to ensure that legal scholars and empirical researchers are maximizing their ability to understand and influence this time of change in international economic law, collaboration—not distrust—between these groups will be essential.
The relationship between empirical social scientists, lawyers, and legal researchers often recalls that of Sherlock Holmes (the “sage”) and Dr. Watson (the “know-it-all scientist”). Sherlock Holmes and Dr. Watson were lovers of the outdoors, so come spring one year, it was hardly surprising that they decided to spend a weekend camping on the Sussex downs. As night fell, they pitched their tent, put on nightshirts, nightcaps and bed socks; and after a soothing cup of cocoa, said a cordial goodnight and went to sleep. A few hours later, the following conversation ensued:
Holmes nudged Watson, and said: “Watson, Watson, look at the stars!”
“What, what?” said Watson, roused from a deep slumber. “Ah yes, Holmes, the stars.”
“Well,” said Holmes, “what do you make of them?”
Watson, by then awake, summoned his academic training in the scientific method and said: “Well, Holmes, judging by the position of the stars and the moon, I deduce chronologically, that it is some three hours since we fell asleep; geographically, that the earth has rotated forty-five degrees during that time; astronomically, that the handle of the big dipper is still pointing to the north star; and finally, meteorologically, that we can expect a fine day tomorrow. Will that do?”
“You idiot,” said Holmes. “I meant that someone has stolen our tent!”
As Holmes points out, it is sometimes necessary to look at a situation from a broader perspective to see what is transpiring and not just focus on the details. Like Holmes, we will focus our attentions on the big picture: Where is the “tent”? What is the role of international economic law in a changing world, and how can the work of empirical social scientists complement that of legal scholars in thinking about and responding to that change?
I. The Empirical Turn in International Economic Law
We begin by defining what we mean by “empirical” research. At the most general level, empirical research is anything that is not purely theoretical or purely doctrinal. Selected anecdotes and isolated historical episodes are, in this sense, empirical. But what we intend to focus on in this paper is something more systematic: by “empirical” we mean a systematic examination of observable phenomena from which the researcher explicitly seeks to draw broader conclusions about the way the world—or some part of it—works. As we envision it here, empirical research is about trying to draw conclusions that, to some degree, are generalizable; that is, conclusions that apply to more than one specific historical case. This is the kind of empiricism lawyers should care about because, as scholars and practitioners, their professional raison d’etre is to develop rules with broad applicability.
The empirical turn in legal scholarship generally has been pretty well-documented. Indeed, there is even a law school ranking based on institutional strength in empirical legal studies. In the specific area of international economic law, the trend is less noted, but is on the rise. A Westlaw search of the “journals and law reviews” database shows that the number of articles containing the term “international economic law” and some variant of “empirical” or “statistical significance” has increased almost six-fold since 1998. Almost a third of that increase has been published in the past four years alone. Moreover, one of the leading international economic law journals—the Journal of International Economic Law (JIEL), which first went to press in 1998—is peer-reviewed, something considered “quirky” in the field of law. Its emphasis is on studying “fundamental, long-term, systemic problems and [offering] possible solutions, in the light of empirical observations and experience[,]” and it appears well able to do so: of the 56 members of the editorial board of the JIEL (which reads like a Who’s Who of IEL scholars, many of whom were at the conference), 30 have PhDs or SJDs, and thus are likely to have some training in quantitative methods. And of the 82 or so speakers and moderators at the conference, 32 (or 39%) have PhDs or SJDs.
It would be easy, of course, to overstate the significance of these figures. For example, at the ASIL IELG Conference at Bretton Woods in November 2006 on the state and future of the international economic law discipline, 43% (17 of 39) of the speakers and moderators had PhDs or SJDs—more, percentage-wise, than at the recent conference. Moreover, there is a striking contrast between the composition of the JIEL Editorial Board and this conference—whereas there are a good number of Economics PhDs. on the former, there were few at the recent conference. But it is no stretch to conclude that on top of the wealth of legal knowledge in our field, we also have a good amount of empirical expertise. Indeed, substantial empirical research is taking place in all types of international economic law. International trade law continues to be the main locus of empirical work, but great strides continue to be made to augment doctrinal legal analysis with empirical research in the areas of development, international finance and investment, and international arbitration. Not only is the use of empirical methodology becoming more prevalent in international economic law, but the trend itself is becoming increasingly self-conscious (this paper being but one such example).
II. What Can Empirical Research Really Tell Us?
What do we make of this trend? And what are we likely to learn as a result? Indeed, can we really trust empirical researchers to “get the world right?” Admittedly, empirical social scientists hardly have a stellar record of predicting outcomes. In 1972, the Club of Rome published its Malthusian treatise The Limits to Growth, in which it offered a prediction that the world might run out of oil by 1992. Throughout the 1980s and early 1990s, a familiar refrain was repeatedly heard from historians, economists and economic advisers about the rise of Japan, the absolute and relative decline in the competitiveness of the U.S., and the need to adopt activist industrial policy or forever stand in the shadow of the Rising Sun. In 1990, Professor John Mearsheimer forecasted the imminent decline of NATO (which since his prediction has grown from 16 members to 28) as well as the weakening of the European Community (which, now the European Union, has more than doubled in size, from 12 to 27 members, since his prediction). As for Francis Fukuyama’s “end of history,” since 9/11 it has been postponed indefinitely. And sadly, Keynes’ vision of a leisure society—“three-hour shifts or a fifteen hour week”—seems increasingly fantastical. Even in France.
When it comes to predicting outcomes, it seems the less we empiricists know the better we do. Take, for example, the Supreme Court Forecasting Project, “a friendly interdisciplinary competition to compare the accuracy of the different ways in which legal experts and political scientists assess and predict Supreme Court decision making.” We cannot attest to exactly how friendly it turned out to be, but the idea was quite intriguing: who would make the better prediction of how the Court would decide the next Term’s cases, statistical forecasters or constitutional law experts? The rules of this friendly competition were as follows. The legal experts could read anything, including past court decisions and even the parties’ briefs. They could also consider extra-legal factors if they so desired, such as the Justices’ policy preferences and ideologies. The statistical model, on the other hand, was parsimonious; it took into account only a few bits of information, such as the ideology of the circuit court from which the case was referred, the type of petitioner and respondent, the type of issue and whether constitutionality was at stake. Using these variables, the model coded every case decided by the Supreme Court for the past eight terms prior to 2002—some 628 cases. The results: the model predicted 75% of the 68 cases during the 2002 Term correctly. The legal scholars—with their far more detailed knowledge—made correct predictions approximately 59% of the time. Of particular relevance to our present topic, the statistical model hugely outperformed legal experts regarding economic issues before the Supreme Court; it correctly predicted how the Court would rule about 85% of the time. The legal scholars might have done better with a coin flip; they were right just less than half the time.
Prediction, of course, is not understanding. We can predict that the sun will rise tomorrow morning, but we attorneys have not even the most basic understanding of the laws of astrophysics. In the Supreme Court example, the empirical social scientists involved could not have cared less about understanding the contents of the law; they were concerned only with predicting the outcomes of discrete cases. Not all empirical social scientists, however, are so easily satisfied. Like lawyers and legal scholars, they too want to understand the law. To do so, empirical researchers have developed increasingly sophisticated methods that allow us to analyze broad patterns in complex legal texts, and to determine how varied actors understand and react to the existence of and changes in particular legal rules and institutions.
One powerful new tool allows researchers to build computer models that “search out” not only key words in texts, but also word clusters that take into account word-order meaning. This sort of methodology can be a particularly useful tool to understand very broad patterns in a very large number of comparable documents. Professor Arthur Spirling, for example, applies this methodology to a central issue in U.S. history: the 600 or so treaties and other agreements negotiated between the federal government and various Native American tribes. What distinguishes these treaties over time, he finds, is one major linguistic distinction: the harshness of the terms. Without having to read 600 treaties, then, Spirling finds a systematic pattern that Native Americans’ bargaining position deteriorated vis-à-vis the U.S. at specific times during the 19th Century.
One more trend in empirical legal research is also worth mentioning, and it involves a move to the micro-level. Increasingly, empirical researchers are interested in how law matters to the way people behave. The problem is that real world behavior is subject to scores of influences we social scientists cannot control. To ameliorate this problem, researchers increasingly are turning to surveys to find out how people think about international law.
As part of the Minnesota conference, we administered just such a survey, which was distributed as participants registered. It asked participants to give their subjective assessment about whether one hypothetical country was a relatively risky place to invest, and whether another hypothetical country should be considered a good trade partner. Not everyone received the exact same information, however; only half of respondents were given the italicized and bracketed information in the questions set forth below. The variation in survey results produced by this small variation in information proved astounding:
Question #1: We first asked participants to assess the information provided and answer the following question: How risky would you say it is for a manufacturing company in 2010 to invest in a country with the following characteristics?
- Has a history of capital controls, but has not interfered with the repatriation of profits within the past year,
- Has not had a fair, competitive national election in the past decade,
- [Has a bilateral investment treaty with an arbitration clause with your home country],
- 90% of urban areas have access to electricity 23.4 hours per day over the past year,
- Scores well (above the global average) on the World Bank’s “rule of law” scale,
- Has had moderate growth (4-7%) over the past decade.
Responses (participants were asked to please check one):
_____Risk is likely to be very low
_____Risk is likely to be moderately low
_____Risk is likely to be moderately high
_____Risk is likely to be very high
Results. Even at a conference of international economic lawyers and legal scholars, the existence or non-existence of a bilateral investment treaty (“BIT”) was seen as very significant to the riskiness of investment. Comparing the results of the surveys, we observed the following:
Just by exposing respondents to the hypothetical “fact” that a country had a BIT with the participant’s home country, the proportion of responses that categorized the risk as moderately to very high dropped from 31% to 21%. No one exposed to the BIT information thought the country posed a “very high” risk for investors. The sample size is admittedly small, but it is interesting that we could see a ten-percentage-point shift from a high-risk to a low-risk category with political, policy, economic and other legal conditions held constant, simply by changing the information about the existence of a BIT. Thus, a not insignificant number of conference attendees believed that the existence of such a commitment may matter to the relative risk of an investment.
Question #2: We also asked each conference participant to answer the following question, this time in the area of international trade:
- How desirable would you say a country with the following characteristics would be as a potential partner in a new bilateral preferential trade agreement? The country. . .
- Is a moderately-sized country classified as “upper-middle income” by the World Bank,
- Is known currently to protect agriculture through moderately high tariffs and modest export subsidies,
- Is a member of the WTO,
- [Last year failed to comply with a decision of an appellate panel regarding health and safety measures],
- Is of no particular strategic or political importance to you or your country,
- Has had moderate growth (4-7%) over the past decade.
Responses (respondents were asked to please check one):
_____Desirability as a candidate for a new preferential trade agreement is likely to be very low
_____Desirability as a candidate for a new preferential trade agreement is likely to be moderately low
_____Desirability as a candidate for a new preferential trade agreement is likely to be moderately high
_____Desirability as a candidate for a new preferential trade agreement is likely to be very high.
Results. Once again, there was an experimental treatment embedded in the question: only half of participants were exposed to the information that the country had failed to comply with an appellate decision of the WTO in the past year. Evidently this is unforgivable, at least in the short term, to international economic law experts. This information alone had a profound impact on participants’ responses, as we can see by comparing the answers of those who received the non-compliance information with the responses of those who did not:
Maybe this is the result of experimenting on a conference of international lawyers, but this time the results are extreme. Sixty-three percent of participants thought the country would make a very or somewhat desirable partner for a preferential trade agreement based on economic and political conditions, when unaware of its recent non-compliance with an appellate decision of the WTO. Evidently, this information is crucial, because when it is revealed, a meager 17% of participants thought this country would make a “somewhat desirable” trading partner, and none thought that this country should be considered “very desirable.” The sample group—attendants at an ASIL IEL conference—quite clearly view non-compliance with the WTO appellate body as a very serious matter indeed. If this is how this group reacts to non-compliance, what might this mean for decision-makers? While we might expect the survey participants to weigh law violations more heavily than the average policymaker does, the results of our experiment do suggest that a reputation for non-compliance could have serious consequences for new agreements down the road. In other words, we have some evidence of why, despite its inability to enforce its decisions, the dispute settlement mechanism of the WTO has some bite.
Young high-tech researchers are salivating at the prospects of converting document series into databases or conducting surveys of elite legal actors. We write neither to condemn nor condone this work—though we agree that these methods have interesting possibilities and great risks. The risk is that, in our excitement to “know more” and to “see further” using fancy new techniques, we will run roughshod over genuine expertise. Any associated downside risks will be minimized and upside payoffs maximized only to the extent that social scientists listen carefully to legal scholars to hone their programs and interpret their results. Once legal scholars get over being appalled, there will be critical ways in which their insights will help empirical researchers who perform legal or textual analysis avoid the most egregious errors and make the most of their work.
III. The Importance of Empirics in International Economic Law
Forecasting, electronic textual analysis, experiments embedded in surveys—these new, improving research methods might seem useless to legal researchers. But somehow we need to get a systematic understanding of the way our world operates. How can we assess and reassess our legal theories, doctrines and make policy prescriptions without more than a perfunctory look at the changing world we are trying to address? Moreover, how can we possibly talk about policy without some idea of the conditions under which legal innovations have “worked” in the past? David Trubek’s words are apropos in this context: “law cannot be defined [or evaluated] other than by the difference it makes in society, and empirical inquiry is necessary to determine what that is.” It is preposterous to imagine we can understand how law works and how to design policy without empirical legal studies. We can use assumptions to build models and theories, but we need facts to build the world we want.
Make no mistake—we are all empiricists. Each of us carries a model in his or her mind about the way the world works: lifting trade barriers increases trade, increased trade increases global welfare, level playing fields generate “fair” results. Realists, in turn, would believe that states renege on their legal agreements opportunistically and that hegemonic law leads to unfair distributive consequences, bailouts, court adverse selection, and moral hazard. What scholars must ask is whether these models are “right.” Have the right lessons been learned? A very interesting book summarizing the findings of decades of psychology research, The Science of Fear, would suggest we do not: people consistently overestimate the likelihood of sensational outcomes, partly because our brains are wired to beware of unlikely but deadly risks, but also because the media feeds the market for disastrous news. As a result, we grossly overestimate the likelihood of falling victim to catastrophic events or developing breast cancer in one’s 40s. The “facts” many of us carry in our heads just do not reflect the facts on the ground. Thus, we cannot simply rely on our own understandings of the world—systematic empirical knowledge must be substituted for biased worldviews.
This sort of cognitive bias creates acute challenges (and, indeed, opportunities) for researchers and scholars. One scholar has encountered some of these challenges in writing her recent book about the positive impact that the ratification of multilateral human rights agreements has had on domestic politics, litigation, and demands on human rights outcomes. In short, her empirical research shows that, in some cases, governments that have ratified the Convention Against Torture (CAT) or the International Covenant on Civil and Political Rights (ICCPR) actually do reduce torture, allow religious freedom, and provide fair trials more than comparable countries that have not ratified such treaties. Why are complex quantitative and qualitative methodologies needed to demonstrate this convincingly? Because it is somewhat hard to believe. These findings cut against the biased information we are fed each day: News headlines scream “torture of Iraqi detainees by the Iraqi authorities,” using electrocution, electric drills, and even execution, or reveal images of abuse by Indian officers of youth in Kashmir. We never see headlines about CAT ratifiers that proclaim: “Niger Eschews Torture” or “Way to Go Uruguay!” so we conclude that it is naive to think that any of the treaties lawyers have carefully crafted over the last several decades could possibly be effective; we must be wasting our time.
Biases pervade international investment law as well. Pick up (or, more likely, navigate to) a mainstream financial news source, and we read about the progress of the growing network of bilateral investment treaties in protecting foreign investments—to everyone’s advantage. Recently, the Wall Street Journal glowed with enthusiasm for a U.S.-India BIT. High on Obama’s to-do list in India, declared one op-ed, should be “seek[ing] a broad expansion of bilateral trade and investment, beginning with a long-delayed Bilateral Investment Treaty.” Another article put it thus: “Now the question is how quickly commercial cooperation can move forward, and whether traditional barriers to cross-border investment can be removed.” The Economist Intelligence Unit, reporting on Ecuador’s recent decision to “tear up” a number of its bilateral investment treaties as violative of its national Constitution, stated that “[t]he government’s decision to pull out of investment treaties comes at a time when some officials have been seeking to improve perceptions of Ecuador’s business climate and will deter foreign investment.”
Systematic empirical research provides a quite different picture. Overall, research suggests that the ability of BITs to attract foreign direct investment is minimal. The studies that find some positive effects to BIT ratification caution that BITs are beneficial in countries that already have strong property protection regimes in place. In explaining the rush to negotiate and ratify BITs in the 1980s and 1990s, Jose Alvarez wrote a fascinating account based on his State Department experience during the height of the BITs-signing frenzy. Developing countries often entered into BITs without much of an understanding about the legal consequences. This is consistent with the empirical work that was conducted with Andrew Guzman and Zachary Elkins on the pattern of BIT signings over time. Using a statistical model to analyze the probability that any two states would conclude a BIT agreement, the study found patterns that suggested the BIT cascade of the 1980s and 1990s was a competitive scramble by developing countries to impress on creditor nations that the latter’s investment will be secure. As evidence, developing countries tended to ratify BITs when those countries with the most similar product profile, the most similar infrastructure, and the most similar workforces (i.e., their closest competitor nations) did so. To this, one might add a degree of economic desperation; Professor Simmons is doing some work that reveals that developing countries were much more likely to ratify BITs during periods of economic downturn. Evidence that BITs actually attract significant foreign direct investment is weak at best. There is little doubt about what they systematically do attract: international arbitration. The more BITs a country signs, the more likely they are to show up on the list of cases on the International Center for Settlement of Investment Disputes website. It is no wonder, then, that developing as well as developed countries are having second thoughts about BITs, insisting on renegotiation, declaring moratoria on new agreements, and, in some cases, even terminating existing agreements. We have also seen a recent spike in desperate attempts at arbitration award annulments. The empirical work that tries to “take the pulse” of the legal regime for protection of FDI suggests that the patient is not all that well.
And yet the fanfare continues. Even if researchers might know that, as an empirical matter, BITs rarely meet their stated goals, practitioners—politicians and businesses, in particular—still believe otherwise. Associations of American businesses regularly call for the U.S. to negotiate BITs, particularly with Brazil, Russia, India, and China (where the bulk of foreign direct investment already goes). Also, when BIT talks with China were announced in June of 2008, for example, “American business interests reacted positively” to the news. Enthusiasm is not limited to the developed world—developing countries continue to negotiate BITs amongst themselves and with the North.
In this time of crisis and change, researchers are inundated with sensationalized and unsophisticated accounts of the world. It is, therefore, particularly important that as we try to sort out what has happened and how to move forward, we do so with the strongest empirical basis we can. Reactionary policies grounded in sensational views of the recent crisis are the likely result of a failure to merge empirical research with legal reform.
IV. Linking Empirics with Law and Policy
This takes us to the heart of the matter—in what ways does empirical research help international economic lawyers and legal scholars in their capacities as policy makers? Two avenues, distinct but intimately related, immediately come to mind.
First and foremost, empirical studies give researchers the ability to systematically evaluate legal institutions in light of their goals. As discussed briefly above, researchers proceeding empirically have confirmed that the energies and resources expended on negotiating and ratifying multilateral treaties about trade and human rights, for example, are not for naught—these legal texts do promote many of their stated goals. One of the first systematic treatments by Andrew Rose, a professor in the business school at UC Berkeley, came to what for many was a surprising and highly counterintuitive conclusion: he found little evidence that countries joining or belonging to the GATT/WTO have different trade patterns from outsiders, though the Generalized System of Preferences seems to have a strong effect. Empirical work of this kind is wonderfully provocative, and scholars got to work to determine whether Rose’s findings were robust. Empirical political scientists Judith Goldstein, Michael Tomz, and Doug Rivers looked at the way “members” were coded in Rose’s study and found that there were actually lots of countries who were not full members (and therefore not coded as such in Rose’s dataset), but had formally been extended all the benefits of membership. After reflecting in the data that these states do in fact “participate” in the regime, Goldstein and her collaborators found a strong and positive effect to the liberalizing rules of the GATT/WTO. In another study, however, examining the distributive consequences of international economic law, Subramanian and Wei found that the WTO promotes trade strongly but unevenly. If we were sure it was otherwise, it would be high time to determine where else to focus our attention. This might be the case in the BIT context, where there are certainly questions about whether the BIT regime has achieved its goals.
As an obvious corollary, empirical research might give us reason to support the revision of our legal rules. Consider Article 28(2) of the GATT, which allows developing countries to implement infant-industry protections. Empirical data has somewhat consistently shown that infant-industry protections do not work to spur long-term growth or development. If the goals of the WTO are to promote trade liberalization, global welfare and development, the disconnect between empirics and reality might appear disconcerting.
Second, empirical studies provide facts on which to base legal doctrine and public policy. The most famous example of this in American law is probably footnote 11 in Brown v. Board of Education, where the U.S. Supreme Court cited to social psychology studies to support its conclusion that “separate but equal” segregated educational facilities are inherently unequal. Certainly doing so was not necessary to the Court’s analysis—it could have relied on the moral and doctrinal judgment that discrimination per se is injurious—but the persuasive power of such research is at times undeniable.
In the area of international trade, it is becoming clear that legal structures and processes are having important impacts on outcomes—and that we should remember these lessons when considering how these agreements might be modified in the future. An important body of empirical work by a number of international trade scholars—beginning in earnest with Robert Hudec—determined that developing countries, particularly Least Developed Countries, initiated GATT/WTO disputes significantly less frequently than they “should,” and even then garnered only mixed results in trade litigation. Especially troubling was Marc Busch’s revelation that LDCs were one-third less likely to file complaints against developed states under the WTO than they were under the post-1989 GATT regime. This research also pointed to reasons for this trend: the WTO’s increased legal and procedural complexity resulted in higher costs and required greater legal capacity. Most striking are not these conclusions, which developing countries and their advocates already knew well, but their results—these findings gave credence to calls from developing countries for the establishment of an Advisory Centre on WTO Law (the Advisory Centre), which was finally formed in 2001. The Advisory Centre’s “mission is to provide developing countries and LDCs with the legal capacity necessary to enable them to take full advantage of the benefits and opportunities offered by the WTO.” After the better part of a decade, the Advisory Centre has assisted developing countries initiate some 19 complaints—the same number of complaints that the U.S. itself initiated in the same time period.
Consider as well the rise and fall of Strategic Trade Policy (STP). In the 1980s, a group of economists—Nobel Prize winner Paul Krugman among them—developed a powerful theoretical critique of comparative advantage in high-technology industries characterized by increasing returns to scale. Always provocative, Krugman at first questioned whether free trade had become passé, but when he saw his theories being co-opted by protectionist policy-makers he was relentless in demanding thorough empirical research before making this theory operational. In the U.S. at least, the push for STP has diminished, and seemingly for the best: aside from the complex political economy problems posed by STP, empirical research on each major attempt at STP has shown that the policy imposed net costs each time. The STP saga demonstrates the limits of theoretical models (and reactionary policy-making) and the importance of empirical study to effectively translate theory into policy and practice.
Recent BITs research is also instructive. Notwithstanding the problems with BITs, researchers Tobin and Busch recently analyzed annual data on pairs of developing and developed countries between 1960 and 2004, concluding that BITs raise the prospects of getting a North-South preferential trade agreement, at least up to a point. Thus, we might encourage negotiation of BITs as an indirect way to increase trade and investment flows. Viewed in this light, the recent commentary on the occasion of Obama’s trip to India may not seem so flawed.
Empiricism, of course, is just one tool in the woodshed. Legal institutions are political compromises between competing goals, not real-world manifestations of state-of-the-art statistical analysis. Although empirical methodologies are important in discovering how our legal institutions operate and how they impact our world, we cannot begin to truly understand them simply by running regressions. In translating empirics into policy, the role of the lawyer, the political economist, and the politician are paramount, given their intimate knowledge of the institutional and normative constraints that operate on the ground. Thus, even if some empirical studies demonstrate the weakness of the theories underlying particular international economic laws, we might not actually expend resources on their revision for a number of reasons—in the infant-industry context, for example, because of what Article 28(b) represents to developing countries: the recognition of their historical experience and the importance of policy space as a general principle. Thus, when we are evaluating our world and the legal institutions that create it, empirical studies are undeniably important, but they can never tell the whole story.
The ASIL IELG brought scholars together to understand and influence a time of change in international economic law. This is no easy task, and will require in-depth study by researchers from a broad range of disciplines. It is our firm belief that an important—indeed, an indispensable—tool of international economic law going forward will be systematic empirical research. Given its unique nexus with economics, political economy, international relations, and domestic and international law, the empirical turn in international economic law is no coincidence, and should be embraced. We do not mean to argue that it is essential for legal scholars to run out and get a Ph.D. in statistics or economics or social science. In most cases that would be an utter waste of a lawyer’s time and training. One does not need a Ph.D. in social sciences to make appropriate use of empirical research—one only needs a friend or colleague with empirical training. Interdisciplinary collaboration unlocks synergies; this is the university’s version of comparative advantage.
Notwithstanding increases in the amount of empirical international economic law research and advances in the quality of empirical methodologies, however, controversy remains as to whether the empirical trend is a good thing for the study of international economic law. On the one hand, there are those who push back on empiricism’s own terms. What does this empirical data actually tell us? Why is it important? Are the conclusions robust, and why do social scientists turn up so many inconsistent answers? Everyone should be asking these questions of empirical research—this is not a sore spot felt by the legal academy alone.
But we sense the anxieties among some legal scholars run deeper. Some are concerned that empirical research is not, and certainly should not be, what the legal academy is all about. Social science and legal scholarship are distinct enterprises with inherently different purposes, it is easy to believe, and each discipline is taught to put its intellectual firepower to different use. Science is always a hypothesis from which we can advance in the face of better evidence or more convincing theory. The key is absolute transparency in data methods—publicness and replicability are primary values. As Keynes famously wrote in his 1933 Essays in Biography, “There is no harm in being sometimes wrong—especially if one is promptly found out.” (Which, to our occasional embarrassment, we all can potentially be!) More important than the “stance” a researcher takes today is her commitment to the scientific process, to the treatment of data as public goods. Social scientists can critique, bicker, and collaborate, but we don’t have to settle anything by the end of the day. The normal science model is all about edging toward the truth in the long run, and generally doing so in a cooperative, as opposed to adversarial, way.
Legal scholars are trained to use their intellectual skills differently: to make the best case to win the point. At the end of the day, a decision must be taken in a case or a policy must be chosen and implemented. The need to decide pushes legal scholars into their respective positions. The pressure of a decision also makes legal scholars less tolerant of ambiguity and uncertainty, both of which are rife in the “scientific” enterprise. Social scientists draw conclusions with varying degrees of certainty, measured by confidence intervals, but advocates cannot be wafflers. Lee Epstein and Gary King aptly described the nature of the problem thus: “An attorney who treats a client like a hypothesis would be disbarred; a Ph.D. who advocates a hypothesis like a client would be ignored.”
This viewpoint overstates the gulf between social scientists and lawyers to some extent. As our opening tale about Holmes and Watson suggests, and as we have intended to convey, there undoubtedly are important synergies that can be achieved via partnerships between legal scholars and empirical researchers. Of particular benefit to lawyers, the factual groundwork to evaluate legal institutions and formulate policy prescriptions will be better laid. For empirical researchers, the benefits are immense as well: lawyers and legal scholars can focus us on questions that actually need answering, can help us understand why things are the way they are and what possibilities there are for the future, and are the conduits by which data and doctrine are translated into policy. At the most fundamental level, moreover, empirical researchers and lawyers are engaged in the same impossible task: a search for the truth. By working together, international economic lawyers and empirical researchers can focus on the most important variables and the most important questions—while using legal scholars’ doctrinal, philosophical, and public policy knowledge to confirm and explain empirical findings—thereby improving the process by which data is accumulated and distributed. This, in turn, will increase the value of empirical research to international economic lawyers. There is hope yet that we can see eye to eye, and in doing so, improve our vision.
We hope that our readers—lawyers and legal scholars in particular—will consider the extent to which their work depends upon empirical claims about our world. We find ourselves in a period of crisis and change—consider too whether we are relying on sensationalized worldviews in formulating our beliefs and policies. And consider, finally, the comparative advantages you possess in analyzing your world. The possibilities for collaboration are many, and we should all think about whether it would enrich our scholarship to work together to an even greater degree than we already have.
* This paper is adapted from the Keynote Address given by Professor Simmons on November 18, 2010 at the American Society of International Law International Economic Law Interest Group conference held at the University of Minnesota Law School, “International Economic Law in a Time of Change: Reassessing Legal Theory, Doctrine, Methodology and Policy Prescriptions.” In this paper, the authors have deliberately adopted a tone that lies somewhere between the strained formality of a law review article and the feigned nonchalance of an academic address.
** Clarence Dillon Professor of International Affairs; Director, Weatherhead Center for International Affairs, Government Department, Harvard University.
*** J.D. candidate, May 2011, Harvard Law School; M.Sc. International Political Economy, London School of Economics & Political Science, 2007.
. See Merriam-Webster’s Collegiate Dictionary 408 (11th ed. 2003).
. See generally Elizabeth Chambliss, When do Facts Persuade? Some Thoughts on the Market for “Empirical Legal Studies,” 71 Law & Contemp. Probs. 17 (2008); Tracey E. George, An Empirical Study of Empirical Legal Scholarship: The Top Law Schools, 81 Ind. L.J. 141 (2006); Theodore Eisenberg, Why do Empirical Legal Scholarship?, 41 San Diego L. Rev. 1741 (2004); David M. Trubek, Where the Action Is: Critical Legal Studies and Empiricism, 36 Stan. L. Rev. 575 (1984).
. See George, supra note 2.
. Using Westlaw’s Journals and Law Reviews (JLR) Database, the authors conducted the following searches and obtained the following results on January 27, 2011: (1) date(bef 1998) & “international economic law” & “statistic! signific!” or “empiric!” – 149 results; (2) date(1998) & “international economic law” & “statistic! signific!” or “empiric!” – 48 results; (3) date(aft 1998) & “international economic law” & “statistic! signific!” or “empiric!” – 830 results. (830 + 48) / 149 = 5.892.
. A search of (date(aft 2005) & “international economic law” & “statistic! signific!” or “empiric!”) returned 406 results. 1027 (the total number of results over all years) / 406 = 2.529.
. Information for Authors, J. Int’l Econ. L., http://www.oxfordjournals.org/our_journals/jielaw/for_authors/index.html (last visited Jan. 28, 2011).
. Michael J. Madison, The Idea of the Law Review: Scholarship, Prestige and Open Access, 10 Lewis & Clark L. Rev. 901, 909 (2006) (“([A]side from a few quirky journals) there is no peer review [of legal scholarship].”).
. About the Journal, J. Int’l Econ. L., http://www.oxfordjournals.org/our_journals/jielaw/about.html (last visited Jan. 28, 2011).
. This figure does not include the editorial advisory board members. See Editorial Board, J. Int’l Econ. L., http://www.oxfordjournals.org/our_journals/jielaw/editorial_board.html (last visited Jan. 28, 2011).
. This number was arrived at by canvassing the resumes of the editorial board members. Id. See also, http://www.ijiel.com/boed.htmlhttp://www.ijiel.com/boed.html
. This number was arrived at by canvassing the resumes of the conference participants, including panel members and other speakers. Symposium, International Economic Law in a Time of Change: Reassessing Legal Theory, Doctrine, Methodology and Policy Prescriptions Brochure (Nov. 18–20, 2010), http://asil.org/files/2010/asil_IEcLIG_brochure _111010.pdf.
. Again, the authors canvassed the resumes of the participants at that meeting, drawn from the schedule of proceedings, www.asil.org/pdfs/ielgconf0606.pdf.
. See, e.g., Juscelino F. Colares, A Theory of WTO Adjudication: From Empirical Analysis to Biased Rule Development, 42 Vand. J. Transnat’l L. 383 (2009) (conducting empirical assessment of complainant win-rates at the WTO to demonstrate that the interpretation of WTO agreements via dispute settlement has fostered a normative free trade vision, indicating biased rule development and some judicial lawmaking); Marc L. Busch & Krzysztof J. Pelc, Does the WTO Need a Permanent Body of Panelists?, 12 J. Int’l Econ. L. 579 (2009) (using statistical analysis to determine, inter alia, that that rather than constituting a permanent body of panelists, the WTO would be better served by establishing a pool of permanent chairs); Meredith Kolsky Lewis, The Lack of Dissent in WTO Dispute Settlement, 9 J. Int’l Econ. L. 895 (2006) (demonstrating with empirical data that WTO dissents can and do make a difference in WTO jurisprudence).
. See, e.g., Michael Trebilcock & Paul-Erik Veel, Property Rights and Development: The Contingent Case for Formalization, 30 U. Pa. J. Int’l L. 397 (2008) (surveying empirical literature to argue for a reconception of the link between property rights and development); Lee G. Branstetter, Do Stronger Patents Induce More Local Innovation?, 7 J. Int’l Econ. L. 359, 359 (2004) (demonstrating that “benefits of stronger IPRs—to the extent that they exist at all—are more likely to come instead from an acceleration in the domestic deployment of advanced technology by the affiliates of foreign firms.”).
. See Gary Hufbauer & Daniel Danxia Xie, Financial Stability and Monetary Policy: The Need for International Surveillance, 13 J. Int’l Econ. L. 939, 939 (2010) (concluding that “[e]mpirical evidence for the USA, other [OECD] countries, and a few emerging countries lends strong support for the connection between exceptionally fast growth of DFM [de facto money] and subsequent financial instability.”); Panagiotis Delimatsis & Pierre Sauvé, Financial Services Trade After the Crisis: Policy and Legal Conjectures, 13 J. Int’l Econ. L. 837, 843–47 (2010) (suggesting empirical means to gauge the fallout from the financial crisis); Jason Webb Yackee, Bilateral Investment Treaties, Credible Commitment, and the Rule of (International) Law: Do BITs Promote Foreign Direct Investment?, 42 Law & Soc’y Rev. 805 (2008) (finding that stronger BITs that provide access to arbitration are not associated with increased investment).
. See, e.g., Susan D. Franck, Empirically Evaluating Claims About Investment Treaty Arbitration, 86 N.C. L. Rev. 1 (2007).
. See International Economic Law: The State and Future of the Discipline (Colin B. Picker, Isabella D. Bunn & Douglas W. Arner eds., 2008) (containing essays from a prior conference convened by the ASIL International Economic Law Interest Group discussing the role of empirical research in international economic law).
. See Treading Lightly: Does Mankind Need More than One Planet?, Economist, Sept. 19, 2002, available at http://www.economist.com/node/1337251.
. See generally Paul R. Krugman, Pop Internationalism (1994) (discussing the popular concern over U.S. competitiveness and, inter alia, comparing the myth of Asian economic growth with the reality of Japan’s more modest recent growth rate).
. John J. Mearsheimer, Back to the Future: Instability in Europe After the Cold War, 15 Int’l Security 5 (1990), available at http://smp.fsv.cuni.cz/Mearsheimer.pdf.
. See NATO Member Countries, http://www.nato.int/cps/en/natolive/nato_countries.htm (last visited Feb. 2, 2011).
. See “The Author Replies” in Stanley Hoffmann, Robert O. Keohane & John J. Mearsheimer, Correspondence: Back to the Future, Part II: International Relations Theory and Post-Cold War Europe, 15 Int’l Security 191 (1990), available at http://mearsheimer.uchicago.edu/pdfs/A0015.pdf.
. See Member States of the EU, http://europa.eu/abc/european_countries/eu_members/index_en.htm (last visited Feb. 2, 2011).
. Francis Fukuyama, The End of History and the Last Man (Free Press 2006) (1993).
. A. J. Veal, The Elusive Leisure Society, 17–18 (University of Technology Sydney, School of Leisure Sport and Tourism Working Paper No. 9, 2009) available at http://www.leisuresource.net/service3.aspx (follow link) (quoting Keynes’ 1931 essay Economic Possibilities for Our Grandchildren).
. See Supreme Court Forecasting Project, Wash. U. Sch. L., http://wusct.wustl.edu/ (last visited Feb. 2, 2011).
. Andrew D. Martin, Kevin M. Quinn, Theodore W. Ruger & Pauline T. Kim, Competing Approaches to Predicting Supreme Court Decision Making, 2 Persp. on Pol. 761, 761 (2004), available at http://wusct.wustl.edu/media/man1.pdf.
. The statistical model took into account: “(1) the circuit of origin for the case; (2) the issue area of the case, coded from the petitioner’s brief using Spaeth’s protocol; (3) the type of petitioner (e.g., the United States, an injured person, an employer); (4) the type of respondent; (5) the ideological direction of the lower court ruling, also coded from the petitioner’s brief using Spaeth’s protocol; and (6) whether or not the petitioner argued the constitutionality of a law or practice” (internal footnotes removed). Id. at 762.
. Id. at 763.
. Martin et al., supra note 27, at 765 Figure. 3.
. See Arthur Spriling, Bargaining Power in Practice: US Treaty-Making with American Indians, 1784–1911, Working Paper, Harvard University, 2010, available at http://www.people.fas.harvard.edu/~spirling/documents/SpirlingPolmeth2010.pdf.
. Id. at 30.
. Id. at 30–31.
. The work of Michael Tomz, who conducts empirical research in this way and also designs software to allow other researchers to conduct such research, is particularly important in this regard. See, e.g., Michael Tomz, The Foundation of Domestic Audience Costs: Attitudes, Expectations, and Institutions (Apr., 2009), in Kitai, Seido, Gurobaru-shakai (Expectations, Institutions, and Global Society) (Masaru Kohno and Aiji Tanaka eds., 2009), available at http://www.stanford.edu/~tomz/pubs/Tomz-AudCost-Foundations-2009-04-14b.pdf. For statistical software information, see, for example, http://politicalscience.stanford.edu/faculty/tomz.html.http://www.stanford.edu/~tomz/pubs/Tomz-AudCost-Foundations-2009-04-14b.pdfhttp://www.stanford.edu/~tomz/pubs/Tomz-AudCost-Foundations-2009-04-14b.pdf.
. Some of Professor Simmons’s graduate students, for example, presently are using Arabic language programs to analyze several decades of Egyptian clerics’ fatwas, as well as Japanese language programs to analyze thousands of Japanese electoral platforms.
. David M. Trubek, Where the Action Is: Critical Legal Studies and Empiricism, 36 Stan. L. Rev. 575, 581 (1984).
. See generally Daniel Gardner, The Science of Fear (2008).
. Id. at 57.
. Id. at 157–59.
. See, e.g., Beth A. Simmons, Mobilizing for Human Rights 273 (2009).
. See id.
. See Huge Wikileaks Release Shows US ‘Ignored Iraq Torture’, BBC News (23 October 2010, 05:42AM), http://www.bbc.co.uk/news/world-middle-east-11611319; see also Wikileaks, www.wikileaks.org (last visited Jan.26, 2011).
. See Indian Authorities Must Investigate Online Video of Kashmir Detainee Abuse, Amnesty Int’l (September 12, 2010), http://www.amnesty.org/en/news-and-updates/indian-authorities-must-investigate-online-video-kashmir-detainee-abuse-2010-09-13.
. Richard L. Armitage & R. Nicholas Burns, A To-Do List for Obama in India, WSJ.com (November 4, 2010), http://online.wsj.com/article/SB10001424052702304155604575581721120234484.html.
. Amol Sharma, Five Ways U.S.-India Trade can Improve, The Wall St. J. (November 2, 2010), http://blogs.wsj.com/indiarealtime/2010/11/02/five-ways-us-india-trade-can-improve/.
. Ecuador Economy: Bilateral Investment Protection Treaties Torn Up, Economist Intelligence Unit (Sep. 27, 2010), http://www.eiu.com/index.asp?layout=VWArticleVW3&article_id=77456592®ion_id=&country_id=1790000179&channel_id=190004019&category_id=500004050&refm=vwCat&page_title=Article&rf=0.
. See José E. Alvarez, The Return of the State, 20 Minn. J. Int’l L. (forthcoming 2011).
. See id.
. See Zachary Elkins, Andrew Guzman & Beth Simmons, Competing for Capital: The Diffusion of Bilateral Investment Treaties, 1960–2000, in The Backlash Against Investment Arbitration: Perceptions and Reality 369 (Michael Waibel, Asha Kaushal, Kyo-Hwa Liz Chung & Claire Balchin eds., 2010).
. See, e.g., Jason W. Yackee, Do Bilateral Investment treaties Promote Foreign Direct Investment? Some Hints from Alternative Evidence, 51 Va. J. Int’l L. 397, 434 (2011). (conducting empirical analysis to conclude that “[w]hile BITs are routinely described as important tools for attracting FDI, and while certain empirical studies claim to have isolated huge causal impacts, my own examination suggests that, at best, BITs spur investment only irregularly, inconsistently, and with generally unassuming impact”); Emma Aisbett, Bilateral Investment Treaties and Foreign Direct Investment: Correlation Versus Causation, in The Effects of Treaties on Foreign Direct Investment, 395 (Karl P. Sauvant & Lisa E. Sachs eds., 2009) (identifying a number of serious methodological challenges largely ignored in BIT studies, including problems of endogeneity, autocorrelation, and omitted variables; finding that once these problems were addressed using appropriate statistical methods, significant correlations between BIT ratification and FDI inflows disappeared); U.N. Conf. on Trade & Dev., Bilateral Investment Treaties in the Mid-1990s, U.N. Doc. UNCTAD/ITE/IIT/7, 105 (1998). (concluding that BITs could be expected to “marginally increase” foreign direct investment, but such an effect is usually small).
. See Christina Knahr, Annulment and Its Role in the Context of Conflict Awards, in The Backlash Against Investment Arbitration: Perceptions and Reality 151, 151 (Michael Waibel, Asha Kaushal, Kyo-Hwa Liz Chung & Claire Balchin eds., 2010).
. See Bush Urged to Pursue BRIC Pacts, Globe & Mail (Dec. 5, 2007), http://www.theglobeandmail.com/report‑on‑business/bush‑urged‑to‑pursue‑bric-pacts/article139409/.
. See Steven R. Weisman, U.S. and China Agree to Ease Foreign Investment, NY Times (Jun. 06, 2008), http://www.nytimes.com/2008/06/19/business/worldbusiness/19trade.html?_r=1&scp=1&sq=US%20and%20China%20agree%20to%20ease%20foreign%20investment%202008&st=cse.
. See Simmons, supra note 44.
. Andrew Rose, Do We Really Know that the WTO Increases Trade?, 94 Am. Econ. Rev. 98, 98 (Mar. 2004).
. See Judith Goldstein, Douglas Rivers & Michael Tomz, Do We Really Know That the WTO Increases Trade? Comment, 97 Am. Econ. Rev. 2005, 2005 (Dec. 2007).
. See Arvind Subramanian & Shang-Jin Wei, The WTO Promotes Trade, Strongly but Unevenly, 72 J. Int’l Econ. 151 (2007).
. See Alvarez, supra note 51.
. See General Agreement on Tariffs and Trade, Oct. 30, 1947, 61 Stat. A‑11, 55 U.N.T.S. 194, available at http://www.wto.org/english/docs_e/legal_e/gatt47_02_e.htm (stating that parties implement protective or other measures such as flexibility in their tariff structure, grant tariff protection and quantitative restrictions for balance of payments purposes).
. See, e.g., Howard Pack, Industrial Policy: Growth Elixir or Poison?, 15 World Bank Res. Observer 47, 48, 60 (2000) (surveying past empirical evidence disparaging import-substitution and conducting survey to determine in part that industrial policy in Japan and Korean at most “may have been a minor hormone.”); Steven Radelet & Jeffrey Sachs, Asia’s Reemergence, Foreign Affairs, Nov./Dec. 1997, at 51 (Some modest infant-industry successes have been noted in large countries, though balanced by high costs, and its record elsewhere “is one of almost unremitting failure”); Anne O. Krueger & Baran Tuncer, An Empirical Test of the Infant Industry Argument, 72 Am. Econ. Rev. 1142, 1142 (1982).
. See Brown v. Bd. of Educ., 347 U.S. 483, 494 (1954); see also Edmond Cahn, Jurisprudence, 1955 Ann. Surv. Am. L. 655, 655–58 (1955) (discussing the meaning and dangers of footnote 11).
. See Marc L. Busch & Eric Reinhardt, Developing Countries and General Agreement on Tariffs and Trade/World Trade Organization Dispute Settlement, 37 J. World Trade 719, passim (2003) (demonstrating empirically that the WTO “only serves to reinforce th[e] tendency [of developing countries to exact few concessions from defendants in trade litigation], given both the incentives to litigate as well as developing countries’ lack of capacity to push for early settlement.”); Robert E. Hudec et al., A Statistical Profile of GATT Dispute Settlement Cases: 1948–1989, 2 Minn. J. Global Trade 1, 6 (1993); see also Andrew T. Guzman & Beth A. Simmons, Power Plays and Capacity Constraints: The Selection of Defendants in World Trade Organization Disputes, 34 J. Legal Stud. 557 (2005) (conducting an empirical analysis of disputing behavior in the GATT and concluding that poorer countries chose their targets strategically so as to conserve their legal and other resources, thus choosing only the biggest targets, and only initiating disputes rarely); see generally Marc L. Busch & Eric Reinhardt, Testing International Trade Law: Empirical Studies of GATT/WTO Dispute Settlement, in The Political Economy of International Trade Law: Essays in Honor of Robert E. Hudec 457, 466–68 (Daniel L. M. Kennedy & James D. Southwick eds., 2002) (explaining why LDCs are not as active in the WTO dispute settlement system).
. Eric Reinhardt, Aggressive Multilateralism: The Determinants of GATT/WTO Dispute Initiation, 1948–1998, Emory University, 15–21 (Feb. 16, 2000), userwww.service.emory.edu/~erein/research/initiation.pdf.
. See Marc L. Busch & Eric Reinhardt, Testing International Trade Law: Empirical Studies of GATT/WTO Dispute Settlement, in The Political Economy of International Trade Law: Essays in Honor of Robert E. Hudec 457, 467 (Daniel L. M. Kennedy & James D. Southwick eds., 2002) (noting the various reasons put forward by empirical researchers to explain under-participation by developing countries under-participation in WTO dispute settlements).
. See Kim Van der Borght, The Advisory Center on WTO Law: Advancing Fairness and Equality, 2 J. Int’l Econ. L. 723, 724–26 (1999) (describing various developing country proposals for a capacity-building institution and advocating its creation).
. See The Agreement Establishing the Advisory Centre on WTO Law, Dec. 1, 1999, available at http://www.acwl.ch/e/documents/agreement_establishing.pdf (noting in its opening preamble that “developing countries, in particular the least developed among them, and the countries with economies in transition have limited expertise in WTO law and the management of complex trade disputes and their ability to acquire such expertise is subject to severe financial and institutional constraints . . . .”).
. The ACWL’s Mission, Advisory Ctr. on WTO Law, http://www.acwl.ch/e/about/about_us.html (last visited February 1, 2011).
. Advisory Ctr. on WTO Law, Report on Operations: 2009, 8 (2009) (emphasis added), available at http://www.acwl.ch/e/about/reports.html (“[B]etween 2001 and 2008, the [Advisory Centre] (nineteen times) has worked legally on behalf of the complainant members in more disputes than any WTO member acted as a complainant in its own disputes except for the United States (also nineteen times) and the EC (twenty-one times). Put differently, if the ACWL were not an intergovernmental organization but instead were itself a WTO member country, it would be considered as the third most frequently active complainant litigant.”).
. Paul R. Krugman, Is Free Trade Passé?, 1 J. Econ. Persp. 131, 131 (1987).
. See generally Krugman, supra note 17.
. See id. at 113; Paul R. Krugman, Making Sense of the Competitiveness Debate, 12 Oxford Rev. Econ. Pol’y 17, 23–24 (1996) (contrasting the “punditry on the semiconductor industry” with the realities of intervention in that sector); See generally Wilfred J. Ethier, Modern International Economics 274–767 (3rd ed. 1995) (describing European and Japanese semiconductor policies); Paul Krugman & Alasdair Smith, Introduction, in Empirical Studies of Strategic Trade Policy 1, 7 (1994) (“It is also true that the research generally provides little support for a drastic rethinking of trade policy. Nobody has yet provided empirical evidence that would suggest large gains from protection or export subsidy.”).
. Jennifer L. Tobin & Marc L. Busch, A BIT is Better Than a Lot: Bilateral Investment Treaties and Preferential Trade Agreements, 62 World Pol. 1, 31 (2010).
. See supra notes 41–42 and accompanying text.
. John Ruggie might describe this as a manifestation of “embedded liberalism.” See John Gerard Ruggie, International Regimes, Transactions, and Change: Embedded Liberalism in the Postwar Economic Order, 36 Int’l Org 379, 379–80 (1982).
. John Maynard Keyes, Essays in Biography 317 (W.W. Norton & Co., Inc. 1933).
. Lee Epstein & Gary King, The Rules of Inference, 69 U. Chi. L. Rev. 1, 7 (2002).