The World Bank SEPTEMBER PREMnotes know what we deliver 2013 No. 5 Special Series on Implementing Right to Information Anupama Dokeniya Over the last two decades, several developing countries have adopted right to information (RTI) laws, bringing the number of countries with such laws to more than 90. But empirical evidence on how effectively RTIs have worked, whether they have been effectively enforced, and if they have had any impact on improving accountability, is limited. This note examines how RTI laws have worked in different country contexts, drawing from an analysis of their implementation in eight countries spanning different regions, income levels, political forms, and administrative traditions. The experiences of these countries show that implementation has faced challenges across countries, but has been especially difficult in countries where the broader governance environment is weaker on dimensions such as the rule of law, government effectiveness, voice and accountability, civil liberties, and political rights. During the implementation phase, with the erosion of the initial political momentum behind the law, the incentives for officials to comply with the new legislation also tended to erode, resulting in efforts to undermine the law. The evidence suggests that as countries with challenging governance environments and capacity limitations adopt RTI laws, they will need to both devise implementation solutions that can adapt to these limitations and undertake complementary initiatives to strengthen their broader governance environment. Overview But at the same time, some experts have A number of developing countries—in Eastern cautioned that RTI implementation could prove Europe, Latin America, Asia, and Africa—have considerably more challenging in countries recently adopted RTIs, bringing the total number with weaker governance environments. For of countries with such laws to more than 90, up instance, Roberts (2006) points out that the from just 13 in 1990. RTIs, which give citizens1 efficacy of these laws has yet to be tested in the the right to have access to information about the new adoptee countries that have poorer records functioning of their governments, were seen for on key governance dimensions such as political much of the 20th century as operationalizing an and civil liberties, rule of law, and accountable intrinsic democratic right, a right that had been governance. Kreimer (2008) points out the incorporated into international human rights “ecology of transparency”—which comprises conventions2 and in the constitutions of several dedicated oversight and monitoring groups, countries. But as RTIs have been adopted by more accountability and media groups that use RTI developing countries, they have been cast also as a to uncover information, and media and press tool for furthering development objectives and as laws that encourage free expression and free fundamental to enhancing citizen participation in association—is essential to the effective function- governance and improving citizens’ understand- ing of RTIs. ing of public policy by enabling them to assert This note highlights the findings from an as- their claims on service entitlements, scrutinize sessment of how RTI laws have actually worked public officials and public expenditures, and ex- in eight countries spanning different regions, ercise a more direct form of social accountability. income levels, political forms, and administrative FROM THE POVERTY REDUCTION AND ECONOMIC MANAGEMENT NETWORK traditions - Albania, India, Mexico, Moldova, Peru, Lessons from Better- Romania, Uganda, and the United Kingdom.3 The Performing Countries analysis confirms that the effectiveness of RTI has Of the case study complement, implementation indeed been more challenging in countries weaker on was more effective in the three countries stronger in governance dimensions such as the rule of law, govern- terms of both their internal institutional capacity and ment effectiveness, accountability, civil liberties, and their governance environment—the United Kingdom, political rights (figure 1). In these countries, it has Mexico, and India. These countries rank higher on been difficult to establish implementation institu- key governance indicators as shown in figure 1. The tions and build capacity for RTI, and civil society and United Kingdom was not included in these graphs citizen groups have also not been very successful in because its scores were significantly higher than all leveraging the law as an instrument to hold govern- the other cases, making comparison within the re- ments accountable. maining cases difficult.4 These three countries show On the other hand, countries that ranked better better scores on the World Governance Indicators on these governance dimensions have more effectively (WGI), such as government effectiveness and rule established institutions to oversee and promote the of law dimensions of, which measure characteristics law, build capacity, monitor compliance, and provide such as quality of the civil service, quality of policy redress for noncompliance. They are also the countries formulation and implementation, the credibility where civil society and media groups have been more of the government’s commitment to such policies, successful in leveraging RTI to extract information to and the extent to which civil servants are likely to hold governments accountable, suggesting that several follow and implement the letter of the law. These of these governance dimensions might be crucial to characteristics might explain better performance on creating an enabling environment for the effective procedural compliance with RTIs and the establish- implementation of the law. ment of implementing institutions. Experiences of the more effective implementers India, Mexico, Romania, and the United Kingdom provide useful lessons on institutional design and set up both independent oversight agencies as well as capacity-building measures for countries adopting nodal agencies within the executive branch to promote RTI. s But as countries with more challenging enabling awareness about the law within the government and environments adopt the law, the case study examples among the public; oversee and monitor compliance; also suggest that they need to devise implementation and support technological and human resource ca- systems more in tune with their capacity and insti- pacity across government. Independent information tutional environments. The next section highlights commissions set up as grievance redress mechanisms lessons on effective implementation from the more for noncompliance—such as the Federal Institute for effective implementers, and the subsequent section Access to Public Information (IFAI) in Mexico, the looks at the challenges that countries with more dif- Central Information Commissions (CIC) in India, and ficult environments faced. the Information Commissioner’s Offices (ICOs) in the At the same time, several challenges to implemen- United Kingdom—were important champions of the tation have emerged across both better and poorer law. They pushed for compliance through both their performing countries. In particular, resistance to adjudicatory and promotional roles. implementation from both officials and political In Mexico, although the rule of law indicator is actors—manifested in attempts to pass amendments weaker than in the other three countries between 2000 to restrict the scope of the law, refusal of officials to and 2010, the high profile of the 2002 Transparency help requesters, and even harassment of information Law and championship by IFAI were critical factors requesters by officials—suggests that the passage of in its effectiveness. IFAI, which is regarded as the gold RTIs is a limited step in the direction of accountable standard of independent oversight agencies, aggres- governance, and continued vigilance by both indepen- sively promoted access by disseminating information dent oversight and civil society organizations (CSO) about the law at the state level, persuading governors is necessary. The section on “Challenges” discusses to pass similar laws, encouraging citizens to make the forms that such resistance to implementation information requests, and disseminating information can take and some suggestions on how these could about the Mexican experience internationally. Its be addressed. regular publication of data on compliance and other 2 PREMNOTE SEPTEMBER 2013 statistics collected from implementing agencies put capacity for responding to requests, with ministries, pressure on poorly performing agencies. departments and agencies (MDAs), and other enti- Relative operational autonomy and significant ties under the law, designating information officials resources to develop capacity enabled these agencies and information response units, as per the law’s to act as effective arbiters against noncompliance with provisions. However, because these implementing the law. For instance, both the Mexican IFAI and the entities were not allocated additional budget, but British ICO had a high degree of political support, were were expected to find funding for RTI from existing provided substantial resources, and were staffed with budgets, their relative investment in this function commissioners with strong technical expertise, some varied, driven by the resources they could dedicate, drawn from civil society. IFAI’s rulings on several high- the commitment of agency leadership to transpar- profile cases—such as the diversion of public health ency, the specific functions of the MDAs, and the resources allocated to pro-life organizations and the extent of public demand. embezzlement of resources from large government In the United Kingdom, where agencies have trust funds—also signaled its willingness and ability considerable autonomy in determining their ar- to take on politically challenging cases and established rangements, smaller departments tended to delegate the seriousness of the new transparency reform (Fox, RTI compliance to their legal teams or corporate Jiménez, and Haight 2009). The UK ICO was respon- services divisions, whereas larger departments and sible for improving the compliance of several ministries those receiving a high volume of requests set up dedi- and agencies. cated teams. In India, departments focused more on In India, the CIC also championed implementa- formulating policies at the central level put in place tion, providing progressive rulings on several cases, leaner implementing structures with fewer informa- such as the disclosure of assets by election candidates, tion officers, given that most of their programs are although the organizational challenges it faced – for implemented at the state and local government lev- instance, several posts lying vacant – reflected the more els, while departments such as public works, which generic inefficiencies of India’s public sector. are heavily engaged in program implementation, Although Romania did not have an independent appointed a higher number of information officers. oversight agency, the judiciary’s role was very im- Experiences of the implementing MDAs highlight portant in several landmark judgments that pushed an important constraint: although MDAs with larger the envelope on transparency in the early years of budgets might be able to absorb the additional costs implementation. More recently, however, pressures of RTI implementation, this is rarely the case for on judicial capacity have resulted in a weaker record poorer entities, and RTI implementation plans might on adjudication, highlighting the importance of an need to factor in allocation of necessary resources to independent information commission (IPP 2011). these poorer entities. Although much emphasis for RTI implementation The record on the use of the law is also stronger is placed on independent information commissions, in these four countries. The law was extensively used the examples of these countries show that a nodal by both private citizens for accessing information of agency within the government is also critical, especially personal interest (such as benefits and entitlements in the early years, to steer change-management efforts under government programs), and by civil society and build up the requisite capacity. The Secretaria de and media groups to extract information related to la Funcion Publica (SFP) in Mexico, the Department of mismanagement of public funds, poor performance Personnel and Training in India, and the Ministry of be agencies, and instances of fraud, collusion, cor- Justice in the United Kingdom championed training, ruption, or nepotism. But as a way of highlighting capacity building, and monitoring efforts. In Romania, how RTI works as an instrument of accountability, the RTI nodal agency within the government, which it is interesting to note that even in these countries, was initially the Ministry of Justice, monitored compli- the percentage of population using the law was ance statistics in the early years. However, as the RTI quite small, in the range of .03–.04 percent of the nodal agency became smaller and more marginalized population, not very far from the levels of use in with budget cuts, Romania’s monitoring function more developed countries with a longer history of became weaker, highlighting the long-term challenge implementation.5 Clearly, the significance of RTI lies of sustaining implementation efforts. not in the pervasiveness of its use, but for the kinds India, Mexico, the United Kingdom, and Roma- of information it can bring to light when leveraged nia were also more effective in setting up frontline effectively. SEPTEMBER 2013 PREMNOTE 3 As with government effectiveness and rule of law Challenges in Countries with dimensions, the United Kingdom ranks much higher Weaker Governance Environments than the other case study countries on voice and ac- In Albania, Moldova and Uganda, countries that have countability, political rights and civil liberties—dimen- weaker scores on both the governance environment and sions that might explain the higher responsiveness of rule of law, implementation was more challenging. In the state to civil society pressures in these countries.6 Albania and Moldova, the absence of a nodal agency India and Mexico also have strong, mature CSOs and within the government to oversee RTI and sustain media groups with a historical track record of engaging capacity development led to fragmented implementa- with the government on tough governance issues, and tion efforts and hindered capacity building, especially these groups were also successful in leveraging RTI to expose major corruption scandals. Further, in Mexico in the face of successive changes in government and the and India, RTIs were used not only by larger, metropoli- reorganization or elimination of ministries. In Albania, tan nongovernmental organizations (NGOs), but also although the People’s Advocate attempted to initiate by grassroots groups and NGOs in the service delivery several measures to strengthen RTI implementation, sectors. In Romania as well, during the post-Soviet such as promulgating implementation regulations, in decades, CSOs have gained voice, leverage and capac- the absence of adequate resources and political back- ity, actively and effectively engaging with the state on ing, its efforts were largely unsuccessful. In Uganda, several laws and policies. the nodal agency faced resource constraints from the These findings from the better-performing outset and was never truly effective in facilitating countries, therefore, suggest the importance of both implementation. independent and executive oversight institutions Adjudication systems for noncompliance in Ugan- to promote and monitor implementation, politi- da and Albania were also weaker. In Uganda, where the cally independent and well-resourced information law provides for direct appeals to the judiciary, judicial commissions as well as adequate resources within record on high-profile cases, such as that of the disclo- implementing agencies to enable them to dedicate sure of contracts awarded to Tullow Oil, highlighted personnel, conduct training, and build capacity. The the lack of technical capacity within the judiciary to limited anecdotal data also point to the important role interpret the RTI, and likely a lack of independence.7 that other accountability institutions—such as parlia- In Albania, the ombudsman, who was charged with mentary committees, courts, audit institutions, and providing redress for noncompliance, was a champion anticorruption agencies—play in ensuring sanctions of the RTI, but inadequate resources and lack of politi- and follow-up actions when corruption or nonperfor- cal support made it a weak adjudicator. mance was uncovered through RTI requests. These poorer performing countries also did not Peru stands between the groups of better and weak- set up robust systems for monitoring RTI compliance. er implementers. It set up implementation institutions As a result, little data are available to help understand and designated a nodal agency within the government, the use of RTI compliance levels or the comparative the Office of the President of the Ministerial Cabinet performance of agencies. Interviews suggested that (PCM), which monitors compliance data. There were although there are dedicated CSOs that routinely made also active CSOs—in partnership with the ombuds- RTI requests, they largely met with either overt refusals man—advocating for RTI implementation. But there is or no response. In Moldova, where a more pronounced little evidence of the law having been used in any major compliance culture did lead to routine releases of victories forcing the release of information or enabling information in response to requests, CSO and media significant gains in accountability. The absence of an interviewees suggested that the information was often independent RTI agency and the resulting limitations incomplete or not useful. The limited data collected in independent oversight might at least partly explain through CSO monitoring efforts shows a similarly why the law’s effectiveness has been limited. The om- dismal picture on compliance. A challenging overall budsman, which performs this oversight role, has been accountability environment, as reflected, for instance, a champion of the law, but it has multiple responsibili- in lower scores on indicators such as political rights, ties, so its resources and attention are fragmented. The civil liberties and voice and accountability (figure 1), record of the judiciary is also discouraging: the courts might explain Moldova’s lower level responsiveness have not pushed aggressively for compliance, especially to CSO pressures, and hence the limited potential of in high-profile cases. instruments such as RTIs. 4 PREMNOTE SEPTEMBER 2013 These challenges—which have been called the the documents in their possession, and to make these implementation gap (Global Integrity 2009)8—are documents available to the public. not unique to RTIs, and are evident in the experience Such flexible and context-specific approaches have with several governance reforms that are adopted with not been tried and tested, but inevitably will need to much promise and fanfare only to become completely be factored into the design of both the law and imple- ineffective during implementation. This has led to mentation plans. increasing focus within development practice on the importance of designing reforms suitable to country Conclusion contexts, with the limitations and the constraints In conclusion, it is useful to discuss some challenges to of the country’s enabling environment in mind—the implementation that were common across countries, “good fit” versus “best practice” approach. suggesting that regardless of income levels and gover- There is a paucity of examples from the limited nance systems, the passage of the RTI law is a nominal experience on RTI implementation in developing step, and implementation requires continued vigilance country contexts on how a good fit approach might be operationalized. On the design of the law itself, from both independent oversight agencies and civil some initiatives have been launched—for instance, by society monitors. the Organization of American States and the African RTI is not politically neutral. Control of informa- Union—to develop model RTI laws that would be more tion provides opportunities for patronage and rent appropriate for their contexts and account for factors seeking for both public officials and political actors, such as the poor state of recordkeeping.9 creating inherent incentives to resist the law. Further, During implementation, experimentation with an RTI entails a fundamental paradigm shift toward sequencing and prioritization of different measures a presumption of transparency and disclosure, away might be a useful approach. For instance, in the short from the “presumption of secrecy” that has tradition- term, it might be useful to concentrate intensive efforts ally characterized bureaucracies (Weber 1992). An on building capacity for information management in RTI law represents a fundamental paradigm shift from those agencies that have high demand for informa- these embedded incentives and bureaucratic norms, tion. Similarly, state-of-the-art advanced technology and inevitably incites resistance, as is evident from solutions for requests, such as Mexico’s Sistema Infor- the long struggles for the passage of the laws in several matizado de Solicitudes de Informacion, might be too countries, and stalled efforts in others. expensive to implement in poorer countries and in In the case study countries, championship of the smaller entities, which might focus instead on creating law by political elites, primarily to establish their a culture of responsiveness through training and other democratic credentials and gain legitimacy from incentives. Departments that have a higher number international partners and domestic constituencies, of requests relating to government performance—the often during political transitions, enabled some more “sensitive” requests that officials are particularly consensus on the passage of the law across the po- cautious about—might need to put more effort into litical spectrum. During the implementation phase, communications, creating incentives that signal the however, with the erosion of political momentum department’s leadership, and other measures to cre- behind the law, the incentives of officials to comply ate a more open culture with moderate, incremental with the new legislation also tended to erode, and investments in technology platforms and a longer-term both officials and politicians attempted to undermine vision for full migration. the effectiveness of the law. For cost-intensive activities, such as strengthening Restrictive amendments to limit the scope of the records management, actions that address the most law ex post were introduced in Albania, India, Mol- challenging constraints could be prioritized rather dova, Peru, and the United Kingdom. In India, for than attempting a full-fledged rollout of advanced instance, the Department of Personnel and Training systems. Initial immediate actions could include, for attempted to restrict “file notings”—the notes of officials instance, harmonizing inconsistent regulatory regimes on administrative files—from the purview of the law.10 and putting in place clear roles, responsibilities, and Moldovan authorities drafted legislation in 2005 to coordination mechanisms as well as implementing limit access to information. The United Kingdom minimum standards for records management, such as attempted to exempt Parliament from the purview of requiring agencies to create updated lists or registers of the Freedom of Information Act, and created restric- SEPTEMBER 2013 PREMNOTE 5 tive provisions about fees for requests. Similar efforts livery NGOs, which help deliver government programs to block implementation have been evident in several and sit on ministerial committees, had easier access developed countries as well.11 In India, the United to information, while requests from accountability- Kingdom and Moldova, civil society opposition and focused NGOs, which are seen as more antagonistic, advocacy efforts thwarted these attempts, but in Peru reported being routinely denied information. and Albania, civil society resistance was less successful A strong accountability environment, active and in preventing restrictions to disclosure. engaged media and CSOs, and an adherence to the Further, although most laws require government rule of law can address some of these challenges, as officials to provide assistance to requesters, requesters the experiences of the better-performing countries often had to make multiple trips to submit and follow show. But all countries need to focus on creating the up on requests, and very rarely got any assistance. In right incentives for public officials to comply with India, for instance, information requesters, especially the law along with training and awareness raising, in rural areas, reported facing harassment and threats and signaling that leadership considers the RTI a (RAAG and NCPRI 2009), and a number of RTI ac- high priority. For instance, providing a higher profile tivists were even reported killed.12 When information to information units within an agency’s structure, was released, it was often incomplete or contained and factoring in information-responsiveness in per- terminology impossible to decipher for either average formance evaluations could be ways of indicating citizens or CSOs. leadership commitment. Moreover, although RTI laws are intended to democratize access to information, in practice, access About the Author continued to be influenced by personal relationships Anupama Dokeniya is a Governance Specialist at the and informal contacts. Hence, in Uganda, service de- World Bank. Figure 1. Rankings of Case Study Countries on Key Governance Indices a. Rule of lawa b. Government effectivenessa 2010 2010 Uganda India 2005 2005 2000 2000 Romania Romania Peru Moldova Mexico Moldova Uganda Mexico Peru India Albania Albania -1.5 -1.0 -0.5 0 0.5 -1 -0.5 0 0.5 higher values correspond to better governance higher values correspond to better governance continued on next page 6 PREMNOTE SEPTEMBER 2013 Figure 1. Rankings of Case Study Countries on Key Governance Indices (continued) c. Civil libertiesb d. Political rightsb 2010 Romania Romania 2005 2000 Mexico Mexico 2010 2005 India 2000 India Peru Peru Albania Albania Moldova Moldova Uganda Uganda 0 1 2 3 4 5 6 0 1 2 3 4 5 6 1 = highest freedom; 7 = lowest freedom 1 = highest freedom; 7 = lowest freedom e. Voice and accountabilitya Notes 2010 1. In many countries, the right extends to noncitizens India 2005 2000 as well. 2. Universal Declaration of Human Rights; Interna- Romania tional Covenant on Civil and Political Rights; African Charter on Human and People’s Rights; European Moldova Convention for the Protection of Human Rights and Fundamental Freedoms; and American Convention on Human Rights. Mexico 3. This note is based on a research project conducted in eight countries; more details and references can be Uganda found in Dokeniya (2013). 4. The World Bank’s Worldwide Governance Indica- Peru tors on rule of law, government effectiveness and voice and accountability, and Freedom House indicators on civil liberties and political rights are good proxies for Albania these dimensions (http://info.worldbank.org/governance/wgi/sc_coun- -1.5 -1.0 -0.5 0 0.5 try.asp). higher values correspond to better governance 5. RTI tends to be used by approximately .03–.4 per- Source: a. http://info.worldbank.org/governance/wgi/sc_country.asp; b. Freedom cent of the population. House surveys, 2000–2010, http://www.freedomhouse.org/report-types/freedom- 6. Freedom House indicators on political rights and world. civil liberties and WGI indicator on voice and account- ability are good proxies for these dimensions. 7. In response to an appeal on nondisclosure of oil exploration contracts with Tullow Oil, the judge ruled SEPTEMBER 2013 PREMNOTE 7 that since the plaintiff—CSOs—had not demonstrated 12. http://en.wikipedia.org/wiki/Attacks_on_RTI_ac- the public interest in the disclosure of the contracts, tivists_in_India. the government’s decision to withhold was valid; this was a considerable misinterpretation of the law. References 8. The 2009 Global Integrity Report pointed out that Dokeniya, Anupama. 2013. “Implementing Right to Infor- countries that are among the largest recipients of mation: Lessons from Country Experiences.” World international donor assistance tend to have the largest Bank, Washington, DC. Fox, Jonathan, Carlos García Jiménez, and Libby Haight. implementation gap, suggesting that aid-dependent 2009. “Rural Democratization in Mexico’s Deep South: countries might establish laws and institutions to Grassroots Right-to-Know Campaigns in Guerrero.” meet donor requirements, but do not necessarily Journal of Peasant Studies 36 (2): 271–98. implement them, leading to a proliferation of legal Global Integrity. 2009. Global Integrity Report: 2009 – Key and regulatory reforms on paper, without actual Findings. Washington, DC. http://www.globalintegrity. benefit. org/documents/KeyFindings2009.pdf. 9. See the “Draft Model Law for Au Member States on IPP (Institute for Public Policy). 2011. “Free Access to Public Access to Information” (http://www.achpr.org/files/ Interest Information 10 Years after the Adoption of the Law—An Assessment of Jurisprudence on the Law instruments/access-information/achpr_instr_draft_ regarding Free Access to Information of Public Interest.” model_law_access_to_information_2011_eng.pdf) Bucharest, Romania. and the “Model Inter-American Law on Access to Kreimer, S. F. 2008. “The Freedom of Information Act and Information” (http://www.oas.org/dil/access_to_in- the Ecology of Transparency.” Public Law and Legal formation_model_law.htm). Theory Research Paper Series, University of Pennsylvania 10. The Government of India argued that this ap- Law, 08–06. proach was justified on the basis that deliberative Palmer, Geoffrey. 2007. “A Hard Look at the New Zealand information is widely recognized as constituting a Experience with the Official Information Act after 25 legitimate exemption. Several legal scholars in India Years.” Address to International Conference of Informa- have disagreed with this position and suggested that tion Commissioners, Wellington, November 27. this information should be subject to disclosure (http:// RAAG (RTI Assessment and Analysis Group), and NCPRI (National Campaign for People’s Right to Information). right2information.wordpress.com/2006/08/15/ 2009. Safeguarding the Right to Information: Report of the juristic-administrative-and-political-views-on-file- People’s RTI Assessment 2008. Executive Summary, http:// notings-and-deliberative-processes/). rti-assessment.org/exe_summ_report.pdf. 11. In New Zealand, research showed that straight- Roberts, Alasdair S. 2006. Blacked Out: Government Secrecy forward, nonpolitically sensitive requests resulted in in the Information Age. New York: Cambridge University on-time, satisfactory compliance, while politically sen- Press. sitive requests were delayed, transferred, and refused Weber, Max. 1922. Economy & Society. Berkeley: University (Palmer 2007). of California Press (reprinted 1979). This note series is intended to summarize good practices and key policy findings on PREM-related topics. The views expressed in the notes are those of the authors and do not necessarily reflect those of the World Bank. PREMnotes are widely distributed to Bank staff and are also available on the PREM Web site (http://www.worldbank.org/ prem). If you are interested in writing a PREMnote, email your idea to Madjiguene Seck at mseck@worldbank.org. For additional copies of this PREMnote please contact the PREM Advisory Service at x87736. This series is for both external and internal dissemination 8 PREMNOTE SEPTEMBER 2013