98717 Right to Information Working Paper Requests and Appeals Data in Right to Information Systems Brazil, India, Jordan, Mexico, South Africa, Thailand, United Kingdom, and United States By Jesse Worker with Carole Excell July 20, 2014 Right to Information Working Paper Series World Bank Stephanie E. Trapnell, Series Editor strapnell@worldbank.org Working Paper No. 1 The Right to Information Series brings forward current and ongoing research on issues related to transparency and the right to information. It aims to provide a range of information on policy, practice, experience, and frontier issues related to public sector openness and transparency, including the underlying functions and outcomes of open government efforts. 1 Right to Information Series The Right to Information Working Papers disseminate the findings of works in progress to encourage the exchange of ideas about issues related to transparency and right to information. Findings are released quickly in order to stimulate discussion and further refine the conclusions of research. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Project Leader and Right to Information Series General Editor: Victoria Lemieux, vlemieux@worldbank.org Working Papers Series Editor: Stephanie E. Trapnell, strapnell@worldbank.org Titles in the Right to Information Series Right to Information: Case Studies on Implementation (2014), edited by Stephanie E. Trapnell Right to Information: Requests and Appeals Data in RTI Systems (2014), by Jesse Worker with Carole Excell Right to Information: Recent spread of RTI legislation (2014), by Toby Mendel Right to Information: Identifying Drivers of Effectiveness in Implementation (2014), by Stephanie E. Trapnell Disclaimer The findings, interpretations, and conclusions expressed herein are those of the author(s) and do not necessarily reflect the views of the Executive Directors of the International Bank for Reconstruction and Development/The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. © 2014 The World Bank, 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000, Internet: www.worldbank.org 2 Requests and Appeals Data in Right to Information Systems Brazil, India, Jordan, Mexico, South Africa, Thailand, United Kingdom, and United States Right to Information Working Paper Series Working Paper No. 1 July 20, 2014 ALL COMMENTS WELCOME vlemieux@worldbank.org strapnell@worldbank.org Contents Executive Summary ............................................................................................................................... 5 Introduction ............................................................................................................................................ 7 Background ........................................................................................................................................... 7 Methodology .......................................................................................................................................... 8 Oversight and Monitoring ....................................................................................................................... 9 Volume of Requests and Rate of Responses ....................................................................................... 12 Data on types of requesters ............................................................................................................. 18 Data on type of information requested .............................................................................................. 18 Data on agencies receiving requests ................................................................................................ 18 Use of Exemptions ........................................................................................................................... 20 Appeals ............................................................................................................................................... 21 Sanctions ............................................................................................................................................. 27 Discussion and Conclusion .................................................................................................................. 28 3 Figure 1: Request Volume and Responses (2012-2013), by Percentage – Treemap View .................. 17 Figure 2: Request Volume and Responses (2012-2013), by Percentage – Bar Chart View ................. 17 Figure 3: Levels of RTI Appeals in Brazil ............................................................................................. 21 Figure 4: Levels of RTI Appeals in India .............................................................................................. 22 Figure 5: Levels of RTI Appeals in Jordan ........................................................................................... 22 Figure 6: Levels of RTI Appeals in Mexico ........................................................................................... 23 Figure 7: Levels of RTI Appeals in South Africa ................................................................................... 23 Figure 8: Levels of RTI Appeals in Thailand ......................................................................................... 24 Figure 9: Levels of RTI Appeals in the United Kingdom ....................................................................... 24 Figure 10: Levels of RTI Appeals in the United States ......................................................................... 25 Figure 11: Brazil: Reasons for appeal .................................................................................................. 27 Table 1: Summary of Available Data .................................................................................................... 11 Table 2: Request Volume Overview ..................................................................................................... 15 Table 3: Disposition (Response) to processed requests ...................................................................... 16 Table 4: Agencies receiving most requests, by country........................................................................ 19 Table 5: Summary of Appeals Data ..................................................................................................... 26 Table 6: Summary of appeals in Brazil, 2013 ....................................................................................... 27 Table 7: Sanctions in law ..................................................................................................................... 28 Acknowledgments The author is deeply grateful to Elizabeth Moses, Shannon Alexander, Patrice McDermott, Saad Filali Meknassi, Somrudee Nicro, and Panicha Vornpien for their help in gathering, translating, and organizing data from websites and annual reports. Thanks also go to the following individuals for their guidance and support in the completion of this report, including the contribution of valuable feedback on various drafts: Victoria Lemieux, Robert P. Beschel, Marijn Verhoeven, Marcos Mendiburu, Luis Esquivel, Toby Mendel, and Anupama Dokeniya. Graham Colin-Jones provided editorial assistance. The generous support of the Nordic Trust Fund is acknowledged as having made this study possible. 4 Executive Summary Without a doubt, it is important to look at the data on requests and appeals to better understand the performance of agencies, shortcomings of the legal framework, and to determine the areas for improvement. Acknowledging high-performing agencies generates positive incentives for better performance, while poor performance can be identified and addressed. Reporting of performance data by oversight bodies is also crucial to the principle of openness, especially in cases where the oversight body is autonomous. Issues of accountability come to the fore when data is not available to all interested parties. This paper provides a summary of data about requests and appeals published by central reporting bodies in eight countries. It was prepared for the World Bank as part of a larger study on the implementation of right to information (RTI) laws. It examines available RTI data about requests and appeals from the most recent year of aggregated data—ranging between 2011 and 2013. It does not evaluate subnational data. It assessed these statistics for Brazil, India, Jordan, Mexico, South Africa, Thailand, The United Kingdom, and the United States. Through this assessment it provides trends in how countries are collecting and publishing these data. Statistics were retrieved from annual reports and through online portals. Online portals were used for Brazil, Mexico, and the United States. The types of data collected include:  Volume of requests and rate of responses  Types of requesters  Type of information requested  Agencies receiving requests  Use of exemptions  Appeals and Complaints  Sanctions It compares the legal requirements for each country with the data that are actually provided. It finds that the length of time that the RTI law has been in force is not related to the level of detail of reporting on RTI requests and appeals, as Brazil, whose law was passed in 2011, discloses as much or more data than the United States (1966), and far more than Thailand (1997). However, legal requirements for data collection and disclosure may have an impact, as Thailand had both the weakest law and practice in this regard. Overall, statistics on requests and responses, the agencies which receive the most requests and the number of appeals were the most commonly published. Oversight agencies rarely report detailed information on why appeals were made and how they were resolved. Statistics on the type of information requested are typically aggregated at the ministry or agency level, revealing only the targets of requests, not the type of information sought. While most countries have time requirements for responding to requests, very few countries publish data on the percentage of responses meeting this requirement. Requesters in Brazil, India and Mexico were the most likely to appeal information refusals. Only South Africa, the United Kingdom, and Brazil made available information on the resolutions to appeals. 5 While Brazil, India, Mexico, South Africa, the United Kingdom, and the United States have legal provisions enabling public officials or authorities to be sanctioned or disciplined for failure to disclose information, only India reported actually applying sanctions during the reporting period. 6 Introduction This paper analyzes data about requests and appeals from Right to Information (RTI) laws in South Africa, the United States, Thailand, Jordan, India, Mexico, Brazil, and the United Kingdom as part of a larger assessment of RTI implementation conducted for the World Bank. It provides background on the national reporting of RTI requests and appeals, including the benefits and limitations, the scope of the data available for the eight countries, and a snapshot of trends in data collection and rates of use. This larger assessment included four in-depth case studies on RTI implementation in South Africa, Thailand, Jordan, and the United States, a refined methodology paper, and this analysis of national data about RTI requests and appeals collected from the eight countries above. Background While Right to Information (RTI) legislation is hardly a new phenomenon--dating back nearly 250 years— two-thirds of national RTI laws have been enacted since 2000.1 In many cases, the effective implementation of RTI legislation has required significant reforms for public agencies, including the creation of new structures, procedures, and systems to enable new practices of records management and information disclosure to take root.2 Disclosing information, maintaining accurate and thorough records, and sharing information with other agencies frequently constitute major challenges to an entrenched bureaucratic culture. In order for RTI laws to achieve their goals of transparency, reduced corruption, greater public participation, better information management, and a more responsive government, it is considered particularly beneficial to designate a central body to collect and publish information, including detailed statistics, about how frequently the law is being used to request information from different public agencies, and how these agencies are dealing with requests. By collecting statistics and tracking implementation of the RTI law, this agency can determine where to prioritize training, or where to enforce compliance through sanctions (if authorized).3 The statistics, if reported accurately and with useful metrics, have the potential to serve as one of the benchmarks for measuring the performance of a country’s RTI system over time. Effective monitoring and oversight requires individual public authorities to submit annual reports that detail what efforts they have made to implement the law and how they have responded to information requests received. However, the types of statistics that are gathered depend upon a variety of factors, including the specificity of the national legal requirements, the performance monitoring system of each agency (and possibly the records management capacity), and the ability of the oversight agency to compel agencies to collect real-time data. In addition, the number and type of agencies that are required to report data is not the same across countries. Some countries may require hundreds of agencies to report, while other countries require less than fifty. Thus, cross-country comparison of data about requests and appeals across countries has limitations to its analytical power. This report will provide a 1 Centre for Law and Democracy, Global RTI Rating Index; Accessed March 10, 2014. http://rti- rating.org/country_data.php 2 Information Commissioners and other oversight mechanisms,”, Right2info.org. http://www.right2info.org/information-commission-ers-and-other-oversight-bodies-and-mechanisms 3 Anupama Dokeniya, “Implementing Right to Information: Lessons from Experience,” The World Bank. 2013 7 snapshot across eight countries of trends in reporting data about RTI requests and appeals and to the extent possible, will make comparisons on the effectiveness of implementation. In some countries including Brazil and South Africa, outside monitoring by freedom of information advocates4 has produced extensive data about requests and appeals, which may illuminate shortcomings in implementation.5 Such data can be invaluable as tools to pressure governments to take action, as was the case recently in Brazil where the Office of Comptroller General announced it would evaluate where implementation was lagging and which institutions were in need of better compliance, following a critical civil society assessment the year before.6 While such compliance testing can be an effective tool to create change, it cannot be expected to substitute for a well-functioning central system for collecting agency data about requests and appeals, as such compliance testing is costly, labor-intensive, and often requires a large scope to produce reliable data. Methodology The categories of statistics that were collected for this report were chosen by project partners based on generally-accepted practice and availability. Data were collected from the designated central body that collects this information, in most cases the national oversight body and in some countries the national nodal agency for RTI,7 but did not include subnational request data. Data collection methods included using web portals to directly download data and the most recent annual reports. Categories include:  Basic overview data, including date of most recent annual report or data collection;  Request and response data, including total annual requests at national level, and responses by type;8  Timeliness of response;  Requester profile information, including type of requestor (individual, business, NGO, media, etc.)  Most requested information types;  Exemption data, including total refusals and most-used exemptions;  Data on appeals, including total numbers, reasons for appeal, and responses;  Sanctions data, including the number of personnel or agencies receiving sanctions for failure to release information, for destroying information, or for inappropriate release of information; and 4 Outside monitoring of requests and appeals often consists of compliance testing. Civil society groups submit and track requests to agencies in order to determine timelines for response, quality of response, and decisions (or lack thereof) by agency officials. In some cases, oversight agencies also conduct compliance testing, but it is a secondary means of collecting data. 5 PAIA Shadow report http://foip.saha.org.za/uploads/images/PCSN_ShadowRep2013_final_20131029.pdf Iv “Brazil making plans to monitor new law,” FreedomInfo; February 17, 2014. http://www.freedominfo.org/2014/02/brazil-making-plans-monitor-new-law/ 7 A nodal agency refers to an agency designated by the government as the agency primarily responsible for overseeing national implementation of the RTI law. Responsibilities typically include capacity building, support and promotion. 8 While requests per capita is sometimes used as a metric to evaluate overall use, its value as an indicator is questionable. First, countries have different standards for what constitutes a request. The U.S., for instance, counts requests for one’s own personal data as an information request, while other countries do not. Additionally, these numbers do not include subnational requests in federal states, which can (in the case of India for instance) far outnumber national requests. The types of agencies that must report also varies across countries, making the pool of reporting bodies not comparable. 8  The number of complaints registered.9 The costs, both incurred and charged to the requester, were not reviewed for this analysis. For the purposes of cross country comparison, this report will rely on data from the most recent collection period, ranging from 2011 to 2013. Oversight and Monitoring The legal responsibility for overseeing compliance with the RTI laws is, in the countries studied, mandated to an existing agency (South Africa, Brazil), an Information Commissioner/Commission (Jordan, United Kingdom, Mexico, India, Thailand), or an Ombudsperson (US Office of Government Information Services). However, in two countries – the UK and the US – the collection of data about requests and appeals has been delegated to a nodal agency rather than the oversight body. The quantity of data collected does not correlate with the length of time that the law has been enacted. The United States (1966) and Thailand (1997) have the two oldest laws of the eight countries but reported less detailed data than Mexico (2003) and Brazil (2011). The following oversight bodies in the eight countries provided aggregated data through annual reports or online portals, as shown in Table 1:  Currently, South Africa’s oversight body is the South African Human Rights Commission (SAHRC). However a new law was passed in 2012 that will create an Information Regulator, which will take over reporting responsibilities from the (SAHRC). This body will have significantly greater authority to enforce compliance, including by hearing appeals against refusals to provide information, something the SAHRC cannot do. Information officers for each public agency are required by law to annually submit data on all requests received, granted in full, partially granted (or refused), or granted in the public interest, requests refused, as well as the exemptions used, and appeals at the agency level (internal) and to a court.10  In Jordan, the Information Commissioner and the Information Council are responsible for collecting and publishing statistics from public agencies. While Jordan’s law was passed in 2007, reports have not been published and some sources suggest data have only been collected since 2012. The requirements on data collection are very general. The law does not even impose a requirement on public agencies to produce annual reports on requests.11  In Thailand, public agencies are required to report on their implementation of the law to the Official Information Commission (OIC), but it is unclear how many agencies are actually doing this, because, as of early 2014, the OIC was not aggregating national statistics and making them publicly available online. Thus while these data should exist, they are not currently being made 9 Appeals are formal requests to a higher authority for a reversal of an official decision. Complaints, on the other hand, are filed for a variety of reasons that require no legal basis, including lack of assistance with filing requests, lack of a designated information officer, denying a fee waiver, destroying a record which had been requested, lack of communication about the status of a request, etc. 10 South Africa Promotion of Access to Information Act 2 of 2000; http://www.dfa.gov.za/department/accessinfo_act.pdf 11 Saad Filali Meknassi,”Assessing the implementation of the Law on Securing Access to Information (no. 47 of 2007) of Jordan,” 2014. 9 publicly available. Additionally, the law is weak on the specifics of reporting standards.12 Perhaps due to these weak standards and practices, the OIC collects the least amount of data of any of the eight countries.  Brazil’s Office of Comptroller General was designated as the oversight agency. It is required by the law to publish detailed annual reports on classified information, information requests and responses, and aggregated statistics on the applicants.13 Of the eight country sample, Brazil’s oversight agency provided the most complete overall data.  In India, the Central Information Commission (CIC) is responsible for collecting and releasing centralized data about requests and appeals. India’ RTI law requires each government ministry or department to compile information and data on RTI requests and submit an annual report to the CIC. The data required included the number of requests and rejection, the exemptions used, fees charged, and sanctions imposed.14  Mexico’s RTI law was widely praised for its establishment of a separate federal institute mandated to oversee RTI implementation. The Federal Institute for Access to Information (IFAI) has many functions including promoting federal compliance, resolving disputes and appeals, capacity- building, and education. IFAI is able to monitor federal agency requests through an online platform called “INFOMEX” through which users can make requests and appeals and which generates centralized data automatically.15 Annual report requirements of agencies include request and response data but do not stipulate how exemptions were used. Exemption data were not available in the IFAI report.16  In the United States, the nodal agency, the Office of Information Policy (OIP) of the Department of Justice, is responsible for central reporting on the law. In 2007, the Office of Government Information Services (OGIS) was formed as part of the National Archives and Records Administration (NARA) to serve as quasi-ombudsperson to support implementation of the law. Chief Freedom of Information Act (FOIA) officers at each agency are responsible for submitting annual reports to the OIP. Agencies are required to submit data on the number of requests, refusals, exemptions used, timeliness in response, and data on requests outstanding.17  The oversight body in the United Kingdom is the Information Commissioner, but this is not responsible for central reporting. Instead, this is done by the Ministry of Justice, currently the nodal agency, although this designation has changed multiple times since the law was passed. The Information Commissioner hears appeals and provides trainings while the nodal agency publishes data about requests and appeals. While statistics are gathered by the Ministry of 12 Centre for Law and Democracy, Global RTI Rating Index; Accessed Mach 10, 2014. http://rti- rating.org/country_data.php 13 Brazilian Access to Information bill approved by the Senate; 2011 14 The World Bank, “Implementing Right to Information: A case study of India,” 2012. 15 A large percentage but by no means all Mexican requests go through INFOMEX. 16 IFAI, 2013. 17 US Freedom of Information Act,7(e)1. http://www.justice.gov/oip/amended-foia-redlined.pdf 10 Justice, there are no legal requirements regarding what must be collected by individual public authorities in the FOIA.18 While over 100,000 public bodies are subject to the Freedom of Information law, statistics are only monitored for 41 major, central government bodies.19 Table 1: Summary of Available Data Note: “N/A” refers to “not applicable”because of the lack of a statutory req uirement for that oversight function. A blank cell indicates that the data were not made available online by the oversight body. Country Year Central reporting body Volume & Agencies Most Appeals Sanctions of Law responses receiving frequently to requests most invoked requests exemptions Brazil20 2011 Office of Comptroller X X X X General India21 2005 Central Information X X X X X Commission Jordan22 2007 Information X X X N/A Commissioner and Information Council Mexico 2002 Federal Institute on X X X Access to Information (IFAI) South 2000 South Africa Human X X Africa Rights Commission Thailand 1997 Official Information X N/A Commission United 2000 Ministry of Justice X X X X Kingdom United 1966 Office of Information X X X X States Policy 18 Freedom of Information Act, 2000. http://www.legislation.gov.uk/ukpga/2000/36/contents 19 Ministry of Justice, Government of UK, “Freedom of Information Statistics: Implementation in Central Government, 2012 Annual”, April 25, 2013. https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/300909/foi-stats-q4-oct-dec- 2012.pdf 20 Brazil Office of Comptroller General; Retrieved March 15, 2014. http://www.acessoainformacao.gov.br/sistema/Relatorios/Anual/RelatorioAnualPedidos.aspx 21 Central Information Commission of India, Annual Report, 2012; http://cic.gov.in/AnnualReports/AR-2011- 12/AR2011-12E.pdf 22 In Jordan, the reports are not made public. This report relies on copies of the 2012 and 2013 reports that were provided to our researchers on an informal basis. 11 Volume of Requests and Rate of Responses The number of requests received by agencies and the rates of response types to those requests constitute basic information about requests. Laws often require specific statistics from public agencies to be reported at least once per year, or an oversight agency that can dictate and enforce reporting requirements. This category of data about requests should contain: 1. Total number of requests 2. Percent granted in full 3. Percent granted in part (or alternatively, refused in part) 4. Percent refused on exemptions 5. Percent refused for non-exemption reasons (with as much detail as possible) 6. Percent outstanding (this may include requests in which the agency has sought clarification from the requester) And additional explanatory data is collected in several other cases: 1. Percent that were transferred to another agency (India, Mexico) 2. Type of information sought 3. Agencies with the most requests 4. Timeliness of response As shown in Table 1, Thailand stands out in this sample for not stipulating the above requirements and not making these basic data publicly available. Brazil, Mexico, the United States and the United Kingdom contained the most categories of data across the most available years. Jordan was the most likely to grant an information request in full, followed by South Africa and Brazil. However, Jordan’s reporting only covers 15 agencies when civil society activists estimate the number of public agencies to be over 120. Furthermore, a large proportion of all of Jordan’s requests were addressed to the Department of Statistics and may not be formal requests under the RTI law.23 India and the United States, which—due not only to the size of their populations but also to high requesting rates—received the most requests and also had the largest backlog of outstanding requests. It is important to note that multiple requests may be made by single individuals—unique requestors were not tracked in any of the countries summarized here. A Note on the presentation of data in this report: The presentation of requests and responses in this report may not match government produced charts and graphs even though all underlying data have been gathered from official government sources. Governments present these data in multiple ways using different and often overlapping categories, without regard for cross-country comparison. In an attempt to standardize the results for comparison across countries, the following adjustments have been made. Wherever possible, “refused requests” 23Saad Filali Meknassi,”Assessing the implementation of the Law on Securing Access to Information (no. 47 of 2007) of Jordan,” 2014. Civil society groups have, for example, questioned whether the Department of Statistics requests really are RTI requests, as they are not made with formal request forms. 12 have been disaggregated to refusals by exemption and all other refusals.24 The category “all other refusals” includes refusals (1) due to the cost of providing the information, (2) because the information does not exist, or (3) for unspecific categories called “unreasonable” or “vexatious” requests. Some countries are unable to process all requests within a year, leading to outstanding requests, or backlogs. This report makes the distinction between requests received during the year and total requests in the queue (including backlogs). Thus, percentages for “granted in full” or “refused on exemption”, are a proportion of the requests processed for that year, not the total requests in the queue. India and Thailand have not been included in the charts on request volume and response because data was either unclear or not available. India’s Central Information Commission only reports total requests, outstanding requests, refusals (unspecific), and transferred requests. Thailand reported no data on requests. For multiple countries, there is a category of “unidentified request results”, which are requests that cannot be accounted for within the official counts of total requests, requests still outstanding, and requests processed. The challenges of interpreting the data for cross-country analysis highlight the uneven reporting and statistical data gathering practices in different countries. Country findings  Brazil’s 2011 law began collecting data in May of 2012. In the first full year (2013), 86,661 requests were reported from 285 agencies, a per capita rate of 0.04%. Of those, a large majority were granted in full (71.6%) or in part (4.0%). Nearly eighteen percent (18%) were refused or unresolved for non-exemption related reasons, including because the information did not exist, the request was incomprehensible, or duplicative requests, among other reasons.25 Requests refused on exemption totaled 6.4%.  Since the passage of RTI legislation in 2005, India’s volume of requests has increased from 24,436 in 2005-2006 to 655,572 in 2011-2012.26 India’s Central Information Commission does not provide data on requests granted; however, in 2011-2012, 8.12% of requests were refused in full and 12.24% were transferred.27 The most striking number may be the backlog of requests, which at the beginning of 2011 stood at 430,425—more than the total requests received during that year. 24 This latter group has been consolidated in a category called “Unresolvable”, or refused for non -exemption reasons. 25 Also included: decisions which were still being made, “unreasonable requests”, requests which were not actually requests or requests to the wrong agencies. (Note: it is unclear if these requests were transferred.) 26 India Central Information Commission (Note: most recent annual report from 2012). These numbers, which just include centralized data from the national government, do not represent subnational requests, which as previous studies have shown, constitute a far greater number. 27 India CIC provides numbers for denials and transfers, leaving the vast majority unidentified. There are no readily available data on granted requests. 13  Jordan’s 2007 law only began reporting data about requests and appeals in 2011-2012. Data in 2013 covered 15 agencies reporting 2,286 requests (0.04% per capita). Of those, 95.6% were granted in full and 4.4% refused. There were no data on why the requests were refused.  Mexico, which has consistently collected request data since the law was passed in 2002, reported 142,766 requests at the national level from January 1st to December 15th in 2013.28 Since 2008, the number of requests has increased 35% overall (from 105,250). At the time of research, 11,193 requests were still outstanding (7.8%). Of those which had been processed, 71.7% of requests were granted in full or the information was already publicly available.29 In 2013, 15.5% of requests were refused for non-exemption reasons30 and 3.7% were refused on exemption. Transferred requests accounted for 9.1% of the total. Mexico had the 2nd highest requests per capita at 0.12%  South Africa’s request numbers stand out because of the high degree of variability in reporting from year to year. In 2012, 185 agencies reported requests, while in 2013, only 108 reported, likely contributing to a 38% drop in reported requests. However, in 2009, with only 104 agencies reporting, the number of requests (24,786) was still marginally higher than 2013 (23,380). In 2013, 85.5% were granted in full, and of those, 3.4% were granted “in the public interest”. Only 1.7% were listed as fully refused, however the rest (12.2%) are not accounted for in the statistics. Of those granted, 3.3% required a time extension. South Africa’s requests per capita stood at 0.05%.  Thailand’s oversight agency did not provide online data on requests at the time of research. The only nationally-aggregated data made available by the OIC were on complaints and appeals.31  The United Kingdom reported 49,464 requests at the national level in 2012 from 41 agencies.32 Of those, 8.2% were still being processed or were outstanding. Of those which had been processed, 43.6% were granted in full and 11.7% were granted in part. Thirty percent were refused because the information was not available, due to cost prohibition, duplicate requests, or other reasons not covered under exemptions. Exemptions caused refusals 14.3% of the time. The UK reported 0.08% requests per capita, however as noted above, only a small slice of agencies who receive requests are monitored.  The United States began FY13 with a backlog of 71,790 requests and reported 704,394 requests across 99 agencies in FY13, making a potential request total of 776,184. However, according to the FOIA.gov online portal, only 678,391 requests were processed in FY13.33 A backlog of 95,564 was reported at the end of FY13, leaving 2,229 with unidentified outcomes. Of those processed, 35% were granted in full and 30% granted in part. While 6.1% were refused for exemption reasons, 28.9% were refused for non-exemption reasons, were deemed unresolvable, were duplicate requests, or were withdrawn.34 While there is a 20-day deadline to respond to requests 28 Request data was provided in late December. Subsequent requests for the final weeks of the year did not receive responses. 29 Of the 94,381 requests which were granted, 5,142 were for information which was already publicly available 30 These included lack of payment or response from applicants (14,074), information which does not exist (5,413), or the category “no proceedings will be taken on request” (848). 31 Thailand Office of the Official Information Commission; http://www.oic.go.th/content_eng/stat.htm 32 Ministry of Justice, Government of the United Kingdom, “Freedom of Information Statistics: Implementation in Central Government, 2012 Annual,” April 25, 2013. https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/300909/foi-stats-q4-oct-dec- 2012.pdf 33 Disposition report, 2013; All agencies. www.foia.gov/data.html 34 Non-exemption refusals include lack of records, improper FOIA requests, duplicate requests, and records held at other agencies. Notably, 10,288 requests were withdrawn (1.5%). 14 by law, the average response times for simple, expedited and complex requests exceeded that deadline by two to sixty-two days.35 Table 2: Request Volume Overview Country Number of Number of Requests Outstanding/backlog % national agencies per capita36 Outstanding requests reporting (year) Brazil 86,661 (2013) 285 0.04% 618 0.7% India 655,57237 66 0.05% 430,425 (April, 2011 – 65.6% March, 2012) Jordan 2286 (2013) 15 0.04% 0 0 Mexico 142,766 247 0.12% 11,193 7.8% (2013)38 South 23,380 (2012- 108 0.05% 0 0 Africa 2013) Thailand N/A39 276 N/A N/A N/A United 49,46440 41 0.08% 757 1.5% Kingdom (2012) United 776,18441 99 0.22 95,564 12% States (2013) Note: N/A stands for “Not Available”, indicating that either the oversight body does not collect these data or has not made them available. In the case of Thailand, data are collected, but OIC does not produce national statistics for public disclosure, other than for appeals and complaints. The CIC in India reports refusal data, but does not provide figures for granted in full or part. Jordan and Mexico do not have a category for granted in part. 35 US Office of Information Policy, “Summary of Annual FOIA Reports Fiscal year 2012,” 2012. 36 Requests per capita: total requests divided by population. Population data based on: http://data.worldbank.org/indicator/SP.POP.TOTL. 37 This number includes backlogged requests from previous years 38 Through December 15, 2013 39 While the OIC reports that 276 agencies submitted data in 2012 in its annual report, it does not make national statistics on requests available 40 Not listed: 26% of the total requests were deemed unresolvable (see Figure 1) 41 While 704,394 requests were received in FY13, 71,790 were outstanding from the previous year when the year began. 15 Table 3: Disposition (Response) to processed requests Country Number of Granted Granted in Refused Unresolved/refused Other processed in full part (%) on for other reasons requests (%) exemptio (year)42 n (%) Brazil 85,705 71.6% 4.0% 6.4% 18% 0 India N/A43 N/A N/A N/A44 N/A Jordan 2,286 95.6% 0 Not Not specified—4.4% specified refused in total —4.4% refused in total Mexico 131,573 71.7% 0 3.7% 15.5% 9.1% South 23,380 85.5%46 0.5% Not Not specified--1.8% 12.2% Africa45 specified- refused in total -1.8% refused in total Thailan N/A N/A N/A N/A N/A d United 48,707 43.6% 11.7% 14.3% 30.4% Kingdo m United 678,391 35% 30% 6.1% 28.9% States Sources: Brazil: Office of Comptroller General; Jordan: Saad Filali Meknassi from Jordan Information Commissioner; Mexico: IFAI, personal communication; South Africa: South Africa Human Rights Commission Annual Report; United Kingdom: Department of Justice Annual Report; United States: FOIA.gov/data 47 42 (Total requests)-(Outstanding requests) 43 Only refusals and transfers are reported in the annual report, making it difficult to ascertain how many were granted and outstanding 44 While the CIC annual report indicates how many exemptions were invoked in total, it is unclear how many refusals were due to exemptions since multiple exemptions could be invoked in one refusal. 45 An additional 12.2% of South Africa’s requests had unidentified dispositions 46 Of this, 3.4% were granted on public interest grounds 47 “Unidentified” refers to those requests in India which have not been refused or transferred, but have not been identified as granted. 16 Figure 1: Request Volume and Responses (2012-2013), by Percentage – Treemap View Figure 2: Request Volume and Responses (2012-2013), by Percentage – Bar Chart View 17 Data on types of requesters Information on the types of individuals or organizations which are requesting information was very limited across the sample. Only Brazil and Mexico, which have online request systems, report this information in national databases. In 2013, 95% of the requesters in Brazil were individuals and nearly 5% were businesses.48 In Mexico, 68% of requesters were individuals, 15% were businesses, 8% were other government agencies, 5% were media, and 4% were civil society organizations. The ability to make a request can vary by law, and is sometimes limited by law to an individual or a citizen. Data on type of information requested Statistics on the type of information being requested can help illuminate how the law is being used and to what effect. In addition, it can help countries identify types of data that it might be useful to proactively disclose. The extent to which data on types of information requested can be useful depends on how the data are collected and presented. In the eight sample countries, these data are highly aggregated and generalized. Only Mexico provided a level of detail in relation to this component, and Brazil also provided some breakdown along these lines. The lack of detailed data about what information was requested limits analysis on what type of information is most frequently requested by country.  Brazil aggregates this information into broad categories. In 2013, financial information was the most requested, followed by government, social security, science, management, preservation, and economics.  Mexico’s most sought after information in 2013 was information such as procedures, statistics and the results of surveys, followed by public institution activities, Personal data, and grant programs. Data on agencies receiving requests Data on the agencies that receive the most requests can indicate that an information request system might require additional staff and resources, and in some cases, may even require a separate unit to process the high volumes of requests. It may also indicate that an agency should investigate which documents can be released as proactively disclosed information in order to reduce the number of requests. In the countries below, central information was provided about the bodies to which requests were directed, but not the type of information requested. Data is captured in Table 4:  India aggregates requests by the body receiving the requests at the Ministry or Department level. In 2012, the Department of Posts received the most requests, followed by Employees Provident Fund Organization, Delhi Police, Bharat Sanchar Nigam Limited, and State Bank of India.  In 2013, Jordan’s Department of Statistics received the most requests, followed by the Royal Jordanian Geographic Centre, the Meteorological Department, the Social Security Corporation, and the National Library. 48 Brazil disaggregates requests into two categories: citizens and companies. 18  In the UK in 2012, requests were not broken down by type of information sought or even in detail by public authority receiving the request; all Departments of State totaled together received the largest number of requests, followed by the Health and Safety Executive, National Archives, Revenues and Customs, and the Charity Commission.  In the United States in 2012 requests were again broken down only by type of public authority, with the Department of Homeland Security receiving by far the most requests, followed by the Department of Justice, Health and Human Services, Department of Defense, and the Social Security Administration. Table 4: Agencies receiving most requests, by country Country Most Second Third most Fourth most Fifth most requested most requested requested sector requested sector requested sector sector sector Brazil N/A N/A N/A N/A N/A India Department of Employees Delhi police Department of State Bank of Posts provident telecommunications India fund (BSNL) organization (social security) Jordan Department of Royal Meteorological Social Security National Statistics Jordanian Department Corporation Library Geographic Centre Mexico N/A N/A N/A N/A N/A South Africa N/A N/A N/A N/A N/A Thailand N/A N/A N/A N/A N/A United Departments Health and National HM Revenue and Charity Kingdom of State Safety Archives Customs Commission Executive United Department of Department Health and Department of Social Security States Homeland of Justice Human Defense Administration Security Services Note: “N/A” indicates that these data were not made available by the oversight agency 19 Use of Exemptions One of the purposes of RTI laws is to change laws, institutions and practice governing information from a “need to know” system to a “right to know” system, emphasizing maximum disclosure.49 However, every RTI law has a set of exemptions which withhold certain types of information from disclosure. 50 Most countries have exemptions for commercial secrets and national security. Other frequent exemptions include privacy, exemptions related to investigations or law enforcement, and information covered by legal privilege. To support the right to information, it is critical that the law defines exemptions narrowly. Procedures may also include harm tests to prove that disclosure will create harm, severance clauses, and clear explanations concerning the legal grounds on which the information is being refused. Even better is to have a provision to balance the public interest against the importance of the interest protected by the exception, particularly in cases where public security, health, and the environment may be impacted. While, Public agencies that have a history of treating information secretively require training, new incentives and possibly structural changes to enable a culture of openness to take hold. Exemption statistics enable an oversight body to better understand, at the very least, how frequently different exemptions are being invoked, and ideally, the types of requests that are triggering exemptions. While appeals may shed light on the misuse of exemptions, they cannot be relied upon if a country does not have a well-established or accessible appeals process. Only four of the eight countries provided nationally aggregated data on the exemptions used. Of those, the most commonly invoked were: privacy, classified information, national security, and finances, which can include commercial confidentiality or management of the economy. Exemption statistics, when provided, typically describe how many times an exemption has been cited, not how many refusals were due to any particular exemption. Multiple exemptions may be cited in one refusal, and exemptions may be cited, but be overruled by public interest tests.  In Brazil, the most frequently used exemption in 2013 was data that was specifically made confidential by a piece of legislation, followed by “unreasonable requests”51, and classified data.  In 2011 in India, the most commonly used exemption was 8(1) or information received in confidence from a foreign government, cabinet papers, or parliamentary privilege. Unfortunately the use of exemptions is not well defined in the annual report.  While Jordan listed 100 refusals based on exemptions in 2013, it did not provide any information on the specific exemptions used.  Mexico refused 2,415 requests based on exemptions in 2013, but did not provide any information on the exemptions.  In the UK in 2012, privacy was the most used exception, followed by “formulation of government policy”, commercial interests, and investigations/proceedings by public authorities.  The most frequently cited exemption in the United States in 2013 was information compiled by law enforcement that could be considered an invasion of personal privacy (7c), followed by personnel and medical files (6)52 and law enforcement procedures and techniques in 49 Anupama Dokeniya, “Implementing Right to Information: Lessons from experience,” The World Bank. 2013 50 “Exceptions” and “exemptions” are used synonymously in the literature 51 In personal communication with the Office of Comptroller General, an example of this was given as architectural plans to a Federal bank. Notably, the use of this exemption increased from one time in 2012 to 434 in 2013. Both of these seem to suggest abuse of the exception and thus illegitimate refusal of the requests. 52 While both exemptions are based on personal privacy, they are distinct exemptions 20 investigations (7e).53 While data provided show how many times exemptions were cited in processing requests, it does not show how many requests were refused by exemption.  South Africa reports exemptions at the local or provincial level, but the data are not aggregated into a national report. Appeals The availability of appeals systems and ways in which requesters can appeal refusals to grant information vary by country. Most countries have multi-stage appeals systems that start with an internal appeal at the agency level. If the requester is not satisfied with the response, the next stage is usually an external appeal at an information commission, a court, an administrative tribunal, or an ombudsperson. These bodies often vary in their authority to enforce binding decisions, with some only having the power to make recommendations. In some other cases, these bodies merely offer mediation services. In addition to the challenges posed by different levels and types of appeals, the statistics on appeals often aggregate appeals numbers and offer little explanation as to the basis for an appeal and how it was resolved. Brazil stood out for having the most disaggregated and detailed data on appeals, including the reasons for the appeals, response time, and numbers disaggregated at the monthly level. Due to the high level of variance in data reported by the countries in the sample, it is difficult to draw conclusions based on the available information. Table 5 below compares the number of appeals to the number of refusals to gain insight into how frequently appeals are being used.  Brazil’s Right to Information Act allows four levels of appeal as shown in Figure 3. Requesters may first appeal to the supervisor of the authority who made the decision within the agency. They can appeal that decision to the highest authority within the agency. After that, they can appeal to Office of the Comptroller General. The final stage of appeal in the Mixed Commission for Information Reassessment. There were 6,177 first appeals in 2013, 5,915 of which were processed. A full table of Brazil’s appeal resolutions for 2013 is provided below. For further details, see Table 6 and Figure 11. Figure 3: Levels of RTI Appeals in Brazil 53 More information regarding US FOIA exemptions can be found here: http://www.foia.gov/faq.html#exemptions 21 Mixed Commission on Comptroller Information General Reassessment Highest authority Within agency within agency  In India, the initial appeal is received internally by an appellate authority within the agency as shown in Figure 4. The next stage is the State or Central Information Commission (depending on the nature of the request). The decisions of these commissions are binding. There were 33,922 appeals in 2011-2012, or 63.5% of total refusals. However, no data have been provided on how they were resolved. An appeal against a decision of the CIC may then be made to the courts. Figure 4: Levels of RTI Appeals in India Supreme Court High Courts State or Central Within Information agency Commission  In Jordan, the Information Council is the first stage of appeals as shown in Figure 5. Jordan recorded 15 appeals in 2013, but no data are available on how they were resolved. Requesters may appeal the decision of the Information Council to the High Court of Justice, but as revealed through interviews with stakeholders, the costs and procedural difficulties have made this avenue impassable so far.54 Figure 5: Levels of RTI Appeals in Jordan High Court of Justice Information Council  In Mexico, as shown in Figure 6, requesters can first appeal an agency’s decision to deny access to the Comite de Informacion located within each agency. Requesters can then submit a second 54Saad Filali Mekassi, “Assessing the Implementation of the Law on Securing Access to Information (No. 47 of 2007)” 22 appeal to the IFAI, which is autonomous and whose decisions are binding. Requesters may appeal the IFAI’s decisions to the federal courts, although federal agencies cannot. There were 7,390 appeals at the national level in Mexico in 2013, totaling 66% of refusals for that year. The IFAI did not report data on the reasons for appeal or their resolution status. Figure 6: Levels of RTI Appeals in Mexico55 Federal courts IFAI Within agency (Comité de Información)  South Africans can appeal to the public body where the request was made and after that, seek to be heard by the courts, as shown in Figure 7. There were 80 appeals in South Africa in 2013 (20% of refusals), nine of which were granted in full (11%). Originally, a requester was required to lodge a complaint in the High Court. In October 2009, rules of procedure were promulgated by which PAIA requests could be enforced in the Magistrates Court. However, because of high court costs, it is rare for an appeal to progress further than the High Court. Figure 7: Levels of RTI Appeals in South Africa Constitutional Court Supreme Court of High Court Appeals Magistrates Courts Within agency  In Thailand, as shown in Figure 8, requesters can appeal an agency refusal to the Information Disclosure Tribunal (IDT) and if unsatisfied with that decision, to the Administrative Court, where the decision is final. In 2012 in Thailand there were 253 appeals made to the IDT, with 51% 55The federal court system in Mexico consists of several levels of the judiciary up through the Supreme Court, with various scopes of jurisdiction to hear cases. These levels of courts are, respectively, Electoral Tribunal, District Courts, Unitary Circuit Courts, Circuit Collegiate Courts, and finally, the Supreme Court of Justice. 23 granted in full and 12% refused. There were no data provided by Thai government for 37% (n=93) of the appeals in 2012.56 Figure 8: Levels of RTI Appeals in Thailand Administrative Court Information Disclosure Tribunal  The United Kingdom has a five-step appeals process, as shown in Figure 9. After the first internal appeal, requesters can appeal to the Information Commissioner’s Office (ICO), then the Information Rights Tribunal, then the Administrative Appeals of the Upper Tribunal and finally the Court of Appeals. However, data retrieved from the Ministry of Justice only reported appeals in 2012 at the first two stages. There were 2,724 appeals for internal review and 351 to the ICO. Of the 2,558 appeals which were processed, internal reviews resulted in the granting of 11% of appeals in full, 14% partially, while 75% were refused. Of the 186 the ICO had processed, it granted 16.7% in full, 15.6% in part, and refused 67.7%. Figure 9: Levels of RTI Appeals in the United Kingdom Court of Appeal Administrative Appeals Information Chamber, Rights Tribunal Upper Tribunal Information Commission Within agency  In the United States, as shown in Figure 10, the first stage of appeals is internal. If that does not resolve the issue, the requester can seek mediation through the Office of Government Information Services—but this is not formally an appeals mechanism. Then, appeals go to District courts. In 2013, agencies started the year with 3,955 unprocessed appeals from previous years and 56 This is based on web-based research and conversations with Panisa Luangvormeth, Director of the Office of Information and Technology Group (OOIP) 24 received 12,028 during the year. Of the combined 15,983, 12,042 were processed in 2013. Of those processed, 5,377 (44.65%) were affirmed, 1,912 were partly affirmed (15.88%) and 1,842 were successfully appealed (15.30%). A remaining 2,911 (24.17%) were given the category “Other reason”, which is not explained on the FOIA web portal. Figure 10: Levels of RTI Appeals in the United States Supreme Court Federal Courts of Appeal Federal District Courts Within agency 25 Table 5: Summary of Appeals Data First level Data Second Data Third level Data Fourth Data of appeal available on level of available of appeal available level of available resolution appeal on on appeal on resolution resolution resolution Brazil Superior Yes Highest Yes Comptroller Yes Mixed Yes authority authority General Comm. (same within on Info agency) agency Reasses sment 57 India Internal No State or No Courts58 No appeal Central Information Commission . Jordan Info No High Court No Council of Justice Mexico Comité de No IFAI Yes Courts59 No Informació n South Internal Yes Courts60 Yes Africa appeal Thailand Info. Yes Administrati (None Disclosure ve Court reported in Tribunal 2012) United Internal Yes Information Yes Courts and None Kingdom Appeal Commission Tribunals61 reported United Internal No Courts62 No States Appeals Notes: “No” indicates that while the oversight agency reported appeals, resolutions to the appeals were not reported. “None reported” indicates that the oversight agency reports appeal resolutions, but did not report any appeals for that stage. 57 Mixed Commission on Information Reassessment 58 After the Information Commission stage, appeals progress through the Courts system of India. These stages are the High Courts, and finally, the Supreme Court. 59 The federal court system in Mexico consists of several levels of the judiciary up through the Supreme Court, with various scopes of jurisdiction to hear cases. These levels of courts are, respectively, Electoral Tribunal, District Courts, Unitary Circuit Courts, Circuit Collegiate Courts, and finally, the Supreme Court of Justice. 60 After the decision at the internal appeal stage, appeals progress through the court system of South Africa. These stages are respectively, Magistrates Court, High Court, Supreme Court of Appeals, and finally, the Constitutional Court. 61 After the Information Commission stage, appeals progress through the Courts and Tribunals System of the United Kingdom. These stages are, respectively, the Information Rights Tribunal, the Administrative Appeals Chamber of the Upper Tribunal, and finally, the Court of Appeals, none of which reported any appeals. 62 After the decision at the internal appeal stage, appeals progress through the court system of the United States. These stages are respectively, Federal District Courts, Federal Court of Appeal, and finally, the Supreme Court. 26 Table 6: Summary of appeals in Brazil, 2013 1st appeal 2nd appeal 3rd appeal 4th appeal Appeals received 6,177 1,934 1,024 99 Appeals 5,916 1,821 966 N/A processed Granted in full 2,534 467 60 N/A Granted in part 490 110 27 N/A Denied 2,762 1,164 424 N/A Ongoing or 130 80 455 N/A dropped Figure 11: Brazil: Reasons for appeal Brazil: Reasons for appeal 12% Information incorrect or incomplete Other 29% 59% Unsatisfactory justification for secrecy Total: 6,177 Source: Brazil Office of Comptroller General, 2013 Sanctions The availability of sanctions for serious failures to comply with the law, whether by individuals or public authorities as corporate entities, has the potential to improve implementation. As shown in Table 7, most RTI laws provide for some sort of sanctions, but experience suggests that sanctions are very rarely imposed in practice.63 Of the eight countries, sanctions of one sort or another are provided for in Brazil, India, Mexico, South Africa, and the United States, but only India reported any sanctions in the most 63 See Thomas M. Susman, Ashwini Jayaratnam, David C. Snowden, Michael Vasquez, Enforcing the Public’s Right to Government Information: Can Sanctions Against Officials for Nondisclosure Work? . Available at: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2295466. 27 recent year of data. Significantly, and unlike in most countries, in India the oversight body can apply sanctions directly, without needing to go to court.  While Brazil’s FOI law provides for sanctions, none have been reported by the Office of Comptroller General.  India recorded 28 sanctions in the most recent year of data. This falls around the median number, though not the mean, because of a record 432 sanctions in 2010-2011 because of 420 officials from Hindustan Copper Limited and STQC who were recommended for discipline under the RTI Act.  In Mexico, disciplinary measures can be taken against officials who obstruct access.  Thailand’s Official Information Act (OIA) does not prescribe sanctions for those agencies that fail to comply with information disclosure, but it does discipline bureaucrats who release information that is deemed to be confidential under exemptions, creating a disincentive to disclose information.  In the United Kingdom, criminal sanctions may be applied for obstruction of the law but the Ministry of Justice does not indicate in its annual report if and how sanctions have been applied. It is believed that sanctions have never been applied in the UK so far.  While the United States FOIA has a sanctions clause, sanctions have never been enforced on an individual employee, although there have been a few cases of departments being required to pay requesters attorney’s fees covered (costs) after losing a court appeal. There were no data on sanctions in the most recent report.  There is no sanctions mechanism in Jordan, Thailand, or South Africa. Table 7: Sanctions in law Sanctions provided in law Scope Brazil Yes Public agent or military64 India Yes Public officials65 Jordan No Mexico Yes Public officials South Africa Yes Public officials66 Thailand No United Kingdom Yes Public Authorities and public officials 67 United States Yes Public Official or employee Discussion and Conclusion The data about requests and appeals presented in this paper demonstrates primarily that the state of data collection and reporting by oversight agencies is far from complete or standard across the eight countries in the sample. In most cases it was impossible to ascertain whether the data collected by central agencies were comprehensive—including all of the government agencies that fall within the 64 Brazil Law 12,527 on Access to Information; 2011. 65 India Right to Information Act, 2005. 66 South Africa Protection of Personal Information Act, 2013. But note that the oversight agency does not have sanctioning authority. 67 United Kingdom Freedom of Information Act, 2000. 28 scope of the law—as this information is not made available in reports or online (e.g. Thailand and Jordan). Brazil provides the most detailed data including tables and charts for comparison, information regarding appeals, their nature, and their resolution. Thailand and Jordan had the least data collected across all the components assessed. While nearly all countries, with the exception of Thailand, made data available on the volume of national requests and the rate of responses, either in an annual report or via an online portal, data on the use of exemptions was less prevalent, as were the reasons for appeals, or how they were resolved. Data about requests and appeals were not available readily online for multiple years and discrete statistics on, for example, the type of agency which received the most requests per year, were not reported consistently in most countries. Data on the type of information requested were almost always aggregated at the agency or Ministry level, with Mexico and to a lesser extent Brazil also providing information about the type of information requested. The length of time since the law has been passed mattered less than the strength of the legal provisions regarding reporting on requests and appeals. The United States and Thailand have the oldest laws, but lagged behind Brazil in regard to the availability of data in several categories. New databases and the development of online portals are a positive improvement in the collection of statistics, and the Brazilian government’s achievement in such a short time is highly commendable. Other noteworthy findings include:  Thailand does not have legal requirements to provide specific statistics to its oversight body;  India’s RTI system is the only one to record sanctions taken against individuals who failed to provide information;  The United States was the least likely to grant an information request in full, and Jordan (with a very small sample size) was the most likely;  Requesters who were refused in Mexico, Brazil, and India, were much more likely to appeal their refusals, possibly indicating a better functioning or more accessible appeals mechanism than in the other countries; and  Brazil’s collection of reasons for appeal was particularly illuminating as it may also be used to highlight areas where exemptions are being misused. Greater transparency around exemption use and appeals may also help improve awareness about the importance of RTI laws and build public trust in using the law. Overall, these statistics, while helpful, are more insightful when reviewed in combination with qualitative data on agency behavior or other surveys on public awareness on RTI. Overly aggregated statistics and lack of clear definitions in reports create limitations for analysis. Oversight and monitoring bodies need to design data monitoring systems and understand the importance of data about requests and appeals to highlight non-compliance and the need for improvement. Oversight bodies in developing systems need adequate systems in order to collect and collate RTI statistics. Overall, the unevenness of reporting in the countries included in this study suggests that a global discussion on RTI reporting could be worthwhile and enable more effective monitoring and implementations of RTI laws within countries in addition to sharing of data on experiences between countries. 29