Exporting American Discovery

“A curious quirk of our law is that American courts are not limited to American disputes.”

–Then–Magistrate Judge Paul Singh Grewal, Northern District of California1

“[T]he obvious question is well, why, don’t they have access to that discovery vehicle in Argentina, why do they need access to this little Reno office[?]”

–Then–Chief Judge Robert C. Jones, District of Nevada2

Introduction

Across the country, federal courts now routinely have a hand in the resolution of foreign civil disputes. They do so by compelling discovery in the United States—typically as much discovery as would be available for a lawsuit adjudicated in federal district court—and making it available for use in foreign civil proceedings3 governed by different procedural rules. In the past decade and a half, federal courts have received and granted thousands of such discovery requests.4 They come from foreign courts and foreign parties.5 They seek discovery for cases ranging from billion-dollar environmental controversies6 to the dissolution of marriages,7 and in countries as varied as Chile,8 Romania,9 Iran,10 and South Korea.11 Since most of these requests are decided in low-profile, unpublished orders buried in federal dockets around the country, they have received little systematic attention from scholars despite their transformative impact on the practice of global litigation.12 Nearly every major law firm and numerous smaller ones now advise clients and strategize around the availability of compelled discovery in the United States for use abroad.13 Practitioners consider this feature of US law “an invaluable tool,”14 “a powerful strategic advantage,”15 and “a back door” for foreign litigants16 that parties ignore “at their peril.”17

I call this growing global role played by US courts and judges the “export” of American discovery.18 In today’s globalized world, disputes increasingly cannot be confined to one legal system alone. The fact that evidence relevant to a foreign dispute might be located in and exported from the United States is a symptom and symbol of this modern reality. The export of American discovery provides an illustrative case study of the institutional challenges that arise when disputes straddle contrasting legal systems. For what is being exported is not just information—typically in the form of witness testimony or the production of documents—that may be submitted as evidence before a foreign tribunal. Along with that information comes the compulsory power of US courts and a set of procedures and litigation values found virtually nowhere else in the world.

American civil procedure is well recognized as being exceptional.19 Discovery in US federal courts is “far broader” in scope than in other countries20 and is primarily conducted and controlled by the parties, rather than by judges.21 Expansive discovery is central to American litigation, and is intertwined with the very mission of government in American society—one that is more “reactive” and plays a smaller ex post role resolving disputes, rather than a larger ex ante role implementing state programs.22 Outside the United States—where litigation may perform a different function in regulating society—American discovery is regarded as excessive and has been approached with skepticism and animosity.23

Operating across different discovery systems offers private actors opportunities for arbitrage. Seeking US discovery is typically straightforward. One files a request with the federal district court where the discovery target is located. That request is usually entertained and granted with minimal judicial activity and on an ex parte basis—without the participation of the foreign opposing party or the foreign tribunal before which the discovery is to be used. The target of the request is then subpoenaed and ordered to produce discovery according to the scope and practices of the Federal Rules of Civil Procedure (FRCP). The target might happen to be the foreign opposing party or might choose to voluntarily alert the foreign opposing party, who might in turn inform the foreign tribunal, but that does not always happen. The foreign opposing party and the foreign tribunal might never know that one side of the case was built using different discovery practices than those governing the remainder of the dispute.

These requests generate complications for both the foreign opposing party and the foreign tribunal. Take the example of a discovery request filed in the Northern District of Alabama by a private actor seeking discovery from a US bank for a contemplated lawsuit in the British Virgin Islands.24 Such prefiling discovery requests are permitted under 28 USC § 178225 —an expansive federal statute written in the 1960s that is now the pivotal law governing the export of discovery. The request was initially granted ex parte and a subpoena was served on the bank, which was not a party to the contemplated suit. Months after the request was granted and the subpoena served, the anticipated opposing party moved to intervene, arguing denial of due process because the company was never informed of the request at the time the court considered it. In fact, similar discovery requests had been made in three additional US district courts without alerting the opposing party.26 The Northern District of Alabama nevertheless declined to vacate the discovery order.27 This fact pattern is common, and there is no effective procedural mechanism for ensuring symmetrical discovery when one side benefits from broad US discovery while the other is limited to the more restrictive procedures of the foreign tribunal.

Consider also the example of the European Commission filing an amicus brief before the US Supreme Court in 2003 to prevent a district court in the Northern District of California from compelling the production of evidence, ostensibly in its aid.28 At the time, the European Commission was investigating an antitrust complaint brought by Advanced Micro Devices (AMD) against its worldwide competitor, Intel.29 When the Commission declined AMD’s suggestion that it seek certain documents from Intel, AMD asked the Northern District of California, the jurisdiction where Intel is headquartered, to subpoena Intel for that same information.30 The request eventually reached the Supreme Court, where the European Commission argued that granting AMD’s request in the United States would be a “direct interference” with the Commission’s own “orderly process” and would undermine its policies, increase its workload, and divert its enforcement resources.31 The Commission protested: “[we do not] want to be used as a pawn by . . . private entities seeking to employ [American] processes . . . to obtain . . . discovery that’s available under no other circumstances.”32

These two cases illustrate the consequences that American discovery can have for foreign courts and foreign parties—consequences that were not intended by Congress or by the Supreme Court. When Congress first enacted § 1782 in its current form, it believed the statute to be “the kind of assistance that is likely to be preferred abroad,”33 and advised federal courts to consider the “character” of the foreign proceeding and the “nature and attitudes” of the foreign country and tribunal.34 When the Supreme Court took up the dispute between Intel and AMD, it further directed district courts to avoid offense to foreign courts and to maintain an appropriate level of parity between foreign litigants.35 Based on factors set out by the Court,36 the Northern District of California denied AMD’s discovery request on remand, given the European Commission’s expressed rejection of US discovery.37

Yet, the export of American discovery has also been described as “legal imperialism”38 and “officious intermeddl[ing].”39 And this Article’s empirical and doctrinal analyses confirm that federal judges are not able to apply the Supreme Court’s instructions in practice. Some pinpoint the problem as § 1782’s conferral of broad discretion on federal judges,40 though there is much disagreement over how that discretion should be narrowed.41 Others contend that the export of American discovery violates the separation of powers in the United States,42 and that it is a unilateral fix for a set of issues that call for a multilateral solution.43 While a treaty governing the international exchange of evidence—the Hague Convention on the Taking of Evidence Abroad in Civil or Commercial Matters (“Hague Evidence Convention”)—was signed in 1970,44 it is widely considered to be ineffective,45 leading some to demand further efforts to forge procedural uniformity across nations46 and others to bemoan the futility of seeking convergence given the vast procedural differences worldwide.47 Needless to say, there is no consensus on whether the export of American discovery is working, and, if not, where the problems lie and what an alternative system for sharing evidence would look like.

I argue that the export of American discovery is in need of reform. Its most pressing shortcoming is its failure to include and to provide due process to the appropriate actors. When discovery is exported, it has ripple effects for foreign adversaries against whom the evidence is to be used, and for the foreign tribunal overseeing the dispute. Yet, there is no established mechanism for informing or involving those entities. Instead, federal judges have remained at arm’s length from the very foreign proceedings they are influencing, and have engaged in the impossible task of abstractly weighing foreign interests with no foreign input. The problem, therefore, is not an excess of discretion, but a shortage of information. This problem can be addressed in the immediate term through shifts in judicial practice, and in the medium term through changes to the FRCP and amendments to § 1782. In the long term, a broader reconceptualization of the global role played by federal judges—and the global space in which they operate—is needed.

This Article proceeds in four Parts. Part I describes the centrality of § 1782 to the export of discovery. The statute allows foreign tribunals and foreign private actors to seek US discovery directly from federal district courts. In interpreting the statute, the Supreme Court instructed district courts to consider a foreign tribunal’s need for and receptivity to US discovery, and to maintain an appropriate measure of parity between the parties.48 The statute’s implementation by lower courts is now fractured along many lines, exposing confusion surrounding the statute’s purpose and scope, as well as its intended effect on foreign tribunals and the parties before them.

Part II provides a comprehensive, nationwide, descriptive account of how foreign discovery requests have operated in district courts. I compiled a dataset of over three thousand foreign discovery requests filed under § 1782 between 2005 and 2017, approximately two thousand of which are for use in civil proceedings abroad. Relying on the dataset, I show that the foreign civil demand for US discovery has approximately quadrupled in that time, and that their countries of origin have diversified. Demand from foreign parties now overshadows demand from foreign tribunals, with private actors making more complex and creative strategic uses of US procedures across borders. Overall, the grant rate is very high (94 percent in 2015) while the contestation rate is relatively low (22 percent in 2015), calling into question whether US judges are serving as effective discovery gatekeepers for disputes in foreign tribunals.

Part III performs a doctrinal evaluation of the export of American discovery and concludes that US judges are at a severe informational disadvantage when fielding requests from foreign parties. It is assumed that the machinery of domestic discovery can be extended to exported discovery, but the discretion and expertise exercised by federal courts in the domestic context cannot be transferred to discovery requests from foreign parties due to the absence of critical actors and the lack of relevant information. Consequently, federal judges have devised shortcuts for the analyses they conduct, which in turn place a heavy thumb on the scale for granting discovery. Those shortcuts not only undercut congressional intent and the Supreme Court’s interpretation of § 1782, but also pose larger normative problems by eroding comity, adverseness, and basic notions of due process and fairness. By contrast, the discretion and expertise exercised by federal courts in the domestic context is inapposite for discovery requests from foreign tribunals, the vast majority of which are governed by treaty.

Part IV proposes several reforms. I advocate for more active judicial management of § 1782 requests that systematically seeks out the participation of foreign opposing parties and foreign tribunals. I also recommend restructuring § 1782 requests so that each foreign discovery request is no longer filed as a stand-alone US case—a feature that makes these requests uniquely difficult to administer.

Finally, I conclude that this study of § 1782 and its unaccounted-for foreign impacts suggest broader challenges that courts face when disputes straddle multiple legal systems. Litigants and attorneys are adapting to the transnational nature of litigation by strategizing across borders and deploying national procedural tools to their global advantage. Meanwhile, judges are increasingly at an informational disadvantage as they continue to operate within national silos, adjudicating disputes, only parts of which are before them. These challenges call for institutional solutions in the form of court-to-court information sharing and coordination across borders, as well as a shift toward reconceptualizing federal judges as global actors who share overlapping authority with foreign judges and arbitrators.

I. The Mechanics of Export

This Part sets out the mechanics, governing laws, and institutional actors engaged in the export of discovery. The United States has been offering some form of exported discovery since the mid-nineteenth century. Today, 28 USC § 1782 is the key governing statute. It provides the broadest and most direct access to US discovery, and it is also used internally by the Department of Justice to execute discovery requests made through two indirect routes—the traditional system of letters rogatory and the Hague Evidence Convention. This Part describes the three existing paths to US discovery, highlights the centrality of § 1782, and examines the Supreme Court’s interpretation of the statute.

A. Three Paths to US Discovery

1. Letters rogatory.

The traditional “letter rogatory”—a formal request to perform a judicial act sent from the court of one country, typically through diplomatic channels, to the court of another country49 —is the oldest mechanism for obtaining discovery assistance in the United States.50 It continues to operate today, with the Department of State and Department of Justice acting as intermediaries for executing them.51 Internationally, letters rogatory are fulfilled on a discretionary basis and as a matter of international comity.52 Domestically, they are accorded the same more favorable treatment as requests made under the Hague Evidence Convention,53 described below.

Because letters rogatory originate with a request from a foreign court (rather than a party to a foreign suit), the receptivity of that court to US discovery assistance is assured. Letters rogatory have been criticized for being unwieldy, time-consuming, and costly.54 They typically travel from the requesting court in a foreign country, to that country’s ministry of foreign affairs, then to that country’s embassy in the United States,55 then to the US Department of State, and then to the Office of International Judicial Assistance (OIJA) in the Department of Justice. OIJA screens the request for straightforward technical requirements, attempts to secure the requested discovery voluntarily, and in the absence of voluntary compliance, submits the request to a district court,56 which then relies on § 1782 for execution. Once compelled, the discovery travels back to the requesting court via the same path. The State Department estimates that this process can take a year or more and recommends use of more “[s]treamlined procedures,” such as an international treaty or a direct petition to a court, when possible.57

2. 28 USC § 1782.

During the years following World War II, a surge in international business and other cross-border activities led to a “flood of litigation” in the United States with international elements.58 American lawyers, frustrated by procedural differences between the United States and the civil law countries of Europe and Latin America, called for “a modernization of international legal procedure.”59 With the adoption of the FRCP in 1938, American lawyers were accustomed to a liberal, party-driven system of discovery aimed at uncovering the “fullest possible knowledge of the issues and facts before trial.”60 Yet, in other countries they were forbidden from taking evidence directly,61 and had to rely on letters rogatory, which they found to be “inefficient, time consuming, and costly.”62 Some jurisdictions, like Germany and the Netherlands, would not compel the testimony of unwilling witnesses even when a letter rogatory was issued.63 And testimony secured through a letter rogatory might not be usable in the United States due to noncompliance with domestic requirements such as examination under oath and oral questioning by an attorney.64

Congress enacted a number of measures to address these challenges. To facilitate the import of discovery, Congress authorized federal courts to subpoena and to hold in contempt an American citizen or resident in a foreign country.65 The FRCP were revised to specify how testimony could be taken abroad,66 and permitted district courts to order the production of documents, regardless of location, so long as they are in the “possession, custody, or control” of a party to a US proceeding or a nonparty witness over whom the court has jurisdiction.67

To facilitate the export of discovery, Congress passed 28 USC § 1782 in 194868 and rewrote it in 1964.69 The statute reads:

The district court of the district in which a person resides or is found may order him to give his testimony or statement or to produce a document or other thing for use in a proceeding in a foreign or international tribunal. . . . The order may be made pursuant to a letter rogatory issued, or request made, by a foreign or international tribunal or upon the application of any interested person and may direct that the testimony or statement be given, or the document or other thing be produced, before a person appointed by the court. . . . The order may prescribe the practice and procedure, which may be in whole or part the practice and procedure of the foreign country or the international tribunal, for taking the testimony or statement or producing the document or other thing. To the extent that the order does not prescribe otherwise, the testimony or statement shall be taken, and the document or other thing produced, in accordance with the Federal Rules of Civil Procedure.70

In 1996, the statute was revised to include discovery for use in criminal investigations71 —a subject beyond the scope of this Article.

At the time of the 1964 revision, the statute was considered “a major step in bringing the United States to the forefront of  nations.”72 It was thought to provide “equitable and efficacious procedures” for foreign tribunals and litigants, and to meet the “requirements of foreign practice and procedure.”73 Congress hoped that foreign countries would reciprocate by providing US litigants with easy access to foreign discovery.74 Despite these aspirations, many have observed that foreign countries have not in fact reciprocated.75

3. Hague Evidence Convention.

While these domestic policies were being put into place, it was also recognized that the problems arising from international litigation needed an international response. During the late 1960s, the United States led the negotiations for the Hague Evidence Convention, which was adopted in 1970 and entered into force for the United States in 1972.76 It operates through “Letters of Request,” which are issued by the judicial authority of one contracting state to the designated “Central Authority” of another, and then to the proper domestic authority for execution.77 Within the United States, OIJA serves as the Central Authority.78 Once a request reaches OIJA, it is carried out in the same way regardless of whether it is a request under the Hague Evidence Convention or a letter rogatory from a non-Convention country.79 OIJA screens the request, attempts to secure the evidence voluntarily, and then forwards the request to the appropriate federal district court for compelled discovery under § 1782.80 Since requests must originate from a foreign judicial authority (rather than from a party), receptivity of that authority to US discovery is assured.

The Convention was initially presented to the US Senate and to bar associations as requiring other countries to make concessions while not necessitating significant changes domestically given the existence of the more powerful § 1782.81 It is now in force between the United States and fifty-four countries.82 Many have commented that it has failed to achieve its intended purpose of bridging the gap between the United States and civil law countries.83 All but a handful of contracting states have adopted a declaration that they will not execute letters of request seeking pretrial discovery of documents as is common in the United States.84 Meanwhile, observing that the Convention’s procedures “would be unduly time consuming and expensive, as well as less certain to produce needed evidence than direct use of the Federal Rules,” the Supreme Court concluded in 1987 that the Convention is not the exclusive or even the required first resort procedure for American litigants seeking evidence abroad.85 Instead, a US court may continue to unilaterally compel extraterritorial discovery from those subject to its jurisdiction86 —a practice resented by foreign countries.87 There remains no effective international agreement governing discovery across borders, and some suggest that the United States should denounce the Hague Evidence Convention.88

B. The Centrality of 28 USC § 1782

These three paths to US discovery now overlap. A foreign tribunal seeking discovery in the United States may either use letters rogatory, the indirect path of the Hague Evidence Convention, or the direct path of § 1782. Similarly, a foreign private actor may either indirectly ask the relevant foreign tribunal to send a request, or directly request it from a federal district court under § 1782, bypassing the foreign tribunal and other intermediary national authorities. Within the United States, all requests are ultimately executed by federal district courts under § 1782.89 Section 1782’s statutory language does not differentiate between direct requests and indirect requests, or between requests from foreign tribunals and those from foreign private actors.

Direct requests from foreign private actors have caused the most complications and are the source of nearly all appeals in the past decade.90 Some scholars argue that permitting private actors to access US discovery was a “dramatic departure” from the traditional notion that judicial assistance is provided by one court to another.91 In particular, courts have struggled with the question of how they should weigh a foreign tribunal’s receptivity to US discovery. Section 1782’s statutory language does not shed light on this question, but the statute’s legislative history notes that federal district courts should consider “the nature and attitudes of the government of the country from which the request emanates and the character of the proceedings in that country” when granting or denying a § 1782 application.92

In 2004, the Supreme Court considered whether sought-after discovery must be discoverable abroad for it to be discoverable under § 1782—a shorthand for determining receptivity that had been adopted by some courts.93 The Supreme Court rejected the foreign discoverability requirement in Intel Corp v Advanced Micro Devices, Inc,94 a case in which the discovery sought by a private actor under § 1782 was not discoverable abroad and expressly not wanted by the foreign tribunal at issue: the European Commission.95 The Commission filed an amicus brief explaining that the discovery would give AMD access to documents it is not permitted to review under European law, would undermine the Commission’s policies on confidential information, and would increase its workload and divert its enforcement resources.96 Section 1782, the Commission argued, could “become a threat to foreign sovereigns if interpreted expansively.”97 Several industry associations filed amicus briefs expressing concern that compelling discovery under § 1782 that is not discoverable abroad could produce unfair outcomes when US-style discovery benefits one side of a dispute but not the other.98 The Department of Justice filed an amicus brief supporting a broad interpretation of § 1782 since it relies on the statute to execute letters rogatory and letters of request under the Hague Evidence Convention.99

The Supreme Court reasoned that a foreign discovery restriction does not necessarily translate into an objection to exported discovery from the United States.100 Instead, the Court enumerated four discretionary factors for district courts to consider: (1) whether the requested evidence is available without § 1782; (2) whether the foreign government or court is receptive to US federal court judicial assistance; (3) whether the request “conceals an attempt to circumvent foreign proof-gathering restrictions or other policies of a foreign country or the United States”; and (4) whether the request is unduly burdensome or intrusive.101 The Supreme Court explained that these factors, as well as the exercise of judicial discretion more generally, could safeguard comity. Similarly, parity between the foreign adversaries could be maintained either by the district court conditioning its grant of discovery, or by the foreign tribunal conditioning its acceptance of US discovery, on a reciprocal exchange of information.102 On remand, the Northern District of California denied the discovery request given the European Commission’s amicus brief expressing resistance to US discovery.103

Since the Supreme Court’s decision in 2004, lower courts’ implementation of § 1782 have continued to fracture along many lines. Most prominently, the Courts of Appeals now disagree on whether the statute can be used to compel discovery in aid of foreign commercial arbitrations. The Seventh Circuit recently held that § 1782 does not extend to private international commercial arbitrations, placing it in agreement with the Second and Fifth Circuits and in conflict with the Fourth and Sixth Circuits.104 Lower courts also disagree on whether § 1782 can be used to compel documents physically located abroad but under the possession, custody, or control of a US entity.105 Further complicating the matter, entrepreneurial litigants now use discovery obtained through § 1782 in multiple proceedings before multiple tribunals once the statute’s requirements are deemed satisfied with respect to one foreign proceeding.106 These are just a few of the developments that expose persistent confusion surrounding the statute’s purpose and scope, as well as its intended effect on foreign tribunals and the parties before them.107

II. Evidence from Federal Dockets

This Part presents the findings of the first nationwide study of foreign discovery requests in federal district courts. Despite anecdotal reports that foreign discovery requests have experienced “a groundswell of popularity,”108 there have been no attempts to systematically investigate recent trends in the overall number of requests, or their nature, origins, and outcome. I fill this gap by compiling and analyzing the most exhaustive existing dataset of requests for discovery to be used in civil disputes abroad.

The number of discovery requests for use in foreign civil proceedings received by district courts has indeed surged, approximately quadrupling between 2005 and 2017. Their countries of origin have diversified, suggesting that the historical concern about procedural differences between the United States and the civil law countries within Western Europe and South America may not be as central as it used to be. Meanwhile, there is a growing need to understand legal systems in Asia and Eastern Europe.

The vast majority of requests fall into two categories: indirect requests from foreign courts and direct requests from foreign parties. In other words, foreign parties now have a more direct relationship with US district courts than do foreign tribunals. Moreover, demand from foreign parties now overshadows demand from foreign tribunals in both number and complexity of requests. Whereas foreign tribunal requests are fairly straightforward, homogenous, and most frequently connected to family law matters, foreign party requests are more sophisticated, varied, and most frequently connected to commercial matters. These divergences suggest that there are different dynamics at play in these two sets of requests. Whether US judges are serving as effective discovery gatekeepers in each type of request is examined in the next Part.

Analysis of federal dockets is also illuminating for what it cannot reveal. Requests are typically considered ex parte, without informing or including the foreign tribunal or foreign opposing party, though occasionally notice is given in haphazard ways. For reasons explained below, it was not possible to systematically track when notification was provided to the foreign tribunal or foreign opposing party and therefore when a foreign tribunal was aware that a US discovery request had been made. Because I could not track foreign tribunal awareness of a US discovery request being made by a foreign party, it was also not possible to determine how often and which foreign courts tended to welcome or resist sought-after discovery. Finally, docket analysis provides an incomplete picture of when multiple US discovery requests are made for the same foreign proceeding, and no information about what happens after a request is granted. Federal judges make foreign discovery decisions in the absence of this information.

A. Data Overview

I compiled a dataset composed of over three thousand discovery requests, filed between January 1, 2005 and December 31, 2017, seeking compelled discovery under § 1782 for use in foreign proceedings. Since § 1782 authorizes discovery requests for foreign civil and criminal proceedings, this initial dataset included both. Appendix A describes the process I used to build the dataset. Briefly, I relied on the Public Access to Court Electronic Records (PACER) system—a mandatory electronic docketing system that provides access to all actions filed in federal district courts nationwide.109 I devised a text search for identifying § 1782 requests that maximized sensitivity without significant sacrifices in specificity. I then ran the search on Bloomberg Law, which renders PACER text searchable. I manually eliminated false positives and performed several checks to ensure that the dataset is close to exhaustive and lacking in bias. I selected the time period 2005 to 2017 to capture recent trends in foreign discovery requests: 2005 is the first full calendar year after the Supreme Court decided Intel, and 2017 is the final full calendar year prior to commencement of this study. The dataset is summarized year by year in Table 5 of Appendix C. I drew a random sample of over one thousand discovery requests for more detailed analysis, also summarized year by year in Table 5.

Analysis of the sample shows that approximately one-third (919) of all requests were connected to criminal proceedings abroad (“criminal requests”), while approximately two-thirds (2,070) were connected to civil proceedings abroad (“civil requests”). Breaking down the requests by year shows that the number of civil requests has grown rapidly, approximately quadrupling between 2005 (49 requests) and 2017 (208 requests). While it is not possible to determine from docket analysis the causes of this rapid rise, there are several possible reasons for it: (1) an increase in cross-border activity leading to more disputes abroad for which evidence may be gathered in the United States;110 (2) an increase in foreign substantive laws with extraterritorial reach such that more transnational activity is subject to civil suits abroad;111 and (3) an increase in awareness and use of § 1782 by law firms, attorneys, and parties.

There is an inverse trend for criminal requests brought under § 1782,112 though this does not necessarily reflect an overall contraction in criminal requests due to the enactment of an overlapping federal statute in 2009.113 There is also a small number of cases every year—cumulatively more than one hundred over the study period—that are refiled under another case number due to confusion concerning the proper case type designation for foreign discovery requests.114 The rising number of civil requests, along with upper and lower bounds representing 95 percent confidence intervals, are visualized below in Figure 1 and summarized numerically in Table 6 of Appendix C.

Figure 1: Estimated Number of Civil Requests, 2005–2017115

Approximately half of civil requests are sent to the Southern District of New York, the Southern and Middle Districts of Florida, and the Northern and Central Districts of California,116 with the rest distributed across the country. Over 60 percent of district courts nationwide received at least one foreign civil discovery request during the study period.117 The remaining analyses below rely on in-depth coding of civil requests from the randomly drawn sample. Appendix B describes the methodology I used for coding.

B. Request Analysis

I examined basic characteristics of civil requests: who requested them, who they targeted, the nature of the foreign tribunal, the nature of the foreign proceeding for which they were requested, and the country of origin.

1. Requestor.

The vast majority of foreign discovery requests come either indirectly from foreign tribunals through OIJA (approximately 40 percent) or directly from foreign parties under § 1782 (approximately 55 percent). For the most part, foreign tribunals continue to seek judicial assistance indirectly through the Hague Evidence Convention and letters rogatory despite having direct access to federal courts.118 Additionally, a tiny number of requests originate from a broader class of “interested persons” (approximately 0.77 percent) who are not a party to, but have some procedural rights in, a foreign proceeding.119 These findings are summarized in Table 8 of Appendix C. Figure 2 below visualizes the increase in tribunal and party requests over time, and shows that the number of party requests has exceeded the number of tribunal requests in most recent years.

Figure 2: Estimated Number of Civil Requests from Foreign Tribunals and Foreign Parties, 2005–2017120

 

2. Target.

The vast majority of civil requests seek discovery from nonparties to the proceeding abroad. This is true for requests from foreign tribunals (approximately 85 percent) and foreign parties (approximately 88 percent), and remains consistent over the study period. Nonparty targets include banks, Internet and social media companies, as well as law firms. The prominence of nonparties as targets is in part because nonparties located in the United States cannot be reached by the foreign court where the case is pending, and in part because discovery targeting nonparties is favored under Intel, so it is advantageous to target a US nonparty even if a foreign party holds the same information.121 Table 9 of Appendix C provides a breakdown of requests from foreign tribunals and foreign parties by target. 

3. Nature of foreign tribunal and proceeding.

Overall, requests from foreign tribunals are more homogeneous than those from foreign parties. Virtually all requests from foreign tribunals seek discovery for use in one pending litigation before a foreign court. Requests from foreign parties are more varied. The vast majority are for use before foreign courts (approximately 90 percent), but a steady number are for use in commercial arbitrations (approximately 9.9 percent) and a smaller number are for use before foreign regulatory agencies (approximately 4 percent) and investor state arbitrations (approximately 2.5 percent).122 Most party requests are for use in one foreign proceeding (approximately 72 percent) or in only pending foreign proceedings (approximately 84 percent). But nearly a third are for simultaneous use in multiple proceedings worldwide (approximately 28 percent) and a steady number are for use in contemplated foreign proceedings that are yet to be filed (approximately 15.9 percent). Tables 10 and 11 of Appendix C summarize these findings. The number of foreign party requests seeking discovery for contemplated prefiling proceedings—as well as those for multiple parallel proceedings worldwide—are increasing over time.123

There is significant breadth in the substantive merits issues in dispute in the foreign civil proceedings. Requests from foreign tribunals are concentrated primarily in family law (approximately 52 percent), followed by contract (approximately 15 percent) and employment law (approximately 12 percent). These substantive areas have remained consistently prevalent for tribunal requests over the study period.124 Requests from foreign parties are concentrated primarily in contract law (approximately 27 percent), followed by intellectual property and trade secret law (approximately 19 percent), and corporate law (approximately 12 percent). While the prevalence of contract law disputes in foreign party requests has remained consistent over the study period, the number of intellectual property and corporate law disputes has grown over time and the number of family law disputes has diminished.125 These shifting tides suggest that the private usage of § 1782 is increasingly driven by corporations and increasingly involves the types of cases that are likely to result in voluminous discovery requests.126 Table 12 of Appendix C summarizes these findings.

4. Country of foreign proceeding.

As the number of civil requests has increased over the years, so too has the diversity of countries from which they originate.127 This is true both for all requests taken together and for tribunal and party requests taken separately.

Breaking down the countries by region, legal system type, and Hague Evidence Convention status128 further illustrates the range of countries with which district courts are interacting. The majority of tribunal requests come from the Americas (approximately 62 percent), and the most significant region of growth during the study period is Eastern Europe. The majority of party requests come from Western Europe (approximately 61 percent), and the most significant region of growth for party requests is Asia.129 Tribunal requests predominantly seek discovery for use in civil law countries (approximately 93 percent) and in countries for which the Hague Evidence Convention is in force with respect to the United States (approximately 86 percent). Party requests are again more varied. About as many come from common law countries (approximately 44 percent) as civil law countries (approximately 46 percent), and a steady number comes from mixed or other legal systems (approximately 17 percent).130 Most party requests come from countries for which the Hague Evidence Convention is in force with respect to the United States (approximately 62 percent), but a significant number also come from other countries (approximately 43 percent).131 See Tables 1 and 2 below summarizing these findings.

Table 1: Estimated Number of Civil Requests by Region of Origin, 2005–2017132

	Requests from Foreign Tribunals		Requests from Foreign Parties
	Number 95% CI	>Percentage 95% CI	Number 95% CI	Percentage 95% CI
Americas	565483–660	62% 53%–73%	273 219–343	23% 18%–29%
Caribbean	0 0–11	0% 0%–1.2%	60.7 40–96	5.1% 3.3%–8%
Western Europe	166 126–222	18% 14%–24%	732 639–837	61% 53%–70%
Eastern Europe	112 81–158	12% 8.9%–17%	53.9 35–87	4.5% 2.9%–7.2%
Middle East	38.3 23–67	4.2% 2.5%–7.4%	74.2 51–112	6.2% 4.2%–9.3%
Asia	25.5 14–50	2.8% 1.5%–5.5%	209 163–271	17% 14%–23%
Africa	3.2 1–17	0.35% 0.11%–1.9%	20.2 11–43	1.7% 0.92%–3.6%

Table 2: Estimated Number of Civil Requests by Legal System Attribute133

	Requests from Foreign Tribunals		Requests from Foreign Parties
	Number 95% CI	Percentage 95% CI	Number 95% CI	Percentage 95% CI
Legal System Type (2005–2017)134
Common Law	28.7 17–54	3.2% 1.9%–5.9%	525 445–620	44% 37%–52%
Civil Law	846 759–940	93% 83%–98%	553 470–650	46% 39%–54%
Mixed/Other	35.1 21–63	3.9% 2.3%–6.9%	210 163–273	17% 14%–23%
Hague Evidence Convention Status (2005–2017)135
In Force	779 >690–878	86% >76%–97%	745 652–851	62% 54%–71%
Not in Force	131 97–181	14% 11%–20%	519 441–613	43% 37%–51%

Breaking down the countries by rule-of-law rating,136 which was examined for the three-year period 2015–2017 due to limitations in the rule-of-law data used,137 more party requests are for use in countries with relatively high rule-of-law scores than are tribunal requests. Approximately half (57 percent) of party requests came from countries with rule-of-law scores in the top quartile compared to a third (33 percent) of tribunal requests. At the other end of the spectrum, approximately one-tenth (9.4 percent) of party requests came from countries with rule-of-law scores in the bottom quartile compared to approximately a quarter (24 percent) of tribunal requests. Note that a far greater proportion of party than tribunal requests (approximately 22 percent versus 0.95 percent) came from countries for which no rule-of-law rating is available, suggesting that more party requests are coming from countries that have not traditionally received as much attention in the US legal community. Table 3 below summarizes these findings.

These findings suggest that US courts are now interacting with a larger spectrum of legal systems worldwide, as opposed to the relatively narrow focus on the civil law systems within Western Europe and South America when the Hague Evidence Convention was negotiated during the 1960s.138 Since most legal scholarship on comparative law concentrates on these regions, it is unclear how conclusions from that literature extend to the rest of the world. There is a growing need to understand a more diverse set of legal systems worldwide, and the types of discovery to which they may be open.

Table 3: Estimated Number of Civil Requests by Rule-of-Law Score139

	Requests from Foreign Tribunals		Requests from Foreign Parties
	Number 95% CI	Percentage 95% CI	Number 95% CI	Percentage 95% CI
Quartile 4	101 >73–142	33% 24%–47%	224 180–279	57% 46%–71%
Quartile 3	95.3 68–136	31% 22%–45%	59.2 39–92	15% 9.9%–23%
Quartile 2	31.8 19–57	10% 6.3%–19%	28 16–53	7.1% 4%–13%
Quartile 1	72.2 50–108	24% 16%–36%	37.4 23–65	9.4% 5.8%–16%
No Rule-of-Law Score	2.9 1–15	0.95% 0.33%–4.9%	87.2 61–126	22% 15%–32%

C. Outcome Analysis

Finally, I examined the outcome of requests. I used two crude proxies to gauge the complexity of requests: the number of docket entries and the number of orders.140 I also tracked whether the request was ultimately granted.141

Foreign party requests are more complex and require more judicial activity to resolve than those from foreign tribunals. But both sets of requests are resolved relatively quickly and with minimal judicial activity. Requests from foreign tribunals are typically resolved with one order—the order granting or denying the request—and such requests were granted approximately 98.1 percent of the time.142 Requests from foreign parties typically take about three orders to resolve, and were granted approximately 86.6 percent of the time.143 The overall grant rate for all requests over the study period was approximately 91.9 percent.144 Excluded from these grant rate calculations are requests that did not reach resolution, often because the foreign proceeding had settled or otherwise reached a conclusion before the US discovery request reached completion, which in turn led the applicant to withdraw the request. Requests with no resolution outnumber denials in both tribunal and party requests, suggesting that the lack of timing coordination between the US discovery request and the foreign plenary proceeding limits the potential use of US discovery abroad.145 Figure 3 below visualizes the estimated grant rate for all requests, tribunal requests, and party requests over the study period. Tables 13, 14, and 15 in Appendix C summarize the numerical outcome data.

While it would be illuminating, it is not possible to compare these grant rates to those of domestic discovery requests. This is first and foremost because domestic discovery is typically negotiated between the parties at the outset and does not generate a record in the docket unless there is a disagreement and a motion to the court to resolve it. Additionally, I have not found any empirical research quantifying the rate at which motions to compel or motions for protective orders are granted domestically.

Figure 3: Estimated Grant Rates of Civil Requests

For 2015 only, I tracked the outcome and contestation status—meaning whether the request was challenged146 —for every request. Table 4 below summarizes the findings. Tribunal requests were granted 98.9 percent of the time and contested only 3.3 percent of the time. Party requests were granted 90 percent of the time and contested 37 percent of the time. The grant rate was higher for uncontested party requests (93.9 percent) than contested party requests (82.4 percent). The overall grant rate for both sets of requests was 94 percent and the overall contestation rate was 22 percent.

Table 4: Grant and Contestation Rates for 2015147

	Requests from Foreign Tribunals	Requests from Foreign Parties	All Requests
Overall Grant Rate148	98.9%	90%	94%
Contestation Rate149	3.3%	37%	22%
Grant Rate for Contested	67%150	82.4%	81.1%
Grant Rate for Uncontested	100%	93.9%	97.3%

Logistic regression consistently indicated a strong negative correlation between a request being contested and its likelihood of being granted.151 This correlation was strongly indicated even in the presence of other variables significantly correlated with contestation.152 With respect to the remaining variables, logistic regression yielded inconsistent results.153

D. Unknowns

While docket analysis can shed light on the nature of foreign discovery requests and their outcomes in federal district courts, there are several critical pieces of information that docket analysis cannot reveal. First, it cannot show whether the foreign opposing party was notified about discovery requests coming from foreign parties. Requests are typically considered ex parte at the outset.154 If an ex parte request is granted, as most are, a subpoena and order are served on the discovery target, which is when the target learns of the discovery request. If the target is, or is related to, the foreign opposing party (itself difficult to ascertain in a systematic manner), then the latter learns of the request at the same time. If the target is not related to the foreign opposing party, notification cannot be systematically tracked because, as discussed in Part III, it is provided haphazardly and sometimes in the absence of a written record.155 Without knowing when notification was provided, I was unable to examine the extent to which the low contestation rate was due to lack of notice versus failure to contest. If the lack of notice is depressing the contestation rate, then it may also be raising the grant rate.

Second, for similar reasons, and also discussed in Part III, it was not possible to systematically track whether the foreign tribunal received notification of the request. Without this information, I was unable to differentiate lack of awareness that the discovery request had been made from lack of resistance to US discovery. Consequently, it was not possible to look for any overarching patterns in the types of foreign tribunals that are more receptive or opposed to US discovery.

Third, docket analysis provides incomplete information about when multiple discovery requests are made in the United States for the same foreign proceeding. Unlike domestic discovery requests that are a part of a larger plenary dispute and overseen by the same judge, each foreign discovery request is its own stand-alone action. Consequently, a single proceeding abroad can generate many discovery requests in different US district courts, each before a different judge. Sometimes, these requests are consolidated and other times they are not.

Finally, docket analysis does not provide systematic information about what happens after discovery is compelled—including whether the requested discovery was ultimately produced and submitted to the foreign tribunal, whether the foreign tribunal took it under consideration, and whether it affected the outcome of the foreign proceeding. In Intel, the Supreme Court asserted that district and foreign courts could, respectively, condition grants and acceptances of discovery on reciprocal exchanges of information between the foreign parties to maintain an appropriate measure of parity.156 District courts very rarely impose such a condition,157 and it is not possible to know from docket analysis if such a condition is imposed downstream by the foreign tribunal. In short, the impact of discovery compelled in the United States on foreign proceedings and foreign parties is largely unknown—to scholars relying on docket analysis as well as to US judges deciding foreign discovery requests, and, in most cases, granting them.158

III. Evaluating American Discovery in the World

Having characterized the rise in foreign demand for US discovery, the differences between demand from foreign tribunals and foreign parties, the high rates at which both types of requests are granted, and the key unknowns, this Part now evaluates doctrinally whether US judges are serving as effective discovery gatekeepers for foreign proceedings. All foreign discovery requests for use in civil disputes abroad are executed under § 1782, regardless of whether they began as a letters rogatory, a request under the Hague Evidence Convention, or a § 1782 application brought directly by a foreign party.159 Section 1782, in turn, has been described as a “screen” or “threshold determination” of whether to allow a foreign actor access to US discovery as it operates domestically.160 Once that threshold is overcome, § 1782 “drops out” and the “ordinary tools of discovery management” under the FRCP take over.161 It is assumed that the FRCP can be seamlessly translated from the domestic to the transnational context, and that district courts can weigh the interests of affected parties in foreign countries just as they do in the United States.

Drawing from the empirical findings above as well as district court proceedings and appellate court decisions, I argue that the machinery developed for domestic discovery is both improperly applied to—and ill-equipped to manage the challenges of—exported discovery. The blueprint for domestic discovery falls short in distinctive ways when applied to requests from foreign tribunals. In entertaining requests from foreign tribunals, federal courts have a greatly reduced justification for exercising the discretion they typically wield under the FRCP or under Intel’s discretionary factors. That is because tribunal requests are not adversarial, and there is no uncertainty surrounding whether the foreign tribunal is receptive to US discovery, given that the tribunal is making the request.

By contrast, requests from foreign parties are—and should be—adversarial between the two contending parties in the foreign dispute. But much of the time they are not, or they involve some but not all of the relevant stakeholders, resulting in a distorted adversarialism and missing information. I identify who the relevant stakeholders are and the information that each uniquely possesses, in the absence of which district courts are unable to undertake the analysis required under the FRCP or under Intel’s discretionary factors. Because judges are at a severe informational disadvantage when fielding requests from foreign parties, they have developed more manageable heuristics that place a heavy thumb on the scale for granting applications, which in turn threatens to undermine foreign litigation as well as universal and American litigation values.

Part III.A examines the underlying tenets of domestic discovery. Parts III.B and III.C look, respectively, at how foreign tribunal requests and foreign party requests deviate from this framework.

A. A Domestic Procedure for Foreign Use

At the core of American civil litigation is a litigant-driven system for obtaining information from adverse parties that aims to give both sides “the fullest possible knowledge of the issues and facts before trial.”162 That system is considered fundamental to fair adjudication because it narrows the issues, promotes settlement, and reduces surprises during trial.163 It relies on contending adversaries to negotiate a mutual exchange of information that reveals the strengths and weaknesses of each party’s case, and a judge to manage and resolve discovery disagreements. Although this vision of active judicial management of discovery has not been fully realized in the domestic context,164 setting out this ideal highlights the grave problems posed by discovery in the international context.

The FRCP give litigants broad authority to obtain from each other “discovery regarding any nonprivileged matter that is relevant to any party’s claim or defense.”165 This scope is subject to the requirement that discovery be “proportional to the needs of the case,” which requires a consideration of “the importance of the issues at stake in the action, the amount in controversy, the parties’ relative access to relevant information, the parties’ resources, the importance of the discovery in resolving the issues, and whether the burden or expense of the proposed discovery outweighs its likely benefit.”166 The adversaries direct discovery requests to each other and negotiate the level of compliance.167 If they are unable to reach an agreement, a litigant can ask the court to compel discovery or for a protective order to forestall discovery.168

The current regime is the result of changes made over the past few decades in response to criticisms that discovery has been used abusively.169 The ongoing debate on the need for discovery reform in the United States is beyond the scope of this Article,170 except to note the measures courts have adopted to address discovery abuse. Most notably, the Supreme Court has reinterpreted the FRCP’s pleading standard, raising the bar for surviving a motion to dismiss as an indirect way to narrow discovery.171 The FRCP have also been revised to encourage “more aggressive judicial control and supervision,”172 both at the outset through pretrial scheduling conferences, and later on through the proportionality requirement. The adversaries and the court “have a collective responsibility to consider [ ] proportionality.”173 Participation from all parties is needed to elucidate the proportionality factors since each party holds different information, and that information may change or clarify over time.174 When parties disagree, the court’s role is to consider all the factors given the information provided by the parties, and to arrive at a case-specific determination, which is in turn reviewed for abuse of discretion.175 Some scholars have questioned whether judges can effectively apply the proportionality requirement, given the vagueness of the standard and judges’ relative lack of in-depth knowledge about the facts of the case.176 Some of the proportionality factors are difficult to quantify or require forecasting the value of sought-after information to the underlying dispute.177

The FRCP also allow parties to seek compelled discovery from nonparties178 —a process that maintains the core adversarial relationship between the parties by involving all parties to the dispute as well as the court presiding over the action. Notice and a copy of the subpoena must be served on each party to the dispute before it can be served on the nonparty target,179 so that other parties have an opportunity to object, to monitor the discovery, and to seek access to the information produced or make additional discovery requests of their own.180 When the nonparty is not subject to personal jurisdiction in the district where the case is pending, two district courts may be involved. The district court where the case is pending issues the subpoena,181 and the district court where the nonparty is found manages compliance and hears subpoena-related motions182 —a measure designed to protect nonparties through local resolution of disagreements. The judge in the compliance court is encouraged to consult with the judge in the issuing court on subpoena-related motions, since the latter is more familiar with the underlying case.183 Motions can also be transferred back to the issuing court so as not to disrupt the issuing court’s supervision over the underlying litigation, as might occur if the same discovery questions are likely to arise in many district courts or if the issuing court has already ruled on questions implicated by the motion.184 The FRCP recognize that the participation of the judge presiding over the case may be necessary due to her knowledge of the case and in order to consistently manage discovery requests across the case.

B. Use by Foreign Tribunals

When a district court receives a discovery request from a foreign tribunal, its role bears little resemblance to discovery within the United States. Under long-standing custom, these requests are typically considered on an ex parte basis.185 This practice is characterized by Professor Jim Pfander and Daniel Birk as an exercise of “non-contentious” Article III jurisdiction, which gives federal courts power to consider nonadversarial applications asserting a legal interest under federal law.186 While the target of a foreign tribunal request may oppose the discovery sought, leading to litigation that creates “a measure of adverseness,”187 tribunal requests are not adverse between the two contending parties in the underlying plenary dispute. For this reason, discovery requests from foreign tribunals have been likened to administrative subpoenas that federal courts enforce on behalf of agencies188 —a limited judicial role that has been described as “adjunct.”189

Without adversity between the two contending parties to the foreign dispute, district courts cannot exercise their usual broad discretion in evaluating domestic discovery disputes. There are no party needs or interests to weigh, and no disagreements over particular discovery requests to resolve. As noted in Part II, approximately 93 percent of foreign tribunal requests come from civil law countries,190 where discovery is primarily a judicial function.191

Nor is a district court entertaining a foreign tribunal request playing its usual screening role under § 1782. The statute requires that the requested discovery be “for use in a proceeding in a foreign or international tribunal.”192 Three out of the four discretionary factors enumerated by the Supreme Court in Intel—whether the discovery is available to the foreign tribunal without US court assistance, whether the foreign tribunal will be receptive to US court assistance, and whether the discovery request is an attempt to circumvent foreign discovery restrictions—weigh the likelihood that granting the request will offend a foreign tribunal.193 When the request is made by the foreign tribunal, it can be inferred that the discovery is for use in the proceeding it is adjudicating, and that the three comity-oriented Intel discretionary factors are met. Some courts acknowledge that these analyses collapse when the request comes from a foreign tribunal, while others parrot standard conclusory language that the Intel factors weigh in favor of granting the request.194

Not only are the bases for a district court’s usual exercise of jurisdiction under the FRCP and § 1782 inapposite, but for the approximately 86 percent of tribunal requests coming from countries for which the Hague Evidence Convention is in force with respect to the United States,195 district courts are further limited to a handful of permissible reasons for denying requests.196 This restriction under international law is not altered by the Convention’s internal execution through a preexisting general-use statute that is discretionary.197 For all of these reasons, the very low contestation rates and very high grant rates observed—typically with minimal judicial activity, the order granting the request being the only order issued by the court—are, for the most part, justified. These observations suggest that judges are highly deferential to the Hague Evidence Convention and, by extension, foreign tribunals, granting their requests more or less as a matter of course.198 Conversely, judges occasionally exceed their discretion by denying tribunal requests in violation of international law.199

C. Use by Foreign Parties

When a district court receives a discovery request directly from a foreign party, it bears some resemblance to discovery within the United States. The request is coming from a party to a foreign proceeding against an adverse party, which parallels the bipolar structure of domestic discovery disputes. Perhaps because of this analogous structure, some courts are confident that their “substantial experience controlling discovery abuse in domestic litigation” prepares them for “similarly root[ing] out sham applications under § 1782.”200 It is assumed that district courts are best positioned to weigh the needs and interests of parties affected by a foreign discovery request, just as they are in domestic discovery requests,201 and that the FRCP’s safeguards are well-suited to prevent foreign misuse.202 Consequently, the same abuse of discretion standard of review for ordinary discovery rulings is applied to § 1782 rulings.203

Yet, foreign discovery requests are distinctive in two key respects. First, there are two courts involved in a foreign discovery request—the US court entertaining the discovery request, and the foreign court presiding over the action. Since the plenary suit is necessarily abroad, it is governed by a different set of procedural rules. When a discovery request comes from a foreign party, it cannot be guaranteed that the foreign court or tribunal will accept US discovery under the FRCP as would another district court governed by the FRCP. Precisely for this reason, Intel set out three discretionary factors aimed at discerning whether the foreign court is receptive to exported discovery. Second, unlike domestic out-of-district discovery requests targeting a nonparty and involving two district courts, there is no clear requirement for informing or involving other parties to the foreign dispute or for consulting with the foreign court on subpoena-related motions. A foreign discovery request “stands separate from the main controversy” in a heightened way.204 There is both a heightened need for information given procedural differences between countries, and a reduced supply of information due to the absence of procedures for consulting the foreign opposing party and the foreign court. Consequently, requests from foreign parties cause more complications than those from foreign tribunals.

1. Missing stakeholders and information.

There is an acute lack of clarity as to who should be informed, involved, or consulted when a district court receives a discovery request from a foreign party. Following precedents concerning discovery requests from foreign tribunals, many courts have held that it is proper for § 1782 applications to be made on an ex parte basis even when that application comes from a foreign party.205 The rationale is usually that no prejudice will result because the target of the discovery will eventually have an opportunity to contest it once served with the subpoena.206 This reasoning does not distinguish between a discovery request that targets a party and one that targets a nonparty. It is the latter scenario that leaves the foreign adversary in the dark, preventing it from objecting to, monitoring, or seeking access to the requested discovery, as a domestic adversary would be able to do.207

Other courts have not condoned ex parte proceedings. But even when notification is required, courts do not agree on the legal basis for, or the components of, the requirement. Some have applied FRCP 45’s requirement that all parties be notified of discovery requests targeting nonparties.208 Others have applied local rules concerning ex parte orders—for instance, the Central District of California’s rule mandating a memorandum explaining why a matter was brought ex parte.209 Yet others have required notice as a matter of judicial discretion since § 1782 does not prescribe ex parte applications,210 or requested briefing on whether notice is needed.211 Some have even treated foreign discovery requests as if they are full cases or controversies, extending FRCP Rule 4’s requirement that a plaintiff serve a summons and a copy of the complaint on the defendant.212 The specific notification requirement has also varied: courts have ordered applicants to notify the target of the discovery request,213 the adverse party in the foreign proceeding against whom the evidence is to be used,214 and the foreign court itself.215 Sometimes notice is not required but given as a matter of courtesy, with or without a written record.

The confusion goes deeper: whether a foreign discovery request is a case or controversy is itself a question that has caused widespread discord across district courts, revealing uncertainty about the basic structure of these requests as well as the due process and personal jurisdiction requirements attending them.216 As noted in Part II, more than a hundred cases were recategorized by district courts either from a miscellaneous case to a civil case or vice versa during the study period of 2005 to 2017.217 Miscellaneous matters are typically ancillary or ex parte proceedings such as an out-of-district motion to compel or motion to quash, or the registration of a judgment from another district court.218 Civil matters are typically full cases or controversies between adversarial parties that invoke all the protections of the FRCP, including the requirement for a summons and service when a complaint is filed. That there is no case or controversy in the United States attached to foreign discovery requests has befuddled courts.

The result of this confusion and the accompanying erratic notification requirements are missing parties and stakeholders that ultimately deprive federal courts of the information they need to conduct rigorous analyses. Because foreign discovery requests are frequently decided without the knowledge and input of the foreign adversary or the foreign court, the range of basic information that judges struggle to ascertain is staggering. They include: whether the foreign proceeding is civil or criminal;219 whether the foreign proceeding is on appeal;220 whether the requested discovery is relevant to the foreign dispute;221 whether the foreign defendant has been served;222 the whereabouts of the foreign proceeding;223 the scope of discovery that is available in the country where the proceeding is being adjudicated;224 and whether a similar discovery request has already been denied in that country.225 The remainder of this Section examines how these missing stakeholders and this missing information impacts foreign litigation, basic notions of due process and fairness, and US litigation values.

2. Undermining foreign tribunals and litigation.

When the Supreme Court considered § 1782 in Intel, the Court stated that comity is “important as [a] touchstone[ ] for a district court’s exercise of discretion in particular cases.”226 The Supreme Court laid out four discretionary factors for district courts to consider, three of which are directed toward ensuring deference to and avoiding interference with foreign tribunals.227 The first factor is whether the foreign tribunal can itself order production of the evidence sought, or if it is unobtainable without US court assistance.228 Because the underlying discovery request in Intel sought evidence from a party to the foreign proceeding, the Supreme Court focused on the party status of the discovery target, writing that “when the person from whom discovery is sought is a participant in the foreign proceeding . . . , the need for § 1782(a) aid generally is not as apparent as it ordinarily is when evidence is sought from a nonparticipant in the matter arising abroad.”229 The second factor is whether US discovery assistance is desired abroad, and the Supreme Court instructed district courts that they may consider “the nature of the foreign tribunal, the character of the proceedings underway abroad, and the receptivity of the foreign government or the court or agency abroad to U.S. federal-court judicial assistance.”230 The third factor, related to the second, is whether the foreign discovery request “conceals an attempt to circumvent foreign proof-gathering restrictions.”231

Due to the lack of input from foreign tribunals and foreign opposing parties in § 1782 proceedings, and the open-ended nature of these factors,232 district courts have evolved simplified tests that lead to reflexive grants of foreign discovery requests. These simplified tests reflect Professor Maggie Gardner’s observation that the complex inquiries required in transnational cases encourage judges to develop analytical shortcuts that can lead to systemic bias favoring what is known (US parties and US law) over what is not known (foreign parties and foreign law).233 In the § 1782 context, the simplified tests systematically tip the scale toward granting foreign discovery requests while failing to properly apply Intel’s discretionary factors.

The first factor concerning whether the foreign tribunal can obtain the requested evidence without US assistance is often simplified to ask whether the target of the discovery request is a party or nonparty to the foreign proceeding. This analysis is easier to manage judicially, leading many courts to recite standard language that discovery is favored because it is sought from a nonparty.234 In fact, many discovery requests strategically target a token nonparty although the same evidence is also held by a party to the foreign proceeding. These token nonparties include American corporate affiliates of the foreign opposing party235 and American law firms that have represented foreign clients in US litigation.236 Most recently, the Second Circuit reversed a lower court’s grant of a § 1782 petition ordering Cravath, Swaine & Moore LLP to turn over documents in aid of litigation in the Netherlands.237 The reversal was based not on fears of interfering with the foreign litigation but rather on concern that granting the request would undermine attorney-client communications in the United States as well as confidence in protective orders.238 In the absence of information from foreign courts, it is easier to locate a domestic rationale than a foreign one for limiting a doctrine whose primary effect is abroad.

Focusing on the nonparty status of the discovery target leads to a particularly absurd result when the discovery that is sought is in fact located abroad. While the drafters of § 1782 did not anticipate the statute to be used extraterritorially,239 some courts have compelled discovery from the very country where the foreign dispute is being adjudicated, because the FRCP reach documents and other tangible things “in the possession, custody, or control” of the discovery target.240 Such extraterritorial discovery is typically obtainable by the foreign court, and seeking it in the United States should raise strong suspicions of the applicant sidestepping foreign discovery restrictions.

The second and third factors concerning receptivity and whether a discovery request is an attempt to circumvent foreign proof-gathering restrictions are often considered in tandem and have resulted in a number of analytical shortcuts that effectively write these factors out of existence. The most prominent among them is burden shifting, since Intel did not specify burdens. Many district courts have held that the target resisting discovery must provide “authoritative proof” that the foreign court is not receptive.241 Authoritative proof of a negative is hard to come by, particularly where the discovery target is a US nonparty to the foreign litigation, who likely has no information about the foreign tribunal adjudicating the dispute abroad. Other analytical shortcuts that district courts have taken include: inferring that membership in the Hague Evidence Convention signals receptivity to US discovery242 despite the fact that nearly all Convention members have submitted a declaration objecting to pretrial discovery as such discovery is mandated by the FRCP;243 and relying on prior federal court decisions concluding that a foreign country is receptive without looking more deeply at how those courts arrived at their conclusion.244

The overarching result of these simplified tests is that district courts are reflexively granting foreign discovery requests because the comity-based discretionary factors are not gauging whether exported US discovery is assisting or interfering with foreign proceedings. Instead, appellate courts have instructed district courts that it is preferable to modify a request on the basis that it is too burdensome rather than deny or quash a request altogether.245 Information about the burden imposed by a discovery request is more readily available since it can be furnished by the local US target of the discovery, providing another example of how the absence of foreign courts and foreign parties in § 1782 proceedings results in domestic rationale driving a doctrine whose primary effect is abroad. If all else fails, courts reason that the foreign tribunal can simply disregard the discovery compelled by a US court,246 neglecting the judicial and private resources wasted, as well as the fact that foreign countries lacking broad discovery provisions typically do not have admissibility rules. That judges usually do not have any information about what happens to discovery after they compel it for foreign use amplifies this problem.

3. Undermining universal and American litigation values.

Not only are there inadequate safeguards for protecting against compelling discovery that interferes with foreign litigation, the way in which foreign discovery requests are considered also undermines universal notions of due process and fairness as well as deeply held American litigation values. Federal courts are typically permitted to hear “definite and concrete” controversies affecting “the legal relations of parties having adverse legal interests.”247 While ex parte proceedings can be a legitimate exercise of Article III power, they pose potential risks to the rights of absent parties.248 Accordingly, federal courts must be particularly vigilant about protecting those parties, and due process requires that absent parties receive notice of proceedings that concern them, as well as an opportunity to participate.249 Noncontentious jurisdiction ends, or at least must be moderated, where a judgment encroaches on the rights of parties not before the court.250 Moderation may require judges to play a more active role, for instance by conducting their own factual investigation and framing the legal issues, since they cannot rely on an adverse party to do so.251

When foreign discovery requests proceed ex parte, they raise all the alarms that ex parte proceedings usually do. The presence of a nonparty target does not assuage these concerns, as nonparty targets do not have the same interest in resisting discovery as might the opposing party and often cooperate with the requestor, agreeing to a joint protective order that protects the confidentiality of the discovered materials instead of opposing the request itself.252 Sometimes the nonparty target even brings the application on behalf of the foreign applicant.253 The foreign adversary against whom the requested discovery is to be used is an obvious absent party whose rights are affected. The adverse-party requirement articulated by the Supreme Court,254 the Due Process Clause of the Fifth Amendment, and universal conceptions of fairness require that foreign adverse parties be notified of the proceeding and provided an opportunity to participate. Were the adverse parties located in the United States, these requirements would be set out in FRCP 45. But because the adverse parties and the plenary dispute is abroad, courts have not consistently applied any notification requirement, and, on the contrary, have even debated whether the opposing adversary has standing to participate.255

Although a foreign tribunal is not an absent party, compelled discovery in the United States can also alter a foreign tribunal’s ability to manage litigation before it. Notifying and involving the foreign tribunal is justified on this ground. Moreover, it is the type of factual investigation a US judge should undertake when faced with an ex parte § 1782 request. Were the foreign tribunal a different district court in the United States, consultation with the tribunal—and potentially also transfer of subpoena-related motions back to it—would be, respectively, encouraged and permitted under FRCP 45. But because the tribunal is located abroad, courts have, on occasion, precluded its participation.256

Without adversity and the typical due process accorded to parties whose legal interests might be harmed, § 1782 proceedings are often characterized by unfairness and a lack of parity between the parties to the foreign dispute. In Intel, the Supreme Court instructed district courts to consider parity a “touchstone[ ]” for its exercise of discretion, and noted that a district court could condition its grant of a discovery request on the applicant’s reciprocal exchange of information.257 Yet, compelling discovery in the United States conditioned on a reciprocal exchange of information poses more problems than it solves. For one thing, that reciprocal discovery is usually located abroad outside of the jurisdiction of US courts, and requires district courts to effectively order extraterritorial discovery that the foreign court could itself order and thus is likely to be perceived as interference.258 For another thing, when US discovery is sought from a nonparty to the foreign dispute, there is no way for the district court to ensure parity since the nonparty has no use for reciprocal discovery and the adverse party is not before the court.

Moreover, district courts typically do not consider parity at all, occasionally leading to inconsistent and unfair results within the United States. In a set of three related foreign discovery requests spanning nearly a decade, two were granted to Heraeus Kulzer, a German company, for use against its competitor Biomet with which it was embroiled in litigation in both Germany and Switzerland.259 Years later, a district court in the Eastern District of Pennsylvania denied a discovery request brought by Biomet for use against Heraeus Kulzer in related litigation, a decision that was later vacated in part due to parity concerns.260 In fact, foreign parties regularly make not one but numerous discovery requests in the United States, and since there is no requirement for applicants to inform district courts of related requests, it is difficult to ensure consistency even across a single foreign proceeding.261 This is particularly so given the number of splits among the courts on issues of law now plaguing § 1782.

Aside from undermining adverseness, due process, and parity between the foreign parties, the lack of information about the underlying foreign proceeding also frustrates the usual discovery analyses judges are expected to conduct. In particular, judges often have no reliable way of ascertaining whether the requested discovery is “relevant to any party’s claim or defense.”262 They do not usually attempt any proportionality analysis. Many of the proportionality factors are already difficult to gauge in the domestic context, and entirely impossible to gauge in the international context without input from the foreign court and the foreign adversary. For instance, a US federal judge has no basis for understanding “the importance of the issues at stake in the action,” which means the social, philosophical, or institutional significance of the substantive issues in the case263 —matters bound up in the social fabric of the foreign country. Nor do they have much visibility into the parties’ comparative resources or access to relevant information, or the importance of the discovery requested in resolving the case. The fear that federal judges do not have nearly as much facility with the facts as do the parties in a domestic discovery dispute264 is amplified many times over when discovery is requested for a foreign dispute. Whether requested discovery is relevant in foreign discovery requests is further complicated by the lack of information about whether it is possible for US discovery to be put to any sort of use abroad. For all of these reasons, some courts have noted that ex parte foreign discovery proceedings undercut “evenhanded justice and a sense of fair treatment”265 while making it more difficult for judges to make decisions.266

IV. Reimagining Global Discovery

The export of American discovery needs reform. Foreign discovery requests have been rapidly on the rise over the past decade and a half, and foreign party requests, in particular, have exceeded tribunal requests in most recent years.267 The previous Part laid out the many problems that arise when extending the FRCP to the international context. The problems are mild when district courts compel discovery at the request of foreign tribunals, and severe when they compel discovery at the request of foreign parties. Foreign parties now have a more direct relationship with federal courts than do foreign tribunals, and they have made more creative, heterogeneous, and dynamic use of § 1782.268 Meanwhile, federal judges are increasingly at an informational disadvantage as they adjudicate foreign discovery requests with little information about the plenary dispute abroad. This Part offers several proposals aimed at improving the operation of party requests.

Beginning with the most straightforward and achievable reforms, I call for more active judicial management of § 1782 requests so that Intel’s discretionary factors can be sincerely applied rather than merely paid lip service. To that end, I suggest that federal courts systematically invite the participation of foreign tribunals and foreign opposing parties. Because these changes are needed to follow the Supreme Court’s directives in Intel, they can be adopted immediately by federal judges—as some already have. But their consistent application across the country will require the addition of a new Federal Rule of Civil Procedure and moderate amendments to § 1782. This is particularly important given that over 60 percent of federal district courts received foreign discovery requests during the study period.269

Next, I suggest restructuring foreign discovery requests such that they are no longer treated as stand-alone actions in US courts. Instead, requiring that all discovery requests related to the same foreign proceeding be brought as a single unified US action before the same federal judge will improve administrability, facilitate active judicial management, and reduce internally inconsistent and unfair results.

A. Seeking the Participation of Foreign Tribunals

The Supreme Court held in Intel that comity is a policy concern that district courts should address in their exercise of discretion. Accordingly, three of the four discretionary factors set out in Intel are aimed at avoiding offense to foreign tribunals. Yet, the European Commission opposed Intel’s prescription of case-by-case judicial discretion on the basis that a district court “can only weigh fairly the complex interests of a foreign sovereign in aiding or blocking a Section 1782 discovery request if it is made aware of those interests.”270 “[S]o far as the Commission is aware,” it argued, “there is no system for providing it with notice of Section 1782 cases in which its interests are at stake, much less any regular procedure through which the Commission might appear and make those interests known.”271 The analysis above confirmed the European Commission’s suspicion that district courts would not be able to discern the interests of foreign tribunals without their input.

Meaningfully implementing Intel’s discretionary factors requires more active judicial management of foreign discovery requests by federal courts and more active participation from foreign tribunals. Federal judges can accomplish this by systematically providing written notice to foreign tribunals when § 1782 requests are received and sua sponte seeking the tribunals input. District courts may specifically ask foreign tribunals to weigh in on the Intel factors, to provide information needed to apply the FRCP’s proportionality analysis, or to share scheduling information so that US discovery can be produced in time for it to be taken into consideration abroad. As noted in Part II, many foreign discovery requests are currently filed while the foreign litigation is pending but reach no resolution because the foreign litigation is resolved before the application is decided.272

Notification and consultation of foreign tribunals is even more critical in those instances where evidence is requested for multiple parallel proceedings occurring around the world. Recall that approximately 28 percent of foreign party requests are for evidence to be used in multiple foreign proceedings, which may include a US proceeding.273 Even when additional foreign proceedings are not identified in the initial request, entrepreneurial litigants now use discovery obtained through § 1782 in multiple proceedings before multiple tribunals once the statute’s requirements are deemed satisfied with respect to one foreign proceeding.274 Each of those proceedings and their respective courts or tribunals would be affected by compelled US discovery, and their participation is needed to avoid confusion, duplication, and abuse. In addition to whether US discovery should be compelled and on what timeline, courts will likely need to consult with each other on issues of privilege and protective orders that would subsequently limit the use of the materials produced. A joint protective order agreed to by the parties and approved by a court in one jurisdiction will have externalities in other jurisdictions.

Some district courts have already begun taking a more active stance toward foreign discovery requests. They have, for instance, provided notice to foreign tribunals,275 corresponded directly with foreign judges regarding compelled evidence,276 delayed their decision pending foreign court action,277 ordered and received status reports on foreign proceedings,278 examined documents submitted to foreign tribunals,279 and sought rulings from foreign courts.280

Some foreign tribunals have also started taking more active part in § 1782 proceedings either directly or through private parties. The Korean Fair Trade Commission, for example, filed letters to the court in seven foreign discovery requests in the Northern District of California.281 These letters asserted that the Korean Fair Trade Commission was not receptive to US discovery related to disputes between Qualcomm and Apple. Similarly, the European Commission sought to intervene and to file a memorandum in support of a motion to quash a § 1782 subpoena in the District of Massachusetts.282 In more than one instance, German authorities weighed in against the production of evidence in the United States,283 while in another instance, a German court not only expressed receptivity to but postponed its own hearing to permit a litigant to pursue discovery requests in Indiana and Pennsylvania.284 A Swiss arbitrator has conveyed nonreceptivity to US discovery,285 while an Israeli arbitrator has expressed receptivity to US discovery.286 Foreign litigants have also submitted letters from judges in Monaco and Germany noting that they would consider discovery compelled in the United States.287

These existing ad hoc attempts at fostering foreign tribunal participation in § 1782 proceedings reveal that the needs of transnational discovery are not easily generalizable, for they are too specific and fine-tuned to be addressed by blanket rules. A German judge may find US discovery assistance helpful in one proceeding but not in another. A broad-stroke solution covering an entire country, or a type of foreign proceeding such as private commercial arbitration, is unlikely to further the goal of facilitating dispute resolution across borders. Instead, case-by-case participation is needed.

They also suggest that processes for notifying and including foreign tribunals need to be systematized. A system of notification and consultation can be implemented immediately through shifts in judicial practice that federal appellate judges can enforce through abuse of discretion review. To ensure their consistent application across the board, such notification and consultation practices need to be codified through the addition of a new Federal Rule of Civil Procedure specifically governing transnational discovery requests.

Moreover, a system of notification and consultation requires accompanying amendments to § 1782. The statute is currently interpreted to permit discovery requests related to suits that are being contemplated and have yet to be filed.288 It should be amended to preclude prefiling discovery, because coordination is not possible when there is not yet a foreign tribunal presiding over the foreign dispute. There are two rationales for prefiling foreign discovery requests, neither of which is compelling. One is that prefiling discovery is necessary due to many foreign legal systems having higher pleading standards than the United States, and so prefiling discovery is needed to file the suit abroad in the first place. But a higher pleading standard does not justify uncoordinated prefiling discovery because the pleading standard may be a policy choice that reflects a desire to control discovery, as it does in the United States.289 The other rationale is that prefiling discovery is necessary to maximize the chances that US discovery will be compelled in time for it to be used in the foreign tribunal. The challenge of timing US discovery to suit the needs of a foreign proceeding is a reason for better coordination, not for broadening the reach of US discovery in hopes that some of it is useful. Finally, extraterritorial discovery under § 1782 should be explicitly precluded because no coordination is needed in that scenario. The foreign tribunal can order the discovery itself and a US court should refrain from doing so, ostensibly in the tribunal’s aid.

B. Seeking the Participation of Foreign Parties

The Supreme Court additionally held in Intel that parity is a policy concern that district courts should address in their exercise of discretion. Yet, most § 1782 requests target nonparties to the foreign proceeding and are considered ex parte, without the participation of the foreign opposing party. As noted in Part II, only 37 percent of party requests in 2015 were contested either by the foreign opposing party or the nonparty target, and uncontested requests were more likely to be granted than contested requests. Whether the low contestation rate was due to lack of notice versus failure to contest could not be determined from docket analysis.

Some courts no longer allow § 1782 requests to be considered ex parte, but notification requirements are currently erratic and need to be systematically applied.290 The foreign opposing party should always be notified and given an opportunity to intervene. The input of the foreign opposing party is needed to determine whether parity is lost by the grant of US discovery, to uphold universal notions of due process and fairness, and to ascertain information regarding the foreign proceeding that is needed to apply the FRCP. This information includes whether the requested information is relevant to any party’s claim or defense, the importance of the requested discovery to resolving the case, and the parties’ comparative resources and access to relevant information. If granting the § 1782 request will unfairly benefit one party at the expense of another, then the foreign tribunal’s involvement is also needed to enforce a reciprocal exchange of information.

Like the requirement to notify and consult the foreign tribunal, the requirement to notify and include the foreign opposing party can be achieved initially through changes in judicial practices. To ensure consistency across the many district courts that now receive § 1782 requests, this requirement should be codified in the new Federal Rule of Civil Procedure governing transnational discovery requests.

C. Improving Administrative Ease

In Part II, I noted that docket analysis provides incomplete information about when multiple discovery requests are made in the United States for the same foreign proceeding because each foreign discovery request operates as its own stand-alone action. A single foreign proceeding might generate many discovery requests in different district courts and before different federal judges who are not aware of the other requests. This splintered structure makes foreign discovery requests uniquely difficult to administer, including making it difficult to assess whether discovery requests are abusive, overly burdensome, or proportional to the demands of the case.

Take, for example, the Chevron litigation and arbitration in Ecuador concerning oil contamination in the Amazon. Chevron filed over twenty-three § 1782 requests in district courts across the United States,291 resulting in at least fifty federal court orders and opinions.292 The indigenous plaintiffs and the government of Ecuador filed many more. In one instance, the Fifth Circuit overturned the district court’s denial of a discovery request made by the government of Ecuador on the basis that the foreign arbitration for which the discovery was sought was not a tribunal covered by § 1782.293 Although the ruling was consistent with Fifth Circuit precedent, the appellate court relied on judicial estoppel to overturn the denial. Chevron had previously successfully argued before district courts in other circuits that the foreign arbitration was a tribunal covered by § 1782, and then taken advantage of a circuit split to argue the opposite in the Fifth Circuit. The Fifth Circuit became aware of these inconsistent positions because the Chevron foreign discovery requests were highly publicized and most were opposed. But § 1782 requests typically are not high profile and internally inconsistent results could easily escape notice.

I recommend that foreign discovery requests related to the same foreign proceeding be restructured to more closely resemble domestic discovery requests. The unit of analysis should not be a singular discovery request but rather a singular foreign proceeding. Like domestic discovery requests that are initially all directed to the same federal district court that oversees the plenary dispute, foreign discovery requests related to the same foreign plenary dispute should also, at least initially, all be directed to the same federal district court. If some discovery is needed from another district, the out-of-district request can be managed by a compliance court, as it is in the domestic context under FRCP 45.294 Having a central presiding judge is critical to seeing the overall effect of US discovery on the foreign proceeding, even if that judge is not presiding over the plenary dispute. Restructuring foreign discovery requests in this way also reduces the burden on foreign tribunals and foreign opposing parties who can participate in a centralized way rather than in separate actions across many district courts.

Restructuring will require § 1782 to be amended. A more achievable but less effective short-term solution would be to require § 1782 applicants to report all other § 1782 discovery requests linked to the same foreign proceeding so that they may be consolidated.

Conclusion

Despite their rapid growth in recent years, foreign discovery requests remain a small fraction of the overall federal docket.295 Yet, they provide a lens for understanding the broader shift toward disputes that straddle multiple legal systems and the challenges such disputes pose for courts and judges. At other procedural junctures, too, private actors are planning strategically across borders while federal judges are operating at an informational disadvantage and engaging in the futile task of abstractly divining foreign interests. For instance, scholars have criticized federal courts for problematically dismissing transnational cases in the name of comity through doctrines such as forum non conveniens and comity abstention296 even when foreign courts and governments have not welcomed such dismissals.297

As private actors become more savvy and transnational cases become more meaningfully connected to multiple fora, national judges can no longer operate in a silo. They need to seek out the participation of foreign parties because their decisions alter the procedural rights of private actors abroad. They need to seek out the participation of foreign tribunals because they share overlapping authority with foreign judges and arbitrators over certain cases. Ultimately, a reconceptualization of US judges as partners in global governance,298 engaged in the shared task of governing a common transnational litigation space, is needed.

Global governance is defined as “the collective management of common problems at the international level.”299 Transnational litigation and discovery are now common international problems, governed by many national and international courts and tribunals whose authorities are overlapping and nonhierarchical.300 When a foreign dispute leads to a § 1782 request, the foreign court adjudicating the case and the US federal court receiving the discovery request both have authority over the case, potentially leading to confusion and discord. Recognizing that federal judges are engaged in global governance reveals the forest from the trees: individually, these foreign discovery requests are routine and low profile; together, they give rise to a system of global governance marked by institutional conflict and chaos.

This Article serves as a case study of a set of institutional challenges that will only grow in the coming years, and an example of the type of institutional solutions that will be needed. It calls for scholarship on judicial information sharing and coordination across borders,301 as well as a deeper understanding of the diverse legal systems worldwide with which US courts will need to coordinate.

Appendix A: Data Collection Methodology

I relied on the Public Access to Court Electronic Records (PACER) system to compile a dataset of incoming foreign discovery requests for use in civil proceedings abroad.302 Other scholars have noted the sampling bias stemming from empirical efforts that rely on Westlaw or Lexis searches. Because Westlaw and Lexis contain more published than unpublished orders, have varying completeness across district courts, and lack clarity on the exact contents of their databases, searches on these services are likely to yield biased results that do not reflect the overall reality on the ground.303 These problems are amplified by different publication practices across district courts,304 and are particularly worrisome for studying low-profile routine matters such as § 1782 discovery orders, which tend to be unpublished.

PACER is a mandatory electronic docketing system within federal district courts that provides access to all filed actions nationwide. The process of moving to PACER’s electronic docketing system began in 1988, and was mostly completed by the mid-2000s.305 Since PACER has limited search functions and does not allow text searching,306 I searched PACER’s contents on Bloomberg Law. Bloomberg Law’s database includes docket coverage for all federal district courts since 1989.307 According to Bloomberg Law representatives, the service contains all of PACER’s content across that timespan, and continues to collect docket information and filed documents on a rolling basis. Bloomberg’s Advanced Dockets Search allows text searches of all docket reports as well as any underlying filed documents that have been rendered text searchable.308

To compile as exhaustive as possible a dataset of incoming foreign civil discovery requests, I took four steps. First, I crafted and tested a variety of text search parameters for identifying foreign civil discovery requests. Since all foreign civil discovery requests and some foreign criminal discovery requests are ultimately executed under the authority of § 1782, the task was to create a dataset of discovery requests brought under § 1782 during that time period, and then to exclude the criminal discovery requests. I tested different text searches on Bloomberg Law by cross-checking results for individual calendar years with cases identified by Westlaw as citing § 1782 for that same time period. When Westlaw only contained an appellate case, I looked for the corresponding district court case in the Bloomberg Law search results. I chose to use Westlaw over Lexis for this step because Westlaw is more inclusive.309 I adjusted the search terms until the only cases that were missing could not be captured without also pulling in domestic discovery disputes. In other words, I selected the text search that maximized sensitivity without significant sacrifices in specificity. The selected text search was:

“letter rogatory” OR “letters rogatory” OR “judicial assistance” OR “discovery in aid of international” OR “discovery in aid of foreign” OR (28 /5 1782) OR (use /s “foreign proceeding”) OR (use /s “foreign proceedings”) OR (aid /s “foreign proceeding”) OR (aid /s “foreign proceedings”)

Second, on May 31, 2018, I ran the above text search on Bloomberg Law on cases filed in all ninety-four federal district courts between January 1, 2005 and December 31, 2017. As discussed in Part II, these dates were selected to capture how courts have treated foreign civil discovery requests during the years since the Supreme Court’s decision in Intel. This initial search produced over ten thousand results.

Third, since these results included many false positives that were not § 1782 requests, they were manually culled, a process that eliminated approximately two-thirds of the results, leaving over three thousand true positives. All available information was used to determine whether the case was a true § 1782 request, including if it was labelled as such in the title or in the “cause” field of the docket report. I did not include as true positives cases that were unambiguously erroneous uses of § 1782.310 Because the search terms I used included “letter rogatory” and “judicial assistance,” many of the initial results were outgoing discovery requests, seeking evidence from abroad for controversies being adjudicated in the United States. This manual culling was completed with the assistance of a team of research assistants. I spot-checked the results for accuracy.

Fourth, to ensure that the dataset is close to exhaustive and unbiased, I cross-checked with cases identified by Westlaw as citing § 1782 for different calendar years than had been used to construct the text search. The results confirmed my expectations: I missed less than 1 percent of cases identified by Westlaw, and I could not have captured them without expanding the search terms such that I would have lost a significant degree of specificity in the overall result. There does not appear to be a pattern in the types of § 1782 requests that are missing. They are missing because none of the search terms were mentioned in the docket report or in the text searchable underlying documents. I could not detect any pattern in either the text labels entered into docket reports or whether underlying filed documents were text searchable. I also confirmed that the dataset is close to complete and unbiased by internally checking those cases that were refiled under a different case number or consolidated.311 Anytime I came across one of these cases, I checked whether the refiled or consolidated case was also in the dataset. Finally, the dataset’s estimated number of tribunal requests coming from OIJA is in line with the number of requests that OIJA itself estimates they are sending to district courts.312

The dataset as a whole likely misses a few cases on the margins, but is close to exhaustive and unlikely to be biased. However, a cautionary note on the 2017 cases is in order: while Bloomberg Law representatives insist that the service regularly updates dockets and pulls in new documents as they are filed, that is not always accurate in my experience and in the experience of research assistants working on this project. Sometimes, it is necessary to manually click an “Update Docket” button to fetch the most recent documents. The result is that the text search was likely run on some docket reports for 2017 that were not fully up-to-date and may undercount the number of § 1782 requests in that year. This problem did not seem to affect earlier years, likely because those dockets had, over time, become updated either due to Bloomberg Law clients manually updating those cases or by the system’s automatic update mechanism. Additionally, some of the 2017 cases may not yet have reached completion when the text search was run on May 31, 2018, again leading to the search being run on incomplete docket reports and possibly undercounting cases in 2017.

Appendix B: Data Coding Methodology

Of the over three thousand true positive § 1782 requests in the dataset, approximately one-third (over one thousand requests) were randomly selected for further coding for the following information. This coding was completed with the help of a team of research assistants. We met weekly to discuss difficult coding decisions and I spot-checked the results for accuracy. Some cases were sealed overall or had underlying documents that were sealed or otherwise unavailable. All available information was used to make determinations, or data was recorded as missing.313

1. Civil or criminal: Some foreign criminal discovery requests were previously executed under § 1782. Criminal discovery requests were coded as such and not examined further.
2. District court: We coded for the district court to which civil discovery requests were made.
3. Requestor: We coded for who brought the request, whether it was a tribunal, party, or interested person. If the request was brought by a tribunal, we tracked whether the request came directly to the district court, through the Department of Justice’s OIJA, or through a party. In some cases, requests were brought by multiple entities.
4. Target: We coded for whether the request targeted a party, a nonparty, or both.
5. Foreign tribunal type: We coded for the type of foreign tribunal before which the requested discovery was to be used—whether it was a foreign court, an international court, a regulatory agency, a commercial arbitral tribunal, or an investor-state arbitral tribunal. In some cases, requests indicated that the evidence would be used in multiple tribunals.
6. Numerosity of foreign proceeding: We coded for whether the evidence was requested for one foreign proceeding or multiple foreign proceedings.
7. Timing of foreign proceeding: We coded for the timing of the foreign proceeding for which the evidence was requested—whether the foreign proceeding was pending, contemplated, or mixed (if the evidence was requested for multiple proceedings).
8. Country of foreign proceeding: We coded for the country or countries of the foreign proceeding(s) for which the evidence was requested. I then also categorized the country in four ways to look for patterns: by region, by Hague Evidence Convention status, by legal system type, and by rule-of-law score. The categorizations are explained below.
   1. Regions—The countries appearing in the dataset were categorized by the following regions:
      1. Americas—Argentina, Bolivia, Brazil, Canada, Chile, Colombia, Costa Rica, Ecuador, Guatemala, Honduras, Mexico, Panama, Paraguay, Peru, Uruguay, Venezuela
      2. Caribbean—Antigua, Bahamas, Barbados, Bermuda, Cayman Islands, Curacao, Dominican Republic, Jamaica, British Virgin Islands, St. Vincent and the Grenadines
      3. Western Europe—Andorra, Austria, Belgium, Czech Republic, Denmark, European Union, France, Germany, Greece, Guernsey, Isle of Man, Italy, Jersey, Luxembourg, Monaco, Netherlands, Northern Ireland, Norway, Portugal, Spain, Slovenia, Sweden, Switzerland, United Kingdom
      4. Eastern Europe—Croatia, Estonia, Hungary, Moldova, Poland, Romania, Serbia, Slovakia, Russia
      5. Middle East—Bahrain, Cyprus, Dubai, Egypt, Iran, Israel, Kuwait, Morocco, Oman, Saudi Arabia, Turkey, United Arab Emirates
      6. Asia—American Samoa, Australia, China, Hong Kong, India, Japan, Kazakhstan, Laos, Pakistan, Singapore, South Korea, Taiwan
      7. Africa—Ethiopia, Ghana, Liberia, South Africa
   2. Hague Evidence Convention status—The countries appearing in the dataset were categorized by whether the Hague Evidence Convention was in force between that country and the United States in the calendar year when the case was filed. This may not be the same as whether the country is a member of the Hague Evidence Convention. For example, although Brazil acceded to the Hague Evidence Convention, the United States has not accepted its accession, so the Convention is not in force between Brazil and the United States. Below is a list of countries for which the Hague Evidence Convention was in force with respect to the United States, including specific calendar years if that statement only holds true for a segment of the study period. If the Convention came into force between a country and the United States midway through a calendar year, I counted it as being in force starting in the following calendar year. Argentina, Australia, Barbados, China, Colombia (since 2013), Croatia (2010), Cyprus, Czech Republic, Denmark, Estonia, France, Germany, Greece, Hong Kong, Hungary (since 2007), India (since 2007), Israel, Italy, South Korea (since 2010), Kuwait (since 2009), Luxembourg, Mexico, Monaco, Morocco (since 2012), Netherlands, Norway, Poland, Portugal, Romania, Serbia (since 2012), Singapore, Slovakia, South Africa, Spain, Sweden, Switzerland, Turkey, United Kingdom, Venezuela.
   3. Legal system type—The countries appearing in the dataset were categorized by legal system type as follows:
      1. Common law: Antigua, Australia, Bahamas, Barbados, Bermuda, British Virgin Islands, Canada, Cayman Islands, Hong Kong, Ireland, Isle of Man, Jamaica, St. Vincent and the Grenadines, United Kingdom
      2. Civil law: Argentina, Austria, Bolivia, Brazil, Chile, Colombia, Croatia, Czech Republic, Denmark, Dominican Republic, Ecuador, Estonia, France, Germany, Greece, Guatemala, Hungary, Italy, Luxembourg, Mexico, Moldova, Monaco, Netherlands, Norway, Panama, Paraguay, Peru, Poland, Portugal, Romania, Russia, Serbia, Slovakia, Spain, Sweden, Switzerland, Taiwan, Turkey, Uruguay, Venezuela
      3. Mixed/other: American Samoa, Bahrain, China, Cyprus, Egypt, Ethiopia, Ghana, Guernsey, India, Iran, Israel, Japan, Jersey, Kuwait, Liberia, Morocco, Oman, Pakistan, Singapore, South Africa, South Korea, United Arab Emirates
      4. Rule-of-law score—The countries appearing in the dataset in 2015–2017 were categorized by rule-of-law score quartile. To derive the quartiles, I used the World Justice Project’s Rule of Law Index,314 and only looked at scores from 2015 to 2017 because the scoring instrument varied with each report during earlier years, whereas the scores are more comparable during this three-year period.315 The index evaluates eight factors: constraints on government powers, absence of corruption, open government, fundamental rights, order and security, regulatory enforcement, civil justice, and criminal justice. Although each country is given a score as well as sub-scores for each factor, I only relied on the overall composite score. The following is a list of the countries in each quartile, with years if the country was not in that quartile for the full three-year period. In any given year, countries in the fourth quartile have the highest rule-of-law scores, while those in the first quartile have the lowest scores. Some countries in the dataset were not scored by the World Justice Project, and so appear in the “no score” category.
         1. 4th Quartile: Australia, Austria, Barbados (2016), Belgium, Canada, Chile, Costa Rica, Czech Republic, Denmark, Estonia, Finland, France, Germany, Hong Kong, Japan, Netherlands, New Zealand, Northern Ireland (2016, 2017), Norway, Poland, Portugal, South Korea, Singapore, Slovenia (2016, 2017), Spain, St. Kitts and Nevis (2017), Sweden, United Kingdom, United States, Uruguay
         2. 3rd Quartile: Antigua (2016, 2017), Argentina (2016, 2017), Bahamas (2016, 2017), Barbados (2017), Belarus (2015), Bosnia and Herzegovina, Botswana, Brazil, Bulgaria, Croatia, Dominica (2016, 2017), Georgia, Ghana, Greece, Grenada (2016, 2017), Hungary, Italy, Jamaica, Jordan, Macedonia (2015, 2016), Malaysia, Mongolia, Nepal (2015), Panama (2015), Philippines (2015), Romania, Senegal, Slovenia (2015), South Africa, St. Kitts and Nevis (2016), St. Lucia (2016, 2017), St. Vincent and the Grenadines (2016, 2017), Trinidad and Tobago (2016, 2017), Tunisia (2015, 2017), United Arab Emirates
         3. 2nd Quartile: Albania, Argentina (2015), Belarus (2016, 2017), Belize, Burkina Faso (2016, 2017), China, Colombia, Cote D’Ivoire (2015, 2017), Dominican Republic (2015), El Salvador, Guyana (2016, 2017), India, Indonesia, Iran (2017), Kazakhstan, Kyrgyzstan, Lebanon (2015), Macedonia (2017), Malawi, Moldova, Morocco, Nepal (2016, 2017), Panama (2016, 2017), Peru, Russia (2015), Philippines (2016), Serbia, Sri Lanka, Suriname (2016, 2017), Tanzania (2015, 2016), Thailand, Tunisia (2016), Ukraine, Vietnam, Zambia
         4. 1st Quartile: Afghanistan, Bangladesh, Bolivia, Burkina Faso (2015), Cambodia, Cameroon, Cote D’Ivoire (2016), Dominican Republic (2016, 2017), Ecuador, Egypt, Ethiopia, Guatemala, Honduras, Iran (2015, 2016), Kenya, Lebanon (2016, 2017), Liberia, Madagascar, Mexico, Myanmar, Nicaragua, Nigeria, Pakistan, Philippines (2017), Russia (2016, 2017), Sierra Leone, Tanzania (2017), Turkey, Uganda, Uzbekistan, Venezuela, Zimbabwe
         5. No score: Andorra, American Samoa, British Virgin Islands, Cyprus, European Union, Isle of Man, Israel, Jersey, Kuwait, Monaco, Switzerland, Taiwan
9. Substance area of foreign proceeding: We coded for the substantive area in dispute in the foreign proceeding—whether it was antitrust, bankruptcy, contract (generic contract that does not fit into another category), corporate (disputes regarding corporate structure or breach of an officer’s fiduciary duty), environmental, employment, family (disputes regarding divorce, child custody, or inheritance), fraud (generic fraud that does not fit into another category), maritime, intellectual property or trade secret, products liability, securities, tax, tort (generic tort that does not fit into another category), and other (any requests not covered by the above).
10. Number of docket lines: We coded for the number of docket lines it took to resolve a request. This is a very rough gauge of case complexity. Where a request was still pending, we did not track this data.
11. Number of orders: We coded for the number of orders it took to resolve a request. This is another very rough gauge of case complexity. Where a request was still pending, we did not track this data.
12. Outcome: We coded for the outcome of the request—whether it was granted to some extent (including those that were granted in part or subsequently quashed in part), denied altogether, or reached no resolution. Where an initial result was later altered, we looked at the final outcome. A request reached no resolution if it was withdrawn or otherwise terminated before a decision was made. Where a request was still pending, we did not track the outcome.
13. Contestation: For 2015 cases only, we coded for whether the request was challenged. A request may be challenged by the target of discovery, whether party or nonparty to the foreign proceeding, or it may be challenged by the nontarget opposing party. It may be challenged at the outset, or after a request is initially granted ex parte. A challenge may take the form of a motion to quash, motion to vacate, motion to stay, motion for reconsideration, or a motion to compel from the requestor that is opposed. I did not count entries of stipulated protective orders or other requests for confidentiality agreements as contestation.

Appendix C: Supplementary Tables and Figures316

Table 5: Foreign Discovery Requests Executed Under § 1782, Data Summary, 2005–2017

Year	Total number of cases	Cases sampled317
2005	290	102
2006	210	62
2007	275	97
2008	267	91
2009	218	70
2010	197	59
2011	212	72
2012	177	73
2013	233	82
2014	293	94
2015	237	98
2016	312	111
2017	239	81
Total	3160	1092

Table 6: Estimated Number of Foreign Discovery Requests Executed Under § 1782318

Year	Civil Requests319	Criminal Requests	Refiled Requests
Number 95% CI	Percentage 95% CI	Number 95% CI	Percentage 95% CI	Number 95% CI	Percentage 95% CI
2005	48.9 34–72	17% 12%–25%	238 218–256	82% 75%–88%	3.3 1–15	1.1% 0.34%–5.2%
2006	63.4 45–89	30% 21%–42%	131 108–155	62% 51%–74%	15.8 8–35	7.5% 3.8%–17%
2007	86.1 66–112	31% 24%–41%	179 156–203	65% 57%–74%	9.9 5–25	3.6% 1.8%–9.1%
2008	108 86–136	41% 32%–51%	144 120–171	54% 45%–64%	14.4 7–32	5.4% 2.6%–12%
2009	127 105–150	58% 48%–69%	87.9 68–113	40% 31%–52%	3.5 1–17	1.6% 0.46%–7.8%
2010	135 115–155	69% 58%–79%	51.1 35–74	26% 18%–38%	10.9 5–28	5.6% 2.5%–14%
2011	171 154–186	81% 73%–88%	34.8 23–54	16% 11%–25%	6.3 3–19	3% 1.4%–9%
2012	151 139–162	86% 79%–92%	23.1 15–38	13% 8.5%–21%	2.6 1–12	1.4% 0.56%–6.8%
2013	204 190–216	88% 82%–93%	17.5 10–33	7.5% 4.3%–14%	11.6 9–26	5% 2.6%–11%
2014	247 228–264	84% 78%–90%	29.6 18–50	10% 6.1%–17%	16.5 9–34	5.6% 3.1%–12%
2015	219 208–228	92% 88%–96%	7.7 4–19	3.3% 1.7%–8%	10.3 6–22	4.3% 2.5%–9.3%
2016	240 219–260	77% 70%–83%	33 22–52	11% 7.1%–17%	39 27–59	12% 8.7%–19%
2017	208 193–222	87% 81%–93%	3.1 1–14	1.3% 0.42%–5.9%	27.6 17–46	12% 7.1%–19%
Total	2070 1993–2147	65% 63%–68%	919 848–996	29% 27%–32%	172 139–214	5.4% 4.4%–6.8%

Table 7: Estimated Number of Civil Requests by Most Common Receiving Court320

Court	Number 95% CI	Percentage 95% CI
New York Southern	272 218–342	23% 18%–28%
Florida Southern	124 91–173	10% 7.6%–14%
California Northern	94.1 67–137	7.8% 5.6%–11%
California Central	84 58–124	7% 4.8%–10%
Florida Middle	47.1 30–78	3.9% 2.5%–6.5%

Table 8: Estimated Number of Civil Requests by Requestor314, 321

Requestor	Number 95% CI	Percentage 95% CI
Tribunal	912 817–1016	44% 39%–49%
Direct	22.4 13–44	1.1% 0.63%–2.1%
Via DOJ	832 741–933	40% 36%–45%
Via Party	57.6 39–88	2.8% 1.9%–4.3%
Party	1142 1038–1255	55% 50%–61%
Interested Person	16 8–35	0.77% 0.39%–1.7%

Table 9: Estimated Number of Civil Requests by Target322

Target	Requests from Foreign Tribunals	Requests from Foreign Parties
Number 95% CI	Percentage 95% CI	Number 95% CI	Percentage 95% CI
Party	132 97–183	15% 11%–20%	101 72–134	8.4% 6%–12%
Nonparty	774 684–874	85% 75%–96%	1059 959–1166	88% 80%–97%
Both	3.3 1–17	0.36% 0.11%–1.9%	40.5 25–70	3.4% 2.1%–5.8%

Table 10: Estimated Number of Civil Requests by Type of Foreign Tribunal322, 323

Foreign Tribunal Type	Requests from Foreign Tribunals	Requests from Foreign Parties
Number 95% CI	Percentage 95% CI	Number 95% CI	Percentage 95% CI
Foreign Court	897 816–982	99% 90%–100%	1078 979–1184	90% 82%–97%
Commercial Arbitration	3.2 1–17	0.35% 0.11%–1.9%	119 87–167	9.9% 7.2%–14%
Regulatory Agency	9.6 4–27	1.1% 0.44%–3%	47.6 30–79	4% 2.5%–6.6%
Investor-state Arbitration	0 0–11	0% 0%–1.2%	30.6 18–57	2.5% 1.5%–4.7%
International Court	0 0–11	0% 0%–1.2%	3.4 1–18	0.28% 0.083%–1.5%

Table 11: Estimated Number of Civil Requests by Nature of Foreign Proceeding324

Nature of Foreign Proceeding	Requests from Foreign Tribunals	Requests from Foreign Parties
Number 95% CI	Percentage 95% CI	Number 95% CI	Percentage 95% CI
Timing of Foreign Proceeding
Pending	907 830–987	100% 91%–100%	1009 909–1117	84% 76%–93%
Contemplated	0 0–11	0% 0%–1.2%	121 89–169	10% 7.4%–14%
Mixed	3.2 1–17	0.35% 0.11%–1.9%	70.6 48–108	5.9% 4%–9%
Number of Foreign Proceedings for Which Was Discovery Requested
One	907 830–987	100% 91%–100%	864 766–973	72% 64%–81%
Multiple	3.2 1–17	0.35% 0.11%–1.9%	336 275–413	28% 23%–34%

Table 12: Estimated Number of Civil Requests by Most Common Area of Foreign Dispute324, 325

Requests from Foreign Tribunals	Requests from Foreign Parties
Area	Number 95% CI	Percentage 95% CI	Area	Number 95% CI	Percentage 95% CI
Family	478 397–574	52% 44%–63%	Contract	327 266–404	27% 22%–34%
Contract	140 101–197	15% 11%–22%	Patent/IP	230 181–296	19% 15%–25%
Employment	106 74–156	12% 8.1%–17%	Corporate	148 111–201	12% 9.2%–17%

Table 13: Estimated Complexity of Civil Requests326

Year	Requests from Foreign Tribunals	Requests from Foreign Parties
Median Docket Lines 95% CI	Median Orders 95% CI	Median Docket Lines 95% CI	Median Orders 95% CI
2005	5 3–32	1 1–9	6 2–40	1 1–10
2006	3 3–18	2 1–4	11 4–40.5	3 1–8
2007	3 3–3	1 1–1	4 3–33	1 1–7
2008	4 2.5–5.5	1 1–2	20 9.5–39.5	4 2.5–8.5
2009	4 3–5	1 1–1.5	17.5 7–22	4 2–7
2010	3.5 2.5–5	1 1–2	21 9–53	4.5 2.5–12
2011	3 2–3	1 1–1	9 6–19	2 2–4
2012	3 3–4	1 1–1.5	11 8–21	4 2–5
2013	3 3–4.5	1 1–1	11 8–19	3 2–3
2014	3 3–4	1 1–1	11 7–18	4 2–5
2015	4 3–4	1 1–1	9 6–15	2 2–4
2016	3 3–4	1 1–1	13 8–21	4 3–5
2017	4 3–4	1 1–2	11 7–20	3 2–4
Overall	3 3–4	1 1–1	11 9.5–13	3 3–4

Table 14: Outcome for Sampled Civil Requests327

Year	Requests from Foreign Tribunals	Requests from Foreign Parties
Granted No. (%)	Denied No. (%)	No Res. No. (%)	Granted No. (%)	Denied No. (%)	No Res. No. (%)
2005	9 (100%)	0 (0%)	0 (0%)	5 (100%)	0 (0%)	0 (0%)
2006	4 (80%)	1 (20%)	0 (0%)	8 (80%)	2 (20%)	0 (0%)
2007	18 (95%)	0 (0%)	1 (5%)	6 (86%)	0 (0%)	1 (14%)
2008	11 (92%)	0 (0%)	1 (8%)	10 (56%)	3 (17%)	5 (28%)
2009	19 (86%)	1 (5%)	2 (9%)	13 (93%)	1 (7%)	0 (0%)
2010	11 (92%)	0 (0%)	1 (8%)	16 (64%)	4 (16%)	4 (20%)
2011	14 (82%)	1 (6%)	2 (12%)	26 (72%)	5 (14%)	5 (14%)
2012	20 (91%)	1 (5%)	1 (5%)	25 (74%)	3 (9%)	6 (18%)
2013	25 (96%)	0 (0%)	1 (4%)	30 (73%)	4 (10%)	7 (17%)
2014	33 (97%)	0 (0%)	1 (3%)	28 (72%)	6 (15%)	5 (13%)
2015	35 (90%)	0 (0%)	4 (10%)	37 (80%)	2 (4%)	7 (15%)
2016	38 (93%)	1 (2%)	2 (5%)	26 (67%)	9 (23%)	4 (10%)
2017	25 (100%)	0 (0%)	0 (0%)	34 (83%)	2 (5%)	5 (12%)
Overall	262 (93%)	5 (2%)	16 (6%)	264 (75%)	41 (12%)	49 (14%)

Table 15: Estimated Grant Rates for Civil Requests328

Year	Grant Rate	95% CI
2005	92.9%	69.6%—97.8%
2006	81.3%	57.1%—95.2%
2007	100%	87.5%—100%
2008	87%	68.7%—96.4%
2009	93.9%	81.9%—98.3%
2010	87.5%	73.5%—95.7%
2011	87%	75.7%—93.9%
2012	91.8%	82.9%—96.9%
2013	93.5%	86.2%—97.8%
2014	91.2%	83.2%—96%
2015	97.3%	92.1%—99%
2016	86.5%	78.4%—92.3%
2017	96.8%	90.1%—99%
Tribunal Requests 2005–2017	98.1%	96.1%—99.3%
Party Requests 2005–2017	86.6%	82.9%—89.6%
All Requests 2005–2017	91.9%	89.7%—93.6%

Figure 4: Shannon’s Entropy, Calculated on Mixture of Countries for Civil Requests329

• 1In re Ex Parte Application of Qualcomm Inc, 162 F Supp 3d 1029, 1032 (ND Cal 2016).
• 2Transcript of Miscellaneous Hearing, Request for International Judicial Assistance from the National Court of First Instance in Labor Matters No 37 of Buenos Aires, Argentina, No 3:12-cv-00662, *3 (D Nev filed Apr 13, 2013).
• 3By foreign civil proceedings, I mean the full range of foreign civil dispute resolution proceedings for which discovery is now being requested in US federal courts, including those before foreign and international courts, foreign regulatory agencies, as well as commercial and investor-state arbitral tribunals.
• 4See Part II.
• 5Id. By foreign parties, I mean any parties to foreign civil proceedings.
• 6See, for example, Memorandum of Law in Support of the Republic of Ecuador and Dr. Diego García Carrión’s Application for an Order Under 28 U.S.C. § 1782(a) to Issue a Subpoena to John A. Connor for the Taking of a Deposition and the Production of Documents for Use in a Foreign Proceeding, In re Republic of Ecuador, No 4:11-mc-00516, *2 (SD Tex filed Nov 28, 2011).
• 7See, for example, Order Granting in Part and Denying in Part Motion to Compel, Kwong v Battery Tai-Shing Corp, No C 08-80142, *1–2 (ND Cal filed Feb 5, 2009).
• 8See, for example, Memorandum of Law in Support of Application for Order, Request for International Judicial Assistance from the 24th Civil Court of Santiago, Chile in Juan Carols Said Kattan v Antonio Miguel Orlandini Said, No 1:17-mc-22657, *1 (SD Fla filed July 17, 2017).
• 9See, for example, Order, In re Request for International Judicial Assistance from The First Instance Court, Calarasi, Romania; Matter of Ionela Camelia Buzduga v Constantin Viorel Buzduga, No 4:16-mc-01439, *3–4 (SD Tex filed July 4, 2016).
• 10See, for example, Ex Parte Application, Without a Letter Rogatory Pursuant to Title 28 U.S.C § 1782 for Discovery Ordered by Islamic Republic of Iran Kish Island, Department 102, Declaration of Counsel, Declaration of Defendant & Proposed Order, Borhani v Ahmadi, No 8:16-mc-00021, *1–2 (CD Cal filed Sept 19, 2016).
• 11See, for example, Ex Parte Order, In re Letter of Request for International Judicial Assistance from the Seoul Central District Court, In the Matter of Subway International B.V. v Kyung Hee Hwang, No 1:13-mc-00059, *1 (SDNY filed Feb 22, 2013).
• 12Scholars writing on the subject have focused almost exclusively on high-profile decisions and circuit splits. See notes 38–43 and accompanying text. There is one limited systematic study of published decisions granting discovery for use in connection with foreign commercial arbitration, but it only examines a small number of unpublished decisions. See Kevin E. Davis, Helen Hershkoff, and Nathan Yaffe, Private Preference, Public Process: U.S. Discovery in Aid of Foreign and International Arbitration *12–14 & nn 75–77 (NYU School of Law Public Law & Legal Theory Working Paper No 15-51, Oct 30, 2015) (identifying twenty-two published opinions of this type).
• 13See, for example, Duane Loft and Joshua Libling, Discovery in Aid of International Proceedings: Recent Developments in Section 1782 (Boies Schiller Flexner, Oct 20, 2015), archived at https://perma.cc/QU8U-M9BL; H. Christopher Boehning, Damiën F. Berkhout, and Peter Hering, The Consorcio Decision and the Need to Account for Discovery Outside the Arbitration Procedure (International Bar Association Legal Practice Division, Feb 2013), archived at https://perma.cc/2W24-EXE3; Claudia T. Salomon, David McLean, Samuel B.C. de Villiers, and Eric M. Broad, US Court of Appeals Revises Opinion: Section 1782’s Use in Arbitration Ambiguous (Latham & Watkins, Feb 3, 2014), archived at https://perma.cc/B5LL-FAL4.
• 14Robert W. Gaffey and Bridget A. Crawford, Developments in U.S. Law Regarding a More Liberal Approach to Discovery Requests Made by Foreign Litigants Under 28 U.S.C. § 1782 (Jones Day, Apr 2009), archived at https://perma.cc/WJL9-Y2CF.
• 15Lyons Flood, Discovery In Aid of Foreign Proceedings, archived at https://perma.cc/767Z-BFCT.
• 16Christopher J. Houpt and Mark G. Hanchet, Section 1782 Discovery: A Back Door for Foreign Litigants (Mayer Brown, Mar 2012), archived at https://perma.cc/AL66-P94S.
• 17Harout J. Samra, US Courts Affirm Expansive Discovery Under 28 U.S.C. § 1782 (DLA Piper, Sept 29, 2015), archived at https://perma.cc/9C6B-PBWF.
• 18As explained below, “export” here occurs on a case-by-case basis, and with the participation of foreign courts or foreign private actors.
• 19See Robert A. Kagan, Adversarial Legalism: The American Way of Law 3 (Harvard 2001) (“Compared to other economically advanced democracies, American civic life is more deeply pervaded by legal conflict and by controversy about legal processes.”). See also generally Amalia D. Kessler, Inventing American Exceptionalism: The Origins of American Adversarial Legal Culture, 1800–1877 (Yale 2017) (recounting the nineteenth-century origins of the idea of American legal exceptionalism); Richard L. Marcus, Putting American Procedural Exceptionalism into a Globalized Context, 53 Am J Comp L 709 (2005).
• 20Heraeus Kulzer, GmbH v Biomet, Inc, 633 F3d 591, 594 (7th Cir 2011).
• 21See Part III.A.
• 22See Mirjan R. Damaska, The Faces of Justice and State Authority: A Comparative Approach to the Legal Process 71–96 (Yale 1986) (describing two orientations of government: the “activist” state, which pursues its vision of the good life through policy implementation and tends toward legal institutions that are hierarchical judicial bureaucracies; and the “reactive” state, which provides a framework for its citizens to pursue their own goals and tends toward legal institutions that are more like that of the United States). See also Kagan, Adversarial Legalism at 3 (cited in note 19) (describing legal process in the United States as “adversarial legalism,” meaning “policymaking, policy implementation, and dispute resolution by means of lawyer-dominated litigation”); Diego A. Zambrano, Discovery as Regulation, 119 Mich L Rev *25–27, 32–34 (forthcoming 2020), archived at https://perma.cc/W8DS-7B7D (arguing that American discovery serves regulatory goals by relying on private litigants to enforce important statutes).
• 23See, for example, James H. Carter, Existing Rules and Procedures, 13 Intl Law 5, 5 (1979) (noting that the “virtually boundless sweep of the pre-trial procedures presently permitted by many American courts is so completely alien to the procedure in most other jurisdictions that an attitude of suspicion and hostility is created”).
• 24See Memorandum of Law in Support of Bracha Foundation’s Application for an Order Granting Discovery Pursuant to 28 U.S.C. § 1782, In re Application of Bracha Foundation Request for Discovery Pursuant to 28 USC § 1782, No 2:15-mc-00748, *1–2 & n 1 (ND Ala filed May 1, 2015) (Bracha Foundation Discovery Application Memo).
• 25Act of Oct 3, 1964, Pub L No 88-619, 78 Stat 995 (1964), codified at 28 USC § 1782.
• 26See Bracha Foundation Discovery Application Memo at *1–2 n 1 (cited in note 24) (explaining that the private actor commenced a related action in the British Virgin Islands and filed similar discovery applications in the United States District Courts for the Southern District of New York, the Southern District of Florida, and the Northern District of Ohio).
• 27See In re Application of Bracha Foundation Request for Discovery Pursuant to 28 USC § 1782, 2015 WL 6123204, *1 (ND Ala) (explaining that “the use of ex parte § 1782 applications is widespread and that granting them is not improper based on their ex parte nature”).
• 28Brief of Amicus Curiae the Commission of the European Communities Supporting Reversal, Intel Corp v Advanced Micro Devices, Inc, No 02‑572, *1–6 (US filed Dec 23, 2003) (available on Westlaw at 2003 WL 23138389) (European Commission Brief).
• 29European Commission, 37.990 Intel *12–19 (2009), archived at https://perma.cc/9MWQ-XTE8.
• 30AMD’s Application for Order Directing Intel to Produce Documents Pursuant to 28 U.S.C. § 1782 for Use in European Commission, Case No. Comp/C3-37.990 – AMD/Intel, Advanced Micro Devices, Inc v Intel Corp, No 01-7033, *1–3 (ND Cal filed Oct 1, 2001) (available on Westlaw at 2001 WL 35995385).
• 31Transcript of Oral Argument, Intel Corp v Advanced Micro Devices, Inc, No 02-572, *18, 21–22 (US filed Apr 20, 2004) (available on Westlaw at 2004 WL 954709).
• 32Id at *21.
• 33Hans Smit, International Litigation Under the United States Code, 65 Colum L Rev 1015, 1018 (1965). Hans Smit, professor at Columbia Law School, served as the reporter to the Commission on International Rules of Judicial Procedure that drafted the current version of § 1782. See id at 1015.
• 34Judicial Procedures in Litigation with International Aspects, S Rep No 88-1580, 88th Cong, 2d Sess § 9 at 7 (1964), reprinted in 1964 USCCAN 3782, 3788.
• 35See Intel Corp v Advanced Micro Devices, Inc, 542 US 241, 261 (2004) (noting that comity and parity are “important [ ] touchstones for a district court’s exercise of discretion in particular cases”).
• 36These factors include whether the foreign court needs and is receptive to US discovery, and whether the request “conceals an attempt to circumvent foreign proof-gathering restrictions.” Id at 264–65.
• 37See Advanced Micro Devices, Inc v Intel Corp, 2004 WL 2282320, *1 (ND Cal).
• 38Molly Warner Lien, The Cooperative and Integrative Models of International Judicial Comity: Two Illustrations Using Transnational Discovery and Breard Scenarios, 50 Cath U L Rev 591, 624 (2001).
• 39Richard L. Marcus, Retooling American Discovery for the Twenty-First Century: Toward a New World Order?, 7 Tulane J Intl & Comp L 153, 193 (1999).
• 40See Brian Eric Bomstein and Julie M. Levitt, Comment, Much Ado About 1782: A Look at Recent Problems with Discovery in the United States for Use in Foreign Litigation Under 28 U.S.C. § 1782, 20 U Miami Int-Am L Rev 429, 431–32 (1989) (explaining that the “nearly unbridled discretion” exercised by district courts under 28 USC § 1782 in deciding whether to aid foreign fact-finding “is the root of the problem[ ]”); Marat A. Massen, Discovery for Foreign Proceedings After Intel v. Advanced Micro Devices: A Critical Analysis of 28 U.S.C. § 1782 Jurisprudence, 83 S Cal L Rev 875, 877 (2010) (noting that most academic criticism of 28 USC § 1782 and the Supreme Court’s interpretation of that statute has focused on granting “judges too much discretion for evaluating” foreign discovery requests).
• 41Proposals include a default rule that requests be denied when they seek discovery for certain types of foreign proceedings, most notably commercial arbitrations, and that requests be granted only when the sought-after discovery would be discoverable in the foreign proceeding in which it is to be used. Courts and academics are split on whether § 1782 permits the export of discovery to commercial arbitral tribunals, and the latter proposal was rejected as a bright-line rule by the Supreme Court. Compare In re: Application to Obtain Discovery for Use in Foreign Proceedings, 939 F3d 710, 723 (6th Cir 2019) (holding that private commercial arbitral tribunals are covered by § 1782); Servotronics, Inc v Boeing Co, 954 F3d 209, 214–16 (4th Cir 2020) (same), with Republic of Kazakhstan v Biedermann International, 168 F3d 880, 881–83 (5th Cir 1999) (holding that private commercial arbitral tribunals are not covered by § 1782); National Broadcasting Company, Inc v Bear Stearns & Co, 165 F3d 184, 191 (2d Cir 1999) (same). See also, for example, New York City Bar Association Committee on International Commercial Disputes, 28 U.S.C. § 1782 as a Means of Obtaining Discovery in Aid of International Commercial Arbitration—Applicability and Best Practices 19–25 (2008), archived at https://perma.cc/FUP4-K2JD (describing the circuit split).
• 42See David J. Gerber, Obscured Visions: Policy, Power, and Discretion in Transnational Discovery, 23 Vand J Transnatl L 993, 1007 (1991) (noting that the excessive discretion exercised by judges in responding to foreign discovery requests inappropriately “shifts the available use of United States power from the executive and legislative branches to the judiciary”).
• 43See Stephen B. Burbank, The Reluctant Partner: Making Procedural Law for International Civil Litigation, 57 L & Contemp Probs 103, 135–39 (1994) (criticizing the United States’ historical preference for a unilateral approach that is “hard to justify, even in purely practical terms, in a world that is increasingly interdependent and in which our economic strength is waning”).
• 44The United States has ratified the Hague Evidence Convention. See Hague Conference on Private International Law, HCCH Members, archived at https://perma.cc/QC9R-7J5Q.
• 45See Burbank, 57 L & Contemp Probs at 132 (cited in note 43) (noting that the Hague Evidence Convention “has been a disappointment to many countries that ratified it and a source of controversy and friction”).
• 46See James A.R. Nafziger, Another Look at the Hague Evidence Convention After Aerospatiale, 38 Tex Intl L J 103, 116 (2003) (noting the “need for uniform procedures or at least procedural standards” given the Hague Evidence Convention’s failure to bridge international differences in discovery procedures).
• 47See Marcus, 7 Tulane J Intl & Comp L at 199 (cited in note 39) (arguing that despite US reforms to constrain discovery in recent decades, “a new world order that fits the American reality and also commands the respect of the rest of the industrialized world is probably a thing of the remote future so far as discovery is concerned”).
• 48See Intel, 542 US at 262–65.
• 49See Gary B. Born and Peter B. Rutledge, International Civil Litigation in United States Courts 1024 (Wolters Kluwer 5th ed 2011).
• 50See Act of Mar 2, 1855, ch 140, § 2, 10 Stat 630 (empowering federal courts to subpoena witnesses under letters rogatory).
• 51See 28 USC § 1781(a)(1):

  The Department of State has power, directly, or through suitable channels . . . to receive a letter rogatory issued, or request made, by a foreign or international tribunal, to transmit it to the tribunal, officer, or agency in the United States to whom it is addressed, and to receive and return it after execution.

• 52See 22 CFR § 92.54 (2019) (noting that letters rogatory requests “rest entirely upon the comity of courts toward each other, and customarily embody a promise of reciprocity”); William S. Dodge, International Comity in American Law, 115 Colum L Rev 2071, 2078 (2015) (defining international comity).
• 53Telephone Interview with Katerina Ossenova, Office of International Judicial Assistance, Department of Justice (Feb 13, 2019) (Ossenova Interview).
• 54See 22 CFR § 22.1 (2013) (listing $2,275 as the fee for processing a letter rogatory); Donald Earl Childress III, Michael D. Ramsey, and Christopher A. Whytock, Transnational Law and Practice 941 (Wolters Kluwer 2015) (explaining that processing is “frequently [ ] delayed by poor diplomatic relations, bureaucratic inertia, and conflicts with public policy”).
• 55See Andreas F. Lowenfeld, International Litigation and Arbitration 1017 (Thomson West 3d ed 2006).
• 56Ossenova Interview (cited in note 53).
• 57Department of State, Bureau of Consular Affairs, Preparation of Letters Rogatory, archived at https://perma.cc/JCE4-Y6V3.
• 58See Proposed Commission and Advisory Committee on International Rules of Judicial Procedure, S 1890, 85th Cong, 1st Sess, in 103 Cong Rec S 5726 (Apr 16, 1957) (noting that cases with “international ramifications” included “cases in which judicial documents must be served abroad, records or witnesses examined within the territory of a foreign state, or in which proof must be offered of the law prevailing in a foreign jurisdiction”).
• 59Commission and Advisory Committee on International Rules of Judicial Procedure, S Rep No 85-2392, 85th Cong, 2d Sess 2 (1958), reprinted in 1958 USCCAN 5201, 5201–02. See also Burbank, 57 L & Contemp Probs at 110–11 (cited in note 43).
• 60Hickman v Taylor, 329 US 495, 501 (1947).
• 61In some countries like Switzerland, it is a criminal infringement of state sovereignty for a private party to take evidence. S Rep No 85-2392 at 2 (cited in note 58) (“In some jurisdictions, notably Switzerland, there is [a] considerable question whether such a procedure might constitute a penal offense on the assumption that it would be regarded as an illegal usurpation of judicial functions.”).
• 62S 1890, 103 Cong Rec at S 5726 (cited in note 58).
• 63See id.
• 64See S Rep No 85-2392 at 2 (cited in note 59); Childress, Ramsey, and Whytock, Transnational Law and Practice at 940 (cited in note 54); Born and Rutledge, International Civil Litigation at 1025 (cited in note 49).
• 65See Act of June 25, 1948 §§ 1783–84, 62 Stat 869, 949–50, codified at 28 USC §§ 1783–84.
• 66See Burbank, 57 L & Contemp Probs at 112–13 (cited in note 43) (describing amendments to Rule 28(b) in 1963).
• 67FRCP 34, 45.
• 68The 1948 version authorized district courts to depose “any witness residing within the United States [for] use[ ] in any civil action pending in any court in a foreign country with which the United States is at peace.” Act of June 25, 1948, § 1782, 62 Stat at 949.
• 69Act of Oct 3, 1964 § 9, 78 Stat at 997.
• 7028 USC § 1782(a).
• 71Act of Feb 10, 1996, Pub L No 104-106, 110 Stat 486, 486.
• 72S Rep No 88-1580 at 2, 13 (cited in note 34).
• 73Id at 2, 7.
• 74Id at 13.
• 75See, for example, Burbank, 57 L & Contemp Probs at 136 (cited in note 43) (lamenting that § 1782 “reflect[s] the naive view that if the United States were generous, other countries would follow its lead”); Lien, 50 Cath U L Rev at 631 (cited in note 38) (noting that “the invitation [to follow the US example in liberal export] has been declined”).
• 76Convention on the Taking of Evidence Abroad in Civil or Commercial Matters, 23 UST 2555, 2555–56, TIAS No 7444 (1970) (Hague Evidence Convention).
• 77Hague Evidence Convention Arts 1–2, 23 UST at 2557–58.
• 78Ossenova Interview (cited in note 53).
• 79Except that there are typically no fees associated with a request coming from a Hague Evidence Convention state party. Id.
• 80Id.
• 81See Burbank, 57 L & Contemp Probs at 132–33 (cited in note 43).
• 82There are sixty-three contracting states overall. The Convention has entered into force between the United States and all but three of the other contracting states. See Hague Conference on Private International Law, Hague Evidence Convention – Acceptances of Accessions (Apr 21, 2020), archived at https://perma.cc/QHB8-SPXE.
• 83See, for example, Lowenfeld, International Litigation and Arbitration at 1052 (cited in note 55) (noting that the “conflict over the Hague Evidence Convention appears as a kind of replay of the overall conflict about American-style litigation, and in particular about American-style discovery”).
• 84See Hague Evidence Convention Art 23, 23 UST at 2568 (“A Contracting State may at the time of signature, ratification or accession, declare that it will not execute Letters of Request issued for the purpose of obtaining pre-trial discovery of documents as known in Common Law countries.”). The Practical Handbook on the Operation of the Evidence Convention explains that Article 23 does not apply to all requests for production of documents from common law countries during the pretrial discovery phase, but only to those that are not “sufficiently substantiated so as to avoid ‘fishing expeditions.’” See Hague Convention on Private International Law, Practical Handbook on the Operation of the Evidence Convention (3d ed 2016).
• 85Société Nationale Industrielle Aérospatiale v United States District Court for the Southern District of Iowa, 482 US 522, 542 (1987).
• 86Aérospatiale left judges to conduct an open-ended, “particularized analysis” to determine whether first resort to the Hague Evidence Convention is required. Id at 543–44. See also Maggie Gardner, Parochial Procedure, 69 Stan L Rev 941, 946–47 (2017) (finding that this particularized analysis has resulted in federal judges almost never requiring parties to use the Hague Evidence Convention to obtain foreign discovery from other parties).
• 87See Restatement (Third) of Foreign Relations Law of the United States § 442, Reporters’ Note 1 (1987) (“No aspect of the extension of the American legal system beyond the territorial frontier of the United States has given rise to so much friction as the requests for documents in investigation and litigation in the United States.”).
• 88See Hans Smit, Recent Developments in International Litigation, 35 S Tex L Rev 215, 227 (1994) (“While the ruling that the [Hague Evidence] Convention is nonexclusive is to be welcomed, and a different ruling would have been calamitous, it would nevertheless be preferable to denounce the Convention.”). See also Walter B. Stahr, Discovery Under 28 U.S.C. § 1782 for Foreign and International Proceedings, 30 Va J Intl L 597, 599 (1990) (indicating that “litigation regarding discovery in the United States for use abroad has turned on the proper interpretation and application of section 1782, not on the interpretation of the Hague Evidence Convention”).
• 89See 28 USC § 1782.
• 90See, for example, Kiobel by Samkalden v Cravath, Swaine & Moore LLP, 895 F3d 238, 240–41 (2d Cir 2018), cert denied, 139 S Ct 852 (2019); Grupo Mexico SAB de CV v SAS Asset Recovery, Ltd, 821 F3d 573, 573–75 (5th Cir 2016); Heraeus Kulzer, GmbH v Biomet, Inc, 633 F3d 591, 593–94 (7th Cir 2011); Suzlon Energy Ltd v Microsoft Corp, 671 F3d 726, 727–28 (9th Cir 2011).
• 91Born and Rutledge, International Civil Litigation at 1066 (cited in note 49) (asking whether it is “wise for § 1782 to provide judicial assistance at the request of foreign litigants”).
• 92S Rep No 88-1580 at 7 (cited in note 34).
• 93See In re Application of Asta Medica, S.A., 981 F2d 1, 7 (1st Cir 1992); In re Request for Assistance from Ministry of Legal Affairs of Trinidad and Tobago, 848 F2d 1151, 1156 (11th Cir 1988). But see Advanced Micro Devices, Inc v Intel Corp, 292 F3d 664, 668–69 (9th Cir 2002) (rejecting a foreign discoverability requirement); In re Bayer AG, 146 F3d 188, 193–96 (3d Cir 1998) (same); Euromepa S.A. v R. Esmerian, Inc, 51 F3d 1095, 1099–1100 (2d Cir 1995) (same).
• 94542 US 241 (2004).
• 95Id at 250–54.
• 96See European Commission Brief at *4, 15–16 (cited in note 28).
• 97Id at *2.
• 98See Brief of the Chamber of Commerce of the United States as Amicus Curiae In Support of Petitioner, Intel Corp v Advanced Micro Devices, Inc, No 02-572, *1–3 (US filed Nov 15, 2002) (available on Westlaw at 2003 WL 23112944); Brief of Product Liability Advisory Council, Inc as Amicus Curiae In Support of Petitioner, Intel Corp v Advanced Micro Devices, Inc, No 02-572, *2–3 (US filed Dec 31, 2003) (available on Westlaw at 2003 WL 23112943). See also Brief Amicus Curiae of the National Association of Manufacturers in Support of Petitioner, Intel Corp v Advanced Micro Devices, Inc, No 02-572, *8–11 (US filed Nov 15, 2002) (available on Westlaw at 2002 WL 32157392).
• 99See Brief for the United States as Amicus Curiae Supporting Affirmance, Intel Corp v Advanced Micro Devices, Inc, No 02-572, *1–2, 11–13 (US filed Jan 30, 2004) (available on Westlaw at 2004 WL 214306).
• 100See Intel, 542 US at 261–62.
• 101Id at 264–65.
• 102See id at 262.
• 103See Advanced Micro Devices, Inc v Intel Corp, 2004 WL 2282320, *2–3 (ND Cal).
• 104See Servotronics, Inc v Rolls-Royce PLC, No 19-1847, slip op at *15 (7th Cir Sept 22, 2020); In re Guo, 965 F3d 96, 104–06 (2d Cir 2020); El Paso Corp v La Comision Ejecutiva Hidroelectrica Del Rio Lempa, 341 F Appx 31, 33–34 (5th Cir 2009); In re Application to Obtain Discovery for Use in Foreign Proceedings, 939 F3d 710, 717–730 (6th Cir 2019); Servotronics, Inc v Boeing Co, 954 F3d 209, 210 (4th Cir 2020).
• 105See Sergeeva v Tripleton International Ltd, 834 F3d 1194, 1200 (11th Cir 2016) (holding that § 1782 reaches “responsive documents and information located outside the United States” so long as it is within the “possession, custody, or control of” the discovery target); In re del Valle Ruiz, 939 F3d 520, 524 (2d Cir 2019) (holding that “there is no per se bar to the extraterritorial application of § 1782”). But see Pinchuk v Chemstar Products LLC, 2014 WL 2990416, *4 (D Del) (quashing a discovery request for documents located abroad).
• 106This development has gained traction since the Second Circuit held in 2017 that § 1782 “does not prevent an applicant who lawfully has obtained discovery under the statute with respect to one foreign proceeding form using the discovery elsewhere unless the district court orders otherwise.” In re Accent Delight International Ltd, 869 F3d 121, 135 (2d Cir 2017).
• 107See Part III.C.
• 108Globalization Spurs Use of USC 1782 for U.S. Discovery in Foreign Disputes (Bloomberg Law, Feb 17, 2011), archived at https://perma.cc/L56T-2TL7. See also Geoffrey Kertesz, Section 1782: American Dream . . . or Nightmare?, 22 Trusts & Trustees 293, 297 (2016) (“If the reported cases are any indication, use of the statute is on the rise.”).
• 109Other scholars have noted the sampling bias stemming from empirical efforts that rely on Westlaw or Lexis searches. These problems are particularly worrisome for studying low-profile, routine matters such as § 1782 discovery orders, which tend to be unpublished. See, for example, David Freeman Engstrom, The Twiqbal Puzzle and Empirical Study of Civil Procedure, 65 Stan L Rev 1203, 1214–15 (2013) (discussing incompleteness in Westlaw’s and Lexis’s databases); David A. Hoffman, Alan J. Izenman, and Jeffrey R. Lidicker, Docketology, District Courts, and Doctrine, 85 Wash U L Rev 681, 686 (2007) (same).
• 110While different indices for globalization differ on this point, the KOF Globalisation Index, which measures globalization along economic, social, and political dimensions, shows a moderate increase in globalization for the United States during the years 2005 through 2016 (the latest year for which data is available). See KOF Swiss Economic Institute, KOF Globalisation Index, archived at https://perma.cc/6RQL-95VP.
• 111Thanks to Zach Clopton for this insight.
• 112See Appendix C, Table 6.
• 113As noted in Part I, § 1782 was amended in 1996 to encompass discovery requests related to criminal investigations. In 2009, Congress enacted 18 USC § 3512, which authorizes federal judges to “issue such orders as may be necessary to execute a request from a foreign authority for assistance” with criminal matters. Foreign Evidence Request Efficiency Act of 2009, Pub L No 111-79, 123 Stat 2087 (2009), codified at 18 USC § 3512. Correspondence with the Department of Justice confirmed that, since 2009, the agency has gradually reduced reliance on § 1782 for executing criminal requests and no longer relies on the statute for that purpose. Correspondence with a member of the Department of Justice’s Office of International Affairs (Feb 20, 2019) (on file with author).
• 114See Part III.C.1 (discussing the significance of these refiled cases).
• 115Since the random sample was drawn without replacement from a finite population of comparable size, I modeled it as a random draw from a hypergeometric random variable and used this distribution to calculate 95 percent confidence intervals.
• 116See Appendix C, Table 7. Concentration of civil requests in these federal district courts is not surprising given that American, European, and Asian financial institutions have historically been subject to personal jurisdiction in New York, Latin American financial institutions have traditionally been subject to jurisdiction in Florida, and many technology companies are subject to jurisdiction in California. Thanks to Kevin Benish for this observation.
• 117Fifty-eight of the ninety-four district courts in the federal system received at least one civil request, while thirty-six district courts did not receive any.
• 118In a tiny number of actions, requests originating from tribunals are filed directly with a district court (approximately 1.1 percent) or are conveyed to the district court by a foreign party (approximately 2.8 percent).
• 119The Supreme Court interpreted “interested person[s]” from the statutory language of § 1782 to “reach[ ] beyond the universe of persons designated ‘litigant.’” Intel, 542 US at 256.
• 120Since the random sample was drawn without replacement from a finite population of comparable size, I modeled it as a random draw from a hypergeometric random variable and used this distribution to calculate 95 percent confidence intervals.
• 121See Part III for discussion.
• 122Requests that sought discovery for use in multiple forums were counted toward each category, which is why the percentages add up to more than 100 percent. I double counted because each forum is independently significant.
• 123Year-by-year data are on file with author.
• 124See Appendix B for a full list of the substantive area categories.
• 125Year-by-year data are on file with author.
• 126Born and Rutledge, International Civil Litigation at 967–68 (cited in note 49) (noting that substantive claims involving antitrust, patent, product liability, and other similar cases often result in “sweeping” discovery requests compared to routine contract or tort disputes).
• 127See Appendix C, Figure 4 showing and explaining a graph visualizing a rise in Shannon’s entropy, calculated on the mixture of countries over the study period.
• 128See Appendix B for a full list of each of these categories.
• 129Year-by-year data are on file with author.
• 130Requests that sought discovery for use in multiple countries with different legal system types were counted toward each category, which is why the percentages add up to more than 100 percent. I double counted because each legal system type is independently significant.
• 131Requests that sought discovery for use in multiple countries with different Hague Evidence Convention statuses were counted toward each category, which is why the percentages add up to more than 100 percent. I double counted because each Convention status is independently significant.
• 132Where a request sought discovery for use in countries in multiple regions, it was counted toward each region, which is why the percentages add up to more than 100 percent. To calculate the 95 percent confidence intervals, I first used the hypergeometric distribution to estimate a 95 percent confidence interval for the number of (for example) tribunal requests in the overall population. I then used the hypergeometric distribution a second time to estimate how many requests of this particular sort were made by tribunals. I took a conservative approach and used the lower bound for the number of tribunal requests to calculate the lower bound for the number of requests made by tribunals, etc. This method errs on the side of wider confidence intervals.
• 133Where a request sought discovery for use in countries that fell into multiple categories of any attribute, it was counted toward each category, which is why the percentages sometimes add up to more than 100 percent. To calculate the 95 percent confidence intervals, I first used the hypergeometric distribution to estimate a 95 percent confidence interval for the number of (for example) tribunal requests in the overall population. I then used the hypergeometric distribution a second time to estimate how many requests of this particular sort were made by tribunals. I took a conservative approach and used the lower bound for the number of tribunal requests to calculate the lower bound for the number of requests made by tribunals, etc. This method errs on the side of wider confidence intervals.
• 134For a list of countries with each legal system type, see Appendix B.
• 135For a list of countries with each Hague Evidence Convention status, see id.
• 136As noted in Appendix B, I employed the World Justice Project’s Rule of Law Index. Like other rule-of-law indexes, such as the Worldwide Governance Indicators, Freedom House’s Freedom in the World Index, and the Heritage Foundation’s Index of Economic Freedom, the World Justice Project’s Rule of Law Index has been criticized for conceptual and methodological problems. Without overlooking or discounting these problems, I use the World Justice Project’s Rule of Law Index as an imperfect way to generate a rough picture of the types of legal systems with which federal district courts are interacting in foreign discovery requests, and whether that picture is different for foreign tribunal versus foreign party requests.
• 137See Appendix B (discussing rule-of-law score categories and data).
• 138See Burbank, 57 L & Contemp Probs at 132 (cited in note 43) (explaining that the Convention was “[i]ntended . . . to bridge gaps between the civil law and common law systems”).
• 139Where a request sought discovery for use in countries that fell into multiple categories of any attribute, it was counted toward each category, which is why the percentages sometimes add up to more than 100 percent. To calculate the 95 percent confidence intervals, I first used the hypergeometric distribution to estimate a 95 percent confidence interval for the number of (for example) tribunal requests in the overall population. I then used the hypergeometric distribution a second time to estimate how many requests of this particular sort were made by tribunals. I took a conservative approach and used the lower bound for the number of tribunal requests to calculate the lower bound for the number of requests made by tribunals, etc. This method errs on the side of wider confidence intervals.
• 140See Appendix B.
• 141See id.
• 142This grant rate is calculated by taking the number of cases over the sum of the number of cases granted and the number of cases denied. Cases that did not reach resolution are not included in the denominator. See Appendix C, Tables 14 and 15.
• 143See id.
• 144See Appendix C, Table 15.
• 145See Appendix C, Table 14.
• 146A request may be challenged at the outset if notice was provided. More likely, a request was initially granted ex parte and then subsequently challenged in the form of a motion to quash, motion to vacate, motion to stay, motion for reconsideration, or a motion to compel from the requestor that is opposed. A request may be challenged by the target of discovery, whether party or nonparty to the foreign proceeding, or it may be challenged by the nontarget opposing party. I did not count entries of stipulated protective orders or other requests for confidentiality agreements as contestation. See Appendix B.
• 147These figures are not estimates, as they are based on all requests in 2015.
• 148The grant rate is calculated by taking the number of granted requests over the sum of the number of granted requests and denied requests. Requests that did not reach resolution are not included in the denominator.
• 149The contestation rate is calculated by taking the number of challenged requests over the sum of the number of challenged requests and unchallenged requests (including requests involving a stipulated protective order or other confidentiality agreement).
• 150There were only three contested tribunal requests and one was denied.
• 151To identify the best logistic regression model, I followed a standard best-subset approach: for each model size, I fitted a model with every possible combination of variables and chose the model with the lowest residual deviance. To select the appropriate model size and to avoid overfitting, I then performed cross-validation and considered two metrics that penalize large models, AIC and BIC. BIC and cross-validation both pointed toward a one-variable model containing only contestation, whereas AIC pointed toward a three-variable model containing contestation, target identity, and numerosity of foreign proceedings for which discovery was requested. I additionally performed Firth logistic regression, a penalized form of regression designed to address the problems associated with separated or nearly separated data. See Georg Heinze, A Comparative Investigation of Methods for Logistic Regression with Separated or Nearly Separated Data, 25 Statist Med 4216, 4224–25 (2006). The Firth regression suggested a three-variable model containing contestation, target identity, and the legal system type of the country from which the request originated.
• 152In the one-variable model containing contestation, the absence of contestation was associated with a 2.17 increase in the log-odds of a request being granted, significant at the level of p = 0.001. In the Firth three-variable model, contestation was associated with a 1.72 increase in the log-odds of a request being granted, significant at the level of p = 0.015.
• 153Given the problematic nature of the 2015 data set, however, this should not be taken as evidence against an association between these variables and the likelihood of a request being granted. The 2015 data set had a strikingly high grant rate (over 94 percent), perfect separation along one variable, and near perfect separation along another. Several covariates strongly correlated with each other as well.
• 154See Part III.C.1.
• 155For example, notice might be given informally as a matter of courtesy or incidentally through communication between the target and the foreign opposing party. Id.
• 156See Intel, 542 US at 262.
• 157Although I did not systematically track whether granted discovery requests were conditioned on a reciprocal flow of information, reciprocity is extremely rare. A few examples are discussed below in Part III.C.3.
• 158To find this information, one would have to interview those to whom compelled US discovery was granted and track down each foreign proceeding abroad.
• 159See Part I.B.
• 160See, for example, Texas Keystone, Inc v Prime Natural Resources, Inc, 694 F3d 548, 554 (5th Cir 2012); Government of Ghana v ProEnergy Services, LLC, 677 F3d 340, 343 (8th Cir 2012); Heraeus Kulzer, GmbH v Biomet, Inc, 633 F3d 591, 597 (7th Cir 2011).
• 161Heraeus Kulzer, 633 F3d at 597. See also 28 USC § 1782 (“To the extent that the order does not prescribe otherwise, the testimony or statement shall be taken, and the document or other thing produced, in accordance with the Federal Rules of Civil Procedure.”).
• 162Hickman v Taylor, 329 US 495, 501 (1947).
• 163See United States v Procter & Gamble Co, 356 US 677, 682 (1958) (“Modern instruments of discovery . . . make a trial less a game of blindman’s buff and more a fair contest with the basic issues and facts disclosed to the fullest practicable extent.”); Hickman, 329 US at 507 (“Mutual knowledge of all the relevant facts gathered by both parties is essential to proper litigation.”); John H. Beisner, Discovering a Better Way: The Need for Effective Civil Litigation Reform, 60 Duke L J 547, 556–63 (2010).
• 164See Paul W. Grimm and David S. Yellin, A Pragmatic Approach to Discovery Reform: How Small Changes Can Make a Big Difference in Civil Discovery, 64 SC L Rev 495, 505–07 (2013) (noting “a lack of active judicial management” despite general agreement that active judicial involvement in discovery “leads to better . . . results”).
• 165FRCP 26(b)(1).
• 166Id.
• 167See Wayne D. Brazil, The Adversary Character of Civil Discovery: A Critique and Proposals for Change, 31 Vand L Rev 1295, 1304 (1978):

  Modern discovery [ ] has removed most of the decisive plays from the scrutiny of the court. Because so many civil cases are settled before trial and because the conduct of attorneys is subject only to fitful and superficial judicial review during the discovery stage, much of the decisive gamesmanship of modern litigation takes place in private settings.

• 168See FRCP 26(c)(1) (providing that a party “from whom discovery is sought” can move for a protective order following an effort to resolve the dispute without court action; protective order can be issued to protect from “annoyance, embarrassment, oppression, or undue burden or expense”); FRCP 37 (providing that a party may move for an order to compel discovery following an effort to obtain the discovery without court action).
• 169See FRCP 26, Notes of the Advisory Committee on Rules—1980 Amendment, Note to Subdivision (f) (“There has been widespread criticism of abuse of discovery.”); FRCP 26, Notes of the Advisory Committee on Rules—1983 Amendment (“Excessive discovery and evasion or resistance to reasonable discovery requests pose significant problems.”). See also Herbert v Lando, 441 US 153, 179 (1979) (Powell concurring) (expressing concern that federal discovery rules are “not infrequently exploited to the disadvantage of justice”).
• 170See, for example, Matthew T. Ciulla, Note, A Disproportionate Response? The 2015 Proportionality Amendments to Federal Rule of Civil Procedure 26(b), 92 Notre Dame L Rev 1395, 1405–09 (2017) (noting wide differences in opinion concerning the need for discovery reform).
• 171See Bell Atlantic Corp v Twombly, 550 US 544, 558–60 (2007) (discussing the cost of discovery); Ashcroft v Iqbal, 556 US 662, 684–86 (2009) (same).
• 172FRCP 26, Notes of the Advisory Committee on Rules—1983 Amendment, Note to Subdivision (g).
• 173FRCP 26, Notes of the Advisory Committee on Rules—2015 Amendment.
• 174See id (noting that the factors may not be fully understood at the outset and that the requesting party may not know about the burden or expense, while the requested party may not know about the importance of the sought-after discovery for resolving the underlying issues).
• 175See, for example, Moore v Ford Motor Co, 755 F3d 802, 808 (5th Cir 2014) (explaining that a judge’s discovery decision is reversible only if it is “arbitrary or clearly unreasonable” and results in “prejudice”).
• 176See, for example, Ciulla, Note, 92 Notre Dame L Rev at 1402 (cited in note 170) (noting that the impact of amendments has not been as great as expected); Scott A. Moss, Litigation Discovery Cannot Be Optimal but Could Be Better: The Economics of Improving Discovery Timing in a Digital Age, 58 Duke L J 889, 889–90 (2009) (noting that the “proportionality rules are impossible to apply effectively”).
• 177See Jonah B. Gelbach and Bruce H. Kobayashi, The Law and Economics of Proportionality in Discovery, 50 Ga L Rev 1093, 1114–16 (2016).
• 178See FRCP 45(a)(1)(A)(iii) (stipulating that a subpoena may compel a person to “attend and testify; produce designated documents, electronically stored information, or tangible things in that person’s possession, custody, or control”).
• 179See FRCP 30(b)(1), 45(a)(4).
• 180See FRCP 45(a)(4), (b) (setting out service requirements).
• 181See FRCP 45(a)(2).
• 182See FRCP 45(a)(1)–(2).
• 183See FRCP 45, Notes of the Advisory Committee on Rules—2013 Amendment, Note to Subdivision (f).
• 184See id.
• 185See In re Letters Rogatory from Tokyo District, Tokyo, Japan, 539 F2d 1216, 1219 (9th Cir 1976) (“Letters Rogatory are customarily received and appropriate action taken with respect thereto ex parte.”).
• 186James E. Pfander and Daniel D. Birk, Article III Judicial Power, the Adverse-Party Requirement, and Non-Contentious Jurisdiction, 124 Yale L J 1346, 1355, 1390–91 (2015).
• 187Id at 1391.
• 188See In re Premises Located at 840 140th Ave NE, Bellevue, Wash, 634 F3d 557, 566 (9th Cir 2011) (likening a discovery request from the Russian government to “an order enforcing the subpoenas of independent administrative agencies, an order granting a subpoena in aid of an extradition proceeding, and an order to appear before the Internal Revenue Service”).
• 189Pfander and Birk, 124 Yale L J at 1379 (cited in note 186). See also United States v Markwood, 48 F3d 969, 976–77 (6th Cir 1995) (emphasizing that “a “district court’s role in the enforcement of an administrative subpoena is a limited one” consisting of determining whether the agency has met statutory and judicially created standards for issuing and enforcing the subpoena).
• 190See Table 2.
• 191See, for example, John H. Langbein, The German Advantage in Civil Procedure, 52 U Chi L Rev 823, 827 (1985) (noting that in Germany, “[d]igging for facts is primarily the work of the judge”).
• 19228 USC § 1782.
• 193See Intel, 542 US at 264–65.
• 194Compare In re Clerici, 481 F3d 1324, 1335 (11th Cir 2007) (explaining that the Intel factors for receptivity and noncircumvention supported granting the request “given that the foreign tribunal here is the Panamanian Court and the Panamanian Court itself issued the letter rogatory requesting assistance”), with Order, Request for International Judicial Assistance from the 12th Family Court in Istanbul, Turkey; Matter of Ekmekçi v Ekmekçi, No 1:15-mc-22425, *1 (SD Fla filed July 1, 2015) (“The statutory requirements set forth at 28 U.S.C. § 1782(a) have been met. Furthermore, the additional factors to be considered . . . weigh in favor of granting the request.”), citing Intel, 542 US at 247.
• 195See Table 2.
• 196The permissible grounds for denying a discovery request under the Hague Evidence Convention include if the request does not comply with Convention requirements (Article 5), if the request is for a matter that is not civil or commercial (Article 1), and if the country to which the request is addressed “considers that its sovereignty or security would be prejudiced thereby” (Article 12). Hague Evidence Convention Arts 1, 5, 12(b), 23 UST at 2557, 2560, 2562.
• 197See In re Premises Located at 840 140th Ave, 634 F3d at 568, 570–72 (holding that the US-Russia Mutual Legal Assistance Treaty governing discovery assistance in criminal matters superseded § 1782’s grant of discretionary authority to district courts); Restatement (Fourth) of Foreign Relations Law of the United States § 309(2) (2013) (“When there is a conflict between a self-executing treaty provision and a federal statute, courts in the United States will apply whichever reflects the latest expression of the will of the U.S. political branches.”).
• 198There were five tribunal requests that were denied in the random sample. All but one of the denials were for technical reasons, such as a technical defect in the application, and without prejudice. The last denial is discussed below in note 199.
• 199See, for example, Transcript of Hearing on Petition for Judicial Review, In re Request for International Judicial Assistance from the National Court of First Instance in Labor Matters No 37 of Buenos Aires, Argentina, No 3:12-cv-00662, *4–8 (D Nev filed Sept 16, 2013), (denying a tribunal request from Argentina in part due to the judge’s belief that the country from which the request originated did not honor American judgments and extradition requests—a ground for denial not permitted under the Convention).
• 200Glock v Glock, Inc, 797 F3d 1002, 1009 (11th Cir 2015).
• 201In re Schlich, 893 F3d 40, 46 (1st Cir 2018), quoting Seattle Times Co v Rhinehart, 467 US 20, 36 (1984).
• 202A Bill to Revise, Codify, and Enact into Law Title 28 of the United States Code, Entitled “Judicial Code and the Judiciary”, HR Rep No 79-2646, 79 Cong, 2d Sess A146 (1946).
• 203Kang v Noro-Moseley Partners, 246 F Appx 662, 663 (11th Cir 2007). See also ProEnergy Services, LLC, 677 F3d at 344; Nascimento v Faria, 600 F Appx 811, 812 (2d Cir 2015).
• 204In re Premises Located at 840 140th Ave, 634 F3d at 566 (quotation marks omitted).
• 205See, for example, Gushlak v Gushlak, 486 F Appx 215, 217 (2d Cir 2012) (stating that “it is neither uncommon nor improper for district courts to grant applications made pursuant to § 1782 ex parte” and listing several examples); Order, Elkind v CCBill, LLC, No 2:14-mc-00030, *1 (D Ariz filed May 9, 2014) (granting ex parte request).
• 206See, for example, In re Ex Parte Application of Société d’Etude de Réalisation et d’Exploitation pour le Traitement du Mais, 2013 WL 6164435, *2 n 1 (ED Pa) (explaining that ex parte applications under § 1782 are justified because the parties will be given adequate notice of any discovery taken pursuant to the request and will then have the opportunity to move to quash the discovery); In re Letter of Request from Supreme Court of Hong Kong, 138 FRD 27, 32 n 6 (SDNY 1991) (“[S]uch ex parte applications are typically justified by the fact that the parties will be given adequate notice of any discovery taken pursuant to the request and will then have the opportunity to move to quash the discovery or to participate in it.”); Interbrew Central Europe Holding BV v Molson Coors Brewing Co, 2013 WL 5567504, *1 (D Colo) (finding that “Applicant’s ex parte request is appropriately made and that Respondents may later seek modification of the discovery herein ordered by way of an appropriate motion”).
• 207See Part III.A (discussing that FRCP 45 requires notification to all parties of discovery requests targeting nonparties).
• 208See In re Hornbeam Corp, 722 F Appx 7, 10–11 (2d Cir 2018) (applying FRCP 45(a)(4)’s notification requirements but nevertheless affirming the district court’s denial of a motion to vacate or quash and refusal to sanction the applicant due to lack of prejudice).
• 209See, for example, Order by Magistrate Judge Arthur Nakazato, In re Ex Parte Application of Nokia Corp, No 8:13-mc-00010, *1 (CD Cal filed May 15, 2013) (denying a § 1782 request without prejudice for failure to comply with Local Rule 7-19 governing ex parte applications in the Central District of California).
• 210See, for example, In re Merck & Co, Inc, 197 FRD 267, 270–71 (MD NC 2000) (observing that “[n]othing in Section 1782 states that the application is to be made ex parte, much less that the Court must entertain the application ex parte,” and concluding that “nothing in Section 1782 prevents the Court in any given case from advancing the process by requiring the notification to take place at an earlier time in order to reduce disruption and conserve judicial resources”) (emphasis in original).
• 211See, for example, Request to File Under Seal, In re Application of Lúcia de Araujo Bertolla for an Order Pursuant to 28 USC § 1782 to Obtain Discovery for Use in a Foreign Proceeding, No 1:17-mc-00284, *1 (SDNY filed April 25, 2018) (Araujo Bertolla Request to File Under Seal) (explaining that the court requested the applicant brief the issue of why the federal rules’ subpoena notice requirement should not apply).
• 212See, for example, Order to Show Cause, In re the Court Order of the Romford County Court of Great Britian Dated May 21, 2009, No 6-11-mc-00028, *1 (MD Fla filed May 6, 2011) (ordering the applicant to explain in writing why the target has not been served per a prior order and threatening sanctions as well as denial of the request). See also Certain Funds, Accounts and/or Investment Vehicles Managed by Affiliates of Fortress Investment Group LLC v KPMG LLP, No 1:14-cv-01801 (SDNY filed Mar 14, 2014), (issuing a summons for a § 1782 request); Summons in a Civil Case, Blue Traffic Ltd v VT iDirect, Inc, No 1:08-mc-00031 (ED Va filed July 7, 2008) (same). For the text of this requirement, see FRCP 4(c).
• 213See, for example, Order, In re Application of Halliburton SAS, No 1:14-mc-00004, *2 (ED Va filed Feb 4, 2014) (Halliburton Application Order); In re Ex Parte Application of Apple, Inc, Apple Retail Germany GmBh; and Apple Sales International, No 3:12-cv-00179, *1 (SD Cal filed Feb 1, 2012) (Apple Application Order); Order to Show Cause Why this Court Should Not Grant Ecuadorean Plaintiffs’ Ex Parte Application for Expedited Service and Enforcement of Subpoenas to Conduct Discovery for Use in Foreign Proceeding Pursuant to 28 U.S.C. § 1782, In re Application of Daniel Carlos Lusitand Yaiguaje, No 3:11-mc-80087, *2 (ND Cal filed May 9, 2011) (Yaiguaje Application Order).
• 214See, for example, Halliburton Application Order at *2 (cited in note 213) (ordering that applicant provide notice to a number of relevant parties); Apple Application Order at *1 (cited in note 213) (same); Yaiguaje Application Order at *2 (cited in note 213) (same).
• 215See, for example, Order, In re Application Pursuant to 28 USC § 1782 of Financial Guaranty Insurance Co v Lehman Brothers, Inc, No 1:11-mc-00085, *2 (SDNY filed Mar 29, 2011).
• 216For background on the “case or controversy” requirement and its constitutional roots, see Martin H. Redish and Andrianna D. Kastanek, Settlement Class Actions, the Case-or-Controversy Requirement, and the Nature of the Adjudicatory Process, 73 U Chi L Rev 545, 563–66 (2006).
• 217See Appendix C, Table 6. Compare Notice of New Case Number and Notice of Judge Assignment, In re Application of Hulley Enterprises Ltd, Yukos Universal Ltd, and Veteran Petroleum Ltd for an Order Pursuant to 28 USC § 1782 to Conduct Discovery for Use in a Foreign Proceeding, No 2:17-mc-00088, *1 (CD Cal filed Sept 26, 2017) (case converted from miscellaneous to civil); Order Directing the Clerk of Court to Redesignate this Matter as a Contested Civil Case for Statistical Purposes, In re Application of H.M.B. Limited Pursuant to 28 USC § 1782 to Conduct Discovery for Use in Foreign Proceedings, No 1:17-cv-21459 (SD Fla filed May 1, 2018) (same), with Order, APR Energy Holdings Ltd v Australia and New Zealand Banking Group Ltd, No 1:17-cv-02784, *1 (SDNY filed Apr 24, 2017) (case converted from civil to miscellaneous); Notice Regarding E-filing, In re Application of Akebia Therapeutics, Inc for an Order Granting Leave to Issue Subpoena for the Taking of Discovery Pursuant to 28 USC 1782, No 5:14-cv-04678, *1 (ND Cal filed Oct 21, 2014) (same).
• 218See, for example, US District Court Northern District of Texas, Electronic Case Filing: Opening a Miscellaneous Case 1 (Nov 2016), archived at https://perma.cc/Y9VU-KR2X.
• 219See In re Biomet Orthopaedics Switzerland GmBh, 742 F Appx 690, 697 (3d Cir 2018) (noting that the lower court erred in concluding that discovery was sought for a criminal appeal when in fact it was sought for a civil trial proceeding).
• 220See id.
• 221See Order, In re Application of Raoul Malak, No 2:14-mc-00089, *4 (D Ariz filed Feb 17, 2015) (Malak Application Order) (“[T]he Court lacks any meaningful information with which to determine whether the such [sic] discovery is relevant to the foreign proceeding.”).
• 222See Memo Endorsement, In re Application of Gorsoan Ltd and Gazprombank OJSC for an Order Pursuant to 28 USC § 1782 to Conduct Discovery for Use in a Foreign Proceeding, No 1:13-mc-00397, *1 (SDNY filed Oct 22, 2014).
• 223See Memorandum and Order Regarding Intervenor’s Motion to Quash and to Vacate, In re: Application of Hanwha Azdel, Inc and Hanwha L&C Corp for Assistance Before a Foreign Tribunal, No 3:13-mc-93004, *2–4 (D Mass filed Oct 29, 2013).
• 224See Marubeni America Corp v LBA Y.K., 335 F Appx 95, 97–98 (2d Cir 2009).
• 225See In re Chevron Corp, 633 F3d 153, 162–63 (3d Cir 2011).
• 226Intel, 542 US at 261.
• 227The fourth discretionary factor and parity, which the Court also identified as “important as [a] touchstone[ ] for a district court’s exercise of discretion in particular cases,” will be discussed in the following section. Id. See also Dodge, 115 Colum L Rev at 2105 (cited in note 52) (noting that § 1782 is motivated by “adjudicative comity,” which the author defines as “deference to foreign courts”).
• 228See Intel, 542 US at 264.
• 229Id.
• 230Id.
• 231Id at 265.
• 232See Certain Funds, Accounts and/or Investment Vehicles v KPMG, LLP, 798 F3d 113, 118 (2d Cir 2015) (commenting that the Intel opinion does not provide guidance on “minimum requirements or tests to be met”).
• 233See Gardner, 69 Stan L Rev at 959–64 (cited in note 86).
• 234See, for example, Order Granting Application in Part for Judicial Assistance Pursuant to 28 U.S.C. § 1782, In re Application of Lúcia De Araujo Bertolla for an Order Pursuant to 28 USC § 1782 to Obtain Discovery for Use in a Foreign Proceeding, No 1:17-mc-00284, *2 (SDNY filed Nov 13, 2017) (“[T]he Discovery Targets are not parties in the proceedings in Brazil and are not expected to become parties thereto, thus, the need for this discovery is more apparent.”); Omnibus Report and Recommendations on Motions to Intervene, Vacate, Quash Subpoenas, and for Protective Order, In re Application of H.M.B. Limited Pursuant to 28 USC 1782 to Conduct Discovery for Use in Foreign Proceedings, No 1:17-cv-21459, *17 (SD Fla filed July 2, 2018) (refusing to “look beyond the subpoenaed party to ascertain the true target of discovery”) (emphasis in original). But see In the Matter of a Petition for Judicial Assistance Pursuant to 28 USC § 1782 by Macquarie Bank Ltd, 2015 WL 3439103, *6 (D Nev) (arguing that the first discretionary factor “militates against allowing § 1782 discovery when the petitioner effectively seeks discovery from a participant in the foreign tribunal even though it is seeking discovery from a related, but technically distinct entity”); Order, In re Application of Parmalat Brasil S.A. Industrial de Alimentos and LAEP Investments, Ltd for an Order Pursuant to 28 USC § 1782 to Conduct Discovery for Use in Foreign Proceedings, No 1:11-mc-00077, *6 (SDNY filed July 26, 2011) (Parmalat Application) (concluding that the first factor is neutral because “whether those same documents are obtainable in Brazil is, at this juncture, unknown”).
• 235See, for example, Bravo Express Corp v Total Petrochemicals & Refining USA, Inc, 613 F Appx 319, 320–21 (5th Cir 2015) (seeking discovery from US targets that had a corporate relationship and joint business operations with the entities that were involved in the underlying disputed acts); Application for an Order Directing ASML US, Inc. to Respond to Requests for Documents Pursuant to 28 U.S.C. § 1782 For Use in Foreign Proceedings, and Supporting Memorandum, In re Application Pursuant to 28 U.S.C. § 1782 of Nikon Corp, No 1:17-mc-00142, *8 (SDNY filed Apr 26, 2017) (seeking discovery from the wholly owned subsidiary of the opposing party in the foreign action).
• 236See, for example, In re Application of Schmitz, 259 F Supp 2d 294, 296 (SDNY 2003) (“Application of section 1782 does not involve an analysis of . . . why a respondent has the documents.”); Ex Parte Application for an Order Pursuant to 28 U.S.C. § 1782 Granting Leave to Obtain Discovery from Quinn Emanuel for Use in Foreign Proceedings, In re Ex Parte Application of Apple, Inc, No 5:12-mc-80124, *1 (ND Cal filed May 25, 2012).
• 237See Kiobel by Samkalden v Cravath, Swaine & Moore LLP, 895 F3d 238, 241 (2d Cir 2018), cert denied, 139 S Ct 852 (2019) (noting that the sought-after documents were sent by Shell to the United States “solely . . . for the purpose of American litigation”).
• 238See id at 241, 246–47.
• 239See Hans Smit, American Assistance to Litigation in Foreign and International Tribunals: Section 1782 of Title 28 of the U.S.C. Revisited, 25 Syracuse J Intl L & Commerce 1, 10–12 (1998) (noting that allowing such discovery is likely to interfere with foreign court processes while making the United States the “clearing house[ ]” for information “all over the world”).
• 240See Sergeeva v Tripleton International Ltd, 834 F3d 1194, 1200 (11th Cir 2016) (holding that § 1782 reaches “responsive documents and information located outside the United States” so long as it is within the “possession, custody, or control of” the discovery target); In re del Valle Ruiz, 939 F3d 520, 524 (2d Cir 2019) (holding that “there is no per se bar to the extraterritorial application of § 1782”). See also FRCP 45(a)(1)(A)(iii). But see Pinchuk v Chemstar Products LLC, 2014 WL 2990416, *4 (D Del) (quashing a discovery request for documents located abroad).
• 241There is a multiway split in the courts on the question of who bears the burden of showing receptivity or the lack thereof. Compare Ecuadorian Plaintiffs v Chevron Corp, 619 F3d 373, 378–79 (5th Cir 2010); Chevron, 633 F3d at 162 (holding that relevant evidence is “presumptively discoverable” unless the party opposing discovery shows offense to the foreign jurisdiction); In re MTS Bank, 2017 WL 3155362, *6 (SD Fla) (“[C]ourts look for authoritative proof that a foreign tribunal would reject evidence obtained with the aid of § 1782.”) (quotation marks omitted) (emphasis in original), with Foda v Capital Health, 2010 WL 2925382, *2 (ND Cal) (placing the burden of proof on the applicant); In re Application of Chevron Corp, 762 F Supp 2d 242, 252 (D Mass 2010) (noting that the targets of foreign discovery requests “are often individuals plucked out of their repose who may . . . not necessarily [have] the wherewithal to mount a defense to an application, let alone . . . prove a negative, i.e., a foreign tribunal’s non receptivity to the discovery sought”) (emphasis in original), Department of Caldas v Diageo PLC, 925 F3d 1218, 1223 (11th Cir 2019) (taking a “middle-of-the-road approach” that does not apply a rigid burden-shifting framework); Order, In re Application of Digitechnic, No 2:07-cv-00414, *6–7 (WD Wash filed May 8, 2007) (not placing the burden on either side).
• 242See, for example, In re O’Keeffe, 646 F Appx 263, 266–68 (3d Cir 2016) (affirming district court’s conclusion that Hague Evidence Convention being in effect between United States and Hong Kong was sufficient indication that Hong Kong courts would be receptive to American judicial assistance). See also In re Servicio Pan Americano de Protección, 354 F Supp 2d 269, 274 (SDNY 2004) (“Venezuela has indicated its receptivity to federal judicial assistance by its signature of treaties facilitating such cooperation.”).
• 243See note 84 and accompanying text.
• 244See, for example, Order Granting Ex Parte Application for Order to Obtain Discovery for Use in Foreign Proceedings, In re Ex Parte Application of ANZ Commodity Trading Party Ltd, No 4:17-mc-80070, *6 (ND Cal filed Aug 4, 2017) (relying on cases cited by applicant in which US courts had previously granted foreign discovery requests from Hong Kong to conclude that Hong Kong is receptive).
• 245See, for example, Bravo Express, 613 F Appx at 325 (“[M]odification of a subpoena is preferable to quashing it outright, and a district court abuses its discretion when it does not explain its reasoning, does not allow the applicant an opportunity to cure any defects, and does not attempt to modify the subpoena to cure any overbreadth.”) (quotation marks omitted).
• 246See, for example, Heraeus Kulzer, 633 F3d at 597 (“German judges can disregard evidence that would waste the court’s time.”); In re Ex Parte Application of Nokia Corp, 2013 WL 6073457, *3 (ND Cal) (“[T]he German court can exclude evidence of marginal probative value.”).
• 247Aetna Life Insurance Co v Haworth, 300 US 227, 240–41 (1937).
• 248See Pfander and Birk, 124 Yale L J at 1357 (cited in note 186).
• 249See id at 1358.
• 250See id at 1450 n 490 (noting that “the Fifth Amendment’s Due Process Clause may provide a more effective instrument for moderating non-contentious forms than a strict adherence to an adverse-party rule that would foreclose the exercise of all judicial power over such matters”).
• 251See id at 1446.
• 252See, for example, Order Granting Joint Motion for a Protective Order, In the Matter of Miasto Poznañ v Skarb Panstwa, No 1:15-mc-00179 (D Colo filed Jan 7, 2016). A high degree of cooperation typically occurs when the nonparty target is a bank or social media company that is willing to comply with the request but needs a § 1782 order to justify compliance to the client whose information is being released. See also Neil A.F. Popovic and Shin Hahn, Pursuing and Responding to Discovery Requests Under 28 U.S.C. § 1782 (Lexology, Mar 29, 2019), archived at https://perma.cc/XD6Y-4R5D.
• 253See, for example, Complaint for Declaratory Relief, Comcast Cable Communications, LLC v Hourani, No 1:15-cv-01724, *2–4 (DDC filed Oct 19, 2015).
• 254See Aetna Life, 300 US at 240–41.
• 255Compare In re Kleimar N.V. v Benxi Iron and Steel America, Ltd, 2017 WL 3386115, *4 (ND Ill) (“[T]here is no question that an entity against whom the discovery will be used has standing to challenge an order allowing discovery under § 1782.”), with Order, In re Application of Chevron, No 1:10-mc-00001, *1 (SDNY filed Aug 24, 2010) (noting that the plaintiffs in the foreign proceeding for which discovery was sought, “whose standing in this matter is debatable to say the least,” had moved to strike some of the filings submitted by § 1782 applicant, who happened to be the defendant in the foreign proceeding).
• 256See, for example, Memorandum and Order, In re Application of Microsoft Corp, No 1:06-mc-10061, *6 n 4 (D Mass filed Apr 17, 2006) (Microsoft Memorandum and Order) (denying the European Commission’s motion to intervene on the grounds that its views have already been received and represented by Novell, Inc, the nonparty target of the § 1782 request).
• 257Intel, 542 US at 261–62.
• 258See, for example, In re Application of Consorcio Minero, S.A. v Renco Group, Inc, 2012 WL 1059916, *4 (SDNY) (granting reciprocal discovery in Peru); Minatec Finance S.À.R.L. v SI Group Inc, 2008 WL 3884374, *9 (NDNY) (granting reciprocal discovery located in Luxembourg).
• 259See Heraeus Kulzer, 633 F3d at 595–99; Kulzer v Esschem, Inc, 390 F Appx 88, 89–90 (3d Cir 2010).
• 260See In re Biomet Orthopaedics Switzerland GmBh, 742 F Appx at 699 (refusing to accept Heraeus’s argument that Biomet’s discovery request should not be granted due to potential exposure of trade secrets because “Heraeus [had] gained access to wide swaths of Biomet’s potentially proprietary information through its own 1782 discovery requests in the Northern District of Indiana and the Eastern District of Pennsylvania”).
• 261See, for example, Astronics Advanced Electronics Systems Corp v Lufthansa Technik AG, 561 F Appx 605, 606 (9th Cir 2014) (discussing competing § 1782 requests from adverse parties to the same foreign proceeding). But see Republic of Ecuador v Connor, 708 F3d 651, 658 (5th Cir 2013) (relying on judicial estoppel to prevent an applicant from taking advantage of a circuit split).
• 262FRCP 26(b)(1). See also, for example, Malak Application Order at *4 (cited in note 221) (“[T]he Court lacks any meaningful information with which to determine whether the such [sic] discovery is relevant to the foreign proceeding.”); Order Re: Application for Discovery Pursuant to 28 U.S.C. § 1782, MetaLab Design Ltd v Zozi International, Inc, No 3:17-mc-80153, *6 & n 1 (ND Cal filed Jan 11, 2018) (concluding that the requested evidence “may be relevant” to the requestor’s counterclaims, but noting in a footnote that the court could not determine whether the applicant’s assertion that the discovery would allow it “to defend the Canadian Action” was “in fact the case”); Alexander v Federal Bureau of Investigation, 194 FRD 316, 325 (DDC 2000) (changing the relevance standard to include that which “bears on, or that [which] reasonably leads to other matters that could bear on, any issue that is or may be in the case”).
• 263FRCP 26(b)(1). The proportionality analysis was originally introduced in 1983. The corresponding committee notes explained that “the rule recognizes that many cases in public policy spheres, such as employment practices, free speech, and other matters, may have importance far beyond the monetary amount involved.” FRCP 26, Notes of the Advisory Committee on Rules—1983 Amendment, Note to Subdivision (b).
• 264See Massen, 83 S Cal L Rev at 883 (cited in note 40) (noting that in the United States, “litigants control most aspects of discovery while the judge’s role is limited to defining the outer boundaries of the proof-gathering process while avoiding active participation”).
• 265Merck & Co, 197 FRD at 270.
• 266See Certain Funds, Accounts and/or Investment Vehicles, 798 F3d at 125.
• 267See Part II.
• 268See id.
• 269See Part II.A.
• 270European Commission Brief at *17 (cited in note 28).
• 271Id.
• 272See Appendix C, Table 14 (showing a steady number of cases every year that reach no resolution).
• 273See Part II.
• 274See In re Accent Delight International Ltd, 869 F3d 121, 135 (2d Cir 2017) (holding that § 1782 “does not prevent an applicant who lawfully has obtained discovery under the statute with respect to one foreign proceeding form using the discovery elsewhere unless the district court orders otherwise”).
• 275See, for example, Order, In re Application Pursuant to 28 USC § 1782 of Financial Guaranty Insurance Co v Lehman Brothers, Inc, No 1:11-mc-00085, *2 (SDNY filed Mar 29, 2011).
• 276See, for example, Letter from Paul A. Crotty, US District Judge, to Judge Enrique Claudio González Meyenberg, In re Operación y Supervisión de Hoteles, S.A. de C.V., No 1:15-mc-00172 (SDNY filed Aug 12, 2015); Letter from Paul A. Crotty, US District Judge, to Judge Enrique Claudio González Meyenberg, In re Operación y Supervisión de Hoteles, S.A. de C.V., No 1:15-mc-00172 (SDNY filed Sept 4, 2015).
• 277See, for example, Araujo Bertolla Request to File Under Seal at *3 (cited in note 211); Matter of the Application of Ching Chung Taoist Association of Hong Kong Ltd, 2016 WL 5339803, *1 (ND Cal) (denying discovery as to one witness but stating that the court may revisit the issue after the Australian court decides pending motion about discovery from that witness); Order, In re Ex Parte Application of Banco Safra for an Order Pursuant to 28 USC Section 1782 to Conduct Discovery for Use in Foreign Proceedings, No 1:16-mc-21971, *3–4 (SD Fla filed Sept 8, 2016) (giving respondents opportunity to obtain Brazilian court decision regarding their asserted privileges and instructing them to submit a status report regarding their prospective request for declaratory relief in Brazil); Parmalat Application at *5–8 (cited in note 234) (staying discovery request pending further action from Brazilian court).
• 278See, for example, Order Directing Mentor Graphics Corporation to File Status Report, In re Application of Mentor Graphics Corp, No 5:16-mc-80037, *1 (ND Cal filed June 15, 2016); Ching Chung Taoist Association, 2016 WL 5339803 at *3 (noting that the applicant provided notice to court regarding status of underlying foreign proceeding).
• 279See, for example, Memorandum and Order on Petition for Discovery for Use in Foreign Proceedings Pursuant to 28 U.S.C. § 1782, In re Application of George W. Schlich, No 1:16-mc-91278, *1–4 (D Mass filed Dec 9, 2016) (examining submissions from both sides to the European Patent Office).
• 280See, for example, Order, In re Application of Jurema Dimas de Melo Pimenta and Dimas de Melo Pimenta Filho Pursuant to 28 USC § 1782 For Judicial Assistance in Obtaining Evidence in This District, No 1:12-mc-24043, *1–2 (SD Fla filed Jan 29, 2013) (ordering movant to seek ruling or guidance from Brazilian court).
• 281See, for example, Letter from Korea Fair Trade Commission, In re Ex Parte Application of Qualcomm Inc, No 5:16-mc-80002, *1 (ND Cal filed Jan 26, 2016); Letter From Korea Fair Trade Commission, In re Ex Parte Application of Qualcomm Inc, No 5:16-mc-80008, *1 (ND Cal filed Jan 26, 2016); Letter From Korea Fair Trade Commission, In re Ex Parte Application of Qualcomm Inc, No 5:16-mc-80005, *1 (ND Cal filed Jan 26, 2016).
• 282Motion of the Commission of the European Communities to Intervene, In re Application of Microsoft Corporation, No 1:06-mc-10061, *1 (D Mass filed Apr 6, 2006); Memorandum of the Commission of the European Communities in Support of Novell, Inc.’s Motion to Quash, In re Application of Microsoft Corp, No 1:06-mc-10061, *1 (D Mass filed Apr 6, 2006). The motion to intervene was denied on the basis that the European Commission’s views were already represented by the other parties. See Microsoft Memorandum and Order at *6 n 4 (cited in note 256).
• 283See In re Application of Schmitz, 259 F Supp 2d 294, 297 (SDNY 2003) (describing the intervention by the German Federal Ministry of Justice); In re Winkler, 2005 US Dist LEXIS 46937, *1–2 (SDNY) (same).
• 284See Kulzer v Esschem, Inc, 390 F Appx 88, 92 (3d Cir 2010) (noting that although the German court at issue could not itself order the sought-after discovery due to German procedural rules, the court “does not restrict receipt of the evidence sought and in fact has postponed a hearing scheduled for April 15, 2010 to September 30, 2010, specifically for the purpose of permitting [the applicant] extra time to pursue its discovery requests in Indiana and Pennsylvania”).
• 285See El Paso Corp v La Comision Ejecutiva Hidroelectrica del Rio Lempa, 341 F Appx 31, 32 (5th Cir 2009) (“The arbitral tribunal issued an order expressing its views on the § 1782 application, noting that it was not receptive to these discovery efforts.”).
• 286See In re Hallmark Capital Corp, 534 F Supp 2d 951, 957 (D Minn 2007) (“[T]he Israeli arbitrator has stated his ‘receptivity’ to this Court’s assistance.”).
• 287See In re Application of Accent Delight International Ltd, 2016 WL 5818597, *2 (SDNY) (discussing that a magistrate judge in Monaco wrote that it was permissible for two corporations to seek discovery in the United States and submit it in their proceeding in Monaco); In re Application of Schmitz, 259 F Supp 2d at 299 (explaining that a presiding judge of a Frankfurt district court filed a letter stating that “[i]f . . . documents from a US–American proceeding are attached to a written statement in the case file, the Court will take notice of this submission”). In In re Application of Schmitz, the Frankfurt judge later changed his mind after finding out that the German Ministry of Justice had opposed it. See Schmitz, 259 F Supp at 299.
• 288See Intel, 542 US at 259 (holding that § 1782 “requires only that a dispositive ruling . . . be within reasonable contemplation”).
• 289See Part III.A.
• 290See Part III.C.1.
• 291See Roger P. Alford, Ancillary Discovery to Prove Denial of Justice, 53 Va J Intl L 127, 143 (2012).
• 292See id. See also Chevron Corp v Naranjo, 667 F3d 232, 236 (2d Cir 2012).
• 293See Republic of Ecuador v Connor, 708 F3d 651, 657–58 (5th Cir 2013).
• 294See Part III.A.
• 295Toward the end of the study period (2013–2017), there were approximately two hundred foreign discovery requests filed per year, see Appendix C, Table 5, whereas there were 292,076 civil filings overall in federal district courts in 2017. See US Courts, Federal Judicial Caseload Statistics 2017, archived at https://perma.cc/UUJ2-FY57.
• 296See, for example, Pamela K. Bookman, Litigation Isolationism, 67 Stan L Rev 1081, 1089, 1115–16 (2015) (describing these doctrines as “avoidance doctrines” and criticizing them for undermining US interests while driving plaintiffs to sue abroad).
• 297See Maggie Gardner, Retiring Forum Non Conveniens, 92 NYU L Rev 390, 394 (2017) (explaining that certain cases dismissed for forum non conveniens were not viewed by the foreign tribunal as an act of comity but rather as an expression of protectionism).
• 298Thanks to Bill Dodge for suggesting this turn of phrase.
• 299European Union Institute for Security Studies, Global Governance 2025: At a Critical Juncture *17 (2010), archived at https://perma.cc/K8NQ-WRKH.
• 300Political scientists have observed that particular issue areas are now governed by “an array of partially overlapping and nonhierarchical institutions,” which can in turn lead to confusion and conflict. Karen J. Alter and Kal Raustiala, The Rise of International Regime Complexity, 14 Ann Rev L & Soc Sci 329, 333 (2018). See also generally Karen J. Alter and Sophie Meunier, The Politics of International Regime Complexity, 7 Perspectives on Polit 13 (2009).
• 301I explore this subject elsewhere. See generally Yanbai Andrea Wang, Procedural Coordination Across Borders (unpublished manuscript, 2020) (on file with author).
• 302PACER is a fee-based “electronic public access service that allows users to obtain case and docket information online from federal appellate, district, and bankruptcy courts, and the PACER Case Locator.” See Administrative Office of the United States Courts, Public Access to Court Electronic Records, archived at https://perma.cc/7GAV-95AX.
• 303See, for example, Engstrom, 65 Stan L Rev at 1214–15  (cited in note 109) (discussing incompleteness in Westlaw’s and Lexis’s databases); Hoffman, Izenman, and Lidicker, 85 Wash U L Rev at 686 (cited in note 109) (same). When I inquired about the exact methodology by which Westlaw gathers its contents, a Westlaw reference attorney explained: “We . . . obtain reported decisions from the court, we have employees who are out acquiring non-published cases constantly and we receive request[s] to add cases from customer and sales reps.” Email from Stephanie Zoet, Academic Account Manager, Thomson Reuters (July 28, 2017) (on file with author). In phone conversations with Westlaw and Lexis representatives and reference attorneys, I was repeatedly told that the exact methodologies by which they compile their respective databases are proprietary information that could not be shared. Both services declined to put me in touch directly with members of their data team.
• 304Engstrom, 65 Stan L Rev at 1215 (cited in note 109).
• 305PACER was initially accessible at terminals in libraries and other designated locations and became available on the Internet in 2001. See Bobbie Johnson, Recap: Cracking Open US Courtrooms (The Guardian, Nov 11, 2009), archived at https://perma.cc/PZ8V-MFGX (noting that PACER was initially accessible at terminals in libraries and other designated locations, and became available on the Internet in 2001); Engstrom, 65 Stan L Rev at 1208 (cited in note 109).
• 306The PACER Case Locator allows searches by case number, title, party name, and date range, but does not allow searches to identify cases for which this information is not known. See Administrative Office of the United States Courts, PACER Case Locator Advanced Case Search, online at https://pcl.uscourts.gov/pcl/pages/search/findCaseAdvanced.jsf (visited May 1, 2019) (Perma archive unavailable).
• 307See Bloomberg Law, Bloomberg Law Docket Coverage, online at https://www.bloomberglaw.com/dockets/coverage/detail (visited May 17, 2020) (Perma archive unavailable) (noting that Bloomberg contains records of proceedings for all federal district courts from 1989 to the present).
• 308Not all documents filed on PACER are text searchable (some are scanned as images), and the same goes for Bloomberg. There does not appear to be any pattern in which documents are text searchable.
• 309See Hoffman, Izenman, and Lidicker, 85 Wash U L Rev at 710 n 138 (cited in note 109) (noting that all opinions in Lexis were present in the Westlaw database, whereas some opinions were in Westlaw but not in Lexis).
• 310See, for example, Order, Whitehead Clan Foreign Trust v Conway, No 2:14-mc-00072, *1 (D Ariz filed Nov 21, 2014) (denying the request because the court could not discern what the plaintiff was requesting).
• 311See Part II.A; Part III.C.1 (discussing refiled cases).
• 312Ossenova Interview (cited in note 53).
• 313Neither I nor OIJA is aware of any pattern in which cases become sealed or have missing documents. Ossenova Interview (cited in note 53).
• 314Like other rule-of-law indexes, such as the Worldwide Governance Indicators, Freedom House’s Freedom in the World Index, and the Heritage Foundation’s Index of Economic Freedom, the World Justice Project’s Rule of Law Index has been criticized for conceptual and methodological problems. Without discounting or overlooking those problems, I use the World Justice Project’s Rule of Law Index as an imperfect way to generate a rough picture of the types of legal systems with which federal district courts are interacting in foreign discovery requests, and whether that picture is different for foreign tribunal versus foreign party requests. See note 135.
• 315Current & Historical Data (World Justice Project), archived at https://perma.cc/S99Y-KY44:
  [T]he construction of the indicators and the underlying survey instruments were slightly revised with the publication of each report during those years. For these reasons, we ask all users to use caution in comparing scores over time, though it can be noted that indicator construction and WJP’s survey instruments have remained relatively stable since 2015, so comparisons can be made with more confidence from 2015–2020. A spreadsheet of the current and historical data can be downloaded at the site.
• 316The year of a case is determined by the day on which it was filed.
• 317I drew a random sample across the entire study period for detailed coding.
• 318Since the random sample was drawn without replacement from a finite population of comparable size, I modeled it as a random draw from a hypergeometric random variable and used this distribution to calculate 95 percent confidence intervals.
• 319There is a very small number of requests associated with both underlying civil claims and related criminal proceedings or investigations. These mixed requests comprise 2 percent of the sample and were distributed across years. Because this Article focuses on civil requests, I include these mixed cases in the civil requests group.
• 320Since the random sample was drawn without replacement from a finite population of comparable size, I modeled it as a random draw from a hypergeometric random variable and used this distribution to calculate 95 percent confidence intervals.
• 321Where a request came from both a tribunal and another entity, I counted it as a tribunal request. Where the request came from both a party and an interested person, I counted it as a party request. I applied these rules because the Supreme Court instructed judges to gauge foreign tribunal receptivity, which judges can do as long as the foreign tribunal is one of the requestors. See Intel, 542 US at 264–65.
• 322To calculate the 95 percent confidence intervals, I first used the hypergeometric distribution to estimate a 95 percent confidence interval for the number of (for example) tribunal requests in the overall population. I then used the hypergeometric distribution a second time to estimate how many requests of this particular sort were made by tribunals. I took a conservative approach and used the lower bound for the number of tribunal requests to calculate the lower bound for the number of requests made by tribunals, etc. This method errs on the side of wider-than-necessary confidence intervals.
• 323Where a request came from more than one tribunal type, it was counted toward both categories, which is why the percentages add up to more than 100 percent. I double counted because each tribunal type is independently significant.
• 324To calculate the 95 percent confidence intervals, I first used the hypergeometric distribution to estimate a 95 percent confidence interval for the number of (for example) tribunal requests in the overall population. I then used the hypergeometric distribution a second time to estimate how many requests of this particular sort were made by tribunals. I took a conservative approach and used the lower bound for the number of tribunal requests to calculate the lower bound for the number of requests made by tribunals, etc. This method errs on the side of wider-than-necessary confidence intervals.
• 325Where the dispute in the foreign proceeding touched on multiple substance areas, I counted them toward all applicable categories.
• 326I calculated these 2.5 percent and 97.5 percent quantiles for the median using bootstrap resampling (ten thousand bootstrap samples per interval).
• 327This table represents the outcome of foreign civil discovery requests in the randomly drawn sample.
• 328This table shows grant rates in the sampled foreign civil discovery requests, calculated as the number of granted requests divided by the number of granted requests plus the number of denied requests. To calculate the 95 percent confidence intervals for tribunal and party requests, I first used the hypergeometric distribution to estimate a 95 percent confidence interval for the number of (for example) tribunal requests in the overall population. I then used the hypergeometric distribution a second time to estimate how many requests of this particular sort were made by tribunals. I took a conservative approach and used the lower bound for the number of tribunal requests to calculate the lower bound for the number of requests made by tribunals, etc. This method errs on the side of wider-than-necessary confidence intervals.
• 329Shannon’s Entropy has been used in varied contexts to measure the diversity of populations. It captures both “richness” (the number of different categories within a population) and “evenness” (the equiprobability of randomly drawing any particular category). The more categories (here, countries) there are in a population, and the more evenly the population is divided across those categories, the higher Shannon’s Entropy.