98721 Right to Information Working Paper Right to Information: Identifying Drivers of Effectiveness in Implementation By Stephanie E. Trapnell and Victoria Lemieux December 1, 2014 Right to Information Working Paper Series World Bank Stephanie E. Trapnell, Series Editor strapnell@worldbank.org Working Paper No. 2 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 Right to Information: Identifying Drivers of Effectiveness in Implementation Right to Information Working Paper Series Working Paper No. 2 December 1, 2014 ALL COMMENTS WELCOME strapnell@worldbank.org vlemieux@worldbank.org Contents Introduction .................................................................................................................................................. 7 Considering the outcomes of RTI implementation ................................................................................... 7 Effectiveness as Responsiveness ............................................................................................................ 13 Alternative means of exploring effectiveness ........................................................................................ 15 Methodology............................................................................................................................................... 17 Thematic Synthesis ................................................................................................................................. 17 Sampling.................................................................................................................................................. 18 Limitations .............................................................................................................................................. 19 Conceptual Framework ............................................................................................................................... 21 Codes and Fields ..................................................................................................................................... 21 Drivers of Effectiveness........................................................................................................................... 23 Domains of RTI Implementation ............................................................................................................. 24 Findings ....................................................................................................................................................... 26 Enabling conditions ................................................................................................................................. 26 Legal framework for RTI ...................................................................................................................... 26 Advocacy efforts ................................................................................................................................. 28 Policy prioritization ............................................................................................................................. 29 Demand for information ......................................................................................................................... 31 Public Awareness ................................................................................................................................ 31 3 Accessibility ......................................................................................................................................... 32 Institutional capacity............................................................................................................................... 34 Updated, formal practices .................................................................................................................. 34 Staffing levels ...................................................................................................................................... 37 Staff capacity ....................................................................................................................................... 39 Staff incentives .................................................................................................................................... 40 Oversight ................................................................................................................................................. 42 Monitoring of institutional capacity ................................................................................................... 42 Enforcement of disclosure obligations ............................................................................................... 45 Discussion.................................................................................................................................................... 49 Strategic leadership and incentives ........................................................................................................ 49 Barriers to access .................................................................................................................................... 50 Financial and material resources ............................................................................................................ 53 Institutionalization of RTI practices ........................................................................................................ 54 Monitoring responsibilities and institutional arrangements .................................................................. 55 Enforcement mechanisms ...................................................................................................................... 57 Transformative factors: Collaboration and Technology ......................................................................... 58 State-society collaboration ................................................................................................................. 58 Technology .......................................................................................................................................... 60 Intra-governmental collaboration ...................................................................................................... 61 Assessing the drivers of effectiveness ........................................................................................................ 63 Conclusions ................................................................................................................................................. 65 Areas for further research ...................................................................................................................... 65 References .................................................................................................................................................. 67 Appendix A: Codes and descriptions .......................................................................................................... 73 Appendix B: Inter-coder reliability results .................................................................................................. 79 4 Figure 1: Projected outcomes of degree in RTI implementation.................................................................. 8 Figure 2: Use of RTI Requests (from Michener and Worthy 2013: 9)......................................................... 10 Figure 3: Lines of accountability and responsibility that shape incentives ................................................ 49 Figure 4: The virtuous circle of demand for information............................................................................ 50 Figure 5: Location of most commonly available information released by central government (OECD Government at a glance survey 2010) ........................................................................................................ 52 Figure 6: A sample of civil society organizations working on transparency in sample countries (Sunlight Foundation registry).................................................................................................................................... 59 Figure 7: Primary focus area of transparency organizations in sample of countries (from Sunlight Foundation)................................................................................................................................................. 60 Table 1: Orders of change (levels of reform) in organizational change management (Adapted from Kuipers et al 2014: 3) .................................................................................................................................. 11 Table 2: Volume of requests at the national level (Adapted from Worker 2014) ...................................... 14 Table 3: National data on rate of response to processed requests. (Adapted from Worker 2014) ........... 14 Table 4: Description of Thematic Synthesis (adapted from Cruzes et. al. 2014, Pope 2000, and Thomas and Harden 2008) ....................................................................................................................................... 17 Table 5: Characteristics of the 12-country sample (as of 2014) ................................................................. 19 Table 6: Codes emerging during open coding and qualitative content analysis ........................................ 21 Table 7: Mapping of codes into fields ......................................................................................................... 23 Table 8: Global RTI Rating on specific indicators ........................................................................................ 26 Table 9: Global RTI Ratings for RTI Laws (2014).......................................................................................... 27 Table 10: Supporting legal frameworks ...................................................................................................... 28 Table 11: Most common civil society advocacy efforts on RTI issues ........................................................ 29 Table 12: Expression of support for RTI from various stakeholder groups ................................................ 30 Table 13: Public awareness of RTI............................................................................................................... 31 Table 14: Accessibility of request process .................................................................................................. 32 Table 15: Barriers to the appeals process ................................................................................................... 33 Table 16: Existence of updated, formal practices in request processing, records management, and proactive disclosure .................................................................................................................................... 34 Table 17: Proactive disclosure guidance and portals.................................................................................. 36 Table 18: Electronic records management ................................................................................................. 36 Table 19: Staffing levels .............................................................................................................................. 38 Table 20: Agency points of contact for RTI ................................................................................................. 38 Table 21: Staff knowledge of RTI obligations and procedures ................................................................... 39 Table 22: Training providers ....................................................................................................................... 40 Table 23: Existence of poorly aligned staff incentives ................................................................................ 41 Table 24: Monitoring arrangements for implementing rules ..................................................................... 42 Table 25: Monitoring responsibilities performed by public bodies ............................................................ 43 Table 26: Features of data tracking at the country level ............................................................................ 44 Table 27: Levels of appeals process ............................................................................................................ 45 Table 28: Enforcement methods ................................................................................................................ 47 Table 29: Internet access ............................................................................................................................ 53 Table 30: Institutional arrangements for various monitoring tasks ........................................................... 56 Table 31: Records management oversight arrangements .......................................................................... 62 Table 32: Assessment categories and descriptions for drivers of RTI effectiveness .................................. 63 5 Acknowledgments The authors are deeply grateful to the following individuals for their case studies that served as the basis for this report: Albania: Jolanda Trebicka and Gerti Shella India: Mandakini Devasher Surie and Yamini Aiyar Jordan: Saad Filali Meknassi Mexico: Marcos Mendiburu and Yemile Mizrahi Moldova: Sergiu Lipcean and Laura Stefan Peru: Roberto Carlos Pereira Chumbe Romania: Sorin Ionita and Laura Stefan South Africa: Elizabeth Moses Thailand: Somrudee Nicro, Panicha Vornpien, and Nongpal Chancharoen Uganda: Anupama Dokeniya United States: Shannon Alexander with Patrice McDermott All case studies are available in the publication entitled Right to Information: Case Studies on Implementation (2014), edited by Stephanie E. Trapnell. 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: Robert P. Beschel, Luis Esquivel, Toby Mendel, Marcos Mendiburu, Shayamala Shukla, Johanna Suurpaa, and Marijn Verhoeven. The generous support of the Nordic Trust Fund is acknowledged as having made this study possible. 6 Introduction The adoption of laws establishing the right to information (also referred to as access to information or freedom of information) has been extremely active in the past two decades, with 71 such laws having been enacted in the previous fifteen years (Global RTI Rating 2014). In part, this popularity reflects the demand for information about how governments work, what governments do, and what those governments know about their citizens. The right to information (RTI) serves as a means of challenging government secrecy and abuse of power by addressing information asymmetries between government and citizens. But it also serves as the institutionalization of information access, whereby administrative procedural transparency makes information available to the public about government operations that are relevant to the interests of citizens (Dragos et al 2012; Hood 2006; Dror, 1999). RTI is a facet of government transparency, or open government, which is embedded in the administration operations of the public sector, compelling all public officials to consider their roles as caretakers of information, rather than owners. Transparency has been defined and characterized in many ways, but essentially it is an element in the ecosystem of accountability. As a metaphor, it serves as a public value that citizens demand from government (Ball 2009, Greiling and Spraul 2010). It can be understood as an institutional relation of monitoring and oversight, as information exchange over decisions and actions, and as a means of understanding how governments work and what they achieve. (Meijer 2013: 430; Heald 2006: 30). Right to information spans all metaphors as a value common to democratic governance and a procedural tool for access to information. Implementation of the right to information is an integral, foundational factor in the institutionalization of transparency. The implementation of the right to information is realized in the practical components of government administration, i.e., it is established through the daily activities of organizations, particularly within the public sector. But is also influenced through the activities of a range of stakeholders, including citizens, civil society organizations, media, oversight bodies, and politicians. For this reason, robust implementation of RTI systems is challenging. Neuman and Calland (2007) cite lack of leadership, administrative support, and long-term vision as key variables in poor RTI implementation. There are surely many more factors that either cause or hinder effective RTI implementation. In the absence of reliable, consistent data on RTI implementation in public sector management and its governance environment, however, not enough is known about what works and why. The benefits of transparency in government have been celebrated without robust evidence of effectiveness, in part because the right to information is considered an intrinsic good and a human right that should be enjoyed by individuals regardless of its effect on economic growth or political stability. It is also a reflection of the difficulty in defining and measuring effectiveness in the implementation of the right to information. A good starting point for considering the meaning of effectiveness within RTI systems is discussion of its potential outcomes and how those outcomes reflect the achievement of RTI goals. Considering the outcomes of RTI implementation Because of the complex nature of RTI systems, involving actors from government, civil society and independent bodies, the trajectory from laws to social change is neither time-bound nor linear. As such, it is difficult to determine both the nature and timing of outcomes. As a starting point in identifying a 7 theory of change, it may be more analytically useful to propose conceptually logical outcomes that can be linked back to the system in some demonstrable way. Objectives and projected outcomes can vary depending on the stakeholder, i.e., government, civil society, international donors, etc. Because of the diffuse nature of transparency and accountability mechanisms, RTI systems may well achieve any number of outcomes, whether officially envisioned or not. In this discussion, three types of outcomes are envisioned, rooted in a context where accountability and development outcomes are the focus. Categories of first, second, and third-degree as descriptors of outcomes are not necessarily about timing, but rather, are intended to reflect the degree of contribution from RTI implementation process to its outcomes. Outcomes of degree may also be understood as the deepening of transformation within the culture of information in a particular context (district, country, region, etc.) Figure 1: Projected outcomes of degree in RTI implementation First-degree outcomes Second-degree outcomes Information disclosure: Third-degree outcomes Responsiveness to demand for information (rate, quality, and Information usage for timeliness of responses; amount, acccountability: relevance, and regularity of Strategic use of RTI to establish Institutionalization of information proactively disclosed information) accountability measures and access, even if regularly contested: improve operational efficiency (e.g., Shift in bureaucratic culture of anti-corruption preventative and secrecy investigative mechanisms, improved service delivery, etc.) Improved development outcomes: Increased gender equality, standards of living, education and health outcomes, etc. In this context, first-degree outcomes refer to the outputs of the RTI system, and can be expressed as data on how well the system is meeting its mandate to disclose information, e.g. how responsive is the agency to the demand for information? This includes the rate, quality, and timeliness of responses, as well as the amount, relevance, and regularity of proactively disclosed information. Even though this data is collected, issues with data quality and reliability in government administrative systems, and the small size of samples from civil society compliance testing, make its use problematic for assessing effectiveness. 1 It is possible to view first-degree outcomes as an imprecise estimation of demand for information, but only to a small extent. Data on the composition of the group of requesters is needed before inferring increased demand, as the volume of requests may be driven by a small group of urban elite, or a few CSOs or journalists. Indeed, a recent study shows that requests per capita are below 1% for the eight 1 These issues are discussed in detail in the section on Effectiveness as Responsiveness. 8 countries studied, which includes more established regimes in the United States, Mexico, and India (Worker 2014). Second-degree outcomes are about information usage for accountability purposes, i.e., strategic use of RTI to hold government or public officials responsible for their actions and improve operational efficiency. These outcomes answer questions about the extent to which information is being used to hold government to account. These outcomes concerning accountability are a primary rationale for transparency reforms. Transparency initiatives are, in fact, a means of making government more accountable. Accountability encompasses the idea of being responsive, and responsible, about decision- making and activities. 2 Responsiveness is captured by the concept of “answerability,” which refers to the right of citizens to request a response to questions about government decision-making, as well as the obligation of government to provide a response. Responsibility is related to “enforcement,” which is about the capacity to ensure that an action is taken, and provides for access to mechanisms for redress when accountability measures fail (Schedler 1999). Enforcement of accountability writ-large extends beyond right to information oversight functions into horizontal accountability institutions such as the supreme audit institution and anti-corruption agencies that exert pressure on organizations or individuals within government to address grievances. Vertical forms of accountability are advanced by civil society organizations through various means of protest and advocacy, while diagonal accountability arrangements see citizens engaging directly with horizontal accountability institutions through policymaking, expenditure tracking, and participatory budgeting processes (Fox 2014). RTI implementation forms a small, but important, part of these established accountability mechanisms. Basing their understanding of the relationship between transparency and accountability on Besley (2006), Florini (2007), and Carey (2009), Bauhr and Grimes characterize the principal-agent problem at hand: Transparency reduces the problem of information asymmetry by shedding light on the extent to which the agent (i.e., the government) is pursuing goals that are in the interest of principals (i.e., its citizens) effectively and efficiently. In other words, transparency should enable citizens to evaluate to what extent their interests are being served by government and encourage accountability and deter abuses and misdeeds by officials (Bauhr and Grimes 2014: 310). But in certain political contexts that are characterized by pervasive corruption, information disclosure may induce resignation and a withdrawal from civic engagement (Bauhr and Grimes 2014). When there is little chance of reliable sanctions or holding agents accountable, we are left with a collective action problem, whereby the public would benefit from changes in behavior (e.g., disclosure of information) but there is a high cost for the few individuals who move against the tide (e.g., informal penalties, ostracization), making it unlikely that any one individual can or will undertake the task alone. Contextual factors play a significant role in identifying the most appropriate path to accountability or reform. Information about government and its activities must be made available, but does it matter what is done with that information? As shown in Table 1, both the nature of the information request (public, 2 Transparency and accountability are elements of good governance – the effective exercise of authority in the allocation of economic and social resources and the management of government functions. The focus of governance reform is on administration and its processes, rather than political systems of government. 9 private) and its level of politicization can determine whether information is used to hold government in check. Indeed, the typology presented by Michener and Worthy suggests that many requests span both private and public uses, and the information may eventually be used for accountability purposes depending on the content of government response and the context in which the information will be used (2013:10-11). Figure 2: Use of RTI Requests (from Michener and Worthy 2013: 9) The civic uses of open government data suggested by Davies (2010) supports the view that information (even in the form of quantitative data) can be used for a variety of purposes beyond scrutiny/accountability, including campaigning and lobbying, collaborative production of information, planning, and services, and improving the quality of supply and demand of information for economic development. But in a developing country context, particularly when linked to public sector reform, transparency policies are often justified based on their political relevance, i.e., the practicality of information for accountability mechanisms employed by citizens (Michener and Worthy 2013; Darch and Underwood 2010; Fung et al 2007; Curtin and Meijer 2006). It is important to remember that information can be used for a multitude of purposes, regardless of the potential for its role in holding government or public officials in check. Third-degree outcomes include the institutionalization of information access (disclosure as business-as- usual) and development outcomes. Both of these outcomes are often long-term, requiring much more than mere transparency to facilitate positive results. Nearly all countries face ongoing challenges associated with the bureaucratic culture of secrecy that hampers efforts to obtain information. In some countries, officials are apprehensive about sharing information that was once considered secret or that may attract criticism to the agency upon close inspection. In other cases, civil servants believe that the information that they generated is, in fact, their personal possession and will not even share it with colleagues. In many instances, particularly within the political system, information is viewed as power, not as a public good or part of the knowledge 10 commons. Well before the introduction of right to information, these attitudes can become entrenched, making it difficult to instill the values of openness, even in the context of a clear right to information. But often, it is not stubborn reluctance to disclose information that hinders RTI implementation. In many countries, the incentive structures are complex and riddled with inconsistencies, leaving public officials with little guidance as to how they should approach information disclosure. This is the case in Thailand, where there are no incentives to release information when punishments for disclosure are severe (Nicro et al. 2014). In Uganda, officials take an oath of secrecy upon taking office, and are penalized for disclosing secret information (Dokeniya 2014: 291). In these cases, both the personal and institutional incentive structure must be re-arranged to encourage disclosure and signal senior-level policy prioritization for RTI implementation. As with transparency and cultural shifts, the link between transparency and development outcomes is long-term and likely influenced by a number of factors beyond an RTI system. This makes measurement of whether outcomes are achieved a challenging process. Very often, studies depend on correlations and regressions to estimate the relationship between development outcomes and transparency, with indicators for transparency ranging from the existence of an RTI law to open budgeting measurement, to constructed indices of transparency. Theories of change about these processes have yet to be supported with reliable evidence. However, Bellver and Kaufman (2005) have shown that human development outcomes such as life expectancy, female literacy and vaccination coverage are associated with transparency (as an index constructed from a variety of sources), even when controlling for per capita incomes. Fukuda-Parr et al. (2011) show that the open budget index correlates significantly with the human development index, gender-related development index, gender empowerment measure, primary education, and water and sanitation access. However, when differences in per capita income and region are held constant, budget transparency retains a significant statistical association with only a few variables, namely infant and child survival, the percentage of the population using improved drinking water, and public health expenditure levels (Fukuda-Parr et al. 2011: 17-18). The intended outcomes of degree in RTI implementation dealing with public sector reform parallel the orders of change (or levels of reform) found in organizational change management literature, shown in Table 1. Table 1: Orders of change (levels of reform) in organizational change management (Adapted from Kuipers et al 2014: 3) Order Description st 1 : Sub-system • Adaptation of systems or structures • Occurs within part of an organization or sub-system • Is incremental nd 2 : Organization • Transformational change • Movement in core organizational paradigms • Organization-wide • Whole systems change rd 3 : Sector • Identity change • Cross-organizational change • Change spans specific organizational boundaries • Affects many organizations/sector-wide change 11 Applying the orders of change to RTI implementation suggests that first-degree outcomes are reflected in sub-system change, where formal practices for request processing, records management, and proactive disclosure are implemented into administrative operations within departments of organizations. Change is incremental, depending on the resource-constraints facing organizations and the degree of RTI policy prioritization. This kind of sub-system change bleeds into organizational change as information disclosure becomes an accepted practice across the organization. Second-degree outcomes about accountability are more transformational. These outcomes associated with the use of information for accountability purposes have to do with holding government in check and effecting change within organizations and systems. Participatory budgeting is an example of a transparency initiative that can transform core paradigms at the organizational level about funding allocations and policy priorities. The use of information obtained through RTI processes can also be used to unearth corrupt practices, improve service delivery at the municipal or sub-national level, and enhance the efficiency of internal organizational operations. Third-degree outcomes reflect the wide scope of change – identity change that spans organizational boundaries. The institutionalization of information access is about changes in administrative culture that overcome practices in government secrecy, information hoarding, and perverse incentive structures. While development outcomes extend well beyond the borders of the public sector, they can still be reflected in changes at the inter-organizational and sectoral levels. 12 Effectiveness as Responsiveness Second-degree outcomes of information usage, both intended and unpredictable, vary across a spectrum of practical uses and theoretical models, while third-degree outcomes are diffuse and long- term in nature. Assessing the effectiveness of RTI implementation against these kinds of outcomes requires reliable, rigorously-collected empirical data, which is beyond the scope of this study. First- degree outcomes of information access seem the most accessible or practical means of evaluating effectiveness since information access is a logical, if not basic, outcome of RTI implementation (Fox et al 2011). The question then becomes how responsive are government agencies to demand for information? The answers may include a range of variables: rate, quality, and timeliness of responses, as well as the amount, relevance, and regularity of proactively disclosed information. Much of the past and ongoing work in measuring the performance of right to information systems has focused on this responsiveness of the government to information requests (e.g., Calland 2010). It is a viable basis for advocacy work by civil society organizations and interest groups. There have been several cross-country studies and country-specific research on the responsiveness of governments to information requests (Dokeniya 2013; Worthy et al. 2011; Hazell et al.2010; Open Society Justice Initiative 2006; RaaG 2009; Global Integrity Report 2004-2011). These studies employ a variety of methodologies, including case studies and local expert assessments with scoring criteria, both which may involve interviews and desk research of policy reports, media articles, and other sources of information. There has also been compliance or field testing by civil society organizations and donors to determine the costs, timing, and quality of responses from different ministries and government agencies. However, the costs of large-scale cross-country compliance testing are considerable. The Transparency in Governance section of the OECD Government at a Glance survey of member governments is one of the few attempts to collect data on the extent of proactive disclosure. But much of its data is concerned with the location of access, rather than the regularity of updates or relevance to demand. Open data assessment and evaluation of government web portals are two separate lines of inquiry that capture partial data on proactive disclosure, but the results are not geared towards establishing links to RTI implementation process, and have yet to prove useful for this purpose (Open Data Barometer 2013, Open Data Index 2013, Luna et al 2013, Holzer and Manoharan 2007). An additional source is the tracking/reporting data that agencies collect on their own RTI implementation processes. Table 2 provides a snapshot of the requests and appeals data from eight countries, demonstrating that comparison of performance data is problematic at best. The number of agencies reporting ranges from 15 to 285 per country, a gap that does not necessarily correspond to population differences. It may, in fact, reflect different institutional arrangements, reporting rules and compliance and/or differences in the scope of the laws. Though the number of requests submitted varies across countries, per capita request statistics are generally similar at less than 1%, albeit with substantial variance among countries. Given the information that is available, backlogs seem to be much higher in countries where number of requests is higher (India, Mexico, US), but India is clearly an outlier in the number of requests outstanding. 13 Table 2: Volume of requests at the national level (Adapted from Worker 2014) Country Number of Number of Requests Outstanding % national requests agencies per capita /backlog Outstanding 3 (year) reporting Brazil 86,661 (2013) 285 0.04% 618 0.7% India 655,572 66 0.05% 430,425 65.6% (April, 2011 – March, 2012) Jordan 2286 (2013) 15 0.04% 0 0 Mexico 142,766 (2013) 247 0.12% 11,193 7.8% South Africa 23,380 (2012-2013) 108 0.05% 0 0 4 Thailand N/A 276 N/A N/A N/A United Kingdom 49,464 (2012) 41 0.08% 757 1.5% United States 776,184 (2013) 99 0.22% 95,564 12% Table 3 shows that more than half of information requests are being granted in full or in part (range of 55%-85%), with less than 15% of refusals based on exemptions, although only half the countries reported this statistic. In absolute numbers, South Africa’s disclosure rate is higher than that of the United Kingdom and United States. But unqualified comparison of the numbers is not appropriate, as the complexity of requests varies across countries. Detailed data on the content of requests is extremely difficult to obtain, and in fact, is not collected in a majority of countries. 5 By and large, data is disaggregated at the level of ministry, not by the content of request or identity group of requester, e.g., academic, media, private sector, etc. The most commonly collected request data the total number of requests, and the number of requests granted, refused, outstanding, and transferred. 6 Table 3: National data on rate of response to processed requests. (Adapted from Worker 2014) Country Granted in full Granted in part Refused on Unresolved/refused (%) (%) exemption (%) for other reasons Brazil 72% 4% 6% 18% India N/A N/A N/A N/A Jordan 96% 0 N/A Not specified—4% refused in total 7 Mexico 72% 0 4% 16% 8 9 South Africa 86% 0.5% N/A Not specified--2% refused in total Thailand N/A N/A N/A N/A United Kingdom 44% 12% 14% 30% United States 35% 30% 6% 29% 3 Data from sub-national and/or independent agencies is not collected in some countries, leading to substantial differences in the types and number of agencies reporting. 4 While the OIC reports that 276 agencies submitted data in 2012 in its annual report, it does not make national statistics on requests available. 5 The United States categorizes requests on the level of complexity, but does not release data on content or identity group of requesters. 6 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. 7 An additional 9.1% of Mexico’s requests had unidentified dispositions. 8 An additional 12.2% of South Africa’s requests had unidentified dispositions. 9 Of this, 3.4% were granted on public interest grounds. 14 The appeals data collected in Worker’s paper demonstrates more variation but offers much less explanatory information. 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. Some patterns in the appeals data show that data is generally collected only at level of agency and external appeals body. Though appeals are not only about refusals to provide information, the percentage of refusals that are appealed ranges from 2%-66%, and the percent of appeals upheld ranges from 12%-75% (Worker 2014). Appeals provide even less information about the extent of information disclosure, as the nature of appeals is complex, and does not reflect a simple or deliberate refusal to provide information. Decisions depend on the nature of the request, basis for appeal, and case precedent. There is no reason to believe that data on appeals can provide a clear assessment of more or less information disclosure. Analyzing the data on requests and appeals helps provide an understanding of the performance of agencies, shortcomings of the legal framework, and points to areas that need improvement. But this kind of descriptive data does not provide enough information to determine whether information disclosure process is effective. In order to get a clear picture of the reasons for variation in the data, it is necessary to investigate contextual factors in the enabling environment and administrative factors at the agency level. All data must be interpreted in a way that is appropriate for the context, and often this requires additional data collection and analysis. Alternative means of exploring effectiveness Overall comparisons about the effectiveness of countries’ RTI systems across the sample is a difficult task, as contextual factors play a large role in the performance and sustainability of RTI systems. Some countries struggle with legacies of administrative secrecy, while others face serious records management challenges that impact the capacity to respond to requests. There is also significant variation in the extent of RTI implementation within agencies or public bodies within one country, rendering de-contextualized conclusions about “country performance” in the realm of RTI somewhat abstract. In general, RTI systems in India, Mexico, the United States, and the United Kingdom are considered robust, but still facing challenges. India has a vibrant civil society that engages with the RTI system regularly and at all levels and sectors of implementation, yet it still struggles with low levels of capacity within the public sector. Mexico is considered a model RTI system because of its independent and well- funded information commission, which succeeds in enforcing disclosure obligations on public bodies using a variety of methods. The RTI system in the United Kingdom has succeeded in implementing RTI throughout public bodies through its professionalized civil service and the monitoring and enforcement capabilities of an independent information commission. Even though it lacks an information commission, the United States has been implementing RTI for nearly five decades, and its RTI system is considered functional, yet characterized by delays in appeals processing. On the other side of the spectrum are new and struggling RTI systems, where implementation is either slow in taking hold, or has suffered setbacks. Jordan is still in the early phases of implementing RTI within the public sector, and many agencies have yet to develop forms or procedures for requesting access. Uganda faces general challenges with levels of staff capacity and resources within the civil service, while the implementation of Moldova’s RTI system has not been supported by any nodal authority or monitoring body. Public officials in Thailand operate with unclear policies on the kinds of information that can be released, but are subject to severe sanction for release of classified information. 15 In the middle of the spectrum are countries that have implemented RTI with varying degrees of progress, but still face serious challenges. Albania’s RTI system is characterized by informal personal networks within the public sector that substitute for formalized practices, but makes obtaining information through RTI procedures problematic for ordinary citizens. Peru and Romania have engaged civil societies, yet grapple with training, enforcement, and monitoring. South Africa has an active human rights commission that conducts regular evaluations and training for civil servants, but also lacks enforcement authority and faces the challenge of low capacity within the civil service. Though these descriptions of country RTI systems are accurate in that they are factual, they are far from comprehensive or contextualized, and they provide little indication of effectiveness. But there is no quantifiable, reliable measurement of RTI implementation yet available to measure effectiveness as the first-degree outcome of extent of information disclosure. Data on the number of requests, responses, and appeals, and proactively disclosed documents provides a general picture of the volume of requests being processed and information released by administrative systems, and may even capture data on timeliness (Worker 2014). But this type of data offers little information on the quality of responses, relevance of proactively released information to demand, or the satisfaction of users. These latter factors are arguably more important to understanding the extent of disclosure than figures on volume and timeliness, as they provide insight into whether there is pro forma adherence to RTI disclosure obligations, or by contrast, a deeper level of commitment from government to the principles of information access. There is also the question about the reliability of the administrative data being published by countries on their performance in RTI systems. This is not necessarily a reflection of intentional obfuscation on the part of governments, but perhaps, the state of institutional capacity within an agency. Inconsistent adherence to internal tracking systems for requests, as well as variance in the quality of performance monitoring systems within public bodies, may contribute to imprecise data. Other types of monitoring, such as compliance testing by civil society organizations or oversight bodies on the rate of response, are often only feasible for a sample of agencies, as are external checks on the quality of government response. Generalization to the entire set of government agencies with unreliable and inconsistently collected data must be done with qualification. Basing conclusions about effectiveness on administrative data or compliance testing generates a partial understanding of how a system is performing, but is far from complete. In an attempt to encompass a wider frame of study than responsiveness, performance, or the mechanics of implementation, this study focuses on the precursors to implementation effectiveness, i.e., what is preventing or facilitating implementation in practice? This is not a question of whether specific outcomes are achieved, but rather, an inquiry into the drivers of implementation that lead to effectiveness. With this exploratory frame, it is important to study so-called “success stories” in RTI implementation, as well as systems that are struggling with different aspects of implementation. The conclusions in this study about the drivers of effectiveness in RTI systems are based on a synthesis of the successes and constraints to a functioning RTI system that were documented in twelve country case studies. Recurring issues, areas of success, and large and small failures were categorized and grouped so as to generate insight into what matters for RTI effectiveness beyond the parameters of one case. The following section on methodology explains the steps that were taken to identify the drivers of effectiveness using a qualitative approach to data synthesis. 16 Methodology Thematic Synthesis The study employs a thematic synthesis of twelve country case studies that examined the quality and extent of implementation of right to information systems. Its primary aim is to identify drivers of effectiveness for RTI implementation, but consideration of important themes that characterize RTI implementation (e.g., innovations, good practices, challenges, etc.) was a secondary goal. Rather than a reorganization and summary of information that characterizes a literature review, thematic synthesis aims to identify the recurring themes or issues in a collection of primary research, and to generate an analytical understanding that extends beyond the conclusions of the individual cases. Themes are identified through iterative coding of the text, and then organized and analyzed for the purpose of interpretation. Reliability is established through intercoder reliability checks on the codes that are generated and the text that is coded, with the aim of assessing the extent to which independent coders reach the same conclusion when evaluating the same text. The purpose of this method is to develop analytical themes through a descriptive synthesis and find explanations relevant to a particular review question (Ring et al. 2011). Information is often tabulated to allow identification of prominent (or higher-order) themes and offer structured ways of dealing with the data in each theme (Tacconelli 2009). Table 4: Description of Thematic Synthesis (adapted from Cruzes et. al. 2014, Pope 2000, and Thomas and Harden 2008) Key features Purpose Progressive identification of themes to form a chain of reasoning Data sources Findings and interpretations of existing studies and relevant theory Data collection Purposive sampling Process Constructing interpretations Product Conceptual maps and interpretations Steps Code data: Identify and categorize interesting Extract data: Extract data from the primary concepts, findings, and results in a systematic studies, including bibliographical information, fashion across the entire data set aims, context, and results Translate codes into themes, sub-themes, and higher order themes Create a model of higher-order themes: Explore relationships between themes and create a model of higher-order themes. Generate analytical themes: Infer barriers and facilitators from the content of the themes and consider implications Coding involves the assigning of a code (category, theme, idea) to part of a text for the purposes of classification or identification. Codes are tags or labels for assigning units of meaning to the descriptive or inferential information compiled during a study. Codes usually are attached to 'chunks' of varying size— words, phrases, sentences, or whole paragraphs, connected or unconnected to a specific setting. They can take the form of a straightforward category label (Miles and Huberman 1994: 56). 17 Coding can be employed without a priori theoretical considerations, so that codes are inductively created while the researcher is reviewing the text (Glaser 1978, 1992). This is the hallmark of the Grounded Theory Method developed by Glaser and Strauss (1967) and further refined by Strauss and Corbin (1998) and Charmaz (2001). 10 But prior knowledge of a phenomenon can also inform data analysis. Literature reviews are, in fact, “vital tools for developing frameworks and paradigms…which work as scaffolding for the researcher” (Fram 2013: 4). Qualitative content analysis is thus a variant of coding that begins with ex ante categories derived from theory or empirics, but allows for those categories to change as coding progresses (Gläser and Laudel 2013: 22). Relevant text is then extracted and simplified, capturing the content rather than the phrasing or framing of the information, allowing it to be cross-tabulated and compared. In thematic synthesis, coding and extracting overlap during the review process, allowing for inclusion of pre-existing empirical considerations that do not crowd out inductive reasoning. Qualitative data undergoes an iterative and inductive process of constant recoding through the Constant Comparative Analysis method, whereby data is repeatedly compared to other data during coding, with the aim of ensuring that the emerging codes account for all the data being reviewed. This reduces the data so that core categories can emerge (Fram 2013; Charmaz 2001; Strauss 1987; Glaser 1978; Glaser and Strauss 1967). Sampling Purposive sampling was used to select cases, as a form of non-probability sampling in which decisions about the sample of cases were strategic and tied to the objectives of the study. The aim of purposive sampling is not prediction, but interpretative explanation that extends beyond the analysis of each individual case (Thomas and Harden 2008; Doyle 2003). Results are not generalizable to an entire population without qualification. Instead, they contribute to a more sophisticated understanding of the phenomena being studied and serve as a basis for further research. In this context, case studies were initially designed and written with the aim of serving as the basis for a larger qualitative synthesis. A first round of eight case studies was completed in 2012 with a focus on identifying factors associated with implementation effectiveness, using a minimal framework for investigating and organizing qualitative data that is reflected in the content of the studies. A second round of four indicator-driven case studies was conducted in 2014. As part of the second-phase project design, indicators were discussed, vetted, and revised by researchers involved in the project so that indicators could serve as practical guides for data collection in the case studies. Criterion sampling was thus employed for this study, as all cases focused on the implementation of right to information systems as shaped by legal frameworks, public sector practices, and enabling environments, using a variety of data collection methods for the purposes of triangulation of data, e.g., interviews with public officials and civil society organizations, analysis of administrative data, third-party compliance testing and analysis, desk research, and where possible, use of indicators to structure research and analysis. 11 10 Grounded Theory Method is an inductive methodology that involves coding for saturation, where no new properties or dimensions are emerging from continued coding and comparison when additional data is added to the sample. This is considered theoretical saturation, and forms the basis for theory-making (Holton 2010). 11 This report is complemented by three other publications: a volume of the 12 case studies that form the basis for this report (Trapnell 2014), an analysis of data on requests and appeals in eight countries (Worker 2014), and a short report on the spread of RTI legislation (Mendel 2014). http://go.worldbank.org/7IP1PWBFV0 18 Table 5 provides details on the sample of cases from a wide variety of contexts. The range of countries studied was limited to some extent by the existence of RTI laws, which are somewhat concentrated in middle-income and high-income countries. Most of the countries in the sample passed their RTI laws in the late 90s and early 00s, leading to a timeline of about 10-15 years for implementation at the time of the data collection. Exceptions to this rule are the United States - 1966, Uganda – 2005, and Jordan – 2007. Table 5: Characteristics of the 12-country sample (as of 2014) Population GDP per Passage of State Government Political system (millions) capita(US) RTI law Albania 3.2 $4,000 Unitary Parliamentary democracy Parliamentary 1999 India 1236.7 $1,503 Federal Federal republic Parliamentary 2005 Jordan 6.3 $4,909 Unitary Constitutional monarchy Parliamentary-Monarchy 2007 Mexico 120.9 $9,749 Federal Federal republic Presidential 2002 Moldova 3.6 $2,038 Unitary Republic Parliamentary 2000 Peru 30.0 $6,796 Unitary Constitutional republic Presidential 2003 Romania 20.1 $8,437 Unitary Republic Mixed 2001 South Africa 52.3 $7,352 Federal Republic Parliamentary 2000 Thailand 66.8 $5,480 Unitary Constitutional monarchy Parliamentary 1997 Uganda 36.4 $551 Unitary Republic Mixed 2005 United Kingdom 63.6 $38,920 Unitary Constitutional monarchy Parliamentary 2000 United States 313.9 $51,749 Federal Federal republic Presidential 1966 Limitations The data collection and frame for analysis for the underlying case studies were structured by a set of parameters (or indicators) that were informed by prior research and practitioner expertise. However, there is no doubt that some areas could have benefited from a deeper level of inquiry, or may have appeared throughout cases even though they were overlooked in the initial data collection instruments. One of the goals of the qualitative coding process is to capture themes and patterns that were not fleshed out in the initial data collection strategy. Because of the wide scope of investigation with the case studies, abstraction from micro-level cases to a macro level of analysis was challenging, as was the aggregation of information into manageable items for systemic analysis. One suggestion for future work in this area is to narrow the field of inquiry to fewer components, so as to encourage a deeper level of analysis that does not require extensive abstraction to a higher-order. This might be achieved through provincial, sectoral, agency-level or even thematic studies. The case studies that serve as the basis for this report were researched and written up by a variety of authors, with different levels of focus and knowledge, but with extensive experience in studying or working with RTI systems. They brought different skill sets to the analysis, and applied their understandings of what matters for RTI systems to the subject matter, albeit within an analytical framework that required triangulation of data for reliability purposes. Further qualitative or quantitative inquiry into the effectiveness of RTI systems may confirm, clarify, or contradict these findings, but will more than likely build off of the substantial work that has already been accomplished in the case studies. 19 Because the study is based on only twelve country case studies, there remains the possibility that other factors matter for the effectiveness of RTI systems, but they were not captured in the studies, and subsequently in this synthesis report. Generalizations to the universe of RTI systems may not be appropriate, but the goal is to present findings from this qualitative synthesis as a first, rigorous step towards understanding the drivers of effectiveness in RTI systems, even if they are further revised as additional research is conducted. 20 Conceptual Framework Codes and Fields In this synthesis report, both open coding and qualitative content analysis were used to categorize data from twelve country case studies that totaled 650 pages of text. This dual approach reflects the current knowledge base of RTI systems, and serves to identify themes that may not have been made explicit in the individual case studies, but regularly appear across the cases. In terms of agency operations and oversight of RTI systems, empirical investigation has been ongoing for several years. Hence, qualitative content analysis was conducted using ex ante variables for investigation that are based on a growing consensus on the administrative practices that form the basis of request processing, records management, and proactive disclosure, as well as the features of monitoring and enforcement of RTI systems (Carter Center 2014; Fox et al. 2011; Neuman 2009). These codes reflect an understanding of RTI systems as embedded in public administration functions at the agency or ministry level, and requiring centralized government oversight of implementation processes and compliance with legal mandates, along with a multi-level, accessible process for appeals and complaints. Because RTI systems operate within a dynamic ecosystem of transparency and accountability actors, there is less agreement on the external factors that drive success within implementation. Thus, open coding served as the primary basis for identification of themes related to the enabling environment for RTI systems, including but not limited to, the role of civil society organizations, political stability, international pressure, and the domestic demand for information. Table 6: Codes emerging during open coding and qualitative content analysis Access to agency CSO – Advocacy 1 16 31 Monitoring – data 46 Records management appeals Implementation Access to external 2 17 CSO – Advocacy Law 32 Monitoring activities 47 Records oversight appeals Access to request Records-RTI 3 18 CSO assistance 33 Penalties for disclosure 48 process collaboration Performance 4 Agency leadership 19 Data protection 34 49 Request informality evaluations 5 Appeals process 20 Equipment 35 Personnel – turnover 50 Request process Public Information 6 Binding decisions 21 Ethical norms 36 51 Requester profile Officers (PIOs) External appeals 7 Budget 22 37 Political support 52 Resistance process Bureaucratic opposition 8 23 Implementing rules 38 Politicization 53 Responsible units to law 9 Bureaucratization 24 Intelligibility 39 Private sector 54 RTI law Proactive – 10 Civic space 25 Internal tracking - data 40 55 Sanctions dissemination 11 Compensation 26 International funding 41 Proactive system 56 Staff knowledge 12 Complaints 27 International pressure 42 Public consultations 57 State secrets Compliance – informal Intra-agency State-society 13 28 43 Public knowledge 58 tactics cooperation collaboration 21 14 Corruption 29 Judiciary 44 Public outreach 59 Technology Costs – Appeal 15 30 Liaisons 45 Public use of RTI 60 Training deterrent 61 Whistleblower Coding produced an average of 260 references (instances of coded text) per country case. Reliability checks were conducted independently on 5% of all coded text with two coders that were not involved in the project, using the set of codes in Table 7. Inter-coder reliability tests resulted in 92% matching among the coders, with 30% of external coding suggesting additional nuances in interpretation, but no new codes. Once prominent themes were confirmed through an iterative process of reviewing and coding, a conceptual mapping was produced in order to capture the relationships among codes. This mapping of codes into higher-order fields reflects the barriers and facilitators to RTI effectiveness that were captured in the content of the codes. 22 Table 7: Mapping of codes into fields Field: Advocacy efforts Field: Formal practices Field: Policy prioritization Code: Civic space Code: Appeals process Code: Agency leadership Code: CSO – Advocacy Implementation Code: Bureaucratization Code: Bureaucratic opposition to law Code: CSO – Advocacy Law Code: Proactive system Code: Political support Field: Accessibility Code: Records management Field: Public awareness Code: Access to agency appeals Code: Request process Code: Private sector Code: Access to external appeals Field: International intervention Code: Public knowledge Code: Access to request process Code: International pressure Code: Public outreach Code: Corruption Code: International funding Code: Public use of RTI Code: Costs – Appeal deterrent Field: Intra-governmental collaboration Field: Staff capacity Code: CSO assistance Code: Intra-agency cooperation Code: Staff knowledge Code: Intelligibility Code: Liaisons Code: Training Code: Proactive – dissemination Code: Records-RTI collaboration Code: Politicization Code: Request informality Code: Records oversight Field: Staff incentives Code: Requester profile Field: Legal framework Code: Ethical norms Code: Resistance Code: RTI law – implementation Code: Compensation Field: Enforcement Code: Public consultations Code: Penalties for disclosure Code: Binding decisions Code: Whistleblower Code: Performance evaluations Code: Complaints Code: Data protection Field: Staffing levels Code: Compliance – informal tactics Code: State secrets Code: Public Information Officers (PIOs) Code: External appeals process Field: Monitoring Code: Personnel – turnover Code: Judiciary Code: Monitoring activities Code: Responsible units Code: Sanctions Code: Monitoring – data Field: Transformative factors Field: Financial and material resources Code: Implementing rules Code: State-society collaboration Code: Budget Code: Internal tracking - data Code: Technology Code: Equipment Drivers of Effectiveness Fields that contained the highest number of distinct references across the case studies were identified as prominent drivers of effectiveness. They were then organized into domains of RTI implementation that reflect the codes generated through this study, as well as the relationships among those codes and their higher-order fields. Figure 2 presents the drivers of effectiveness as identified in this synthesis report. It is important to point out that the findings from the study suggest that government action and administrative processes are only some of the factors that facilitate effective RTI implementation. Drivers of effectiveness span the range of actors involved (citizens, activists, policymakers, civil servants, political appointees, ministers, civil society organizations, media, firms), and often require collaboration among these actors for successful functioning. 23 Figure 2: Drivers of effectiveness in RTI implementation 1 2 3 4 Enabling Conditions Demand for Information Institutional capacity Oversight Legal framework Public Awareness Updated, formal practices Monitoring of institutional Request processing capacity Advocacy efforts Accessibility of RTI Proactive disclosure processes Records management Enforcement of disclosure Policy prioritization obligations (appeals, Staffing levels sanctions) Staff capacity (training and resources) Staff incentives The drivers also extend across local and national levels of implementation. The domains of both enabling conditions and oversight focus on national-level features of RTI systems, while demand for information bridges national, provincial, and local levels, and institutional capacity centers on the functioning of individual agencies. In fact, all drivers pertain to implementation at various levels and interact with each other in ways that reflect local conditions , e.g., demand for health and medical information may be influenced by accessibility barriers at a central agency that are in turn fueled by a lack of national-level policy prioritization. This suggests that sequencing of reform should be based on an adequate assessment of the drivers and a subsequent interpretation of their relationships to each other in specific circumstances. Sequencing necessarily varies by country context and the presence or absence of drivers of RTI implementation effectiveness. It is beyond the parameters of this synthesis paper but warrants future study. Domains of RTI Implementation The analysis of codes and fields produced a conceptual mapping of four domains of RTI implementation that encompass the drivers of effectiveness. They encapsulate the legal and civic environment in which RTI is implemented, the demand and supply of information, and oversight of the RTI process through monitoring and enforcement. Figure 3: Domains of RTI Implementation 24 1 Enabling Conditions 2 4 Demand for Oversight Information 3 Institutional capacity Enabling conditions inhere both within and outside the public sector, but are not only relevant to right to information. They include a well-designed legal framework, as well as a functioning civil society with the capacity to engage with government and to advocate for reform. Also included are political stability and ongoing policy prioritization from executive or legislative leadership that signals the importance of RTI policies within government agencies. Demand for information is driven by user knowledge of the right to information, the understanding of bureaucratic procedures that serve as gateways to accessing information, and barriers to accessibility that originate within the public sector. It includes rule-based and accessible government processes that allow individuals and groups to engage with the government through regular channels. Institutional capacity refers to the non-specialist functions of bureaucracies, including strategic planning, personnel management, and monitoring progress towards institutional goals. Administrative practices that are grounded in operational transparency lay the foundation for open government. Poor operational performance in RTI is often the result of a combination of factors: resource-constrained environments, combined with a lack of training and employment incentives, resulting in poor performance. Oversight has to do with monitoring activities and enforcement capabilities at the system level. This includes support to the implementation process across government through the provision of guidance, implementing regulations, and training. Enforcement concerns appeals and sanctions, but also encompasses other methods to compel regular and consistent information disclosure by public bodies. 25 Findings Enabling conditions Legal framework for RTI All the country cases confirm that the implementation of RTI is critically influenced by the quality of the laws that establish the entitlement to information. In many cases, the regulatory framework establishing RTI and ensuring its implementation consists of several laws, decrees, or ministerial orders, some of which contradict the principle of disclosure, and some of which support RTI though implementing rules. The latter is particularly relevant if the RTI law exhibits flaws, but would be difficult to amend. Weaknesses in the RTI law can, in fact, lead to a number of implementation problems. Absence of deadlines for responding to information requests, or legally specified deadlines that extend beyond the standard 20-day limit, allow delays to characterize the RTI system across public bodies. In fact, most countries in the sample have a 30-40 day deadline with the option to extend. Table 8 captures scores for the RTI law in areas that were cited as obstacles for implementation in the case studies. Of particular concern were secrecy provisions in other laws and weak or limited exceptions in RTI laws, which allowed information officers to exempt a great deal of information from disclosure. This was noted as a problem in Uganda, Moldova, and Jordan, where state secrets laws trump RTI provisions (Dokeniya 2014: 290; Lipcean and Stefan 2014: 162; Meknassi 2014: 378). In Peru, recently revised declassification timelines trump RTI disclosure provisions, and broad exceptions in the RTI law in Thailand give officials wide discretion in determining what to disclose (Pereira Chumbe 2014: 194; Nicro et al. 2014: 490). Of course, even well-specified exceptions may be abused, but narrowly defined exceptions limit the latitude of officials in exercising discretion. Table 8: Global RTI Rating on specific indicators RTI law Exceptions Assistance in Assistance trumps other consistent w/ 20-day formulating for special No cost or secrecy international deadline to and clarifying needs Fee waivers provisions standards respond requests requesters available* Max Score 4 10 2 2 2 Albania 0 0 0 0 0 Neither India 4 10 1 2 2 Fee waivers Jordan 0 7 1 0 0 No cost Mexico 0 8 1 2 2 No cost Moldova 2 10 1 2 0 No cost Peru 0 10 1 2 0 No cost Romania 1 9 2 1 0 No cost South Africa 4 8 1 2 2 Fee waivers Thailand 0 6 0 0 0 Fee waivers Uganda 2 9 1 2 2 Neither United Kingdom 0 6 1 2 2 Neither United States 2 7 1 1 2 Neither Average 1.3/4 7.5/10 1/2 1.3/2 1/2 * Does not include appeals 26 Table 9 demonstrates that the laws establishing RTI in the sample countries range from a low of 55 to a high of 119, where the maximum score is 150. Countries score relatively well in the categories of right to access, scope, and request procedures, but struggle with the quality of legal provisions governing exceptions, sanctions, appeals, and promotional measures. Exceptions to disclosure obligations include indicators on internationally-accepted exemptions, the harm and public interest tests, severability clauses that allow portions of records to be released, and reasons for refusals. Table 9: Global RTI Ratings for RTI Laws (2014) Promo- Right to Request Scope Exceptions Appeals Sanctions tional TOTAL Access procedures measures Max Score 6 30 30 30 30 8 16 150 Albania 4 27 11 3 18 2 4 69 India 5 25 27 26 29 5 13 130 Jordan 0 25 7 10 8 0 5 55 Mexico 6 22 25 22 26 2 16 119 Moldova 5 28 23 23 17 4 10 110 Peru 4 29 19 17 14 4 8 95 Romania 5 29 17 13 4 6 9 83 South Africa 6 25 21 25 14 6 14 111 Thailand 4 24 14 13 14 2 5 76 Uganda 6 26 23 22 11 5 5 98 United Kingdom 2 25 20 12 23 7 10 99 United States 4 18 19 16 14 4 14 89 Average 4 25 19 17 16 4 9 Table 10 demonstrates that half of the countries either have state secrets laws that supersede RTI laws or have RTI laws with broad exceptions to disclosure, making it difficult for officials to determine what kinds of information can be disclosed, particularly if they will be penalized for violating exemptions requirements. Even when state secrets laws do not contradict or dominate RTI laws, there are still questions about the appropriateness of some exemption policies that maintain secrecy. This pertains both to more advanced RTI systems, such as the United Kingdom and the United States, as well as struggling systems such as Uganda, Jordan, and Thailand. 27 Table 10: Supporting legal frameworks Competing state Public Whistle-blower Data protection/ secrets law or broad consultations protections Privacy exceptions in RTI law Albania No No Yes Yes India Yes No Yes No Jordan No No No Yes Mexico Unknown No Yes No Moldova No No Yes Yes Peru Yes Yes Yes Yes Romania Yes Yes Yes No South Africa Unknown Yes Yes No Thailand No No Yes Yes Uganda Unknown Yes No Yes United Kingdom Yes Yes Yes No United States Unknown Yes No No Prevalence 4 out of 12 6 out of 12 9 out of 12 6 out of 12 At least one-third of countries fail to legislate the requirement for public consultations, whistle-blower protections or data protection. These laws are complementary and supportive of the right to information by allowing civil society input on RTI policies, protecting public officials from releasing information that is in the public interest and exposes wrongdoing, and by protecting individuals from release of sensitive, personal data. Advocacy efforts All case studies cited the efforts of civil society organizations, media groups and journalists, and academics as instrumental in the formation and/or implementation of RTI laws. In fact, lack of involvement in passage of the RTI law in Albania, Jordan, Moldova, and Thailand has resulted in lack of ownership by citizens, and very slow implementation progress due to lack of demand ( Trebicka and Shella 2014; Meknassi 2014; Lipcean and Stefan 2014; Nicro et al. 2014). This is true even in countries with more active post-communist civil societies, such as Albania and Moldova. In over half of the countries, civil society conducts compliance testing on the rates and quality of response from government, which can be compared to self-reported data from administrative system to provide a more accurate understanding of agency performance. 28 Table 11: Most common civil society advocacy efforts on RTI issues Governance Lobbying Involved in Strategic Monitoring of Environment for/against passage of law litigation implementation (CIVICUS amendments 12 score) Albania - + - - 0.58 India + + + + 0.54 Jordan - + + + 0.37 Mexico + + + + 0.51 Moldova - + + - 0.49 Peru + + + + 0.59 Romania + + + + 0.60 South Africa + + + + 0.64 Thailand - + - - 0.41 Uganda + + - + 0.39 United Kingdom + + + + 0.86 United States + + + + 0.82 8/12 12/12 9/12 9/12 0.57 Average CSOs are comparatively less involved in RTI implementation in Jordan, Thailand, and Uganda, which correlates with a lack of civic space for this kind of engagement, indicated by the lower CIVICUS scores highlighted in Table 11. These countries also heavily regulate their NGO communities, which serves as a deterrent to the establishment and activities of civil society organizations. The average CIVICUS score for the sample of cases is 0.57, indicating that the governance environment need not reach the scores of advanced economies, such as the United Kingdom and the United States, in order to predict sustainable advocacy efforts on RTI. Generally speaking, once a country passes a score of 0.50 on the Civicus Index, advocacy efforts are seen to span the range of most common options. Policy prioritization Policy prioritization is comprised of signaling events by high-level officials within government, which consequently impact the strength of strategic and operational leadership of senior public officials engaged in the management of the public sector. Findings from the country cases suggest that political support is a major driver of sustainable and effective implementation. 12 Content of Governance Environment Sub-Index: Civil society infrastructure, Policy Dialogue, Corruption, Political Rights and Freedoms, Associational Rights, Rule of Law, Personal Rights, NGO Legal Context, Media Freedoms. Underlying scores for the aggregated score would be helpful in identifying which areas are more likely to contribute to increased CSO engagement. 29 Table 12: Expression of support for RTI from various stakeholder groups OGP Politicians Ministers Civil Society International aid membership* Albania - - + + + India +/- - + + - Jordan - - + + + Mexico + + + + + Moldova - - + + + Peru +/- +/- + ? + Romania ? - + + + South Africa - - + + + Thailand - - +/- - - Uganda - - + + - United Kingdom +/- + + N/A + United States +/- + + N/A + 5/12 4/12 12/12 8/10 9/12 *OGP = Open Government Partnership +/- = mixed support Table 12 demonstrates that public expressions of support for RTI by politicians and ministers are uncommon across the sample of countries. However, in countries that are considered more robust, e.g., Mexico, the United Kingdom, and the United States, there is not only public support for RTI implementation, but there is also clear evidence of sustained, well-funded initiatives supported by high- level political figures, such as presidents and members of parliament (Mizrahi and Mendiburu 2014; Trapnell 2014; Alexander 2014). In nearly all cases, civil society organizations and media are active in either advocacy or collaborative activities. International support in the form of funding or technical assistance seems to correlate closely with level of GDP, rather than with types of implementation activities. There is a variety of reasons for lack of political support for the right to information, leading to ineffective or absent initiatives that prioritize RTI implementation. Political instability has hindered implementation in Albania and Moldova, and by extension, limited the prioritization of RTI within the public sector (Trebicka and Shella 2014: 11; Lipcean and Stefan 2014: 163). The push for accession into the European Union initially drove implementation in Romania, helped in particular by the publication of league tables that assessed implementation efforts, but political interest has tapered off. In the absence of a monitoring body, implementation has stalled (Ionita and Stefan 2014: 244, 250). The case in South Africa is similar, except that it was post-Apartheid reforms that drove RTI and have now lost steam (Moses 2014: 458). Politicians and bureaucrats in Thailand view RTI as a tool for investigating corruption and ousting officials from government, and thus there is little political support (Nicro et al. 2014: 482, 518). 30 Demand for information Public Awareness Public awareness of RTI influences not only the ability of requesters to access information, but also their understanding of the information rights accorded to citizens. Table 13 identifies countries in which the general public demonstrates an operational understanding of RTI. In country cases where lack of public knowledge of RTI was cited as a barrier to effective implementation, conclusions were based on surveys carried out by government monitoring bodies or civil society organizations (Trebicka and Shella 2014: 29; Devasher Surie and Aiyar 2014: 75; Meknassi 2014: 388; Lipcean and Stefan 2014: 167; Pereira Chumbe 2014: 206; Moses 2014: 426; Dokeniya 2014: 283-84). Table 13: Public awareness of RTI General public demonstrates an operational understanding of RTI Albania - India - Jordan - Mexico + Moldova - Peru - Romania + South Africa - Thailand - Uganda - United Kingdom + United States + 4/12 Public outreach and promotional measures are an overlooked aspect of government responsibility, even when mandated by law. It is important to make citizens aware of their rights if their participation in the information access regime is to be considered a measure of the effectiveness of the RTI system. In the sample of cases, only four countries - India, Mexico, Peru, and South Africa– have mandated government agencies that regularly conduct activities aimed at RTI education and outreach (Devasher Surie and Aiyar 2014:62-64; Mizrahi and Mendiburu 2014: 118; Pereira Chumbe 2014: 197; Moses 2014: 447-48). However, Mexico is the only country of these four with a general public that demonstrates an understanding of RTI. This suggests that public outreach is either not being conducted effectively by government in other countries, or that public outreach activities have less of an effect on public awareness than previously thought. All country cases described CSO engagement (and sometimes media) in public education and awareness-raising on RTI, in some cases this included assistance with formulation and submission of their information requests. 31 Accessibility Beyond awareness of information rights and processes, accessibility of information encompasses structural factors that block access to the RTI request and appeals process and the dissemination of proactively disclosed information. Findings from the case studies revealed a wide scope of instances where accessibility was compromised, with factors ranging from petty corruption to informal networks, lack of information officers, and language differences. One of the most entrenched obstacles to access to information is the legal provision that excludes non-citizens from making requests for information. The legal frameworks in Uganda, Jordan, and Thailand allow only citizens to benefit from the right to information system, excluding the international community, migrants, and immigrants from exercising a right to information (Dokeniya 2014; Meknassi 2014: 379; Nicro et al. 2014: 484). Table 14: Accessibility of request process Accessibility to request process is weak/poor Albania + India + Jordan + Mexico - Moldova ? Peru Varies Romania - South Africa + Thailand - Uganda + United Kingdom - United States - 6/12 Table 14 indicates that requesters in half the countries in the sample face other obstacles to accessibility, which can be compromised by a variety of factors. Petty corruption was cited as a concern in Albania and Uganda, where soft money is used not to expedite requests, but rather to ensure that requests are accepted, with no guarantee of a quality response (Trebicka and Shella 2014: 34; Dokeniya 2014: 301). Personal connections are important again in Albania, but also in Jordan, where informality in request processing is a result of the low number of requests, further exacerbating the problem (Trebicka and Shella 2014; Meknassi 2014). Moreover, in Jordan, there is simply no form available to request information in most ministries, even though it is mandated by law, and a reason for use of the information must be provided (Meknassi 2014: 377). Language difficulties are a prevailing factor in the low demand for information in South Africa, as requesters struggle with not only formulating a request, but also understanding the rules for submitting requests (Moses 2014: 452). Despite the high volume of requests in India, with remarkable successes in securing accountability in some areas, citizens still face a lack of information on the filing process, and they are regularly unable to find contact information for submission at district and local government levels (Devasher and Surie 2014: 75). Providing assistance to requesters is mandated by law or included in policy guidelines for Mexico, Moldova, South Africa, United Kingdom, and the United States, however, the practice of assisting is not 32 uniform or regular (Mendiburu and Mizrahi 2014: 118; Lipcean and Stefan 2014: 164; Moses 430; Trapnell 2014: 329; Alexander 2014: 558). In some countries, such as Albania and Thailand, assistance may be provided as specified by non-RTI laws, but there is no guarantee that RTI requesters will benefit from this practice (Trebicka and Shella 2014: 18; Nicro et al. 2014: 506). Table 15: Barriers to the appeals process Lack of external appeals Lack of accommodation in Lack of internal appeals authority judicial process for non- process (e.g. public body) (e.g. Information Commissioner) expert filers Albania - + + India - - + Jordan + - + Mexico - - + Moldova - + + Peru - + + Romania - + + South Africa - + + Thailand + - - Uganda - + + United Kingdom - - + United States - + + 2/12 7/12 11/12 Nearly all the countries in the sample provide for first instances of appeals at the level of the public body. Because this level of appeal is at the same organization as the initial request, users have a higher likelihood of receiving assistance from officials in the appeals submission process. As shown in Table 15, only five of the countries in the sample specify that appeals (either first or second) are filed through an independent, non-judicial authority, such as an information commission or council. Appeals to these kinds of authorities do not require substantial fees or expertise, the processes are simplified for maximum accessibility, and decisions are often binding. Several countries in the sample face obstacles to an accessible appeals process, for reasons having to do with both accessibility and problematic judicial authority. In Peru, a legal suit is file through the constitutional action of habeas data, which must be maintained over time, and even then, decisions may not be applied as general rules (Pereira Chumbe 2014: 198-99). Appeals in the South African courts are complex and expensive, deterring most requesters from pursuing information disclosure through this option (Moses 2014: 452). Judicial review is such a complex deterrent in Albania that few appeals have ever been filed (citation). The High Court in Jordan often refuses to hear RTI cases, and it is not clear if decisions by the appeals authority are binding on agencies (citation). The judiciary in Uganda is considered to be prone to political influence, and lacks the technical capacity to address RTI issues (Dokeniya 2014: 297-8). In order benefit from the investigatory authority of the public protecter (ombudsman) in South Africa, violation of due process must be established, which is not always the reason for failure to disclose information (Moses 2014). 33 Institutional capacity Updated, formal practices Internal rules and administrative regulations are important for incorporating legal obligations into agency processes and activities. They are present in most or all agencies in only three countries in the sample (Mexico, United Kingdom, United States). These rules lay the groundwork for institutionalization of RTI practices. Formalization here refers to the extent to which agencies have institutionalized their legal obligations into formal practices, procedures, and institutional arrangements that support the basic functions in an RTI system at the agency level. It does not refer to the boundary-setting of case law on appeals, such as is the case in Romania, where laws are operationalized through the courts and non- compulsory guidance is ignored (Ionita and Stefan 2014: 255). Table 16 indicates that most agencies do not benefit from updated, formal practices in all three areas of RTI-related operations. Records management is by far the least formalized or updated of all practices, likely because it is behind the scenes. Proactive disclosure practices are also poorly formalized, and only half of the sample cases enjoy some measure of updated and formal practices in information request processing. Table 16: Existence of updated, formal practices in request processing, records management, and proactive disclosure Request Proactive Records process disclosure management Albania - - - India - - - Jordan - - - Mexico + + - Moldova - - ? Peru ? ? - Romania + - - South Africa + - - Thailand + ? ? Uganda - - - United Kingdom + + + United States + + + 6/12 3/12 2/12 Request process Informal requests for information have always been prevalent among some user groups, such as journalists and civil society organizations, as they are expedited means of obtaining information. There is no reason that informal requests of this nature pose a problem for formalized information disclosure. However, some countries struggle with overcoming a tradition of informal networks in the information request process, or petty corruption in front-line offices, that served to exclude and reinforce 34 established power dynamics. In this sample, Albania, Jordan, Uganda face challenges of that kind (Trebicka and Shella 2014: 34; Meknassi 2014: 388; Dokeniya 2014: 300). Formalization of the information request process is important as a means of providing access beyond a specialist group of users. Informality in the requesting process precludes reliable tracking of activities, as requests are not registered, departments holding information may not be consulted, and the basis for appeals is pre- empted. Even though formal requesting procedures may lengthen the process of obtaining information, they are essential for institutionalizing the process in the organization, rather than with a few individuals. But care must be taken not to overly bureaucratize the process of information request. Without proper guidance on the methods to evaluate information requests, public officials fail to distinguish between RTI-based requests and routine information inquiries. Requests that are treated as routine information inquiries are not subject to RTI deadlines and do not benefit from the protections under law, including appeals and assistance. In other cases, such as in Albania, Moldova, Romania, and Uganda, the formal request process involves several levels of management, and official disclosure often requires the approval of the secretary-general of the ministry (Trebicka and Shella 2014: 34; Lipcean and Stefan 2014: 164; Ionita and Stefan 2014: 252; Dokeniya 2014: 305-6). This over-bureaucratization motivates public officials to treat requests unofficially, as the time and effort involved with processing an official request is overwhelming. Although this tactic may be useful in terms of efficiency of response, it can be abused and result in discarded requests or informal refusals that do not provide a basis for appeal. The combination of extensive informality in practices or a lack of clear distinction between normal business operations and RTI requests leads to poor tracking of requests, such as in Moldova and Jordan. Most countries allow electronic means of submitting formal requests (online and email), but require an official form to be completed and sent by requesters. Electronic submission (as opposed to email requests) is available in five countries in the sample – United States, United Kingdom, Thailand, Mexico, and India – but it is not equally available across agencies. In the case of the United States and Thailand, electronic submission of requests is developed by the agencies and available on agency websites (Alexander 2014; Nicro et al. 2014). In contrast, Mexico and India have a centralized portal for the submission of requests that is used for access to information for any agency in the country (Mizrahi and Mendiburu 2014; Devasher Surie and Aiyar 2014). Proactive disclosure In Romania, it is estimated that only 50 percent of central institutions release information on their websites, with only 25 percent of local institutions doing so (Ionita and Stefan 2014: 258). Albanian ministerial and cabinet decisions, as well as legal amendments and policies, are sometimes not even published in paper form, even though electronic dissemination has been available since 2005 (Trebicka and Shella 201; 25-26). Only two countries in the sample use centralized portals for proactive disclosure (Mexico, Peru), and it is not clear whether this approach is optimal. Centralization of information disclosure avoids problems with less user-friendly agency websites and low technical skills of local agencies. But coordination and technical costs for uploading materials from various parts of the country may outweigh any benefits to citizens for central access, particularly if technical support for the centralization process is not sustainable (Pereira Chumbe 2014: 200). 35 Table 17: Proactive disclosure guidance and portals Guidance provided / Central Web Portal for Categories of documents proactive disclosure to be disclosed Albania - - India + - Jordan - - Mexico + + Moldova - - Peru + + Romania - - South Africa + - Thailand - - Uganda - - United Kingdom + - United States + - 6/12 2/12 Online submission allows for automatic recording and tracking of requests, but country cases noted that investment in design and maintenance are required to ensure sustainability. Mexico’s portal also serves as internal tracking of all requests and responses. This has enabled the information commission and others to identify patterns and trends in information flows that contribute to improved practices, and in particular, performance monitoring. In Thailand, the ministry of environment tracks requests for information and proactively releases information with high request rates (Nicro et al. 2014). Mexico has created a searchable database for all requests and responses, allowing for tracking of highly popular information and identification of information that should be proactively released (Mizrahi and Mendiburu 2014). Records management All but three countries in the sample have issued some guidance on electronic records management, but the level of detail and compliance in this guidance is not clear. Table 18: Electronic records management Guidance exists on electronic records management Albania - India + Jordan - Mexico + 36 Moldova + Peru + Romania + South Africa + Thailand + Uganda - United Kingdom + United States + 9/12 Electronic document management systems are inconsistently implemented even in the strongest RTI systems. In the United States, for example, emails are still printed out in hard copy and scanned into the system, rather than filed electronically directly into the system (Alexander 2014: 585). In weaker and less-resourced systems, emails are not even considered manageable, because of the lack of resources to print and scan. Records management in Romania suffers from a lack of integrated data management, making it difficult to access even current information within the filing systems (Ionita and Stefan 2014: 251-252). E-government processes are implemented by IT staff without concerns of RTI or integrated records management, and information cannot be linked across systems for appropriate categorization and storage. Records management authorities provide much needed guidance on the introduction of electronic management systems within the public sector. In cases where records management is not given high priority, or resources are constrained, manual filing processes are still employed, and storage facilities are poorly organized and insecure. Staffing levels Table 19 indicates that four cases specifically noted that staffing levels were a concern for effective RTI implementation. In Jordan, Moldova, Thailand, and Uganda, demand for information is so low that there the number of staff assigned RTI responsibilities is sufficient to respond to requests (Meknassi 2014; Lipcean and Stefan 2014; Nicro et al. 2014; Dokeniya 2014). Albania faces the challenge of the wide use of political appointments in the civil service, which has lessened the degree of commitment to RTI objectives within an organization, as appointees were more focused on a limited and time-bound political agenda (Trebicka and Shella 2014: 20). In other cases, turnover of staff with specialized knowledge of RTI obligations resulted in a loss of institutionalized knowledge (Pereira Chumbe 2014: 197; Trapnell 2014: 334). 37 Table 19: Staffing levels Low staffing levels are an obstacle to RTI implementation Albania + India + Jordan - Low demand Mexico - Moldova - Low demand Peru - Romania - South Africa + Thailand - Low demand Uganda - Low demand United Kingdom - United States + The placement of RTI functions within separate, dedicated units varies across the countries in the sample. As noted in Table 20, only two countries require the establishment of dedicated units or centers, while four other countries may see dedicated units in agencies with a high volume of requests or complicated schemes for exemption. Table 19 also indicates that at least three countries have agencies without information officers to receive information requests (i.e., Jordan, South Africa, Uganda). Table 20: Agency points of contact for RTI De facto unit / Information officers Unit with RTI responsibilities if committee appointed in most or all not separate RTI unit solely for RTI agencies Albania Varies by agency Public Relations / Communication + India Varies by agency Varies + Jordan - Public Relations / Communication - Mexico + RTI Liaison Committee + Moldova - Public Relations / Communication ? Peru - - + Romania - - + South Africa - - - Thailand + One-stop service center + Uganda - - - United Kingdom Varies by agency FOI Unit + United States Varies by agency FOIA Unit + 6/12 8/12 8/12 38 Staff capacity Staff capacity is driven by adequate training and resources that allow information officers to meet their legal obligations for information disclosure. All countries in the sample faced training constraints and low levels of RTI awareness by public officials (Trebicka and Shella 2014: 29; Devasher Surie and Aiyar 2014: 66-67; Meknassi 2014: 376-7; Lipcean and Stefan 2014: 164; Pereira Chumbe 2014: 207; Moses 2014: 440-41; Dokeniya 2014: 301). The exceptions to this pattern are Mexico, the United Kingdom, and the United States, as shown in Tables 21 and 22. In these three cases, individual agencies have either taken on compulsory training responsibilities for their own staff, or RTI training has become embedded in local training programs, indicating that line ministries are institutionalizing RTI into regular agency functions (Mizrahi and Mendiburu 2014: 103; Trapnell, ed. 2014: 333; Alexander 2014: 574). Staff knowledge was not cited as an obstacle in Romania, where an initial push for training was initiated by the government and sustained by international funding and local NGOs, and resulted in a manual produced for information officers. However, the monitoring and training efforts have since lost steam and are nearly defunct (Ionita and Stefan 2014: 253). Table 21: Staff knowledge of RTI obligations and procedures Poor staff knowledge of RTI is obstacle to RTI implementation Albania + India + Jordan + Mexico - Moldova + Peru + Romania - South Africa + Thailand + Uganda + United Kingdom - United States - 8/12 The large number of public information officers required to implement RTI systems in government bodies, at the national and sub-national levels, makes sustained centralized training difficult to implement in federal systems. This is particularly the case in India, where central and state governments do not have the resources or incentives to train all personnel. In Mexico, the well-funded enforcement body has collaborated with human resource departments with the aim of incorporating RTI trainings into agencies (Torres and Esquivel 2011). But in cases where individual agencies are expected to initiate and fund ongoing training, there are often few resources to do so. Such is the case in South Africa (Moses 2014: 447-8). In over half of the countries in the sample, RTI training is supplemented by civil society organizations, or in some cases like the United Kingdom and the United States, by non- 39 governmental organizations specializing in data protection and information rights. 13 Even with supplemental training provided by civil society, information officers in most countries without compulsory training were operating with low levels of technical knowledge that impeded their ability to disclose information in response to information requests. Table 22: Training providers Additional training provided by Training is Government training providers civil society organizations or compulsory private sector Training Institute for Public Albania Administration (ITPA) - - Central and state governments India Institute of Secretariat and - + Management (ISTM-DoPT) Jordan - - - IFAI (Federal Institute for Access to Information) Mexico SFP (Ministry of Public Administration) + - Individual agencies Moldova - - + Public Administration Secretariat, Peru Ombudsman - + Romania - - + South Africa Ombudsman, DOJ Justice College - + Thailand Office of the Information Commission - - Uganda - - - Information Commissioner's Office, United Kingdom Individual agencies + + Office of Information Policy, Individual United States agencies + + 8/12 3/12 7/12 Staff incentives A pervasive failure to comply with disclosure obligations because of a lack of incentives, or the existence of perverse incentives, affects most of the sample. Even in countries with generally positive incentive structures governing the disclosure of information (i.e., Mexico, United Kingdom, and the United States), poor incentives in records management is nearly universal across the sample. As shown in Table 23, nine countries cases cited a lack of incentives, or the existence of disincentives, as a major obstacle to a functioning RTI system. There are generally no incentives to comply with RTI rules 13 The Campaign for Freedom of Information runs courses on data protection, freedom of information, and other information rights in the UK. American Society of Access Professionals provides various training and educational programs for FOIA professionals and the public in the United States. 40 in Albania, as sanctions are rarely imposed and informal practices dominate daily operations (Trebicka and Shella 2014: 24, 28). Responsiveness to requests is not included in personnel evaluations in India, and there is an ongoing lack of ownership of proactive disclosure because of uncertainty as to whether the departments of technology or communications should take responsibility (Devasher Surie and Aiyar 2014: 66-67). Public officials in Jordan, Thailand, and Uganda lack clear guidance as to the kinds of information that can be disclosed and the manner of disclosure (Meknassi 2014: 382, 479). Penalties for improper disclosure exist in India, Thailand, and the United Kingdom, with a lower threat of sanctions in the remaining countries. However, it is not clear that there is a complete absence of penalties for improper disclosure in any of the countries in the sample. In Romania, civil servants do not consider the RTI law relevant to their interactions with citizens, possibly because there are few monitoring efforts to reinforce the importance of their disclosure obligations (Ionita and Stefan 2014: 253). Table 23: Existence of poorly aligned staff incentives Poorly aligned staff incentives serve as obstacles to RTI implementation Albania + India + Jordan + Mexico - Moldova + Peru + Romania + South Africa + Thailand + Uganda + United Kingdom - United States - 9/12 41 Oversight RTI systems benefit from effective management during implementation that facilitates adequate resources and resolution of administrative delays. Although legal frameworks may clearly specify the enforcement mechanisms for a RTI system, the institutional arrangements for monitoring of implementation may be set through de facto practices in the public sector. Monitoring tasks include oversight over nation-wide implementation, and guidance on the design of policies at the national, sub- national and agency-level, while enforcement responsibilities include decision processes on appeals and complaints, and enforcement of appropriate sanctions for non-compliance. Monitoring of institutional capacity In nearly all countries except for Jordan and Moldova, a public body has been mandated with monitoring responsibilities, has been created specifically for these tasks, or has assumed monitoring tasks as an extension of existing responsibilities for public sector modernization. There is a wide range of public bodies that are responsible for supporting RTI implementation processes within the public sector, including ombudsmen and human rights commissions, ministries of public administration or justice, and special departments within the cabinet or prime minister’s office. Table 23 captures the monitoring arrangements around implementing rules. In two-thirds of the sample countries, the presence of implementing rules across ministries is not consistent (Trebicka and Shella 2014: 15; Meknassi 2014: 388-392; Lipcean and Stefan 2014: 163-64; Nicro et al. 2014: 505-6; Dokeniya 2014: 289-90). In nearly all cases where implementing agencies are responsible for developing implementing regulations or rules, including Albania, Jordan, Moldova, South Africa and Uganda, these rules have been inconsistently developed and/or adopted. The lone exception to this pattern is the United States. Table 24: Monitoring arrangements for implementing rules Responsible body for Internal RTI regulations / developing implementing rules / policies adopted by regulations most or all agencies Albania Implementing agencies - Central and state India governments + Jordan Implementing agencies - IFAI (Federal Institute for Access to Information) Mexico SFP (Ministry of Public + Administration) Moldova Implementing Agencies - Public Administration Peru Secretariat in the Cabinet - Ministry of Public Romania Information (no longer ? exists) 42 South Africa Implementing Agencies - Office of Information Thailand Commission in the Prime - Minister's Office (OIC) Uganda Implementing Agencies - United Kingdom Ministry of Justice + Implementing Agencies (as United States administrative law) + 4/12 Table 25 captures the range of RTI monitoring responsibilities that are assumed by public bodies. Only Mexico performs all these tasks, overlapping between the IFAI (Federal Institute for Access to Information) and the SFP (Ministry of Public Administration) (Mizrahi and Mendiburu 2014). In Albania, Jordan, Moldova, Romania, and Uganda, there are no public bodies that perform monitoring tasks, with the exception of the Albanian Training Institute for Public Administration. In the United Kingdom and the United States, the Ministries of Justice perform a variety of monitoring tasks, but neither country assigns public outreach to any one public body (Trapnell 2014; Alexander 2014). Of countries with an information commission or external appeals body, only Thailand assigns both monitoring and enforcement to one public body – the Official Information Commission (Nicro et. al 2014). The remainder of countries – Mexico, India, and the United Kingdom – split monitoring tasks between the information commission and public sector bodies, while Jordan assigns no monitoring responsibilities at all. Table 25: Monitoring responsibilities performed by public bodies Training Issuance of Issuance of best Publishing Performance provision / Public implementing practice models/ recommendations monitoring Training outreach rules guidance to policymakers oversight Albania - - + - - - India + + + + + - Jordan - - - - - - Mexico + + + + + + Moldova - - - - - - Peru + ? + + ? + Romania + - - - - - South Africa - + + + + - Thailand + - + - - + Uganda - - - - - - United Kingdom + + + - + + United States - + + - + + 6/12 6/12 8/12 4/12 6/12 5/12 43 Table 26 outlines the features of data tracking in the sample countries. In India, Mexico, South Africa, the United Kingdom, and the United States, data from agencies is used to analyze weaknesses in the system and to highlight good practices across agencies. In some countries, data is not collected at all, as in Albania and Moldova, where there is no reporting requirement. Despite a legal obligation in Uganda to collect and present monitoring data to Parliament, no public body has ever done so (Dokeniya 2014: 296). In Jordan and Thailand, no data on requests and appeals is released publicly (Meknassi 2014; Nicro 2014). Table 26: Features of data tracking at the country level Detailed agency- Data collector Manner of Mandatory Consistent level data for requests and agency-level reporting reporting by released to the appeals data release requirements agencies public Albania - - - - - Central Central India Information + Information + - Commission Commission Information Jordan Council - - + - IFAI (Federal IFAI (Federal Institute for Institute for Mexico Access to + Access to + + Information) Information) Moldova - - - - - Coordination Coordination Peru Secretariat in the + Secretariat in the + - Cabinet Cabinet Department for Individual Romania Governmental + agencies + - Strategies (DSG) Human Rights Human Rights South Africa Commission + Commission + - Official Thailand Information - - + ? Commission Uganda Parliament - - + - Information Ministry of United Kingdom Justice + Commissioner’s + + Office Office of Individual United States Information + agencies + + Policy 7/12 10/12 3/12 44 Enforcement of disclosure obligations While monitoring efforts may be comprised of reviewing compliance, providing training, and issuing guidance or internal rules on implementation, enforcement has to do with ensuring compliance of RTI laws. Enforcement includes the hearing of appeals, investigation of complaints, issuance of binding resolutions, and recommendations for sanctions. Appeals are formal requests to a higher authority for a reversal of an official decision or to require a decision to be made. In RTI systems, they can also be filed for other reasons, such as failure to respond within a time limit or charging unreasonable fees, but this depends on the conditions specified in the RTI law. If the requester continues to seek a reversal of the original decision, appeals can progress through several levels of authority until they reach the judiciary. Decisions on appeals are usually binding. 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 or an administrative tribunal. These bodies can vary in their authority to issue binding decisions, but many are able to compel the release of information from public bodies. When external appeals authorities do not exist, or when lack of access to information cannot be appealed by law, requesters can file complaints with grievance redress bodies. These bodies offer mediation services, but cannot compel disclosure through binding adjudication. Table 28 captures the levels of appeals process within each country in the sample, demonstrating that some countries have a very limited appeals process before the final appeal in the courts. It must be noted that grievance redress through the ombudsmen does not result in binding decisions, but rather, channels complaints through a mediation process in the hope that public bodies will disclose information in lieu of appeals. Table 27: Levels of appeals process Internal appeals External appeals Grievance redress Appeals submitted process at the authority for RTI complaints to courts agency level (binding decisions) (mediation) People's Advocate Albania + - + (Ombudsman) Central Information Central Information Commission / State Commission / State India + Information + Information Commissions Commissions Information Council Jordan - (possibly not binding) + - Information Information Mexico + Commission (IFAI) + Commission (IFAI) Moldova + - + - Office of Public Peru + - + Defender (Ombudsman) People's Advocate Romania + - + (Ombudsman) 45 Human Rights South Africa + - + Commission Office of the Information Thailand - Disclosure Tribunal + Information Commission Human Rights Uganda + - + Commission Information Information United Kingdom + Commissioner’s + Commissioner’s Office Office Office of Government United States + - + Information Services (OGIS) 10/12 5/12 12/12 10/12 Very few countries benefit from public bodies with the power to issue binding decisions or recommend sanctions for violations. India, Mexico, Thailand and the United Kingdom benefit from independent information commissioners that can issue binding decisions, while Jordan has a non-independent council that may not have the authority in practice to issue compulsory decisions (Devasher Surie and Aiyar 2014; Mizrahi and Mendiburu 2014, Nicro et al. 2014; Trapnell 2014; Meknassi 2014). These bodies are tasked with ensuring compliance with RTI law, through either binding adjudication or compulsory decisions, but they may not be authorized to apply sanctions. In fact, the only country with an independent commission that can impose sanctions is India. The Mexican IFAI can recommend sanctions to the Ministry of Public Administration, and does so sparingly (Mizrahi and Mendiburu 2014). In the United Kingdom, the decisions are registered with courts and effectively become court decisions after the time for appeal has passed. In some countries, there is also the potential for civil service oversight committees to enforce administrative sanctions on public officials who violate disclosure obligations that have been incorporated into employment rules, but among the sample this is a rare occurrence, particularly when the law is not clear on what constitutes a violation of this sort. The remainder of countries relies on the court system to obligate disclosure, set precedents, and impose sanctions when laws are violated. All countries in the sample allow appeals to be heard in the court system, although this is often a last resort, given the time, expense, and expertise required to move cases through courts. However, courts have been effective in establishing case precedents that influence good practice. In Romania, high-profile strategic litigation is credited for the high response rate by government agencies (Ionita and Stefan 2014: 262). United States case law has expanded and defined the scope of RTI in the absence of any other appeals authority (Alexander 2014). As noted above, legal expertise is often a prerequisite for pursuing disclosure through the courts. 46 Table 28: Enforcement methods Effective Binding Possibility of Persuasion / Public release appeals process decisions by sanctions or fines Monitoring / of data on poor (including enforcement (by IC or courts)* Investigation performance courts) body Albania - - - - - India + + + + + Jordan - ? - - - Mexico + + + + + Moldova - - - - - Peru + - + + + Romania + - + - - South Africa + - - - + Thailand + + - - - Uganda - - - - - United Kingdom + + + + + United States + - + + + 8/12 4/12 6/12 5/12 6/12 * IC = Information Commission or other external public body. Table 28 provides an overview of the various enforcement methods that are utilized across the sample of countries. An effective appeals process is the most prevalent form of enforcement in the sample of countries. The lack of an external appeals agency with the power to perform binding adjudication, combined with a lack of judicial capacity in Albania, Moldova, and Uganda, discourages requesters from pursuing appeals (Trebicka and Shella 2014: 24; Lipcean and Stefan 2014: 165; Dokeniya 2014: 297). The High Court in Jordan has so far refused to hear most RTI cases on procedural grounds, and has ruled against the few cases that have been heard (Meknassi 2014: 382). In fact, there are no enforcement methods employed in practice in Albania, Jordan, Moldova, or Uganda, with the possible exception of Jordan’s Information Council, which may have the authority to perform binding adjudication on appeals. South Africa, Romania, and Thailand benefit from some enforcement methods operating in practice, with Thailand the only one with an external appeals agency. India, Mexico, and UK employ all five methods of enforcement, but this is not surprising, given their relatively well-resourced enforcement bodies with the authority to fulfill their mandates. Outside of the appeals process, enforcement bodies may also resort to less punitive measures to ensure compliance, such as threat of inspection, closed-door meetings, notice of inquiry, or calling senior officials as witnesses (labelled as Persuasion/Monitoring/Investigation in Table 28). In some cases, simply the notice that a complaint has been filed prompts reparative action on the part of agencies, such as in Mexico (Mendiburi and Mizrahi 2014). Public release of poor performance data is employed by half of the countries in the case studies, but it is the weakest method of enforcement, dependent on civil society or other government actors to initiate discussion and/or change. 47 Credibility of sanctioning threat depends on a variety of factors, including nature of appeal, extent of media coverage, and inclination of specific courts to rule against the government. Financial penalties and attorney fees are often imposed through court decisions when cases are on appeal in Romania and the United States (Ionita and Stefan 2014: 255; Alexander 2014: 549). In India and Peru, the threat of sanctions is credible but not consistently applied (Devasher Surie and Aiyar 2014: 71; Pereira Chumbe 2014: 198). 48 Discussion Strategic leadership and incentives As noted above, policy prioritization is determined to a large extent by the level of support expressed by political actors. Administrative capacity is in turn influenced by levels of political support, facilitating prioritization of RTI at lower levels of government. This is operationalized though strategic leadership in public bodies, properly aligned incentive structures, and accountable relationships among oversight bodies, agency head, managers, and information officers In terms of an RTI system, principal-agent relationships between managers and staff can be strengthened through formalized practices and training, in addition to performance appraisal systems that include disclosure obligations. Detrimental behavior patterns can be overcome through changes in the incentive structures rewarding, permitting, or punishing that behavior. 14 But incentive structures are complex and influenced by any number of cultural norms or institutional practices. In particular, obstructive informal practices, inadequate staffing levels, and low staff capacity might need to be addressed before incentives can be properly aligned. Figure 3: Lines of accountability and responsibility that shape incentives Incentive structures that constrain and encourage behavior are shaped by lines of accountability within agencies and across the RTI system, as shown in Figure 3. Relationships between monitoring entities and individual agencies establish channels of communication among information officers, managers, and monitoring bodies, as shown on the left side of the diagram. This includes guidance and training provided by monitoring bodies, and the reporting of requests and appeals data by individual agencies. But these relationships lack the threat of consequences unless disclosure obligations are included in 14 Fox et al. (2011) refer to issues of values on the one hand, and interests and power on the other. This paper takes the view that addressing interests and power within administrative contexts can contribute to changes in the attitudes of public officials. 49 monitoring responsibilities. Because accountability relationships in public sector operations rarely move diagonally from information officers to enforcement bodies, as shown on the right hand side of the diagram, incentives must be shaped through the horizontal accountability relationships between agency heads and information officers, and between agency heads and enforcement bodies. 15 Although chief information officers may be responsible for the operational activities associated with RTI, agency heads are the locus of accountability in RTI implementation, making leadership at all levels a key component in shaping attitudes and behavior, and prioritizing RTI policy. Three country cases (Mexico, United Kingdom, and United States) are robust RTI systems that made no mention of staff incentives as contributing to effective information disclosure. 16 It is likely that there is a significant and positive relationship between a professionalized civil service and properly aligned staff incentives that affect RTI implementation. This implies that appropriate incentives employed in the public sector may have a significant effect on RTI effectiveness at the agency level. These incentives include appropriate time for additional duties, functioning personnel evaluations with accurate measurement, and a lack of formal or informal penalties for disclosing information, e.g., penalties for improper disclosure. Training specific to RTI obligations, including records management, also contributes to encouraging appropriate behavior. The fact that no country mentioned the need for special incentives such as cash bonuses or administrative honors suggests that properly aligned incentives matter more for motivating behavior than special rewards. Barriers to access Demand for information is a critical factor in the effectiveness of RTI systems, as underutilized systems tended to be underdeveloped and exhibited poor performance. The case studies highlighted both knowledge and accessibility as constraints on access to information. These two drivers co-exist in a virtuous circle of demand, shown in Figure 4. The accessibility of RTI systems is significantly influenced by the extent of public knowledge about information rights and RTI processes. In return, knowledge of RTI processes is enhanced through repeated encounters with agencies over information disclosure. Figure 4: The virtuous circle of demand for information Public Accessibility awareness 15 Fox et al. (2011) note that the internal audit units embedded within Mexican public bodies are responsible to the supreme audit institution, establishing a diagonal relationship between audit officers and audit oversight body. This may be an option to insert additional means of accountability within RTI systems. 16 Mexico, the United Kingdom, and the United States are also the only members of the OECD in the sample of twelve countries. 50 An understanding of rights, as well as having accessibility to processes, matter for marginalized groups in very practical ways. Language differences, literacy levels, familiarity with bureaucratic procedures, and internet access are all major factors that prevent large numbers of people from accessing information request processes. There are good practice examples in the country cases where consideration of specific user groups was addressed at the agency or program level, particularly at sub-national and municipal levels, so that information, forms, assistance, and mediation processes are made accessible to marginalized communities. Public awareness of RTI influences not only the ability of requesters to access information, but also their understanding of the information rights accorded to citizens. Knowledge of what is available to be requested, and how information can assist with certain goals (which motivates request making), is also part of this equation. The success of information requests is based upon the technical knowledge needed to formulate requests, as well as the broader knowledge of the kinds of information that can, and should, be requested. As a result, there is concern that requesters comprise a potentially narrow social base of specialist information requesters centered in urban areas (Fox and Haight 2011: 156). Request data collected by governments indicated that ratio of requests to population was less than 1% across a sample of eight countries (Worker 2014). One implication is that the group of requesters that make up the bulk of requests may constitute an extremely small proportion of the population. Questions of representativeness and inclusion abound, potentially compromising the universality of access to information initiatives. Information requests are the primary channels for individuals to communicate their specific demands for information to government bodies. As noted above, the is a learning curve associated with obtaining satisfactory responses to information requests, as requesters must know how and what to request from government bodies. In terms of providing access to information, information intermediaries such as civil society organizations and media play a critical role in their analysis and dissemination of information, but they also serve as filters through their own sets of lenses. Lack of access based on gender, race, and class are also fundamental issues compromising the inclusiveness of RTI systems, though they were only weakly addressed in the country cases. In a separate study from Liberia on gender and right to information, perceptions of mixed-gender groups confirmed that women access information much less often than men, a finding that was also confirmed through observations of RTI units within government agencies. Aggregate data from interviews demonstrated that the barriers preventing women from exercising their right to information included illiteracy, fear of asking, and not knowing how to ask or where to go to find or request information. Responsibilities to household and childcare, as well as mobility and distance to government agencies, were also cited as significant barriers (Carter Center 2014: 25-28). These findings suggest that further study of the relationship between RTI systems and minority groups is essential to making the right to information equally, and fairly, accessible to the general population. Innovative approaches to making information more accessible to marginalized communities combine functions such as assistance to requesters, communication over content of the request, negotiation of what is an acceptable and meaningful response, and complaints processes. They are located strategically near populations of potential information requesters. The Oportunidades program in Mexico has a specific citizen attention window through which most of the information requests and complaints are managed (Mizrahi and Mendiburu 2014: 121; Fox and Haight 2011: 157). It is important to note that both requests and complaints are addressed through this unit. Citizens are engaging and negotiating with government over the kinds of information that they are seeking. In Thailand, there are 51 one-stop service centers with the mandate to provide information to citizens, but there is little negotiation over the kind of information that can be released. In fact, some agencies do not even register information requests, as the information requested does not fall within the very specific parameters laid out in the RTI law (Nicro et al. 2014: 506-510). Because of its focus on citizen uptake, proactive disclosure of information, data, and documents should be based on citizen interest. However, very few countries have procedures to identify information for proactive disclosure beyond a standard list of categories of information to be disclosed. In order to identify information that is most relevant to the demands of the public, agencies should at the minimum be tracking downloads or hard copy requests for proactively disclosed information. There is also a need for consultations with citizens and civil society groups about the content of information that should be released (Darbishire 2010: 27). Information that is not relevant to users or does not meet the needs of information consumers meets only a pro forma adherence to disclosure obligations. Proactive release of information should not become a passive form of resistance to disclosure. Figure 5: Location of most commonly available information released by central government (OECD Government at a glance survey 2010) Central portal Ministry/agency/other website Central portal & ministry/agency/other website Not published online Budget documents Annual ministry reports, including accounts Audit reports All government policy reports Commercial contracts over a stipulated threshold List of public servants and their salaries Administrative data sets Information describing the types of records systems and their contents and uses Information on internal procedures, manuals and guidelines Description of the structure and function of government institutions Annual report on freedom of information law Freedom of information procedural information 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Percentage of 32 OECD countries responding to the survey The accessibility of proactively disclosed information is becoming more of a concern as internet technology is used to disseminate documents, even in areas where internet penetration rates are quite low, shown in Table 29. Online disclosure is the most common form of proactive disclosure of information across all cases and agencies, but disorganized and poorly constructed websites make 52 information extremely difficult to find. As a means of countering this challenge, public officials make registers of information available online in one document, such as in South Africa (Moses 2014). Requesters then contact agencies directly for access to the information, but this creates another layer of work for staff, reducing the efficiency that online proactive disclosure is supposed to offer. Registers of information are not just for use in proactive disclosure of information but also used to facilitate access to records under RTI law using RTI request procedures. When registers of information are not updated, even when mandated by law, the functions of both records management and proactive disclosure can suffer. Table 29: Internet access Percentage of Individuals using the Internet* Albania 60% India 15% Jordan 44% Mexico 43% Moldova 49% Peru 39% Romania 50% South Africa 49% Thailand 29% Uganda 16% United Kingdom 90% United States 84% *International Telecommunication Union (ITU), UN, 2014 In contexts where internet penetration is low, shown in Table 29, alternative methods of dissemination are sometimes employed to facilitate access to proactively disclosed information. Print media and print publications are common methods to disseminate information, and all agencies cited in the cases used this method of disclosure. But low literacy levels may also require radio or television transmission, local bulletin boards, along with townhall or community gatherings. As stated above, agencies in Thailand have opened One-Stop-Service centers of information, staffed by knowledgeable employees who are tasked with assisting information requesters with their searches. Information and documentation are available for inspection as specified by various provisions in the law (Nicro et al. 2014: 506-510). Financial and material resources Although the absence of a specific budget line item for RTI is often considered the primary obstacle to effective RTI implementation, particularly in low income contexts, the issue of resources is much more diffuse. RTI functions tend not to be housed in separate units dedicated to RTI except in agencies that receive large volumes of requests, such as ministries of social welfare or pensions, or agencies that are faced with complicated classification and exception schemes, such as the ministry of defense. Thus, RTI functions are embedded in administrative functions that face common resource failures. In the sample countries, insufficient material resources manifested in broken, missing or outdated equipment that 53 impacted the extent to which officials could identify and locate information, and as noted above, inadequate financial resources led to low levels of staffing. Much of the breakdown in supply of resources is not the failure of individuals or departments and is not related to RTI. Perhaps because of this, or in ignorance of it, most countries expect agencies to cover RTI costs within their existing financial envelope. An exception is the funding that is normally provided to centralized enforcement bodies, such as information commissioners. A response to the issue of lack of resources for RTI should be rooted in a strategy that assigns greater priority for RTI funding within the existing budget, and in this way, is much more likely to be sustainable in the long term. This approach calls for the inclusion of RTI in regular, strategic decision-making and an institutional incentive system to encourage public bodies to do so. Institutionalization of RTI practices No RTI system will function adequately without clear, formalized procedures that remove opportunities for abuse of discretion, set appropriate incentives for civil servants and managers to support its implementation, and encourage leadership across levels of government that signals commitment to the regular practice of information disclosure. These formalized practices are particularly important for records management, request processing, and proactive disclosure, as these areas form the backbone of an RTI system. Records management is the unseen, underappreciated foundation of an RTI system. It comprises the control of processes by which digital and paper information about organizational activities is created, organized, stored, and accessed by employees (Shepherd 2010, Colquhoun 2010). It includes the use of information and communication digital technology for the purposes of storage, retrieval, and dissemination of information. In many circumstances, ranging from high to low-income countries, the issue of records management is overlooked, and as a result, agencies rely upon informal methods of information creation, organization, storage and retrieval. Informality in record-keeping means that that standards cannot be maintained. In fact, poor records management practices produce poor records. This practice results not only in delays in responding to information requests, but also in lower-quality and even unreliable information being distributed about government activities (Shepherd 2011; Luyombya 2011). It also prevents effective collection of data on government functioning, posing a significant obstacle to the success of proactive disclosure of information and open data initiatives. Issues with records management significantly influence the potential success of a RTI system, as well as policymaking and decision-making within the public sector. In the context of an absence of or ineffective records management, it is extremely difficult to determine whether weak RTI performance data indicates an inability to find records or an unwillingness to disclose information. Responding to an information request is rarely straightforward, and there are several stages where bottlenecks can occur: finding and compiling responsive information, consulting with third parties, assessing and applying exceptions (i.e. redacting, e.g. private information), and considering appeals (Shepherd 2010, Colquhoun 2010). One of the most oft-cited practices for lowering the administrative burden and financial costs of request processing is proactive disclosure. Categories of relevant information include administrative data, expenditures and accounts, policy reports, internal procedures and functions, human resources management (salaries and positions), procurement/contracts, declarations of conflicts of interest, and income and assets of public officials. While open data is a part of the proactive disclosure function, it is focused on releasing datasets about government operations and data that the government collects and uses to make policy decisions. The assessment of proactive disclosure is a difficult task, as there are no agreed standards as to the amount, regularity, or content of 54 information that should be disclosed, aside from general recommendations about regular postings, and serious concerns over the relevance and comprehensibility of information to citizens. 17 Specialized ICT tools are intended to streamline the process of online disclosure in order to keep costs down. But they are often a major investment, and without appropriate training and skilled IT staff to deal with interoperability difficulties among information databases, these initiatives often fail. The sustainability of these tools are also at the mercy of staffing levels, unless steps are taken to ensure that institutional knowledge will remain with the organization rather than the individual. Both Peru and Mexico have launched centralized online portals for agencies to upload proactively disclosed information, but questions of sustainability and stakeholder support persist. Based on country experiences, potential good practices for proactive disclosure include: • tracking of high volume requests and releasing popular information (Thailand) • a “rule of three,” whereby information is released if an agency anticipates that it will be requested three times (United States) • an online repository of requests and responses (Mexico) Each practice requires effective data tracking and RTI usage statistics gathering processes and in the case of Mexico, a sustainable level of IT expertise. As with any organization, hiring strategies for an agency-level RTI system must anticipate training needs and turnover. External demand for information, through both information requests and proactive disclosure, will determine the number of staff necessary to meet the needs of the system. Many agencies in the sample of cases do not devote sufficient staff time to RTI tasks, citing a lack of commitment from management or a lack of human resources. There is no one model that will suit all agency contexts and demands. In all cases, however, a designated information officer should exist in at least each agency, perhaps with additional officers in larger administrative units within agencies. Moreover, compulsory training should be accompanied by coordinated efforts to incorporate RTI training into agency training programs, or less commonly, through centralized training efforts by national schools of public administration or enforcement bodies. Updated, formal practices contribute to the functioning of many other drivers within the domains of RTI implementation. They are a necessary, but not sufficient, feature of RTI implementation at the agency level. Engagement with civil society over the development of formal practices for request processing and proactive disclosure is an often overlooked value to government. Collaboration among agencies, oversight bodies, civil society organizations, and media on agency-level policy-making and rule-making processes would provide opportunities to shape practices for internal efficiency and external relevance to users. Monitoring responsibilities and institutional arrangements For effective and consistent implementation of RTI across ministries and agencies, it is important that at least one national-level authority is made responsible for supporting the implementation process, often by articulating a framework for RTI implementation. This body is often referred to as a “nodal 17 There are suggested recommendations for proactive transparency standards in Darbishire (2010). 55 authority,” and its role is to assist agencies in setting up appropriate organizational structures and administrative procedures that facilitate compliance with the law. When nodal bodies are not resourced appropriately, with sufficient authority and capacity, public bodies may fail to develop implementing rules, such as with Thailand and Uganda (Nicro et al. 2014: 505-6; Dokeniya 2014: 289-90). Without guidance and logistical supervision, there is significant likelihood that implementing rules and regulations will not be developed or will be radically inconsistent, particularly when responsibility for developing these rules is left to individual agencies. Because ombudsmen sit outside of the government, they should only be charged with official responsibility for implementation oversight when they have sufficient authority, resources, and capacity to perform the task. In the cases of Peru and South Africa, the ombudsmen are highly organized and resourced, with considerable expertise, and maintain mutually respectful relations with government entities and other RTI monitoring bodies (Pereira Chumbe 2014: 196-204; Moses 2014: 433-436). Table 30: Institutional arrangements for various monitoring tasks Issuance of Training provision Issuance of best Publishing Performance implementing / Training Public outreach practice models/ recommendations monitoring rules oversight guidance to policymakers Training Institute for Public Albania - - - - - Administration (ITPA) Central and state governments, Department of Department of Department of Central and state Institute of India Personnel and Personnel and Personnel and - governments Secretariat and Training Training Training Management (ISTM-DoPT) Jordan - - - - - - IFAI (Federal Institute for Access to Mexico Information), SFP IFAI IFAI, SFP IFAI IFAI, SFP IFAI (Ministry of Public Administration) Moldova - - - - - - Public Administration Public Secretariat in the Administration Office of Public Office of Public Peru Office of the ? Secretariat, Office Defender ? Defender President of the of Public Defender (Ombudsman) (Ombudsman) Ministerial (Ombudsman) Cabinet Ministry of Public Romania Information (no - - - - - longer exists) Human Rights Commission, Public Service Human Rights Commission (PSC), Commission, Human Rights Human Rights South Africa - Department of - Department of Commission Commission Justice and Justice College Constitutional Development (DOJCD) Thailand Official - Official - - Official 56 Information Information Information Commission Commission Commission Uganda - - - - - - Information Ministry of Justice, United Commissioner's Information Ministry of Justice Ministry of Justice - Ministry of Justice Kingdom Office, Individual Commissioner's agencies Office Office of Office of Office of Office of Information Policy Information Policy Information Policy Government United States - - in the Department (OIP), Individual in the Department Information of Justice (OIP) agencies of Justice (OIP) Services (OGIS) 6/12 5/12 8/12 4/12 5/12 5/12 In all of the cases in the sample, and shown in Table 26, data reporting is not considered a priority unless a nodal agency, or enforcement body, requires the submission of an annual report (Worker 2014). It is not surprising that RTI data is not reported when agencies claim to not be aware of their own reporting obligations under the RTI law. Reporting data involves the tracking of responsiveness to requests and appeals processing. It is an ongoing process of collecting information (or data) on how well an RTI system is on track to meet its objectives, and it is extremely important for both real-time and end-of- year evaluations that feed into budgets, resource allocation, and investments in technology for improved processes. However, a major challenge to incorporating RTI indicators and targets into existing administrative systems is the weakness of existing performance monitoring at the agency level. Capacity building may be required to design and implement agency-wide performance monitoring systems that can accurately and efficiently collect data. Electronic submission of requests allows for real- time tracking, although there are issues with the exclusion of marginalized groups without access to, or understanding of, the internet. In the absence of resources (or mandate) to recommend policy changes, issue good practice models, and provide more training, it is likely that data will not be collected regularly or reliably. In resource-stretched circumstances, data is released to the public in the hope that civil society organizations will use the data to highlight gaps and areas of weakness. The credibility of RTI data reports from agencies is also a constant struggle for performance monitoring purposes. Even in strong RTI systems such as Mexico, studies have shown that figures of effectiveness are overestimated (Fox et al. 2011). Several independent studies on compliance with RTI obligations (some conducted by in-country civil society organizations) have demonstrated that agencies fail to meet deadlines, keep requesters informed of status requests, and provide high quality responses (Dokeniya 2013; Worthy 2011; Hazell 2010; Open Society Justice Initiative 2006; RaaG 2009; Global Integrity Report 2004-2011; South African Human Rights Commission Golden Key Awards 2008-present). Findings from this study revealed that compliance testing has become more common across the sample, with the expanded role of CSOs in monitoring of implementation, and the development of more sophisticated monitoring mechanisms by enforcement agencies. Enforcement mechanisms Countries vary in the complexity of the appeals process, and assessment of effectiveness in appeals depends very much on the practices and of individual bodies and the nature of information requests. Delays in appeals at the judicial level characterized most of the countries in the sample. Combined with a lack of faith in judicial capacity or rule of law, delays contributed to a reluctance to pursue appeals through the courts. 57 The existence of an information commissioner or comparable enforcement body in an RTI system allows for a second level of appeal outside the public body or the courts, and which does not require legal representation. In addition, information commissioners, if well-resourced and mandated with appropriate authority, can employ alternative methods of enforcement besides appeals and sanctions. As noted above, enforcement bodies can conduct closed-door meetings, issue notices of inquiry with threat of inspection, publish compulsory guidelines or analyses of specific RTI challenges, or call senior officials as witnesses in investigations. The mission of ombudsmen often does not extend to this type of intervention, and existing nodal bodies may not have the capacity or authority to conduct these kinds of activities. It is clear that credible threat of sanctions is not prevalent, suggesting that the possibility of penalties is not necessarily the primary means of enforcement in RTI systems. Aside from the need for an effective appeals process, which is a necessary but not sufficient means of enforcement in RTI systems, no clear pattern emerged from this analysis about what works best for enforcement. It may well be that a combination of methods works most effectively to enforce disclosure obligations, particularly when capitalizing on the stronger institutions and norms of individual countries. Transformative factors: Collaboration and Technology During the qualitative analysis of the case studies, it became apparent that some features of RTI systems extended across the four domains of RTI implementation, yet were not always included in the narratives of success and failure. They fall into no single domain exclusively, but instead, influence the path of each driver in different ways. These transformative factors can enhance aspects of the drivers significantly by increasing their efficiency, efficacy, and inclusiveness, thus magnifying their impact. Below is brief discussion of the factors identified in the analysis. Findings from the cases are highlighted wherever relevant. State-society collaboration Advocacy efforts refer to contestation over the institutionalization of rights, which by their nature are subject to challenge and transformation by stakeholders. Human rights, in fact, stand in opposition to the arbitrary use of power and privilege (Stammers 2009: 105). Civil debate over the shape of legal instruments or the path of implementation allows for the balance of interests from both government and society, and is a natural part of democratic governance (Cohen and Arato 1994). However, once rights become embedded in structures of administrative power, there is also space for collaboration over the direction of policies and implementing rules (Booth 2012). State –society collaboration includes any kind of engagement between civil society and the government that is not considered adversarial. This includes working groups, stakeholder consultations, and participation in committees and councils. It ranges from joint efforts to produce outcomes, which may include civil servant training and promotional activities for the public, to solicitation of civil society feedback, such as on the relevance of proactively disclosed information. Formalized means of collaboration are important for sustainability purposes, but governments must be able to acknowledge the value-added of joint efforts in RTI systems for institutionalization of these practices to occur. 58 Figure 6: A sample of civil society organizations working on transparency in sample countries (Sunlight Foundation registry) Transparency Organizations, by country United States 87 United Kingdom 17 Uganda 3 Thailand 4 South Africa 4 Romania 5 Peru 4 Moldova 3 Mexico 5 Jordan 2 India 17 Albania 5 0 20 40 60 80 100 Figure 6 presents information on the number of civil society organizations operating on transparency issues within each country in the sample. 18 Civil society organizations (CSOs) and media play a major role in all aspects of implementation in the sample countries, even if their absolute numbers are small. These activities range from training to compliance monitoring and public outreach, with CSOs more substantially represented in these activities. Nearly half of the cases cited civil society involvement in training of public officials (e.g., India, Moldova, Peru, Romania, South Africa, United Kingdom, and United States), suggesting that civil society groups are in close collaboration with government in the planning of training, or assisting government-sponsored RTI training efforts (Devasher Surie and Aiyar 2014; Mizrahi and Mendiburu 2014; Pereira Chumbe 2014; Moses 2014; Trapnell 2014; Alexander 2014). In India and Mexico, civil society organizations are regularly invited to provide input into policies and legal amendments (Devasher Surie and Aiyar 2014; Mizrahi and Mendiburu 2014). Civil society organizations in Jordan have proposed that non-governmental organizations be given seats on the information council (Meknassi 2014: 387). There were also instances in which the civic space did not yet allow for meaningful collaboration between public officials and civil society. In Albania, there are few opportunities for civil society groups or citizens to engage with the highly-centralized decision-making process (Trebicka and Shella 2014: 13). The civil society sector in Moldova lacks institutional capacity and, often, basic equipment, and the polarization of the media has called its independence called into question (Lipcean and Stefan 2014: 168). Civil society organizations in Uganda are often not included in decision-making processes (Dokeniya 2014: 304-5). The small community of civil society organizations and journalists working on RTI in Jordan operates in a politically conservative environment with few opportunities for substantive engagement (Meknassi 2014: 395). Collaborative efforts between civil society and government are predicated upon a shared concern for establishing sustainable and meaningful aspects of open government, from training to public awareness, accessibility, and responsiveness. 18 Data based on self-registration with the Sunlight Foundation by international organizations, and may be missing data: http://sunlightfoundation.com/blog/2013/05/29/transparency-advocates-from-all-around-the-world/ 59 Figure 7: Primary focus area of transparency organizations in sample of countries (from Sunlight Foundation) Issue areas for transparency organizations 25 21 19 20 20 17 16 15 12 13 12 10 6 7 5 2 3 0 As shown in Figure 7, civil society organizations and the media focus on a variety of areas regarding transparency, and they play important participatory roles in RTI implementation. The findings from this study indicate that engagement by civil society on issues related to RTI implementation matter for RTI effectiveness, although no particular form of engagement emerged as more or less important to this outcome. However, the effects of participation and collaboration are likely complex and non-linear. A recent study on participation and development outcomes showed that a budget participation sub-index constructed from items on the Open Budget Index did not predict development outcomes or expenditure levels. A possible explanation suggested by the authors is that the indicators used in the sub-index on participation reflected only possible engagement opportunities, not actual engagement (Fukuda-Parr et al. 2011). These findings confirm that questions about quality and extent of civil society engagement remain unanswered, even though there is general agreement that civil society participation is an essential element in policy implementation. 19 At a minimum, public bodies and governments can be engaging in consultative processes with users to identify information that is in high public demand and relevant for users’ needs, which will ultimately enhance the overall functioning of the RTI system. Technology Technology is a tool that augments success, but it does not solve problems without human and material resources to support it. It is often touted as a relatively straightforward solution to issues of efficiency. In terms of transparency, for example, the role of technology in government can serve as a means for collection, organization, and release of open data in machine readable formats for others to capitalize upon. In addition, technology can be employed to ensure that business activities and citizen- government interaction is more efficient and cost-effective through e-government initiatives. But the success of technology is based on a clear understanding of the needs and interests of users, both 19 Recent studies on e-participation reveal that the outcomes of citizen engagement, where individuals feel that their actions ultimately influenced government decision-making, are correlated with higher perceptions of government responsiveness and transparency (Kim and Lee 2012). 60 internal and external to government, and the capacity of those users to employ technology sustainably (Fung et al. 2010). Online submissions of requests allow government to track the status requests and collect data automatically, obviating the need for public officials to log requests. Web portals are the main mechanism for dissemination of proactively disclosed information, providing access to documents to a wide audience. While it is true that technological approaches to transparency can quickly facilitate access and openness, significant human and technical capital is the bedrock upon which technology rests. With few exceptions, technology for transparency and accountability purposes is rarely adopted across-the-board to universal acclaim. It is more likely that successful technological approaches are tailored to the needs of various user communities (Fung et. al 2010). Parents of school-age children, for example, may be very interested in school fees, curricula, and teaching quality. Information on medical treatments and hospital quality may be valuable to patients, doctors, and otherwise healthy individuals. Interventions that are successful on this kind of large-scale tend to organize and display information that is immediately relevant to users’ lives, and made available in easily accessible formats. Other technological interventions may complement mainstream efforts at transparency by releasing specialized information that is relevant for intermediaries, such as journalists or political campaigns, who then make the information accessible to a wider audience. Ultimately, the most successful use of technology for transparency purposes may lie with interventions that are dedicated to advancing the agendas of civil society organizations (Fung et al. 2010). Examples of this type include budget tracking tools, crowd-sourcing of information via mobile phones, and online platforms for submission of complaints and requests. Intra-governmental collaboration Collaboration between records management authorities, technology departments, and nodal authorities or monitoring bodies is an important means of embedding RTI successfully into administrative operations, with the aim of making information disclosure a business-as-usual process. In fact, there was scant information in the case studies on collaboration between technology officers and RTI officials, and in some cases, responsibility for proactive disclosure practices was caught in disputes between individual departments and the heads of public bodies, with technology staff not even part of the discussion (Devasher Surie and Aiyar 2014: 75). Moreover, communities of practice or networks of practitioners across public bodies were rare except for the United States (Alexander 2014). Collaboration between different technical units can facilitate both dialogue and practical decision- making over RTI responsibilities, while communities of practice among RTI officials encourages sharing of good practices and knowledge about RTI responsibilities. 61 Table 31: Records management oversight arrangements Records Management Oversight Collaboration between RM Authority and RTI Monitoring bodies Albania National Archives - Department of Administrative Reforms India and Public Grievances - Jordan Department of the National Library - Mexico National Archives + Moldova Secretariat Administrative Services - Peru General Archive of the Nation - Romania National Archives - South Africa National Archives and Records Service - Thailand National Archives - Department of Records Information Management in the Ministry of Public Uganda Service (records management) - Ministry of Information and National Guidance (retrieval and dissemination) United Kingdom National Archives + National Archives and Records United States Administration + 12/12 3/12 Records management authorities provide much-needed guidance to agencies on the proper destruction of records, electronic records management, archiving procedures, use of resources, and performance monitoring. Table 31 shows the records management oversight authority and RTI collaboration within the sample of cases. It is only in Mexico, the United Kingdom, and the United States that records management authorities collaborate with RTI oversight bodies to harmonize record keeping policies across the public sector (Mizrahi and Mendiburu 2014; Alexander 2014). In these countries, there are ongoing efforts to standardize record-keeping practices for the benefit of RTI. 62 Assessing the drivers of effectiveness Assessment of the drivers of effectiveness will be conducted through a mixed-methods approach. Research teams comprised of international experts, government officials, and/or civil society organizations will be given specific scoring criteria and requirements for supporting documentation. In this way, data collected will be reliable and countries will be consistently scored on the drivers of implementation supporting RTI systems. The levels of analysis for each domain can be assessed at either the national, sub-national level, or sectoral level, except for institutional capacity, which will apply at the level of individual agencies. Aggregation of individual agency-level data is possible, but it is not likely that any sample of agencies will be representative of the country as a whole. Generalization must be considered carefully. Indicators will be developed and revised through stakeholder consultations. Table 32: Assessment categories and descriptions for drivers of RTI effectiveness 1 Enabling conditions Description Sources of data Quality of legal framework assessed Global RTI-Rating, PAM Freedom of A Legal framework for RTI against international standards Information Extent and nature of roles that civil Civicus Civil Society Index, B Advocacy efforts society plays in the shaping of RTI laws Bertelsmann Transformation Index, and policies. Interviews with CSOs Strength of the signaling mechanisms Interviews with government officials, C Policy prioritization from government that set RTI as a critical Media articles, Executive orders, policy initiative decrees, initiatives 2 Demand for information Extent of citizen knowledge about RTI as A Public Awareness of RTI a right, and as a set of procedures for Survey data, Interviews with CSOs access to information Accessibility of request and appeals Evidence of accessibility and B Accessibility process and proactively disclosed complaints resolution, Interviews with information government officials and CSOs 3 Institutional capacity Existence of practices (not rules) that public officials engage request Interviews with government officials, A Updated, formal practices processing, proactive disclosure, and Evidence of formal practices records management. Assessment of whether the number of staff is sufficient to cope with RTI Interviews with government officials, B Staffing levels obligations, and how this number is Evidence of formal practices arrived at. The knowledge of staff about RTI, and Interviews with government officials, C Staff capacity the resources at the disposal for meeting Evidence of formal practices those obligations. Nature of the incentive structure Interviews with government officials, D Staff incentives governing the behavior of staff and Evidence of formal practices management with regard to RTI. 4 Oversight Extent and nature of activities that Interviews with government officials Monitoring of institutional A support RTI implementation across and CSOs, Aggregated performance capacity government. monitoring data, Compliance testing 63 Enforcement of disclosure Strength and nature of methods used to Interviews with government officials B obligations enforce RTI obligations. and CSOs, Appeals and resolution data 5 Transformative factors Extent of opportunities for civil society to engage with government to jointly Interviews with government officials A State-Society Collaboration contribute to RTI implementation. May and CSOs, Evidence of meetings, be incorporated into relevant indicator workshops, strategic plans areas above Types and extent of technology used in RTI implementation. May be Interviews with government officials B Technology incorporated into relevant indicator and CSOs, Evidence of formal practices areas above Extent of collaboration between records Interviews with government officials, Intra-governmental C management, RTI management, and Evidence of meetings, workshops, collaboration technology specialists. strategic plans 64 Conclusions The conceptual mapping of domains of RTI implementation and their related drivers of effectiveness were built up from a rigorous, iterative coding process that incorporated checks on reliability for the purposes of generalization of higher-order elements. The mapping and related drivers are meant to serve as both a guide to the development of implementation strategies, as well as a heuristic for understanding the dynamism and interdependence of implementation components. There is no one strategy or approach to RTI implementation that works best in all contexts, as implementation of reform efforts across the entire public sector will be characterized by variation. Data will require interpretation of context even with pre-existing assessment frameworks or heuristics. In addition, the sequencing of reforms will depend on the contextual factors in specific circumstances, particularly with institutional capacity, as they address interdependent operational functions within individual agencies or departments. Of all domains in RTI implementation, institutional capacity is primary, since without the mechanisms for disclosure, information will not be released effectively. Without functioning operations, demand falls, and there is little need for oversight. However, institutional capacity will not function effectively or sustainably without enabling conditions, demand for information, and oversight. The best approach is to institutionalize RTI within the public sector so that disclosure becomes a business-as-usual practice. When political support wanes or oversight capacity deteriorates, the impetus to continue disclosing information will remain, even if weakened. Demand can also serve a driving force for reform in these circumstances, in addition to helping to ensure sustainability and collaboration. The findings from the study suggest that international pressure for more effective RTI implementation only goes so far. The development of RTI laws with the encouragement, assistance, or insistence of the international community was a prominent theme throughout the case studies, particularly for EU countries during their accession process. But implementation is a less straightforward task, with many interlocking, moving parts, and international support comes in ad hoc fashion as the process unfolds. A strong implication from these findings is that a national coordinating strategy may be valuable for implementation. This kind of strategy document should take the interdependence of the drivers of effectiveness into account when drafting policies and rules for practice, and can serve as a guiding document when deciding on foreign funding priorities. Areas for further research Coordination of RTI and records management Records management is a fundamental contribution to effective RTI systems. It is the underlying structure that facilitates many aspects of RTI functions at the agency level: information requests, proactive disclosure, internal appeals, and performance monitoring. Records management encompasses the recording, storage, retrieval and archiving of information, and with the introduction of electronic records and e-governance functions within administrative processes, collaboration between RTI monitoring bodies and Records management authorities is just as important to the effectiveness of agency-level RTI systems. • Records management is the cornerstone of RTI implementation; however, little is understood yet about what contributes to an (in)effective records and information management system within agencies and how current practices impact upon RTI system implementation in more 65 than general ways. A series of exploratory case studies similar to those conducted in phase on of this project would support greater understanding of the factors lead to effective records and information management and to the development of rapid diagnostic tools. • The interface between records management, E-governance, technology, and RTI systems is becoming a more prominent issue to contend with, particularly on a very practical level of daily operations. Solutions are likely to be found in administrative practices and through functional collaboration among oversight bodies. Collaboration with stakeholders CSOs offer vital contributions to the implementation of RTI systems as both users of information and as collaborative partners. In cases where monitoring and oversight arrangements are weak, CSOs provide supplementary projects on training of officials, compliance monitoring, and public education, including the provision of assistance to citizens with filing and use of information requests. In cases of strong RTI systems, CSOs are key partners in dialogues with monitoring bodies over enhancements in proactive disclosure and information request processing. Collaborating with CSOs demonstrates a commitment to transparency for the benefit of both government and society. Public outreach for citizens and training events for public officials can be held jointly with civil society organizations to reach as large an audience as possible. For citizens, public awareness is an important support to the effective exercise of their right to information. It is not a simple task for the ordinary citizen that is unaccustomed to bureaucratic practices. Information needs, both complex and basic, must be translated into administrative demands that are acceptable through formalized request procedures and easily answerable by distracted and overwhelmed public officials. Accessibility The accessibility of RTI processes and procedures is of paramount importance to the demand for information, and the extent of inclusiveness of RTI systems. Very little information has been systematically collected on the reach of RTI systems with regard to marginalized communities. It is questionable whether RTI systems can effect meaningful without ensuring that opportunities for participation extend to all citizens, regardless of gender, race, class, or location. Global scope As this report is based on a limited sample of countries, an obvious next step would be to conduct a global survey of countries to assess effectiveness across countries, recognizing that such assessment can only be indicative and not explanatory, nor necessarily sensitive to contextual factors. These kinds of surveys can be conducted as collaborative efforts with in-country oversight bodies and civil society organizations, with a view to strengthening country systems rather than comparing countries outright. 66 References Alexander, Shannon. 2014. “Implementing Right to Information: A Case Study of the United States.” Pp. 539–624 in Right to Information: Case Studies on Implementation, edited by Stephanie E. Trapnell. Washington, DC: World Bank. Anon. 2014. Global RTI Rating. Retrieved (http://www.rti-rating.org/). Arbatli, Elif, and Julio Escolano. 2012. Fiscal Transparency, Fiscal Performance and Credit Ratings. Washington, DC: International Monetary Fund. Retrieved May 28, 2013 (http://www.imf.org/external/pubs/cat/longres.aspx?sk=25996). Ball, Carolyn. 2009. “What Is Transparency?” Public Integrity 11(4):293–308. Bauhr, Monika, and Marcia Grimes. 2014. “Indignation or Resignation: The Implications of Transparency for Societal Accountability: Indignation or Resignation.” Governance 27(2):291–320. Besley, Timothy. 2007. Principled Agents?: The Political Economy of Good Government. Oxford: Oxford University Press. Booth, David. 2012. “The Centrality of Collective Action Problems in Governance for Development: New Evidence.” Governance for Development, World Bank. Retrieved (https://blogs.worldbank.org/governance/the-centrality-of-collective-action-problems-in- governance-for-development-new-evidence). Calland, Richard. 2010. Annex 3 Freedom of Information: Review of Impact and Effectiveness of Transparency and Accountability Initiatives. London, England: Institute of Development Studies. Retrieved (http://www.transparency-initiative.org/reports/annex-3-freedom-of-information- review-of-impact-and-effectiveness-of-transparency-accountability-initiatives). Carey, John M. 2008. Legislative Voting and Accountability. Cambridge University Press. Carlitz, Ruth. 2010. Annex 2 Budget Processes: Review of Impact and Effectiveness of Transparency and Accountability Initiatives. London, England: Institute of Development Studies. Retrieved (http://www.transparency-initiative.org/reports/annex-2-budget-processes-review-of-impact- and-effectiveness-of-transparency-accountability-initiatives). Carter Center. 2014. Women and the Right of Access to Information in Liberia: A Mixed-Methods Study. Atlanta, GA: The Carter Center. Retrieved (http://www.cartercenter.org/resources/pdfs/peace/ati/women-and-ati-prepub-07172014.pdf). Charmaz, Kathy. 2001. “Qualitative Interviewing and Grounded Theory Analysis.” Pp. 675–94 in Handbook of Interview Research: Context & Method, edited by Jaber F. Gubrium and James Holstein. Thousand Oaks, Calif: SAGE Publications, Inc. Cohen, Jean L., and Andrew Arato. 1994. Civil Society and Political Theory. MIT Press. Colquhoun, Anna. 2010. The Cost of Freedom of Information. London, England: Constitution Unit, University College London. Retrieved (http://www.ucl.ac.uk/constitution- unit/research/foi/countries/cost-of-foi.pdf). Cruzes, Daniela S., Tore Dybå, Per Runeson, and Martin Höst. 2014. “Case Studies Synthesis: A Thematic, Cross-Case, and Narrative Synthesis Worked Example.” Empirical Software Engineering 1–32. Curtin, Deirdre, and Albert Jacob Meijer. 2006. “Does Transparency Strengthen Legitimacy?” Information Polity 11(2):109–22. 67 Darbishire, Helen. 2010. Proactive Transparency: The Future of the Right to Information? A Review of Standards, Challenges, and Opportunities. Washington, DC: World Bank. Retrieved (http://siteresources.worldbank.org/WBI/Resources/213798-1259011531325/6598384- 1268250334206/Darbishire_Proactive_Transparency.pdf). Darch, Colin, and Peter G. Underwood. 2010. Freedom of Information and the Developing World: The Citizen, the State and Models of Openness. 1 edition. Oxford: Chandos Publishing. Devasher Surie, Mandakini, and Yamini Aiyar. 2014. “Implementing Right to Information: A Case Study of India.” Pp. 49–102 in Right to Information: Case Studies on Implementation, edited by Stephanie E. Trapnell. Washington, DC: World Bank. Dokeniya, Anupama. 2013. Implementing Right to Information: Lessons from Experience. Washington, DC: The World Bank. Retrieved (http://siteresources.worldbank.org/PUBLICSECTORANDGOVERNANCE/Resources/285741- 1343934891414/8787489-1344020463266/RTI-IPP-Web-Final.pdf). Dokeniya, Anupama. 2014. “Implementing Right to Information: A Case Study of Uganda.” Pp. 275–316 in Right to Information: Case Studies on Implementation, edited by Stephanie E. Trapnell. Washington, DC: World Bank. Doyle, Lynn H. 2003. “Synthesis through Meta-Ethnography: Paradoxes, Enhancements, and Possibilities.” Qualitative Research 3(3):321–44. Dragoş, Dacian C., Bogdana Neamtţu, and Bianca V. Cobârzan. 2012. “Procedural Transparency in Rural Romania: Linking Implementation with Administrative Capacity?” International Review of Administrative Sciences 78(1):134–57. Dror, Yebezkel. 1999. “Transparency and Openness of Quality Democracy.” Pp. 25–43 in Openness and Transparency in Governance: Challenges and Opportunities, edited by Michael Kelly. Maastricht: NISPAcee Forum. Erbas, S. Nuri. 2004. Ambiguity, Transparency, and Institutional Strength. International Monetary Fund. Retrieved May 29, 2013 (http://ideas.repec.org/p/imf/imfwpa/04-115.html). Florini, Ann. 2007. The Right to Know: Transparency for an Open World. New York: Columbia University Press. Fox, Jonathan. 2014. “Social Accountability: What Does the Evidence Really Say?” Retrieved (http://democracyspotdotnet.files.wordpress.com/2014/05/fox_social_accountability_evidence _gpsa_logo.pdf). Fox, Jonathan, and Libby Haight. 2011. “Mexico’s Transparency Reforms: Theory and Practice.” Pp. 353– 79 in Mexico’s Democratic Challenges: Politics, Government, and Society, vol. 19, edited by Andrew Selee and Jacqueline Peschard. Retrieved September 24, 2014 (http://www.emeraldinsight.com/books.htm?chapterid=1906522). Fox, Jonathan, Libby Haight, and Brian Palmer-Rubin. 2011. “Proporcionar Transparencia:?` Hasta Qué Punto Responde El Gobierno Mexicano a Las Solicitudes de Información Pública?” Gestión y política pública 20(1):3–61. Fram, Sheila M. 2013. “The Constant Comparative Analysis Method Outside of Grounded Theory.” Qualitative Report 18:1. Fukuda-Parr, Sakiko, Patrick Guyer, and Terra Lawson-Remer. 2011. Does Budget Transparency Lead to Stronger Human Development Outcomes and Commitments to Economic and Social Rights?. 68 Rochester, NY: Social Science Research Network. Retrieved September 20, 2014 (http://papers.ssrn.com/abstract=2211584). Fung, Archon. 2008. Full Disclosure: The Perils and Promise of Transparency. Reprint edition. New York: Cambridge University Press. Fung, Archon, Hollie Russon Gilman, and Jennifer Shkabatur. 2010. Impact Case Studies from Middle Income and Developing Countries: New Technologies. London, England: Institute of Development Studies. Gaventa, John, and Rosemary McGee. 2013. “The Impact of Transparency and Accountability Initiatives.” Development Policy Review 31:s3–28. Gelos, R. Gaston, and Shang-Jin Wei. 2002. Transparency and International Investor Behavior. Cambridge, MA: National Bureau of Economic Research. Retrieved May 29, 2013 (http://www.nber.org/papers/w9260). Gelos, R. Gaston, and Shang-Jin Wei. 2005. “Transparency and International Portfolio Holdings.” Journal of Finance 60(6):2987–3020. Glaser, Barney G. 1978. Theoretical Sensitivity: Advances in the Methodology of Grounded Theory. 1st edition. Mill Valley, Calif.: The Sociology Press. Glaser, Barney G. 1992. Basics of Grounded Theory Analysis: Emergence Vs. Forcing. Mill Valley, CA: Sociology Pr. Glaser, Barney, and Anselm Strauss. 1999. The Discovery of Grounded Theory: Strategies for Qualitative Research. Chicago: Aldine Transaction. Gläser, Jochen, and Grit Laudel. 2013. “Life With and Without Coding: Two Methods for Early-Stage Data Analysis in Qualitative Research Aiming at Causal Explanations.” Forum: Qualitative Social Research 14(2). Retrieved August 28, 2014 (http://www.qualitative- research.net/index.php/fqs/article/view/1886). Greiling, Dorothea, and Katharina Spraul. 2010. “Accountability and the Challenges of Information Disclosure.” Public Administration Quarterly 34(3):338–77. Hameed, Farhan. 2005. Fiscal Transparency and Economic Outcomes. Washington, DC: International Monetary Fund. Retrieved May 29, 2013 (http://ideas.repec.org/p/imf/imfwpa/05-225.html). Hazell, Robert, Ben Worthy, and Mark Glover. 2010. The Impact of the Freedom of Information Act on Central Government in the UK: Does FOI Work?. Basingstoke and New York: Palgrave Macmillan. Holton, Judith A. 2010. “The Coding Process and Its Challenges.” Grounded Theory Review 9(1). Retrieved September 24, 2014 (http://groundedtheoryreview.com/2010/04/02/the-coding- process-and-its-challenges/). Holzer, Marc, and Aroon Manoharan. 2007. “Global Trends in Municipal E-Government: An Online Assessment of Worldwide Municipal Web Portals.” Pp. 178–88 in Foundations of e-government, edited by Ashok Agarwal and V. Venkata Ramana. New Delhi: Global Institute of Flexible Systems Management. Hood, Christopher. 2006. “Varieties of Transparency.” Pp. 25–43 in Transparency: The Key to Better Governance?, edited by Christopher Hood and David Heald. New York: OUP/British Academy | Proceedings of the British Academy 135. 69 Ionita, Sorin, and Laura Stefan. 2014. “Implementing Right to Information: A Case Study of Romania.” Pp. 235–74 in Right to Information: Case Studies on Implementation, edited by Stephanie E. Trapnell. Washington, DC: World Bank. Joshi, Anuradha. 2010. Annex 1 Service Delivery: Review of Impact and Effectiveness of Transparency and Accountability Initiatives. London, England: Institute of Development Studies. Retrieved (http://www.transparency-initiative.org/reports/annex-1-service-delivery-review-of-impact- and-effectiveness-of-transparency-accountability-and-initiatives). Kaufmann, Daniel, and Ana Bellver. 2005. Transparenting Transparency: Initial Empirics and Policy Applications. Rochester, NY: Social Science Research Network. Retrieved May 29, 2013 (http://papers.ssrn.com/abstract=808664). Kim, Soonhee, and Jooho Lee. 2012. “E-Participation, Transparency, and Trust in Local Government.” Public Administration Review 72(6):819–28. Kuipers, Ben S. et al. 2014. “The Management of Change in Public Organizations: A Literature Review.” Public Administration 92(1):1–20. Lipcean, Sergiu, and Laura Stefan. 2014. “Implementing Right to Information: A Case Study of Moldova.” Pp. 151–78 in Right to Information: Case Studies on Implementation, edited by Stephanie E. Trapnell. Washington, DC: World Bank. Luna, Dolores E., J. Ramon Gil-Garcia, Luis F. Luna-Reyes, Rodrigo Sandoval-Almazan, and Abel Duarte- Valle. 2013. “Improving the Performance Assessment of Government Web Portals: A Proposal Using Data Envelopment Analysis (DEA).” Information Polity 18(2):169–87. Meijer, Albert. 2013. “Understanding the Complex Dynamics of Transparency.” Public Administration Review 73(3):429–39. Meknassi, Saad Filali. 2014. “Implementing Right to Information: A Case Study of Jordan.” Pp. 365–418 in Right to Information: Case Studies on Implementation, edited by Stephanie E. Trapnell. Washington, DC: World Bank. Mendel, Toby. 2014. Right to Information: Recent Spread of Legislation. Washington, DC: World Bank. Michener, Greg, and Ben Worthy. 2013. From Fishing to Experimentation: Transparency as Information- Gathering a Typology and Framework for Analysis. Rochester, NY: Social Science Research Network. Retrieved August 7, 2014 (http://papers.ssrn.com/abstract=2366707). Miles, Matthew B., and A. Michael Huberman. 1994. Qualitative Data Analysis: An Expanded Sourcebook(2nd Edition). 2nd edition. Thousand Oaks: SAGE Publications, Inc. Mizrahi, Yemile, and Marcos Mendiburu. 2014. “Implementing Right to Information: A Case Study of Mexico.” Pp. 103–50 in Right to Information: Case Studies on Implementation, edited by Stephanie E. Trapnell. Washington, DC: World Bank. Moses, Elizabeth. 2014. “Implementing Right to Information: A Case Study of South Africa.” Pp. 419–74 in Right to Information: Case Studies on Implementation, edited by Stephanie E. Trapnell. Washington, DC: World Bank. Neuman, Laura. 2009. Enforcement Models: Content and Context. Washington, DC: The World Bank. Retrieved (http://wbi.worldbank.org/wbi/Data/wbi/wbicms/files/drupal- acquia/wbi/Neuman%20-%20ATI%20Enforcement%20Models.pdf). 70 Neuman, Laura, and Richard Calland. 2007. “Making the Law Work: The Challenges of Implementation.” Pp. 179–213 in The Right to Know: Transparency for an Open World, edited by Ann Florini. New York: Columbia University Press. Nicro, Somrudee, Panicha Vornpien, and Nongpal Chancharoen. 2014. “Implementing Right to Information: A Case Study of Thailand.” Pp. 475–538 in Right to Information: Case Studies on Implementation, edited by Stephanie E. Trapnell. Washington, DC: World Bank. Open Society Institute. 2006. Transparency and Silence: A Survey of Access to Information Laws and Practices in Fourteen Countries. New York. Open Society Institute: Open Society Justice Initiative. Retrieved (http://www.opensocietyfoundations.org/sites/default/files/transparency_20060928.pdf). Pereira Chumbe, Roberto. 2014. “Implementing Right to Information: A Case Study of Peru.” Pp. 179– 234 in Right to Information: Case Studies on Implementation, edited by Stephanie E. Trapnell. Washington, DC: World Bank. Pope, Catherine, Sue Ziebland, and Nicholas Mays. 2000. “Analysing Qualitative Data.” BMJ : British Medical Journal 320(7227):114–16. Rahman, M. Zubaidur. 1998. “The Role of Accounting Disclosure in the East Asian Financial Crisis: Lessons Learned.” Transnational Corporations(7). Ring, Nicola A., Karen Ritchie, Lakshmi Mandava, and Ruth Jepson. 2011. “A Guide to Synthesising Qualitative Research for Researchers Undertaking Health Technology Assessments and Systematic Reviews.” Retrieved September 24, 2014 (http://dspace.stir.ac.uk/handle/1893/3205). RTI Assessment & Analysis Group (RaaG). 2009. Safeguarding the Right to Information: Report of the People’s RTI Assessment 2008. National Campaign for People’s Right to Information (NCPRI). Retrieved (http://rti-assessment.org/). Schedler, Andreas. 1999. “Conceptualizing Accountability.” Pp. 13–28 in The Self-restraining State: Power and Accountability in New Democracies, edited by Larry Jay Diamond, Marc F. Plattner, and Andreas Schedler. Boulder, CO: Lynne Rienner Publishers. Shepherd, Elizabeth, Alice Stevenson, and Andrew Flinn. 2010. “Information Governance, Records Management, and Freedom of Information: A Study of Local Government Authorities in England.” Government Information Quarterly 27(4):337–45. Shepherd, Elizabeth, Alice Stevenson, and Andrew Flinn. 2011. “Records Management in English Local Government: The Effect of Freedom of Information.” Records Management Journal 21(2):122– 34. Shin, Yongseok, and Rachel Glennerster. 2003. Is Transparency Good for You, and Can the IMF Help?. Washington, DC: International Monetary Fund. Retrieved May 29, 2013 (http://ideas.repec.org/p/imf/imfwpa/03-132.html). Stammers, Neil. 2009. Human Rights and Social Movements. London ; New York, NY : New York: Pluto Press. Strauss, Anselm L. 1987. Qualitative Analysis for Social Scientists. Cambridge Cambridgeshire ; New York: Cambridge University Press. Tacconelli, Evelina. 2009. Systematic Reviews: CRD’s Guidance for Undertaking Reviews in Health Care. University of York: Centre for Reviews and Dissemination. Retrieved September 17, 2014 71 (http://www.thelancet.com/journals/laninf/article/PIIS1473-3099(10)70065- 7/fulltext#article_upsell). Thomas, James, and Angela Harden. 2008. “Methods for the Thematic Synthesis of Qualitative Research in Systematic Reviews.” BMC Medical Research Methodology 8(1):45. Trapnell, Stephanie E., ed. 2014a. “Implementing Right to Information: A Case Study of the United Kingdom.” Pp. 317–60 in Right to Information: Case Studies on Implementation. Washington, DC: World Bank. Trapnell, Stephanie E., ed. 2014b. Right to Information: Case Studies on Implementation. Washington, DC: World Bank. Trebicka, Jolanda, and Gerti Shella. 2014. “Implementing Right to Information: A Case Study of Albania.” Pp. 3–48 in Right to Information: Case Studies on Implementation, edited by Stephanie E. Trapnell. Washington, DC: World Bank. Worker, Jesse. 2014. Requests and Appeals Data in Right to Information Systems: Brazil, India, Jordan, Mexico, South Africa, Thailand, United Kingdom, and United States. Washington, DC: World Bank. Retrieved (http://siteresources.worldbank.org/PUBLICSECTORANDGOVERNANCE/Resources/285741- 1343934891414/8787489-1344020463266/8788935- 1399321576201/Requests_and_Appeals_RTI_Working_Paper.pdf). Worthy, Ben, Jim Amos, Robert Hazell, and Gabrielle Bourke. 2011. Town Hall Transparency? The Impact of the Freedom of Information Act on English Local Government. London, England: The Constitution Unit Department of Political Science UCL (University College London). Retrieved (http://www.ucl.ac.uk/constitution-unit/research/foi/foi-and-local-government/town-hall- transparency.pdf). 72 Appendix A: Codes and descriptions Domain: Institutional capacity Field: Financial and material resources Code: Budget Any reference to costs or budgeting. Any of the following: Equipment failures, insufficient Code: Equipment equipment, lack of office space for RTI storage or activities including records management. Field: Formal practices Code: Appeals process Referring to the internal agency appeals process. Excessive layers of approval needed in order for information to be released. Includes the involvement of Code: Bureaucratization multiple layers of authority that appear unnecessarily controlling. Referring to absence or presence of an internal system for Code: Proactive system updating proactively released information and adding new content. Referring to the presence or absence of practices that establish intellectual control of records in all repositories. Code: Records management This includes catalogs of records/central indices of records/registers of information, even if not part of electronic records management systems. Referring to the information request process, including Code: Request process forms, submission methods, and internal procedures Field: Staff capacity Any of the following: Lack of awareness of RTI by staff, lack of knowledge of RTI rules or laws, incapable of doing job correctly or professionally. Does not apply to informal Code: Staff knowledge systems or overall organizational capacity, just staff. See Senior leadership – inadequate for organizational capacity. Any reference to training for public officials, whether Code: Training provided by national or local agency, or CSOs and media. Referring to the level of politicization within the public sector, e.g, excessive politicization through political Code: Politicization appointments, or professionalized civil service supported by meritocratic recruitment. Field: Staff incentives Public officials are motivated to disclose information as Code: Ethical norms part of their role in administering the law, or as an informal norm dictating the role of the public servant. Reference to the presence or absence of financial Code: Compensation incentives for public officials to disclose information in their role as PIOs. 73 Existence of sanctions for release of information, Code: Penalties for disclosure especially in the absence of clear guidelines, or presence of controlling law such as ethics or civil service laws Code: Performance evaluations RTI is incorporated into staff evaluations. Field: Staffing levels Referring to the presence or absence of dedicated information officers that are appointed within agencies. Code: PIOs Includes number of staff handling RTI to deal with workload. Staff turnover is a problem, either through time-bound Code: Personnel – turnover political appointments, frequent restructurings, or employee movement outside public sector. Referring to the presence or absence of units within Code: Responsible units agencies that are tasked with handling RTI requests. Domain: Enabling Conditions Field: Advocacy efforts Any of the following: civil society organizations and media are legally not allowed to operate freely on issues pertaining to RTI or face resistance from government regarding engagement on RTI advocacy issues. Code: Civic space Government actions may limit the ability of CSOs and media to carry out activities, even if not enshrined in law. Does not apply to arrest, detention, professional or personal harassment, and physical assault. See Fear Factor. Civil society organizations and/or Media engage in advocacy activities to improve RTI implementation. This includes research, strategic litigation, articles, domestic Code: CSO – Advocacy and international networking, conferences, etc. Also Implementation includes compliance testing conducted by civil society organizations that serves to complement self-reported data by agencies. Does not include joint efforts with public officials, which is coded State-society collaboration Civil society organizations and/or Media actively engage in advocacy activities to improve RTI legal framework. This Code: CSO – Advocacy Law includes research, articles, domestic and international networking, conferences, meetings with government, etc. Field: International intervention Referring to the presence of international obligations or communities that exert pressure on governments to Code: International pressure establish RTI systems, e.g., Open Government Partnership, European Union, etc. Presence of foreign aid as either direct budget support or Code: International funding strategic funding for specific initiatives related to RTI. Field: Legal framework Refers to strengths or weaknesses of RTI law that affect Code: RTI law - implementation implementation. 74 In practice, agencies hold public consultations on policy Code: Public consultations decisions or implementation processes. Code: Whistleblower There is a whistleblower protection law. Code: Data protection There is a data protection/ privacy law. There is a state secrets law/official secrets act that Code: State secrets substantively compromises the RTI law. Field: Policy prioritization Any of the following: Senior management-level commitment, strategic direction, support is not adequate or is absent; Public Information Officers participate in Code: Agency leadership organizational planning; RTI organizational planning or strategies are developed and revised as needed; Chief RTI officer at level of senior management is responsible for agency-level RTI policy management. Referring to any public expression by bureaucracy in Code: Bureaucratic opposition to law opposition to the RTI law. Includes high-level opposition to the RTI law within the public sector. Referring to public support expressed by politicians and Code: Political support political appointees for RTI laws or implementation processes. Domain: Demand for Information Field: Accessibility Referring to any obstacles to accessibility of the internal appeals process, or the establishment of mechanisms to Code: Access to agency appeals ensure access by marginalized groups. Includes clarity of the internal appeals process as communicated to the public. Referring to any obstacles to accessibility of the external appeals process, or the establishment of mechanisms to Code: Access to external appeals ensure access by marginalized groups. Includes clarity of the external appeals process as communicated to the public. Referring to any obstacles to accessibility of the request process, or the establishment of mechanisms to ensure Code: Access to request process access by marginalized groups. Includes clarity of the request process as communicated to the public. Any reference to petty corruption or bribes that affect the Code: Corruption information request process. The appeals system requires substantial financial Code: Costs – Appeal deterrent resources from appellants, thus discouraging appeals or causing appeals to be abandoned for lack of funding. Referring to CSO practices to educate or assist citizens Code: CSO assistance with filing information requests. Refers to the extent to which responses to information requests are comprehensible and useful, e.g., low quality Code: Intelligibility datasets, legalese language, reams of paper responses, etc. 75 Referring to means of disseminating proactively disclosed Code: Proactive – dissemination information. Informal requests are the norm, rather than exceptions, for media and organizations engaged in activities with certain government offices , e.g., public-private partnerships, CSO collaboration, etc. This also includes an informal request process for any requesters that is Code: Request informality instituted by staff, which by-passes any rules or guidelines (if they exist). Applies to requests that are handled and responded to, indicating that a system exists even if informal. Also includes use of personal connections. For discarded, lost, or ignored requests due to informality, see Cataloging - inadequate. Any reference to the composition of requester groups, Code: Requester profile particularly marginalized communities. Any of the following: bureaucratic resistance to openness; lack of incentives to release information; unwillingness to deal with sensitive or complicated requests; several layers Code: Resistance of managerial approval necessary to release information; centralized approval process requiring head of department or agency to sign off on every information request. Field: Public awareness Any reference to private sector involvement in RTI, Code: Private sector through advocacy, legal drafting, implementation, or usage. Knowledge about RTI by ordinary citizens. May result in Code: Public knowledge low numbers of requests for information. National-level efforts to educate the public about RTI and raise awareness of the RTI system. Agency-level efforts to educate certain publics about the RTI process in that Code: Public outreach agency, including the request and appeals processes. Includes both online and other means of disseminating this kind of information. Referring to the extent of demand for information from Code: Public use of RTI the public, or the use of information by the public, CSOs, media, or private sector. Domain: Oversight Field: Monitoring Any activities that the nodal agency conducts, or fails to conduct, to support the implementation process within and across agencies. Reference to the impact of these Code: Monitoring activities activities should also be included. Does not include CSO monitoring, which should be coded as CSO – advocacy implementation. 76 Referring to the presence or absence of monitoring systems that collect data on RTI requests and appeals at the agency level. Also refers to the quality of the monitoring system (e.g., normal course of business vs ex Code: Monitoring – data post fact gathering) and reliability of the data collected. Also includes good practices in tracking requests and downloads so that popular content is updated frequently and/or operational efficiency is improved (e.g., Mexico online repository of all responses). Any reference to implementation legislation or controlling Code: Implementing rules guidelines issued by monitoring body that explains how the law is to be implemented within the public sector. Internal systems for processing requests are used by officials to track status of request, in order to relay Code: Internal tracking information to requesters and produce accurate reports of request status. Field: Enforcement Referring to the authority of the external body Code: Binding decisions (commission, council, enforcement, etc.) to issue binding resolutions. Also includes outcomes of resolutions. Referring to complaints about RTI issues (not appeals, Code: Complaints which require a basis in law). Also includes outcomes and influence of body to resolve complaints. The national enforcement agency uses non- confrontational measures to oblige agencies to release Code: Compliance – informal information, such as private meetings and consultations, thereby avoiding the use of sanctions, resolutions, and official decisions on appeal. Referring to the process of filing and pursuing appeals Code: External appeals process with the external appeals authority or the courts. The judicial system does not have the capacity to rule effectively on RTI cases, either because the system is Code: Judiciary overburdened, or judges lack knowledge or expertise. The judiciary may also be reluctant to rule on RTI cases due to political or sensitivity concerns. The external body (commission, council, enforcement, etc.) that has official authority to recommend sanctions for RTI violations. Does not apply to agency-level Code: Sanctions sanctions on staff. Also includes whether sanctions were applied by appropriate authority, and frequency of recommendations for sanctions. Transformative factors Opportunities provided by government agencies for collaboration with civil society organizations or media on Code: State-society collaboration activities, for meetings and seminars on issues related to RTI, or for CSOs or media to participate in policy and decision-making on RTI. 77 Referring to online portals for release of proactively disclosed information, either centralized or decentralized. Code: Technology Also includes issues with ICT staff, data transfer and interoperability of databases. Field: Intra-governmental collaboration Referring to either the presence or absence of systems and lines of responsibility regulating transmission of Code: Intra-agency cooperation information between departments, requiring all staff to cooperate with the information officer in accessing information. Referring to the presence or absence of coordinating offices or individuals that provide guidance, liaise with the Code: Liaisons monitoring body, and/or develop communities of practices among PIOs National records management authorities collaborate Code: Records-RTI collaboration with RTI oversight bodies to harmonize record-keeping practices. Referring to the capacity of the national records Code: Records oversight management authority. 78 Appendix B: Inter-coder reliability results Coded Text Coder #1 Coder #2 Index Coder According to the CDDI and Citizens Advocacy Office (CAO) surveys, many public requests for information, whether from individuals or civil society organizations, are treated carelessly 1 and often ignored. There are numerous cases of Resistance Resistance Resistance administrative silence (that is, when the administration does not provide any response even when the time limit for providing information has been exceeded); A CDDI survey found that 87 percent of public employees were PIOs, Staff Staff knowledge, 2 unaware of the law’s existence, and very few institutions had PIOs Knowledge PIOs appointed PIOs. There are no common standards, regulations, or Implementing administrative guidelines followed across institutions to align Implementing Rules, rules, Monitoring 3 their internal rules and procedures with the requirements of Agency leadership Monitoring activities activities, Agency Law 8503; this is left to the discretion and willingness of the leadership administration of each institution. Law 8503 does not specifically oblige the state administration, Ministry of Finance, and/ or public entities to allocate specific funds for its implementation. Still, there has been a trend over recent years for ministries to seek and allocate funds for Budget, Monitoring Budget, Proactive – 4 Budget activities such as Web sites, conferences, workshops, public Activities dissemination awareness campaigns, and other initiatives, to create an environment in which citizens have quicker access to more information. The behavior of officials in their direct interactions with citizens is often described as “not according to the ethical Request 5 Request Informality Request Informality norms,” but rather based on “individual education or informality approach." The role of civil society in lobbing for adopting or rejecting State - Society legal changes to the ATI framework has been also important; CSO – Advocacy CSO - advocacy 6 Collaboration, CSO - an example of this is the recommendations made to the Law law Advocy Law government in 2008 to amend the Law No. 8503. The politically motivated turnover of public employees has continued, intensifying since the formation of the new Personnel government in 2009; this has undermined the competitive Politicization, Personnel – 7 turnover, recruitment procedures required under the Civil Service Law Personnel Turnover turnover Politicization and perpetuated the politicization of the public administration. Records management systems have not yet been fully developed in the line ministries. In general, the record Records Records Records 8 management system is still paper-based; digitalization of Management, management, management, records management system is in a very early stage of Equipment Technology Technology development. Civil society groups monitor implementation, and several CSO - advocacy Implementing Rules, Budget, surveys have been conducted both at the central and local implementation, 9 CSO Advocacy International levels. The Soros foundation, EU, USAID, and other donors International Implementation funding have financed NGOs in this area. funding The legislation governing the People’s Advocate limits its powers to expressing opinions and making recommendations; it can start an investigation into a complaint presented by an entity or citizen and can recommend a criminal investigation, Complaints, Binding Complaints, 10 Complaints court action, or dismissal of officials for serious offenses, but Decisions Binding decisions its decisions are not binding. Therefore the implementation of its recommendations depends on the agency manager’s decision. 79 Table 2 shows that a reasonable number of civil servants have attended ITPA courses, taking into account the relatively small Personnel - Training, number of relevant positions at the central and municipal Turnover, Personnel 11 levels. Unfortunately, these capacity-building efforts have Personnel turnover Politicization, turnover, been difficult to sustain because of high staff turnover and Training Politicization continuous changes in both government administrations and specific ministries. Despite the formal norms requiring the implementation of the right to ATI, informal norms contribute to its poor Ethical Norms, Bureaucratization, implementation, including the high degree of centralization BureaucratizationRes Request Public Knowledge, 12 with which the law is handled in the bureaucracy, a lack of istance, Corruption, informality, Public corruption awareness among citizens of their rights, and the Public Knowledge, knowledge, pervasiveness of petty corruption in the relationships between Request Informality Corruption bureaucrats and citizens. The decision-making process is highly centralized in Albania, the delegation of power is not at a level that could contribute State-society to further empowerment and accountability, and Public Consultation, collaboration, 13 responsiveness towards the public. Despite the progress made Public consultations Resistance Public so far, the government rules of proce- dures for ensuring full consultations access of interested parties or civil society to the policy drafting process are not yet adequately developed. A physical and electronic audit of government departments at the central, state, and district levels found that most of the Section 4 information published was incomplete and out- Monitoring dated. For example, while 65 percent of urban public Proactive System, Proactive 14 activities, Proactive authorities had published details about their respective their Budget dissemination – dissemination organizations on their Web sites, only 45 percent had published PIO information, and only 25 percent had published information on budgets and salaries Approximately 60 percent of both rural and urban PIOs have not received any RTI training, and approximately 40 percent of PIOs cited this as a constraint on their capacities to supply information to RTI applicants. PIOs, Training, Staff Training, Staff Training, Staff 15 The lack of training is reflected in the low levels of awareness Knowledge Knowledge knowledge about the law among PIOs; among the rural PIOs surveyed, 30 percent did not know the provisions of the RTI Act, and are not aware of their roles and responsibilities under it. The number of information officers varies depending on the number of offices, branches, and administrative units within a 16 PIOs PIOs PIOs given ministry or department (as can be seen in the four departments analyzed in table 1). In addition, the information commissions have been empowered to order public authorities to fully comply with the provisions of the RTI Act. Specifi cally, they may order Binding Decisions, Records public authorities to appoint information officers, publish 17 Training, Records Management,Traini Binding decisions specific categories of information, make information available Management ng, PIOs in a particular form, improve records management practices, and enhance the training of officials in the provisions of the RTI Act. The Indian media were early supporters of RTI, with senior journalists lending strong support to the movement. They have also served as watchdogs: in 2006, when rumors of a possible government amendment to the law opposed by the CSO - Advocacy Civic Space, CSO – media began to circulate, the media opposed it. Since its Implementation, CSO Media - advocacy 18 Advocacy passage, national newspapers have regularly featured articles - Advocacy Law, law Implementation relevant to the RTI Act. The RAAG survey of over 60 Public Use of RTI publications in English, Hindi, and other regional languages found that, on average, 65 news items per publication per year deal with RTI. 80 The jurisdiction of each PIO and AA has been clearly defined: officers process information requests and appeals related to Access to Agency Access to request 19 the specific schemes or programs being handled by them. This Appeals, Access to PIOs process, Access to clear allocation of subject area benefits both the officers and Request Process agency appeals citizens who can address queries to specific PIOs or AAs. 82 percent of the public authorities surveyed within the central government reported the need for additional 20 infrastructure to implement the RTI Act; public authorities at Equipment Equipment Equipment the block and local levels of government lack basic infrastructure like photocopy machines and computers. The PwC study concludes that there has been inadequate planning among public authorities for implementing Section 4. Under the RTI Act, it is the responsibility of the public Intra-Agency authority as a whole to proactively publish information, but at Proactive system, Cooperation, Agency Responsible units, 21 a departmental level there is no clarity about whether Section Intra-agency Leadership, Proactive Agency Leadership 4 is the responsibility of the heads of departments or the PIOs. cooperation System This lack of ownership and of clearly defined roles and responsibilities for updating Section 4 information is considered one of the major reasons for poor performance. There are a total of 60 ministries under the central government, each with a number of departments and public authorities under its jurisdiction. Aggregate data on Monitoring - Data, compliance with the RTI Act by these ministries are Proactive - Proactive – 22 unavailable, but information from the government’s national Monitoring data Dissemination, dissemination, PIOs RTI portal suggests that a large number of ministries have Technology complied with the basic provisions of the law, including appointing PIOs and appellate authorities and proactively disclosing information. While the law does not specifically provide for the designation of a nodal implementing agency, institutional arrangements have been made with specific departments to lead implementation at the central and state government levels. At the central level, the DOPT in the Ministry of Personnel, Public Monitoring - Implementing rules, Monitoring 23 Grievances, and Pensions has been designated as the nodal Activities, RTI law activities agency. In this role, the department has the powers and Implementing Rules responsibilities of the central government as outlined under the RTI Act. Notably, the DOPT was closely involved in the formulation and drafting of the RTI Act; it even presented the draft bill in Parliament. There seems to be a wide variation in the seniority levels of PIOs within the central government and across states. In many PIOs, Agency PIOs,Agency 24 instances, junior officers have been designated as PIOs and Agency leadership Leadership leadership AAs, which, according to a 2009 CIC sub committee, is likely to have a detrimental effect on the quality of decisions. Although the RTI campaign did receive some support from progressive bureaucrats (notably, N. C. Saxena and Harsh Agency Leadership, Agency Mander, among others), the bureaucracy’s overall response to Resistance, Civic Space, leadership, 25 the law has been ambivalent. Specifically, this resistance was Bureaucratic Resistance Bureaucratic manifest in repeated attempts to Opposition to Law opposition to law amend the law and a lack of effort at improving the internal capacities of departments to supply information. In each of the four departments analyzed, PIOs handle other portfolios in addition to their responsibilities under the RTI Act. But responsiveness to RTI requests is not considered in the annual personnel performance appraisal of offi cials PIOs, Compensation, Compensation, designated to perform RTI-related functions. Notably, there Performance Performance 26 Performance are no specific incentives that reward PIOs for good Evaluations, evaluations, PIOs evaluations performance in discharging their RTI responsibilities, which is Monitoring Activities one reason for their lack of motivation. Over 10 percent of the PIOs surveyed cited a lack of financial and other incentives as reason for their reluctance to be PIOs. 81 No focal point or Information Agent (IA) is available at public 27 PIOs PIOs PIOs institutions to deal with requests for information (RFI); Access to Request Access to request There are no RFI forms except at the Ministry of Finance, the Access to request 28 Process, Request process, Request Ministry of Justice and the National Library; process Process process The responses received from the IC show that the ministries in question admitted classifying some of their documents as Records Records 29 confidential, but it was noticed that these categories were not Management, Staff Records control management founded on a clear basis for all the ministries. Each ministry Knowledge made its own categorization according to its opinions. There is no specific budget for ATI activities at any level. The ICO and the IC carry out their role without any financial compensation, and the same situation is to be found at the Budget, 30 agency level. Activities of ATI promotion have to be funded by Budget Budget Compensation different lines of the agency budget. There is no specific line item for ATI in agencies’ budgets. Instead, the ATI is funded from agencies’ administrative budgets. There is no coalition or network of CSOs creating synergies CSO - Advocacy Law, and increasing decisively the level of organization and CSO - advocacy 31 CSO - Advocacy CSO - advocacy law advocacy to move forward with amendments and influence law Implementation effectively the parliament and other institutions of the State. ATI is still regarded as a topic to be discussed by experts rather than by citizens who can see the importance of reforms that ATI could bring. There is a need for more efforts to develop Public Knowledge, Public knowledge, 32 Staff Knowledge knowledge around ATI and encourage its use, especially Staff Knowledge Staff knowledge among government departments responsible for its enforcement. CSOs operate in a politically conservative and biased general 33 environment that supports the state’s interventionist role in Civic Space Civic Space Civic space their affairs, which in turn weakens their impact. The Court system proved to have major limitations for the implementation of LSAI. The choice of the HCJ to look into Costs - Appeal Costs – Appeal Costs - appeal 34 complaints is a serious challenge because of the nature of the Deterrent deterrent deterrent Court and the costs that it involves to appeal the decisions of public agencies on ATI. The LSAI does not explicitly promote proactive disclosure and it limits the access to information to the existence of a legiti- mate interest and reason presented by the requester for public information. It states in article 7 that each Jordanian Request Process, Weakness in RTI citizen “is entitled to have access to the information he/she Access to request 35 Access to Request law, Access to requests pursuant to the provisions of this law, if he/she has process Process, RTI Law request process legitimate interest or cause for that.” The LSAI grants access to information exclusively to Jordanian citizens and requires requesters for information to demonstrate what their interest is in accessing the requested information. The ICO has no mandate to request information from other public agencies. He has the mandate of filing the received appeals and presenting them to the IC, for it to resolve the Binding Decisions, case. The ICO then sends the decision to the agency that was Request Process, 36 Binding Decisions Binding decisions the subject of the appeal and the requester. Moreover, the IC Access to Request does not interfere in the classification and categorization Process process. The IC merely receives a list of existing information and its categorization from the public institutions. The Official Communiqués of successive Prime Ministers asking public institutions to implement the LSAI prove that this exercise is incomplete. These executive orders show how Monitoring Monitoring 37 Monitoring Activities difficult it is to enforce the ATI law in the absence of a central activities activities independent agency in charge of the implementation and monitoring of the LSAI. 82 The majority of public institutions have an information center that is supposed to gather the information of the agency and Records organize it so it can be accessed by the public. These Records Management, information centers are not organized the same way in all Records management, 38 Responsible Units, agencies and their resources are very variable depending on Management Proactive Proactive - the priority that each agency gives them. These centers are dissemination Dissemination considered more as libraries than offices and are responsible for organizing the information produced by their own agency. Despite constitutional protections of freedom of information, there remain several laws that hinder free expression and access to information. These include the Jordan Press Associa- tion Law (1998), the penal code (1960), the Defense Law (1992), the Contempt of Court Law (1959), the Protection of Penalties for 39 State Secrets and Classifi ed Documents Law (1971), and the Disclosure, State State Secrets State secrets Press and Publications Law (1999 and its amendments of Secrets 2012). In this context, existing legislation, especially the Archives Law and the Law on the Protection of State Secrets and Documents (LPSSD), will limit signifi cantly the information that may be accessed. A survey of midlevel federal public officials commissioned by the Federal Institute for Access to Information (IFAI) in early 2007 revealed mixed results. Sixty percent of public offi cials Resistance to 40 Resistance Resistance interviewed acknowledge that “officials in some units within disclose their agencies believe they own the information they generate and do not share it, even with other units in the same agency.” The development of an e-platform—the SISI, System for Information Requests—to handle information requests was Monitoring - Data, one of the key innovations introduced by the IFAI. This e- Access to request Technology platform (the expanded version of which was renamed process, 41 Internal Tracking, Technology INFOMEX) both facilitated users’ ability to make information Technology, Access to Request requests, and allowed the IFAI to keep track of government Internal tracking Process agencies’ responses to information requests and supervise their compliance with the law. In 2001 the government’s draft bill was leaked to the press, sparking a debate among the academics, media practitioners, and CSOs who had been advocating for right to information legislation. In May 2001, a number of media executives, CSO - Advocacy Law, journalists, and academics formed a loose coalition—known as Proactive - CSO - advocacy 42 the Grupo Oaxaca—that critiqued the government’s bill and Dissemination, State CSO - advocacy law law drafted an alternative transparency bill, lobbied members of - Society Congress, negotiated with government officials, and organized Collaboration conferences and public forums to disseminate information about the law and generate stronger bases of support, placing the topic on the public agenda. Beyond the number of personnel assigned to each liaison unit, leadership commitment can be evaluated by the profile that liaison units are given within the agency, the reporting lines of Liaisons, Agency Liaisons, Agency 43 Agency leadership authority inside the agency (to whom liaison units report), and Leadership Leadership the engagement of high- level officials in the liaison units and information committees. Both the IFAI and SFP are quite limited in their direct capacity to influence the performance of liaison units and information Liaisons, Monitoring Monitoring 44 committees, and no network exists to allow liaison units in Liaisons Activities activities, Liaisons different agencies to share their experiences and learn from one another. 83 Depending on the number of requests each agency receives, it can appoint as many staff as needed to its liaison unit. For instance, the Instituto Mexicano del Seguro Social (Social Liaisons, Records Security Institute, the federal agency with the highest number Liaisons, Management PIOs, Records 45 of information requests—many of which are requests for Monitoring PIOs, Monitoring control personal data), has appointed a significant number of officials activities Activities to its liaison unit and made substantial improvements in its own records management system to respond to the volume of information requests. More specifi c and detailed implementation provisions were defi ned in the law’s implementing regulations, drafted by the Implementing 46 Implementing Rules Implementing rules Federal Commission for Regulatory Reform (Comision Federal rules de Mejora Regulatoria) in 2003. Federal agencies have incentives to comply with the transparency law because their response to information Monitoring - Data, Monitoring requests is permanently monitored by the IFAI. Through Monitoring - activities, INFOMEX, the IFAI keeps track of all information requests Activities, Internal Monitoring data, 47 received by each agency, the time the agency takes to Compensation Tracking, Compliance Compliance - respond, and the number of requests that are appealed to the – Informal informal, IFAI. Moreover, heads of agencies have an incentive to be well Compensation Technology evaluated, as the IFAI’s evaluations are public. No agency wants to be listed as a poor performer. Agencies might often reply to information requests by issuing information that is difficult—if not impossible—to interpret as Resistance, 48 it is published in PDF format and therefore difficult to reuse. Intelligibility Intelligibility Intelligibility Or, they might respond with pages and pages of legalistic argumentation that is virtually impossible to decipher. Yet the majority of requests are concentrated, both regionally and professionally. Most users (more than 50 percent) are 49 Requester Profile Requester profile Requester profile concentrated in Mexico City’s metropolitan area; 74.6 percent of users have college degrees or postgraduate degrees. To comply with these provisions and make the formats for posting information uniform, the IFAI designed a central Web portal called Portal de Transparencia or POT (Transparency Portal). Each government agency is required to set a Web link Technology, Proactive to the POT on their Web sites. Although some government Proactive - dissemination, 50 Technology officials complain that uploading information can be Dissemination, Technology, sometimes challenging, and some civil society practitioners Proactive - System Monitoring data argue that POT is not always user friendly, it has significantly helped in the verification of compliance with proactive disclosure provisions. Aside from personnel, federal agencies’ budgets are only minimally affected by complying with the law. “For most agencies, implementing an ATI law is possible by diverting existing assets, and as such, is a matter of people, not equipment” (Hernández-Valdéz 2009: 18). Moreover, the study also reveals that relatively few additional personnel are 51 required to respond to information requests. According to Budget, PIOs PIOs PIOs, Budget Alfonso Hernández Valdéz report, in 2007, in small agencies it took only one full-time staff member to respond to all information requests received, on average; in medium- sized agencies, the number rose to 2.5 full-time personnel; and in large agencies the number was 25 (it must be noted that these agencies employ thousands of employees). While the IFAI’s decisions are binding, it does not have sanctioning powers, and thus, it has limited enforcement capacity. In the event a federal agency does not comply with a 52 resolution by the IFAI, the most the IFAI can do is to inform Sanctions Sanctions Sanctions the SFP, the entity responsible for disciplining the federal public servants. The lack of sanctioning powers is considered to be one of the major weaknesses of the IFAI. 84 A new law on state secrets adopted in November 2008 (Access-Info 2008b; 2009a) retained a number of restrictive provisions from the previous law (Banisar 2008). The new law sought to extend the defi nition of a state secret, as well as of types of information that may be protected, lacked any 53 State Secrets State secrets State secrets provisions for whistle-blowers, extended deadlines for protecting secret information were too extensive, and left little room for parliamentary control. As the law does not contain any provision regarding a potential conflict with the RTI act, the secrecy of public institutions continues. Regulations on how the law should be implemented have not been issued, so individual institutions are left to decide if and Implementing 54 Implementing Rules Implementing rules how supplementary regulations should be adopted (some did, rules most did not). There is no dedicated budget allocated for the implementation of the law. Most ministerial accounting departments have no 55 tradition of such collecting fees from requesters and consider Budget Budget Budget the extra trouble of collecting fees not worth its potential benefits. Overall, there is little ownership of the law by political leaders. The issue of the implementation of the RTI law has been at the margins of the public agenda. The political crisis that followed Monitoring 56 the adoption of the law contributed to undermining the Monitoring Activities Political support activities chances of its effective implementation. Implementation of RTI has been at odds with the political agenda of the post- 2001 government. A monitoring report by a local NGO, which looked at 95 public bodies—64 central and 31 local—shows the lack of awareness Staff Knowledge, 57 among civil servants about the law’s main provisions and the Staff Knowledge Staff knowledge Monitoring Activities signifi cant level of confusion on the difference between petitions and proper information requests. Preparing an answer for an RTI request usually involves several persons in any public institution. The fi rst point of contact for a request is usually the offi cials working in the department of public relations, who handle many other tasks. When a request is submitted, it is usually forwarded to the minister or senior offi cials, who decide which department is Bureaucratization, 58 competent to answer the applicant. The response has to get Bureaucratization Bureaucratization Resistance approved by all department heads, the legal department, and the minister. The answers to requests are usually signed by the heads of institutions or other top management offi cials, which shows the existence of multiple fi lters in the provision of information and the lack of autonomy of public servants responsible for this task. Data on ATI requests in many state institutions are still not separated from those of individual petitions. It is usually up to 59 Request Process request process Request process the individual who fi rst receives the document to qualify it as either a petition or a request for information. The judiciary has problems of capacity, integrity, and politicization, as the European evaluation missions attest—a combination of poor material conditions in courts, overburdened staff, and a lack of exposure to international 60 legal thinking, especially in new areas such as modern Judiciary Judiciary Judiciary commercial law, human rights, or freedom of speech. Judges are poorly prepared to rule on the few ATI-related cases that come before them. Moreover, judges tend to stand with public institutions. 85 The SGP, with technical assistance from the PCM’s National Government Offi ce of Electronics and Computing (ONGEI), Proactive developed the Standard Transparency Portal. This portal is Technology, 61 Technology dissemination, designed to resolve the problem of diverse formats, Monitoring - Data Technology inadequate content, duplicate information, and outdated data on the institutional portals referred to in LTAIP. Although several ministries claim that there is a database on Records RTI requests, in response to requests, they have either stated Records 62 Management, Internal tracking that they cannot provide the requested information, or not management Resistance responded. Only those requests that cannot be solved on the spot and are docked in a written form, are in fact registered as ATI requests and counted in the formal statistics. This can be particularly Request 63 mis-leading at the local level, where much of the interaction Request Informality Request Informality informality that happens between local authorities (who are the main service providers for citizens) and their clients is verbal and informal. These studies suggest that there is low awareness about RTI 64 among citizens, and as a consequence, a low number of Public Knowledge Public knowledge Public knowledge requests. After the 2001 elections, the space for civil action also shrunk considerably. CSO efforts focused on monitoring and CSO - Advocacy CSO - advocacy CSO - advocacy 65 assessment projects in most cases, and less direct action Implementation implementation implementation against a specific institution in a flagship case, or strategic litigation. According to IPYS, which has extensive experience making requests to and filing complaints against the government, filing an action of habeas data without a legal defense hurts the plaintiffs. This is because litigation requires a certain level of specialization in the application of procedural norms. Access to External Access to external 66 Judiciary Further, judges tend to require excessively bureaucratic Appeals appeals procedures that are not common knowledge among citizens. Finally, government bodies are defended by public prosecutors, thereby generating a technical imbalance in the litigation, to the detriment of plaintiffs. Another important norm is Law No. 28664, Law of the National Intelligence System (SINA) along with the National Intelligence Directorate (DINI), adopted on January 4, 2006. 67 One negative aspect of this law is that it establishes longer State Secrets state secrets State secrets time limits than LTAIP for the declassification of information: 20 years for secret information, 15 years for reserved information, and 10 years for confi dential information There are key general regulatory norms; in other words, ones that are applicable to all bodies subject to LTAIP. Many public Implementing 68 entities have issued regulations and directives, but these are Implementing Rules Agency leadership rules procedural or organic provisions about the general obligations of each entity. As discussed at the Third National Conference on Access to Public Information (October 2010), most government bodies Request Process, do not have internal policies to ensure compliance with the Access to request 69 Access to Request Request process obligations stipulated in LTAIP for each entity (for example, a process Process regime of internal responsibilities to respond to information requests). Although LTAIP stipulates that noncompliance with its contents constitutes a serious offense and may result in a criminal charge of abuse of authority, compliance ultimately depends on the director of the entity. An example of this is 70 Mayoral Resolution No. 1364- 2010-MPT of December 16, Sanctions sanctions Sanctions 2010, by which the Mayor of the Provincial Municipality of Trujillo suspended a municipal offi cial for 30 days without pay for failing to respond in a timely fashion to a request of access to public information. 86 Because the resources for this work were limited, the German International International International 71 cooperation agency, GTZ provided fi nancial support for the funding, funding, Funding, Technology design and development of the portal. Technology Technology Although signifi cant progress was made in achieving the objectives of the fi rst phase, the implementation of the Standard Transparency Portal is now experiencing some difficulties. In 2011, the change in municipal, regional, and Personnel central government offi cials affected the continuation of this Technology, Technology, 72 turnover, process. Many local and regional government officials trained Personnel - Turnover Personnel Turnover Technology in the implementation of this tool are no longer in office. Moreover, the two offi cials who launched and coached the implementation of this tool in the SGP are no longer employed at the PCM. Between 2009–10, the SGP attempted to train the national offi cials responsible and to build their RTI capacities. There are few available resources for these activities, however, for 73 Budget, Training training, budget Training, Budget which reason the entities requesting the training had to cover the costs themselves, as was the case for the Puno regional government. Finally, the successful exercise of persuasive authority (the magistracy of persuasion) by the Ombudsman’s Offi ce has enabled it to exercise leadership that is reflected in high levels of social acceptance. The persuasive, noncoercive nature of its Monitoring Compliance - decisions, far from being a disadvantage, signifi cantly Compliance – activities, 74 Informal, Monitoring contribute to the achievement of its institutional objectives informal Compliance - Activities because they allow the offi ce to exercise its influence in a informal variety of ways (through public declarations or reports, for example) in its efforts to promote good government practices and respect for fundamental rights. Another factor is that the digitization of information, which increases the capacity for its conservation and dissemination, is limited to specifi c institutions and contexts. The Records Technology, Records 75 government has little experience in the delivery of digitalized Technology management, Management information. The lack of technical assistance on the topic and Technology the costs entailed in undertaking a technological process of that magnitude largely explain this defi cit. Given the limited resources allocated to the General Archive of the Nation to conduct its oversight and technical training 76 Budget, Training Budget, training Records oversight duties, it has a limited capacity to comply with legislation concerning archives. Since went into effect, no national campaign has been launched to disseminate its contents. This affects the Monitoring Activities, Staff Public knowledge, 77 implementation of LTAIP to the extent that there is Public Knowledge, knowledge,Monitor Public outreach considerable ignorance about the law and a lack of awareness Public Outreach ing activities of the scope of RTI. A noteworthy effort of the CPP (Press Council) is the Transparent Municipalities Project, implemented since 2002 in five regions of the country. The project trains public servants of regional, local, provincial, and district governments. This Training, State- Training, State- 78 project works in partnership with the Integrated Financial Training society Society Collaboration Administration System (SIAF) of the Ministry of the Economy collaboration and Finance, the Ombudsman’s Offi ce, the Public Window of the Pontifi cia Universidad Católica del Perú, and the newspapers La Industria, Ahora, El Comercio, and El Tiempo. 87 Governments in candidate states faced substantial pressure from their civil societies to implement the required reforms as soon as possible: joining the EU is regarded as a highly desirable goal due to its demonstrated prosperity.4 In all of the countries, league tables were regularly published and discussed by the media, and government effectiveness judged International International International 79 by how well they performed compared to their peers. In the Pressure Pressure pressure countries that were the least prepared (Romania and Bulgaria), the public was the most enthusiastic about joining the EU (over 80 percent), creating tremendous pressure on governments to demonstrate progress with the negotiations. As a result, new legislation was adopted in both Bulgaria and Romania in the late 1990s and early 2000s. In March 2001, a civil society coalition was forged that CSO - Advocacy Law, effectively facilitated consultations between the government’s Intra-agency CSO - advocacy 80 State-Society Ministry of Public Information (MPI) and the opposition cooperation law Collaboration (mainly the PNL). In retrospect, it can be said that a truly powerful FOIA coordination agency—one that could entrust higher visibility and political salience to the task—failed to materialize in Romania. While the MPI existed, it organized a few rounds of training for civil servants, but there have been no such systematic initiatives since its closure in 2003. Its successors Monitoring Monitoring 81 Monitoring Activities were increasingly marginal institutions—fi rst an agency and— activities activities since 2009—a small department in the government’s Secretariat General. By virtue of law, they continued to collect primary data about FOIA implementation based on the standard fiches filled in by every institution. These data were used for compiling very brief and descriptive annual reports. The tasks related to the FOIA in each institution have no 82 Budget Budget Budget dedicated budget. The designated officers are civil servants with various other duties in the hierarchy who merely dispatch ATI requests to the relevant departments. In most of the local institutions analyzed and in the two sectoral ministries where extensive 83 PIOs PIOs PIOs discussions were held transportation and education), the task of processing the FOIA requests is assigned to a press and public relations office, which registers the requests and monitors their circuit inside the bureaucracy. But IT that is used for communication cannot solve the problem of a lack of integrated data management within the concerned institutions. Since most of the requests concern individual cases, there is a lot of effort invested in manually shuffling the physical archives of ministries Records that have not yet been transferred to e-platforms. Even in the 84 Technology Technology management, case of legislative repertoires where e-databases were created Technology (on the parliament’s Web site or by commercial companies), the staff’s tendency is to recover old pieces of legislation from the ministries’ physical archives, largely because the object of the FOIA requests is the secondary or tertiary legislation that sometimes does not exist in free-of-charge e-repertoires. 88 The fi rst obligation created by the FOIA was for all the public bodies covered by it, depending on their size, to open public relations offi ces or at a minimum to designate civil servants in charge of FOIA compliance within the institution. The process is currently quite formalized and uniform across public bodies: applications are received by the designated service (or person) 85 Request Process Request process Request process and registered; they are then forwarded to the relevant department for a draft response. The draft response is checked by the FOIA offi cer and countersigned alongside the relevant department head’s signature. The offi cial response is sent to the solicitor in the form preferred (for example, by paper or e-mail). In general, there has been relatively little formal training for the concerned civil servants after the law was adopted, although the ministry had initially put a training plan in place. Several rounds of information sessions were organized and a handbook for FOIA implementation was printed and distributed. After the MPI was terminated in 2003, support for Training, State- the task was reduced even further, especially after its Training, Training, society 86 successor agency was also dismantled in 2009. Additional International International collaboration, resources were poured into the system by private operators, Funding Funding International mainly external donors operating directly or through NGO funding projects. These activities, either in the form of capacity building or independent monitoring initiatives were, as a rule, more visible than those organized by the central coordinators, but they were far from suffi cient to cover the entire administration. But introducing meaningful activity reports as a regular practice is an uphill struggle in the Romanian public Monitoring Activities, Monitoring Monitoring data, 87 administration in spite of the legal provisions because this Agency Leadership activities Agency leadership depends on more than the goodwill and determination of the FOIA offi cers. In practice, the courts shaped the interpretation of the law and set the boundaries of the right to free ATI. NGO advocacy programs focused on testing the FOIA were instrumental in bringing strategic litigation to court, allowing judges to pass rulings in this fi eld. Even nontransparent entities were 88 Judiciary Judiciary Judiciary eventually forced by the decisions of the courts to open their doors.This method of making the law operational is an entrenched habit in Romania, where pronouncements are taken into account by other institutions only if the entity generating them has clear enforcement powers. Crucial for the enforcement process—but also for deterrence—is the fact that once a judge has decided against a public institution, the information must be disclosed immediately. If it is not, the judge may impose fi nancial 89 penalties that apply to heads of institution individually or to Sanctions Sanctions Sanctions both the head and the FOIA offi cer as a fraction of their salaries. The fi nes accrue daily until the decision of the court is enforced (that is, until the information requested satisfactorily reaches the applicant). The quality and completeness of information posted online is still weak. A top-level decision was made to reduce compliance costs across the entire public administration with the posting of more relevant information online in the correct Proactive 90 format, reducing the need for active Proactive System Proactive System dissemination, requests; however, there are few signs of this occurring in Proactive system practice. On average, 50 percent of central institutions used their Web sites to publicize information covered by Law 544 in 2008; at the local level, the fi gure was about 25 percent. Applicants also quickly learned how to use the procedures, 91 Public knowledge Public knowledge Public knowledge resulting in a decline in the number of rejected requests. 89 SAHRC’s annual PAIA reports systematically and bluntly highlight the signifi cant implementation and compliance problems for government offi cials, Parliament, and the public at large. Through events such as the NIOF and GKA, SAHRC has repeatedly reached out to other PAIA stakeholders, including Monitoring CSOs, to develop deep and meaningful relationships and work Monitoring activities, State- 92 Monitoring Activities cooperatively on a number of initiatives. Further, as part of its activities society annual submission to DOJCD, the PAIA unit has repeatedly collaboration provided recommendations for the regulatory improvement of PAIA including in 2013 a long list of amendments for strengthening PAIA based on the African Union’s recently passed Model Law on Access to Information. Manuals are often not easy to locate on the institutional websites, and in many cases are outdated, requiring further research to identify the appropriate DIO. The SAHRC has also Proactive noted in the consolidated compliance audit that many dissemination, manuals were not very user friendly and often used Technology, bureaucratic language, thereby defeating the purpose of the 93 Intelligibility Intelligibility Intelligibility, PAIA manual. The majority of institutions did not have Access to manuals readily available in the areas frequently visited by the proactive public. Reliance on internet technology and the failure to disclosure provide manuals to frontline offi ce staff constitute a severe impediment to disadvantaged and illiterate groups in the community who do not have access to such technology. Only 27% of the surveyed public institutions had submitted their reports to the Minister annually as required. SAHRC’s consolidated audit found this number had only increased to 94 Monitoring - Data Monitoring – data Monitoring data 38%. In 2013, the PCNS’s investigation reported that out of the 40 national departments checked, 80% of the bodies surveyed had failed to comply with this requirement at all. Public institutions are also submitting inaccurate information. This may be due in part to poor records management, but Records Records SAHRC has also reported confusion over the specific reporting Records management, 95 Management, Staff dates that should be included. Misinterpretation of the management Staff knowledge, Knowledge requirement to report the number of Monitoring data requests refused in full or granted partially is also common. Unfortunately, after 10+ years since PAIA’s passage, SAHRC and PSC found there were still institutions that had not PIOs, Access to 96 appointed DIOs. In fact, over 60% of the 43 institutions PIOs PIOs request process included in SAHRC’s consolidated audit had not formally designated their DIOs as required under PAIA. In only a little over a quarter of these institutions were 97 Staff Knowledge Staff knowledge Staff knowledge otherofficials aware of the PAIA procedures. A lack of awareness of the obligations faced under PAIA removes the motivation to hire the appropriate number of Staff knowledge, staff for PAIA administration. It also reinforces the lack of Agency Staff Knowledge, Staff knowledge, 98 executive management commitment to PAIA and delivery on leadership, Personnel Turnover Personnel turnover the mandate to provide access to information. Increasing staff Personnel turnover often results in a lack of continuity and in most turnover instances “abandonment of the PAIA portfolio." However, the SAHRC consolidated audit found that only a third of the audited institutions had established internal 99 protocols for dealing with requests for information. In 2011, Request Process Request process Request process PSC found that an average of 62% of departments had procedures in place for access to information. Poor records management is a signifi cant problem at all levels of government and has impacted the capacity of government offi cials to locate and produce requested information. Civil Records society advocates believe that the large percentage of refusals Records Records management, 100 for information requests submitted is directly related to poor Management management Request process, records management. In fact, in the 2013 Shadow Report, the Monitoring data most common ground for refusal cited in 44% of responses was that the records requested do not exist or cannot be found. 90 In general, the lack of fi nancial resources hinders PAIA implementation. PAIA stakeholders all agree that limited fi nancial resources are impacting records management, staffi ng, training, strategic planning and communications, and other key components of effective PAIA compliance. Over 87% of the institutions surveyed in the consolidated audit indicated 101 Budget Budget Budget no specifi c budget for PAIA implementation or compliance. In the institutions with a PAIA budget, allocations reported varied from 0 to R200,000 primarily based on manual printing costs. Most departments audited did not see the need to create a PAIA budget, citing the low number of requests received. Historically, SAHRC training has focused on DIOs in order to improve the poor response levels of government departments to information requests as well as to enhance the implementation of PAIA. The DOJ Justice College also provides 102 Training Training Training PAIA training for public institutions. In 2013, SAHRC conducted 33 workshops and reached out to 1065 DIOs, at least 21 PAIA public institutions, and multiple local communities through trainings and other forums Overall, public institutions engage in very little public outreach on PAIA outside the published Section 14 manuals. In 2012, the SAHRC shifted its emphasis away from the training of government offi cials and began to focus more on supporting Public knowledge, 103 Public Outreach Public Outreach the public’s right to access information. Public outreach This shift was in recognition of the need for the public to hold government to account in the delivery of various social services. SAHRC Civil society has remained an active and forceful participant, engaging academics, international advocates, the media, government institutions, and other South African social justice CSO - advocacy CSOs in a broad range of activities to help strengthen the use CSO - Advocacy CSO - advocacy law, CSO - 104 and implementation of PAIA. Further, they have reached out Implementation implementation advocacy to local communities in an attempt to illustrate how PAIA and implementation access to information can be used as an essential tool in their fi ght for better government service delivery and broader human rights. At present, Thailand is facing a large-scale, nationwide confrontation between two political factions. One faction is CSO – Advocacy Law, CSO – Advocacy 105 calling for political reform. One of the proposed reform areas Political support Political Support Law is elimination of corruption. If this reform materializes, it may have signifi cant and desirable impacts on the OIA. Proactive dissemination, State agencies are required to have information centers, and Proactive Proactive 106 Access to these are proactive in providing information to the public. Dissemination dissemination proactive disclosure The Law relies on officials’ discretion; but officials are Penalties for Penalties for Penalties for 107 reluctant to disclose information for fear of a penalty. Disclosure disclosure disclosure The appeal procedure, guidance on how to appeal, External appeals downloadable appeal forms, appeal examples and the time Access to External Access to external process, Access to 108 required for each step of the process (totaling 46 days) are Appeals Process appeals process external appeals provided online on the OOIC website. There is no fee and no process lawyer required for making an appeal. 91 When an IDT decides that an agency should disclose the information to the requester, the agency is supposed to comply with the decision and disclose the information. However, in case the agency fails to do so, there is no sanction against it. The OIA bestows on the IDT the Binding decisions, 109 authority to make a decision, but does not impose any Sanctions Sanctions Sanctions sanction on the state agencies that do not abide by the decision. The refusal to disclose information may be subject to sanctions under either the criminal law, if deemed to be done in bad faith to damage the complainant, or the civil law for not complying with the IDT’s decision 110 The rulings of the Administrative Court have set a precedent. Judiciary Judiciary Judiciary The understanding of RTI among offi cials might be mainly Staff knowledge, limited to only Sections 7 and 9 of the OIA, which require Staff Knowledge, Staff knowledge, 111 Monitoring proactive disclosure of certain information. Some agencies Monitoring activities PIOs activities have a very limited number of qualifi ed RTI offi cials. In reality, agencies often have problems locating or retrieving Records Records Records 112 the information. Management management management The Cabinet Resolution of January 24, 2012 orders all Ministries, departments, provinces and local administrations to prepare training curricula on the OIA or to integrate the OIA-related issues into the existing training curricula on a continuous basis. In reality, the OOIC, the oversight agency, 113 Training Training Training takes responsibility for providing RTI or OIA training to government offi cials, and to a lesser extent, to the general public. The OOIC conducts four training sessions for state agency officials each year. Information Offi cials at all agencies are supposed to receive training from the OOIC. While CSO work involves making requests for offi cial information, their primary focus is neither on RTI nor on corruption. Indeed, they are beneficiaries of the OIA and their Judiciary, CSO – Judiciary, CSO – requests have increased transparency in the issues that they CSO - advocacy 114 Advocacy Advocacy investigate. Their objectives are to bring justice to the implementation Implementation Implementation impacted communities. In the process of seeking justice, they seek information from state agencies by implementing either the OIA or the Constitution. Private companies do not use the OIA to get information but Private sector, 115 to deny other companies access to their company’s Private Sector Private sector Public use of RTI information. In Thailand, where civil servants are directly under the politicians, with no state workers union to help leverage the voice of offi cials, they are more or less at the mercy of the respective Ministers. The degree to which the OIA will be Political support, 116 Politicization Politicization supported and implemented is therefore dependent on the Agency leadership willpower of those in the Cabinet, the Prime Minister and the ruling political party. The current government has not expressed support for the OIA The OIA (RTI law) is not clear and does not provide specific 117 guidelines on what information can or cannot be disclosed. RTI Law RTI law RTI law This is left to the judgment of officials. Civil society asserts that the exemption for cabinet records, which are accessible only after a minimum of seven years, is a CSO - Advocacy Law, 118 key weakness of the law because policymaking in Uganda is RTI law RTI law RTI Law mainly conducted at the cabinet, ministerial, and sectoral levels. An analysis of the recently published ATIA regulations further highlights other procedural challenges, including the Access to Request Access to request multiplicity of forms (up to 15 different types) that must be Access to request 119 Process, Request process, Request used to access information and the requirement to provide a process Process process name and physical address, eliminating the option of submitting an anonymous request. 92 One of the most signifi cant stumbling blocks to the success of the ATIA was the absence of implementing regulations. The absence of regulations was not a formal barrier to requests for information; citizens could still request information citing the Implementing Implementing Rules, State Secrets, 120 law, despite their absence. However, government officials rules, State State Secrets Implementing rules point out that for public offi cials and civil servants, the secrets absence of the regulations, combined with the Secrecy Law still being in force, created a lack of clarity about their obligations, the procedures, and the functioning of the law. Under the Oaths Act Cap 19, every civil servant, on assumption of offi ce, must take an oath of secrecy against disclosing information received in the discharge of offi cial duties. The Fourth Schedule of the 1995 Constitution also State Secrets, Data Data protection, 121 RTI law requires such an oath. Article 9(1) of the Public Service Protection state secrets Act (1969)35 and Article 22 (12) of the Education Service Act (2002), criminalize the disclosure of information by public servants (Uganda, 2002a). The directorate spearheaded the drafting of the regulations, developed a National Access to Information Program (ATIP) and issued a memo for the appointment of information offi cers within public bodies. An implementation plan was Monitoring implementation developed in 2008 to promote awareness of ATIA, to put in Activitiies, Monitoring 122 rules, monitoring place procedures for accessing information, to cultivate Implementation activities activities a culture of openness, to build the capacity of the public Rules bodies for effective management, to coordinate and disseminate information, and to monitor and coordinate implementation. When the ATIA was enacted, no significant resources were allocated for its implementation. The Directorate of Information, working with other stakeholders, including civil society, has designed a program of approximately 4.5 billion Budget, shillings aimed at implementing the ATIA over a fi ve-year 123 Budget, Training Budget, training Monitoring period through simplifi cation, translation, dissemination of activities the ATIA, awareness-raising for the public, and training of public offi cials, in addition to other activities. But separate resources have not been allocated for this, neither to the directorate nor to individual ministries. In 2009, four years after the implementation of ATIA, fewer than 20 public bodies had appointed information offi cers or public relations offi cers (PROs). In most cases, staff members are assigned the role of IO or PRO in addition to their other PIOs, Agency 124 responsibilities and oftentimes they are not sufficiently PIOs PIOs leadership empowered within the organization to access and disseminate information. Currently, the volume and demand for information is low so IOs and PROs are able to juggle their existing work with process requests for information. Interviews with the offi cials of these various ministries revealed a lack of clarity with regard to the responsibilities of the different institutions and the perception of fragmentation of institutional responsibilities for the management, storage, Records records oversight, 125 retrieval, and dissemination of information.The Ministry of Management, records Records oversight Public Service is considered to be in charge of records Records Oversight management management and storage, while the Ministry of Information and National Guidance is in charge of retrieval and dissemination. Fragmentation of information across departments also makes retrieval difficult. Information on funding is in the finance department; statistics on teachers, schools, facilities, and Records - RTI Records-RTI Records 126 students are in the Department of Education Planning. While Collaboration collaboration management these departments are supposed to provide copies to the resource centers for easy access, information is still kept on office shelves, not at resource centers. 93 Under Section 43 of the ATIA, ministers are required to submit an annual report to Parliament on requests for access to records or information and responsiveness. No public body has come up with a manual or presented an annual report to Records Records 127 Monitoring data Parliament in fulfi llment of the requirements of the ATIA, and Management management Parliament has not asked for either these documents or information. Most ministries do not keep records of the types of information requests received or their responses to them. The lengthy judicial process discourages citizens from using the courts as a means of redress. Several critiques and reports have been issued and many see the judiciary as not being 128 independent of political influence. The funding allocated to Judiciary Judiciary Judiciary the judiciary has been steadily scaled down since 2003, which has forced the courts to scale down their operations up to 60 percent in some respects (IBA 2007). The ATIA provides for sanctions for officials who intentionally denies a citizen with the right of access under the law by destroying, damaging, altering, concealing, or falsifying a record; committing such an offense makes the official liable to 129 Sanctions Sanctions Sanctions a fine not to exceed 240 currency points imprisonment not exceed three years, or both. But to date, no public official has been charged in a court of law for denial of the right to access information. Although government departments are making advances in the management and dissemination of information, as evidenced by the experience of the health and education ministries, systematic disclosure of the documents mandated Proactive by ATIA has not happened. The ministries have not set in place dissemination, 130 systems or mechanisms to be responsive to the 2005 ATIA per Proactive System Proactive system Proactive system, se. In fact, awareness of the ATIA and any measures toward Staff knowledge systematically implementing it have been so poor that the information management and dissemination activities within individual departments and ministries have largely been developed quite separately from the ATIA. Implementation is also facilitated by the fact that public authorities in the United Kingdom are suffi ciently well-funded so that they can usually bear the cost of compliance, and there Budget, Ethical Budget, ethical Budget, Ethical 131 is a strong offi cial culture of taking one’s duty to the law Norms norms norms seriously—even when it might be inconvenient to do so, as is sometimes the case with FOIA requests This historical context meant that debate over the principles and legal structure of the FOIA was already mature well before its drafting. Since the mid-1970s, politicians, offi cials, and CSO - advocacy 132 CSO - Advocacy Law Agency Leadership campaigners produced a string of proposals, white papers, law electoral commitments, and even a few bills that made some progress through the parliamentary process. The introduction of the absolute exemptions is a signifi cant point of difference between the government’s consultation bill and the white paper. This continues to be relevant to the Information Commissioner’s powers. In the case of absolute exemptions, she can determine if particular information falls RTI law, Binding 133 within a particular exemption, but she is not empowered to Data Protection Data Protection decisions order an authority to release information after the application of an exemption has been determined to be correct. In the case of qualifi ed exemptions, the commissioner can conduct a de novo review of whether or not the public interest lies in disclosure and she can order the release of that information. Several procedural provisions were identifi ed by interviewees as having proved particularly conducive to overcoming Access to request bureaucratic resistance and fostering public ATI, including the process, 134 duty to provide advice and assistance to requesters and to Request Process Request Process Resistance, consider any written request for information as an FOIA Request process request, regardless of whether or not the requester explicitly mentions the law. 94 The nodal authority leading implementation of the FOIA within the central government has changed several times since the law was originally passed; it is currently the Ministry of Justice Personnel - Personnel - (MOJ). The nodal agency promoted the law within the Turnover, Monitoring 135 Turnover, Monitoring government and assisted public authorities in their monitoring activities Activities organizational and operational preparations; it also promoted activities cultural change by publicizing good practices and conducting case studies Costs are mainly incurred due to the need to train staff in the legal and strategic requirements of responding to requests and the practicalities of redaction and information management. But other interviews suggest that resource 136 Budget, Training Budget, training Budget implications are highly variable and that many authorities receiving fewer requests have been able to incorporate response processes into their existing legal or public relations functions with minimal disruption Authorities were given considerable leeway in determining their own arrangements. This fl exibility was considered desirable given the range of authorities subject to the law, the differing resources available to each, and the diversity of the relationships they have with the formal political system and 137 Agency Leadership Agency Leadership PIOs with external stakeholders. Smaller authorities have tended to delegate the administrative handling of FOIA matters to their legal teams or to their corporate services divisions; larger authorities and those receiving a high volume of requests have set up dedicated teams However, a lack of widespread support within the cabinet proved particularly signifi cant after the decision to shift to the “big bang.” It introduced an unanticipated delay between the Training, Political Training,political Political support, 138 provision of training and the enforcement of the FOIA’s access Support support Training provisions, demoralizing the supportive constituency of trained offi cers among public authorities that had been built in anticipation of an early entry into force. The interviewees from the National Archives suggested that effective records management tends to exist at agencies dependent on accurate records to conduct core business and at which missing or inaccurate records are likely to be Records Records Records control, 139 identified as a problem in the normal course of business. They Management management Budget also suggested that this view of records management as an overhead cost explains why, historically, only the larger authorities have put in place dedicated records managers. The monitoring of compliance with the FOIA is carried out by the Information Commissioner and the nodal agency. The ICO discharges its duties in a number of ways. Apart from hearing appeals and working with authorities to improve their internal Monitoring handling processes, it maintains a Web site providing Technology, Appeals Technology, 140 activities, Appeals comprehensive documentation, guidance, and model Process appeals process process publication schemes for various kinds of public authorities. The commissioner also supports research into transparency, conducts public information campaigns, and is empowered to report to Parliament on issues of concern on an ad hoc basis. In 2010, the commissioner increased the number of staff dedicated to resolving appeals and reported that the number of open cases dropped signifi cantly, especially among those External appeals that had been open for more than six months. Because it is External Appeals External Appeals 141 process, now routine for cases to be resolved within six months, public Process Process Resistance authorities are under greater pressure to handle requests in a more timely and positive fashion and cannot excuse their complacency by blaming the ICO. 95 Evidence of compliance with proactive publication schemes is patchy, and it is diffi cult to draw any fi rm conclusions from it. There is a high level of compliance with the requirement to adopt a scheme, but evidence of impact or effectiveness is Proactive lacking. This is partly because these schemes are intended to Proactive proactive dissemination, preclude the need for members of the public to formally 142 Dissemination, Public dissemination, Proactive request information, and there are no systematic statistics on knowledge public knowledge guidance, how many people access published documents. The ICO Monitoring data reports that it receives very few complaints about publication schemes, although it is not clear whether this is indicative of general public satisfaction with the schemes or of a lack of awareness of their existence. The existence of a legal rule favoring disclosure allows offi cials to answer informal requests from and engage in discussion with external groups much more freely. This is because all parties know that requesters will probably be able to access State-society information if the matter is put to the test and because the State-Society State-society 143 collaboration, RTI law provides clear standards for justifying the withholding of Collaboration collaboration law genuinely sensitive information if the need arises. In other words, the law does not just empower requesters; it also makes it easier for offi cials engage openly with them when they are so inclined. Litigation has clarifi ed the FOIA’s provisions and provided Implementing Rules, 144 requesters with including disclosure of records and awards of Judiciary Judiciary Judiciary attorneys’ fees; Access to external Litigation is costly and discourages requesters from enforcing Access to External Costs – Appeal 145 appeals, Costs – their rights under the FOIA; Appeals deterrent Appeal deterrent Advocates want a simplifi ed procedure. They would prefer to have a central location where they can make and track requests to any agency and receive responses. Such a portal Request process, does not yet exist for all agencies. FOIAonline accepts requests Request Process, Request process, Access to request 146 for the Environmental Protection Agency (EPA), Department of Access to Request Technology process, Commerce (save the US Patent and Trademark Offi ce), Process, Technology Technology National Archives and Records Administration’s Offi ce of General Counsel, Merit Systems Protection Board, and Federal Labor Relations Authority On the intra-agency level, the Chief FOIA Offi cer (CFO) is responsible for agency-wide compliance with the FOIA. The CFO must hold a senior position at the agency and is mandated to monitor FOIA implementation, designate FOIA liaisons, keep the head of agency and the Attorney General of Intra-agency Liaisons, Intra-agency the U.S. informed of agency performance, and make cooperation, cooperationAgency Liaisons, Intra- 147 recommendations to the agency head about adjustments to Agency Leadership, agency cooperation agency practices, policies, personnel and funding. A second leadership, Monitoring Activities layer at the agency level is the FOIA liaisons. Liaisons serve as Liaisons offi cials with whom requesters can raise concerns. They are responsible for making the FOIA more effi cient by assisting in delay reduction, increasing transparency and assisting in dispute resolution After the E-FOIA required agencies to make records Technology, Proactive – Proactive electronically available, many created FOIA electronic reading 148 Proactive - dissemination, dissemination, rooms and FOIA libraries. Perhaps the most famous is “The Dissemination Technology Technology Vault”, the FBI’s electronic reading room. In assessing if agencies have taken steps to increase proactive disclosure, OIP uses the following indicators: adding materials Implementing Rules, to the website, enhanced usability of the website and other Proactive – Proactive system, Technology 149 efforts to increase proactive disclosures. The USDA, DOJ, EPA dissemination, Technology, Monitoring Data, and NARA all met this milestone in 2012. In fact, most Technology Monitoring data Proactive System agencies did. Every agency, save one, reported an increase in the amount of materials it posted online 96 For their part, CSOs are disappointed in OGIS’s reticence to “name and shame” agencies. Advocates recognize the tension in OGIS’s mandate. They agree the agency should have more Monitoring State-Society Intra-agency resources and many praise OGIS’s mediation efforts. activities, State- 150 Collaboration, cooperation, Moreover, the Office Director regularly connects with society Monitoring Activities Liaisons advocates; the two sides meet to discuss OGIS’s difficulties, collaboration progress and future. Nonetheless, advocates consistently implore the office to do more. In the 37 years since the sanctions provision was established, no sanctions have been imposed on a federal employee for 151 arbitrarily and capriciously withholding a document in Sanctions Sanctions Sanctions violation of the FOIA. In fact, referrals to the Special Counsel or agencies’ disciplinary boards have been extremely rare. The government also keeps careful track of litigation outcomes. The DOJ’s Guide to the Freedom of Information Act is exceptionally comprehensive. The Guide contains nearly Monitoring data, 152 1000 pages of discussion on how litigation, OIP and DOJ Implementing Rules Implementing rules RTI law Guidance, Executive Orders and Presidential Memoranda affect the way agencies should interpret and implement the FOIA. Federal regulations detail how the agency will execute the law. Under the FOIA, agencies must create regulations concerning Implementing Rules, Implementing rules, Implementing fees, the aggregation of certain records, multi-track processing 153 Monitoring Activities, Ethical norms, rules, Public and expedited processing of records. The agency should notify Public Consultations Public consultations consultations the public and solicit public comments before publishing the regulations. Agency appeals Agencies tend to affirm their initial decision or close the Appeals Process, 154 Appeals process process, appeal more often than they partially or completely reverse. Monitoring Data Monitoring data NARA provides oversight and guidance for records Records Oversight, Records oversight, 155 management. Every agency’s records schedules must be Records Records Records oversight approved by NARA. Management management There is no specific line item for the FOIA in agencies’ budgets. Instead, the FOIA is funded from agencies’ administrative 156 Budget Budget Budget budgets. FOIA advocates are split on whether this is a benefit or a detriment. 97