Troubling Transparency: The History and Future of Freedom of Information

THE FREEDOM OF INFORMATION ACT (FOIA) has been called a “leading legal export” of the United States to the rest of the world.¹ But this is increasingly being challenged as more myth than reality. Although there is some evidence that the U.S. FOIA provided a degree of global inspiration for the very idea of adopting an access to information or right to information (RTI) law, it is less clear to what extent other countries actually followed the substantive approach of FOIA in their own reforms.

Consider, by way of analogy, the much-discussed finding that the global impact of the U.S. Constitution has been declining in recent years.² In explaining the decreasing acceptance of the U.S. Constitution as a model abroad, American legal scholars David Law and Mila Versteeg offer five hypotheses, including “the advent of a superior or more attractive competitor.”³ In this connection, we should ask if the U.S. FOIA is similarly losing influence abroad. Indeed, even the initial assumption that the U.S. FOIA did provide a key template for the laws of other countries needs to be assessed critically.

Some authors have suggested that the American experience with FOIA has served as a touchstone for what is needed to ensure “minimally effective” RTI laws.4 For example, the 1980 Basic Press Act of South Korea included an information access clause that was presumably “copied” from FOIA,⁵ and the New Zealand Official Information Act of 1982 has been described as having been shaped in part by the U.S. FOIA.⁶ In recent years, however, the U.S. FOIA has been criticized for various weaknesses, such as its limitation in coverage to “agencies” of the executive branch. When compared with the seemingly superior modern Swedish law, FOIA has a number of disadvantages, including a lack of constitutional support, lengthier processing times, and higher costs.⁷

Scholars have found that the U.S. FOIA, not the Swedish law, has been most influential as a source of inspiration for RTI reform—so much so that some authors have suggested that “the U.S. system is the one normally taken as a model for contemporary FOI laws.”⁸ But what inspiration did other countries actually use when designing the substance of their laws? Even if the United States, as a powerful nation, was well positioned to promote the idea of adopting RTI laws, did other countries embrace FOIA’s specific features? In this chapter, we seek to shed light on these questions, focusing mainly on the extent to which FOIA has or has not been copied in substantive legal terms rather than on how or why the basic FOI model spread.

To date, although numerous comparative and country-specific analyses have been produced,⁹ they contain little reference to whether foreign laws borrowed from the U.S. law in its particulars. We seek to fill this gap in the literature, relying on empirical evidence to assess the extent to which the U.S. FOIA has been influential abroad as a specific legal model.

We approach this issue through the lens of the Right to Information (RTI) Rating.¹⁰ The RTI Rating, developed by the Centre for Law and Democracy (CLD) and Access Info Europe (AIE), allocates points to the legal framework for the right to information up to a possible maximum of 150 based on sixty-one individual indicators, grouped into seven higher-order categories: Right of Access, Scope, Requesting Procedures, Exceptions and Refusals, Appeals, Sanctions and Protections, and Promotional Measures. This chapter looks at how far other countries have followed the U.S. approach over a selection of key indicators from each category. This gives us an externally verified structure and avoids the trap of looking at these laws using FOIA as the starting point, which would naturally generate a skew in favor of the argument that the U.S. approach has been emulated by other countries.

Our analysis is limited to the content of laws, to the exclusion of practice, for two main reasons. First, studying the former is far more achievable. It is extremely difficult to assess practice even in relatively simple areas such as time limits, let alone, for example, how exceptions are used (or abused). Second, we do not believe practice is a robust metric for assessing legal borrowing. For example, strong rule-of-law countries could be expected to have better track records than weak ones in implementing their laws properly. Yet this has little or nothing to do with the extent to which the United States has served as a model in this area.

For reasons of practicality, not all of the world’s 110-plus FOI regimes will be examined. Instead, we identify seven notable FOI countries from different regions of the world: Canada, Hungary, India, Mexico, Nigeria, South Korea, and Tunisia. The methodology, including the reasons for choosing these countries, is elaborated in the text, but factors include geographic distribution, the strength of the legislation, and variance in terms of performance (strength) on the RTI Rating. In addition to assessing the seven chosen laws against the U.S. FOIA, we offer some normative reflections on FOIA and the ways in which we believe it to be better or worse than other approaches.

RTI is more widely recognized today than ever before.11 Sweden was the first country to grant the public a legal right to access government documents. It was not until 1966, two hundred years after Sweden, that the U.S. Congress passed the federal FOIA, becoming the third country in the world (after Sweden and Finland) to do so. As Mendel observed in 2008, over the previous fifteen years there had been “a veritable revolution” in the RTI field:

Whereas in 1990 only 13 countries had adopted national right to information laws, upwards of 70 such laws have now been adopted globally, and they are under active consideration in another 20–30 countries…. In 1990, the right to information was seen predominantly as an administrative governance reform whereas today it is increasingly being seen as a fundamental human right.¹²

As of March 2017, a total of 115 countries had enacted RTI laws.¹³ The strength of these laws varies substantially, although their overall quality has been increasing over time.¹⁴ Evidence also suggests that stronger RTI laws tend to be better implemented than weaker laws.¹⁵

In celebrating FOIA’s fiftieth anniversary in 2016, the United States could be proud of being an early RTI adopter. It does not necessarily follow, however, that the U.S. law has served as a global standard. As Mendel has written: “As an early adopter, and a country that has introduced a significant number of amendments to its law, one might assume that the United States would be a leader in this area. According to the RTI Rating …, that assumption is wrong.”¹⁶ This section focuses on whether the U.S. FOIA has served as a global template and, if so, in what specific ways various countries have emulated it.

We use the RTI Rating to provide guidance for our comparative assessment because, among other things, it is a recognized external reference point for this purpose. The indicators for the RTI Rating are drawn from international standards, and they are continually applied and assessed by researchers at CLD and AIE subject to local expert review.17 The RTI Rating is a methodology for assessing the strength of the legal framework for RTI; it does not look at how any given law is implemented in practice. It is based on sixty-one separate indicators, grouped into seven overarching categories. Each indicator measures whether or not a certain characteristic is or is not present in the legal system as a whole, not just in one specific legal instrument. As an example, Indicator 4 asks whether everyone, or only citizens of the country, has a right to make a request for information.

The RTI Rating is used in two ways in this study. First, it provides guidance regarding the particular qualities or features of an RTI law on which we focus. Due to space constraints alone, it would not be possible to canvass all sixty-one indicators here. Furthermore, some indicators are less useful in terms of tracking reliance on the U.S. approach. For example, almost all access to information laws set a maximum time limit for responding to requests. Comparing the specific time limits in different laws does not necessarily provide much illumination of the extent to which borrowing might have taken place. In contrast, a requirement to provide assistance to requesters is not a necessary or logical requirement of an access law, so a trend in that area is more suggestive that borrowing has occurred. Based on considerations such as “contribution to making an RTI system effective” and “strong ability to track borrowing,” we have focused on twenty key legal features, some of which straddle more than one RTI Rating indicator.

Second, the RTI Rating provides an initial indication of whether or not a legal feature is present in the United States and the seven comparison countries. Although the information from the rating is important and useful, in most cases the study goes beyond the rating to analyze the specific nature of the feature and whether, if it is present in the comparison countries, it appears to be drawn or extrapolated from the U.S. FOIA or to have other origins.

In terms of the choice of countries, the key consideration was to ensure broad geographic distribution. Beyond that, it was deemed important to focus on countries that have a strong rule-of-law tradition because this is integral to an assessment of true legal borrowing rather than mere symbolic posturing.18 It was also deemed important to avoid simply looking at best-practice countries on the RTI Rating because this might generate bias in the assessment. At the same time, it would not be very useful to review very low-scoring countries insofar as they have specifically opted to adopt weak laws (that is, laws not modeled on the U.S. FOIA). The relative rankings (in parentheses) and the overall scores of the seven focus countries in March 2017 are as follows, in descending order of rank: Mexico (1) 136,¹⁹ India (4) 128, Tunisia (10) 120, Canada (48) 90, Nigeria (52) 88, Hungary (53) 87, and South Korea (60) 82.

Finally, we sought to look at laws adopted over a period of time, starting with the relatively early 1982 Canadian law, followed by Hungary (1991), South Korea (1996), Mexico (2002), India (2005), Tunisia (2011), and Nigeria (2011).

RIGHT OF ACCESS The “Right of Access” category of the RTI Rating comprises three indicators, but for this analysis we focus on what we consider to be the most important one: Indicator 1. Is RTI recognized as a “fundamental” right in a country’s constitutional system?²⁰ Strictly speaking, this is an issue that goes beyond the formal question of the extent to which the U.S. FOIA has served as a model given that FOIA is merely a statute. However, the question of constitutional recognition is fundamentally important in its own right. Furthermore, it affects the way we should understand the text of an RTI law inasmuch as constitutional recognition has consequential implications for the way the access to information law should be interpreted.²¹ For example, constitutional recognition of RTI may have crucial implications for how courts and other decision makers address conflicts between an RTI law and other laws. If the RTI law is understood as giving effect to a constitutional right, lexical superiority will tend to flow from that, at least in certain cases.

In assessing this feature, it should be noted that there has been a significant shift in approaches to this issue since the U.S. FOIA was adopted, at which time almost no one recognized RTI as a fundamental human right. Today, dozens of constitutions recognize the right and, in addition, leading courts in many countries have read it into other constitutional guarantees, most commonly the right to freedom of expression.

The United States is increasingly an outlier in this regard. The U.S. Constitution does not explicitly recognize a right to access information held by public authorities. Furthermore, the U.S. Supreme Court has held that this right cannot be read into the First Amendment right to free speech or any other constitutional guarantee as a general matter.22

In contrast, all seven of the comparison countries have at least some constitutional recognition of the right to information. Three of the countries—Canada, India, and South Korea—are similar to the United States in that their constitutions do not textually recognize RTI as a constitutional right. However, the Canadian Supreme Court recognized RTI relatively recently as part of freedom of expression under the Canadian Charter of Rights and Freedoms, albeit limited to cases where access to information is needed for an expressive purpose.²³ As far back as 1982, the Supreme Court of India read a right to access government-held information into the right to freedom of expression under the Constitution.²⁴ Similarly, in South Korea the right to information was read into the constitutional guarantee of freedom of expression in 1989.²⁵ In the other focus countries—Hungary, Mexico, Nigeria, and Tunisia—RTI is explicitly provided for in the constitution.

In terms of persons, some RTI laws are limited to citizens, whereas more expansive laws do not distinguish between citizens and foreigners. Some laws also fail to cover legal persons. The U.S. FOIA is broad in scope here, covering every individual and also legal persons. Five of the focus countries—Hungary, South Korea, Mexico, Nigeria, and Tunisia—are similarly broad in scope. The two exceptions are Canada and India. Both countries’ laws cover legal entities, but the former is limited to citizens and residents and the latter is restricted to citizens.

Understood as a human right, the right to information should cover all three branches of government, and, indeed, all state actors.²⁶ The U.S. FOIA is in this respect an outlier because it does not apply to the legislature or the courts. In terms of the executive branch, its coverage is relatively broad, although it does not cover the president or his or her staff.

From among the seven focus countries, only Canada follows the U.S. approach by excluding both the legislature and the courts. The other six countries cover both courts and the legislature. This may be, in part, due to the fact that these countries have more robust constitutional recognition of a right to information, which then naturally applies to all branches of government.

Five of the focus countries have broad coverage of the executive branch, although Canada and India do less well here. Canada excludes the cabinet and the prime minister’s office as well as some autonomous federal agencies, and India excludes a range of security and intelligence bodies as well as some research and economic bodies.

The U.S. FOIA does not extend to private bodies that receive significant funding or that perform public functions, a position that is replicated in the Canadian law. This is despite the fact that, as a law review commentator pointed out in 2013, “a slowly accelerating trend can be identified where more recent FOI laws recognize some partial aspects of the right to access corporate information.”27 Mexico, Nigeria, and Tunisia cover both categories of bodies, India and Hungary cover private bodies that perform public functions, and South Korea covers private bodies that receive significant public funding.

REQUESTING PROCEDURES This category is a challenge in terms of assessing borrowing behavior by countries because many of the issues measured by the RTI Rating indicators are fairly technical. Nevertheless, four issues are compared here: (1) whether public authorities are required to provide assistance to requesters, (2) whether they are required to transfer requests to other public authorities when they themselves do not hold the information, (3) whether they are required to provide information in the format sought by the requester, and (4) what fees are charged for making and processing requests.

There are two different aspects to the assistance that might be needed by requesters. First, many requesters need help simply to formulate their requests properly. The United States, Hungary, Nigeria, and South Korea do not require that such assistance be provided, whereas Canada, India, Mexico, and Tunisia do.

Assistance also may be needed due to the fact that requesters are illiterate or disabled. In this case, the U.S. FOIA fails to place any obligation on public authorities. Five of the focus countries include specific obligations in their RTI laws to assist disabled or illiterate requesters, or both, but Hungary and South Korea do not.

The second issue considered here is whether there are statutory obligations or procedures for a public authority to follow if the request is misdirected to an authority that does not hold the requested information. Specifically, is the authority required to notify the requester that it does not possess the information in question and to refer the requester to another office or to transfer the request to where it knows the record is located? Neither of these obligations exists under the U.S. FOIA or the RTI laws of Hungary and South Korea. Only India and Tunisia score full points here. Canada, Mexico, and Nigeria score partial points because they impose an obligation to transfer but the grounds for transfer are broad—making it all too easy for authorities to transfer requests even, in some cases, when they do hold the information.

We examined four key standards relating to exceptions: (1) whether the RTI law overrides other inconsistent laws, (2) the primary approach toward how exceptions are elaborated, (3) whether the exceptions are substantively legitimate under international law (that is, whether they protect interests recognized under international law and are harm-tested in the sense that they apply only where disclosure of the information would pose a risk of harm to the protected interest), and (4) whether there is a public interest override for the exceptions.

Indicator 28 of the RTI Rating, one of “the more controversial standards,”²⁹ states that the “standards in the RTI Law trump restrictions on information disclosure (secrecy provisions) in other legislation to the extent of any conflict.” This is not the case with the U.S. FOIA, which, instead, allows certain other laws to maintain disclosure restrictions that go beyond those in the FOIA, through its third exemption, although some qualifications are placed on its reach.³⁰ Furthermore, this is the only FOIA exemption that is not discretionary.

India, Mexico, and Nigeria all have rules in the RTI law that provide for it to override other laws in case of conflict. This is also the presumed position in Tunisia, although it is not stated as clearly as it might be in the law. A fifth focus country, Canada, takes an interesting position on this issue, allowing only those exceptions specifically listed in a schedule to the RTI law to continue to apply over and above the rules of that law. This means that the legislature must specifically approve those exceptions, which is very different from simply preserving all secrecy provisions. Hungary and South Korea fail to provide for any overriding effect of the RTI law.

In terms of the overall approach toward exceptions, the U.S. FOIA contains nine exemptions, one of which, as noted above, refers to other laws, and seven of which describe specific categories of information. However, the remaining (first) exemption authorizes nondisclosure in situations where a record has been classified, pursuant to an executive order issued by the president, on national security or foreign policy grounds. This effectively grants the president broad powers to determine the scope of this exemption.³¹ All of the other countries set out standards for exceptions that do not depend on decisions by political figures.

Many specific types of exceptions—for example, to protect interests such as national security, trade secrets and confidential commercial information, personal privacy, and law enforcement—are found in most RTI laws. Indicator 29 of the RTI Rating sets out ten categories of exceptions that are deemed to be legitimate under international law.32 The scoring under this indicator is unique inasmuch as it allocates 10 points to countries, and then deducts 1 point for each illegitimate or decidedly overbroad exception. Indicator 30 similarly allocates 4 points for harm testing, and then deducts 1 point for each exception that is not harm-tested.

The U.S. FOIA loses 3 points on each indicator for a score of 8 out of a possible total of 14 here.³³ The exemption for geological and geophysical information relating to wells loses 1 point because it fails to identify an interest that needs the protection of secrecy (and hence is not recognized as a legitimate exception under international standards), and other exemptions are too broad or not harm-tested or both (as with the first exception, noted above).

Canada has the dubious distinction of being the only focus country that has a combined score on these exceptions below that of the United States, scoring a mere 6 points. Hungary ties the United States, also with 8 points. The problematic exceptions in these countries are, however, different from those in the U.S. FOIA. The other five focus countries do significantly better than the United States, all scoring at least 38 percent higher on these indicators, starting with Mexico (13 points); South Korea, Nigeria, and Tunisia (all 12 points); and finally India (11 points).

What has commonly been termed the “public interest override” mandates that even if the disclosure of information would pose a risk of harm to a protected interest, that information should still be disclosed if the benefits of disclosure outweigh the harm.³⁴ The U.S. FOIA provides for what amounts to a kind of public interest override only for two exemptions and even then primarily focuses only on individual privacy as the overriding factor.³⁵

Hungary does not have a public interest override at all, and Canada and South Korea both earn 1 point out of a possible total of 4, like the United States, also due to the limited applicability of their overrides. India and Tunisia provide for broadly applicable public interest overrides, earning full points. Nigeria loses 1 point for excluding one exception from the override, and Mexico similarly loses 1 point because the override is only applied at the appeals level.

The U.S. FOIA allows information requesters a right to request an internal review within the agency by appealing to the head of the agency, but it does not otherwise establish an independent administrative review procedure.36 Nigeria also fails to provide for an independent administrative system of appeals for requesters who believe that their requests for information were not dealt with in accordance with the rules in the RTI law.

All six of the other focus countries have in place a system of administrative appeals. A key issue regarding these bodies is whether or not they are independent from government, given that their function is to review decisions that have been made by a public authority. The appeals bodies in South Korea and Hungary fail to meet even the basic standards of independence and are, instead, government-controlled bodies. The other four countries, in contrast, all score at least 5 points out of a possible total of 6 on the RTI Rating for independence.

To be effective, an RTI administrative appeal or oversight body also needs the powers both to investigate claims of a breach of the law and to issue appropriate remedial orders, such as an order to disclose the information requested. In Canada and India, the oversight bodies earn the full 4 points on the RTI Rating for these two categories of powers, and the four other countries—Hungary, South Korea, Mexico, and Tunisia—earn 3 out of 4 points, indicating reasonably extensive powers.

Finally, under this category, the issue of whether the decisions of the body are binding is considered. This is the case in five of the six countries with administrative oversight bodies, with Canada being the outlier. However, the Canadian Information Commissioner has generally benefited from a strong rate of compliance with her decisions,37 and the government has recently made a commitment to give the office binding order-making powers.

SANCTIONS AND PROTECTIONS Two issues are considered in this category: (1) the presence and nature of sanctions for those who willfully obstruct the right to information and (2) the presence of protections for those who, in good faith, release information pursuant to the law. The U.S. FOIA provides for a sanction against agency personnel for “improperly” withholding agency records.³⁸ A separate law prohibits the destruction of government records.³⁹

Of the seven focus countries, five—Canada, India, Mexico, Nigeria, and Tunisia—impose sanctions on RTI violators in one way or another. Canada and Nigeria are limited to criminal sanctions, which experience in different countries shows are rarely, if ever, applied.⁴⁰ However, India, Mexico, and Tunisia all provide for administrative sanctions (fines and disciplinary measures). A controversial innovation in India is that the various information commissioners are given the power to impose fines. These fines, although small, can be applied on a daily basis, providing an ongoing motivation for compliance. The experience in India, in contrast to the criminal sanction countries, is that such measures have been applied fairly regularly.⁴¹

The second issue here, whether immunity from sanction or prosecution is provided to officials who discharge their duties under the RTI law in good faith, is reflected in the laws of only three countries: Canada, India, and Nigeria.

PROMOTIONAL MEASURES RTI laws need a number of measures to help promote proper implementation, which can broadly be grouped together as promotional measures. Overall, the U.S. FOIA does better in the promotional measures category of the RTI Rating than in any other category.⁴² Here, three promotional measures are reviewed: (1) whether public authorities are required to appoint officials with dedicated responsibilities to implement the law, often called information officers; (2) whether there is a central body with responsibility for promoting implementation, including by raising public awareness; and (3) whether a system of reporting by individual public authorities, and then by a central authority, is in place.

The U.S. FOIA and the laws in six of the seven other countries, with the exception of Hungary, require each public authority to appoint officials with dedicated responsibilities regarding implementation of the RTI law.

Looking at the twenty issues we assessed, and counting cases in which five or more countries followed the U.S. FOIA approach as examples of legal “borrowing,” cases in which five or more countries did not follow FOIA as examples of “diverging,” and cases in which the focus countries split three one way and four the other as “inconclusive,” we obtain the following results:

The U.S. Freedom of Information Act has been described as one of America’s “great creations,” epitomizing the country’s risk-taking conception of self-governance and accountability.43 In theory and practice, Americans have enthusiastically embraced the once radical notion that government information belongs to the public. The United States was not the first to recognize the public’s right to know, however, and it has become increasingly uncertain whether the country should be seen as an international leader or a laggard in the field.

There are reasonably strong indications that the very notion of RTI was significantly popularized by the 1966 adoption by the United States of FOIA, although it should be noted that by the end of 1990 there were only fourteen such laws globally and more serious pick-up of the idea only began in earnest around 1997. Beyond popularizing the general idea of a right to access information held by public authorities, the specific approach to RTI taken by the U.S. FOIA reflects the sociopolitical, economic, cultural, and legal values of American society. Among other things, it is reactive rather than proactive in facilitating access to government records. Other countries, in contrast, especially those that have adopted RTI laws more recently, have opted for greater emphasis on proactive publication in their laws.

Our research suggests that there are a number of objectively discernible similarities between the U.S. FOIA and RTI laws from countries around the world, adopted at different times and demonstrating different levels of strength as measured against international standards. We have not assessed the question of causality regarding these similarities, but it seems reasonable to posit that at least some of the similarities must have been due to borrowing from the FOIA model. At the same time, our research indicates that in relation to many of the more important features of RTI laws, as identified through the RTI Rating, cases of divergence from the U.S. FOIA are far more prominent than those of borrowing from it. These results disprove any notion that FOIA has been transplanted abroad in its totality, and they unsettle the often unquestioned assumption that it had at least a major impact on the design of foreign RTI laws. This is hardly surprising, but our results tend to confirm the view that significant innovation and development in RTI practice have been occurring outside the United States.

One of the more important developments noted in this study since adoption of the U.S. FOIA is the trend toward constitutional protection for RTI. This parallels international developments, whereby RTI has been recognized by authoritative international sources as a human right. Another notable development has been the extension of RTI obligations to a much wider range of state actors than under the U.S. FOIA, including not only the executive but also the courts and the legislature, which also can be seen as reflecting the human rights nature of RTI given that human rights bind all state actors. In terms of exceptions, many of the newer laws, again in contrast to the U.S. FOIA, override other inconsistent laws, either fully or at least in part, and the scope of exceptions has been narrowed in many laws to focus on a smaller range of interests and to provide for harm tests and public interest overrides for most exceptions. Finally, an important innovation has been to provide for an administrative level of appeals, with protections for the independence of the appeals body and necessary powers for this body to undertake its oversight functions.

6. K. J. Keith, “Freedom of Information and International Law,” in Freedom of Expression and Freedom of Information: Essays in Honour of Sir David Williams, ed. Jack Beatson and Yvonne Cripps (Oxford: Oxford University Press, 2000), 349–74, at 355–56.

9. See, for example, John M. Ackerman and Irma E. Sandoval-Ballesteros; “Asia Disclosed: A Review of the Right to Information Across Asia,” ARTICLE 19, 2015, www.article19.org/data/files/medialibrary/38121/FINAL-Asia-Disclosed-full.pdf; Colin J. Bennett, “Understanding Ripple Effects: The Cross-National Adoption of Policy Instruments for Bureaucratic Accountability,” Governance 10 (1997): 213–33; Jamie P. Horsley, “China’s FOIA Turns Eight,” freedominfo.org, April 28, 2016, http://bit.ly/1rlOtgk; Toby Mendel, Freedom of Information: A Comparative Legal Survey, 2nd ed. (Paris: UNESCO, 2008); Ann Florini, ed. The Right to Know: Transparency for an Open World (New York: Columbia University Press, 2007); Michener, “FOI Laws Around the World”; Prashant Sharma, Democracy and Transparency in the Indian State: The Making of the Right to Information Act (London: Routledge, 2015); Alasdair Roberts, “India’s Right to Information Act: The First Four Years,” freedominfo.org, January 13, 2010, http://bit.ly/2bO0KrH.

10. See “Global Right to Information Rating,” www.RTI-Rating.org, accessed February 27, 2017. For a discussion of the RTI rating, see Toby Mendel, “The Fiftieth Anniversary of the Freedom of Information Act: How It Measures Up Against International Standards and Other Laws,” Communication Law & Policy 21 (2016): 468–76.

11. For a comprehensive set of resources on freedom of information in international and foreign law, see “The Global Network of Freedom of Information Advocates,” freedominfo.org, accessed December 19, 2016, http://www.freedominfo.org/resources; “Good Law and Practice,” Right2Info, accessed December 19, 2016, http://right2info.org.

15. Daniel Berliner, “Transnational Advocacy and Domestic Law: International NGOs and the Design of Freedom of Information Laws” (unpublished manuscript, n.d.), 8, citing Gregory Michener, “Assessing Freedom of Information in Latin America a Decade Later: Illuminating a Transparency Causal Mechanism,” Latin American Politics and Society 57 (2015): 77–99.

19. The first number represents the position or rank of the country from among the 111 countries (for example, Mexico was in first position and India in fourth), and the second number represents the total score achieved by the country’s legal framework out of the possible total of 150 points allocated through the RTI Rating (for example, the Mexican legal framework achieved a score of 136 points and India earned 128 points).

20. Conceptually, as a “fundamental” right, RTI parallels what Jan Oster of Leiden University notes in arguing that media freedom should be protected as a fundamental right: “Media freedom should not be made subject to legislation, that is, to the will of temporary political majorities which may impose their values on society. Rather, legislation itself has to be justified in light of media freedom.” Jan Oster, Media Freedom as a Fundamental Right (Cambridge: Cambridge University Press, 2015), 52. Oster prefers “fundamental right” to “human right” because it expresses “more adequately” that media freedom applies not only to human beings but also to legal entities such as media companies. Oster, Media Freedom as a Fundamental Right, 52 n. 170.

21. As Pnina Lahav of Boston University has observed in a related context on the value of an explicit constitutional guarantee of a free press: “the judicial awareness of the liberal justifications of a free press and the acceptability of those justifications as part of the legal argument … may depend on and be encouraged by the formal constitutional commitment to press freedom.” Pnina Lahav, “Conclusion: An Outline for a General Theory of Press Law in Democracy,” in Press Law in Modern Democracies: A Comparative Study, ed. Pnina Lahav (New York: Longman, 1985), 343–44 (emphasis added).

30. “This section does not apply to matters that are … specifically exempted from disclosure by statute (other than section 552b of this title), provided that such statute (A) requires that the matters be withheld from the public in such a manner as to leave no discretion on the issue, or (B) establishes particular criteria for withholding or refers to particular types of matters to be withheld.” 5 U.S.C. § 552(b)(3) (2012).

31. A requested record needs to be “in fact properly classified” under the relevant executive order to pass muster in a legal challenge, but this does not affect the power of the president to set the scope of what is secret under Exemption 1. Furthermore, as Pozen notes in his recent study: “Courts have consistently afforded agencies great deference when classified information is at issue. In most Exemption 1 cases, courts grant the government summary judgment without allowing discovery or performing in camera inspection of the requested records, making it ‘virtually impossible for individual litigants to counter the opinions of agency personnel.’” Pozen, “Freedom of Information Beyond the Freedom of Information Act,” 1118 (citations omitted).

33. Under Indicators 29 and 30, three U.S. FOIA exceptions (for internal personnel records, geological and geophysical information and data, and foreign intelligence records) are inconsistent with the international RTI standards due to their breadth and the absence of a harm test. In addition, the harm test does not apply to the exceptions for national security information, reports by agencies in charge of regulating and supervising financial institutions, or inter-agency or intra-agency memoranda or letters.

36. 5 U.S.C. § 552(a)(6)(A)(i) (2012). The Office of Government Information Services, created in 2007 under the OPEN Government Act, provides coordination and mediation services relating to FOIA disputes between the information requesters and the federal government agencies. See “The Office of Government Information Services,” accessed February 25, 2017, https://ogis.archives.gov.

37. For example, in Canada the target set for recommendations that are complied with is 95 percent. Office of the Information Commissioner of Canada, “Report on Plans and Priorities 2016–17,” accessed March 5, 2017, www.ci-oic.gc.ca/eng/rpp-2016-2017.aspx.

38. 5 U.S.C. § 552(a)(4)(F) (2012). Describing this provision, the U.S. Department of Justice states: “Sanctions may be taken against individual agency employees who are found to have acted arbitrarily or capriciously in improperly withholding records. Additionally, the court must award attorney fees and other litigation costs against the government.” “Sanctions for Violating FOIA,” Office of the United States Attorneys, accessed March 27, 2017, http://bit.ly/2n9tXOp.

40. See Thomas M. Susman, Ashwini Jayaratnam, David Snowden, and Michael Vasquez, “Enforcing the Public’s Right to Government Information: Can Sanctions Against Officials for Nondisclosure Work?,” SSRN eLibrary, December 2012, 1, http://dx.doi.org/10.2139/ssrn.2295466.

Borrowing	5
Diverging	10
Inconclusive	5