The World Bank D e c e m b e r PREMnotes 2 0 0 2 n u m b e r 7 6 Public Sector Lessons from Armenia's Institutional and Governance Review Armenia's experience with a new World Bank tool offers valuable guidance for similar efforts elsewhere. Institutional and Governance Reviews Armenia was chosen for several reasons. (IGRs) are a new tool in the Bank's pack- There was a dearth of analytical work on age of analytical and advisory activities (see public sector institutional reforms prior to PREMnote 75). Because they are politically 1998. Moreover, country authorities evinced How can sensitive, the development of these reviews keen interest in an IGR--and were matched involves careful tradeoffs. Though each by strong support from the Bank's coun- Institutional and requires thorough analysis of a country's try unit and team. institutional shortcomings, the final prod- Governance Reviews uct must be acceptable to country author- Consultations ities and other development partners. Before the formal concept review, the task help address To be credible and acceptable, an IGR team discussed the IGR informally with gov- must reflect extensive participation by a vari- ernment authorities and members of civil institutional ety of national stakeholders. In Armenia the society. The discussions focused on poor Bank's IGR team engaged the government public sector performance and on the challenges? (executive, legislature, judiciary), civil soci- development of key public institutions. ety (nongovernmental organizations, polit- During this process a basic question ical parties, trade unions, academics), and emerged: why was Armenia able to gen- other development partners from the out- erate successful policies in some areas set. This approach resulted in widespread (such as defense, foreign affairs, and acceptance of the report's analysis and rec- national security) but not others (such as ommendations within both Armenia and the social sectors)? the Bank. The team obtained valuable information from a variety of sources, including the Objectives of Armenia's review executive branch at the central, provincial, Armenia's IGR was a pioneering effort by and local levels, parliamentarians, NGOs, the Bank's Europe and Central Asia Region academics, and the media. Based on this to systematically evaluate a country's pub- information it was decided that the IGR lic institutions and develop a program of would analyze institutions critical to good reforms supported by follow-up operations. governance, sustainable development, and The IGR had two main objectives. First, it economic growth, and identify priorities was to diagnose institutional dysfunction at for institutional reform (box 1). the national level using quantitative bench- The first preparation mission took place marks of performance. Second, it was to in the fall of 1998. The focus was on checks assess political realities and constraints to and balances in national institutions and reform, to foster the sustainability of Bank the importance of accountability arrange- operations. ments. Taking into account the broader from the development economics vice presidency and pover ty reduction and economic management n e t w o r k political and institutional context, as well as including the prime minister--delayed the exogenous shocks that influence opportu- completion of the IGR. Yet even though nities for reform, the team developed quan- three administrations held office during the titative indicators of public sector IGR process, top-level support for it never performance and explicitly analyzed issues wavered. The IGR was completed in June of political economy. 2000 with collaboration from Armenia's The health sector was chosen as an executive branch, legislature, intellectuals, example for tracing the institutional causes and civil society. of poor outcomes. The team identified When the final draft of the IGR was weaknesses such as unpredictable budgets, shared with government authorities, it was arbitrary policies, and weak contracting lauded as the first attempt to review Arme- mechanisms, civil service incentives, and nia's economic development in the context Top-level support accountability arrangements. The investi- of the country's political and institutional gation served as a launching pad for environment. Apart from minor sugges- for Armenia's IGR examining cross-cutting issues in budget tions, there were no objections to the report. management, civil service and cabinet deci- One high-level policymaker remarked: never wavered sionmaking, and checks and balances on "We, not the Bank, should have reviewed all the executive. This analysis was intended these issues much earlier, as they are at the to identify concrete entry points for heart of the existence and development of reform, to be supported by follow-up dia- Armenia." logue and operations with explicit per- The IGR also had several notable impacts formance benchmarks. during its preparation, including: Discussions were then held with NGOs · The report's analysis of the health sector and academics, and background papers encouraged the government to review its were commissioned from academics, his- strategy for health sector reform. As a torians, and journalists. In addition, house- result the Bank-financed Armenia Health holds and public officials were surveyed Sector Project was restructured during to assess service delivery and public sector the preparation of the IGR. functioning. · The team's comments and participation were solicited on several legislative and Interim impacts capacity-building initiatives, including The tragic October 1999 assassinations of the new Law on Inspections (which several high-ranking Armenian officials-- reduced inspections and other intrusive controls) and the Law on Civil Service. The team's input also led to greater Box 1 Key messages from Armenia's IGR appreciation of the importance of imple- Armenia's IGR argues that the country's two most pressing institutional issues menting a medium-term expenditure are its unbalanced policy development capacity and its underdeveloped insti- framework. tutions for accountability. The first issue reflects shortcomings in the supply of good sector policies; the second, shortcomings in the demand for them. · The IGR supported a government ini- On the supply side the IGR concludes that the capacity to generate good pol- tiative to improve implementation of icy is limited by Armenia's institutional structures and skills--which are unsuited externally financed investment and tech- to the needs of an independent nation-state and a market-oriented economy-- nical assistance projects. The authorities and by the predominance of defense and security concerns. In terms of account- established new institutional arrange- ability the IGR finds that the demand for good policy is hindered by poor ments for project implementation, expenditure controls, pervasive informality, a center of government constrained by the Soviet legacy, and especially by the overarching preeminence of the exec- emphasizing accountability and owner- utive over other branches of government. This weak demand is exacerbated by ship by government agencies. These mea- Armenia's underdeveloped media and civil society. sures were promulgated in the face of The IGR explains why these institutional limitations should matter to poli- opposition from interests who perceived cymakers and others, emphasizing their consequences in terms of inadequate a diminution in their authority and delivery of services to poor people. The report also unbundles these chal- lenges to suggest specific reform actions. resented being made accountable to state agencies. PREMnote 76 December 2002 Dissemination ship of the report. And though Arme- The IGR was disseminated in three stages. nia's civil society organizations are only First, it was distributed publicly. After the starting to develop, very public discus- Bank and the authorities cleared the IGR sions were held on the IGR's analysis and for publication, it was translated into Armen- recommendations. ian and distributed to the executive branch, · Secure assistance from the Bank's country unit parliamentarians, political parties, NGOs, before, during, and after the IGR process. the media, and donor partners. Among other tasks, the country unit must Second, a workshop was held in Yere- determine the IGR's focus and the insti- van in October 2000 to publicly disseminate tutional and governance issues to be the IGR's findings and generate consen- addressed. Moreover, the IGR should be sus on institutional reform priorities, to reflected in the Bank's follow-up coun- underpin a medium-term public sector try strategy, with concomitant provision Institutional reform program. Jointly organized by the of resources. government and the Bank, the workshop · Be clearly defined. Because IGRs are an reforms do not attracted more than a hundred public offi- evolving, country-specific tool, each one's cials from all government branches as well concept, scope, and focus should be come quickly, as representatives of donor agencies and defined as clearly and early as possible. civil society, including political parties, the · Involve an appropriate mix of analytical tools. cheaply, or easily Armenian diaspora, and NGOs. The work- In developing Armenia's IGR, the Bank shop saw vigorous and open debate on the team faced a wide range of possible data issues raised in the IGR--and widespread sources, including official statistics, gov- recognition of the urgency and importance ernance ratings by external agencies, of public sector reform. interviews with domestic stakeholders Third, follow-up consultations were held and external experts, and surveys of with a more focused group of policymak- households, enterprises, and public ers. Several videoconferences were orga- officials. A key lesson was realizing what nized to discuss further follow-up and surveys can and cannot accomplish-- implementation of the IGR's recommen- and the need to focus surveys on essen- dations. Policymakers from the executive tial issues. and legislature and the Public Sector · Foster commitment to institutional reforms. Reform Commission joined the Bank team Institutional reforms do not come for detailed consultations on specific quickly, cheaply, or easily: sustainable fol- reforms. The final meeting decided to low-up requires clarity and consensus on include the IGR workshop recommenda- the strategies, instruments, and time- tions in the reform agenda being developed frames for implementing IGR recom- by the Public Sector Reform Commission. mendations. In addition to indicating specific areas that the Bank and other Lessons and caveats donors intend to support, an IGR could Armenia's IGR offers valuable lessons for propose mechanisms to ensure consen- similar efforts elsewhere. To succeed, IGRs sus between country authorities, differ- should aim to: ent parts of the Bank (both sector units · Ensure country ownership. An IGR requires and the country unit), and other devel- strong support from a country's execu- opment partners. This approach assumes tive and legislative branches. Armenia consensus within the Bank on strategies, provided such support before, during, instruments, and timeframes for public and after the wrenching political turbu- sector institutional reforms, a long-term lence that coincided with the prepara- appreciation by sector units and the coun- tion of its IGR. Moreover, continuous try unit of the human and financial consultations with a wide range of stake- resources required to achieve such goals, holders contributed to country owner- and tie-ins with the Country Assistance PREMnote 76 December 2002 Strategy and Poverty Reduction Strategy delivery. Within the Bank, the IGR bene- Paper processes. fited enormously from close collaboration · Appropriately address sensitive political issues. with other sector units (especially Human Armenia's IGR was relatively frank about Development) and the PREM Network. sensitive issues such as the preeminence of the so-called power ministries (defense, Further reading interior), though the final analysis was Bremmer, Ian, and Cory Welt. 1997. "Arme- somewhat muted in deference to com- nia's New Autocrats." Journal of Democracy ments made during the report's review. 8 (3). In some countries, however, nontrans- Preker, Alexander, and Richard Feachem. parent power structures and extracon- 1995. "Market Mechanisms and the stitutional or informal sources of political Health Sector in Central and Eastern Armenia's IGR power inhibit accountability and trans- Europe." World Bank, Washington, D.C. parency, impede institutional reforms, World Bank. 1997. World Development Report showed the weaken governance, and undermine 1997: The State in a Changing World. New poverty reduction efforts. In such cases York: Oxford University Press. feasibility and value the IGR process needs to examine ------. 2000. "Armenia: Institutional and whether such issues can and should be Governance Review." Europe and Cen- of focusing on analyzed--and if so, the degree of frank- tral Asia Region, Poverty Reduction and ness warranted. If not, it could suggest Economic Management Unit, Washing- political issues in the an alternative mechanism to do so. An ton, D.C. IGR's assessment of institutional arrange- Bank's analytical and ments and sources of dysfunction may This note was written by Amitabha Mukherjee also raise constitutional issues. The extent (Senior Public Sector Management Specialist, for- advisory activities to which an IGR can or should deal with merly with the PREM Unit, Europe and Central such issues--especially if they are at the Asia Region, now with the PREM Unit, East heart of governance and accountability Asia and Pacific Region) and David concerns--may have to be addressed Shahzadeyan (former Public Sector Management more explicitly in future IGRs. Specialist, PREM Unit, Europe and Central Asia Despite the potentially sensitive scope of Region, now with the Office of General Coun- Armenia's IGR, the process followed by the sel, Black Sea Trade and Development Bank, Bank and government teams showed the Thessaloniki, Greece). Helpful comments from feasibility, acceptability, and value of focus- Konstantin Atanesyan, Cheryl Gray, Nick Man- ing on political economy and governance ning, Helga Muller, and Helen Sutch are grate- issues in the Bank's analytical and advisory fully acknowledged. activities. This approach was facilitated by If you are interested in similar topics, consider full support from Armenia's civil society joining the Administrative and Civil Service in addressing previously "unmentionable" Reform Thematic Group or the Public Expendi- issues in a framework that showed how insti- ture Thematic Group. For more information, click tutional dysfunction is linked to weak service on Thematic Groups on PREMnet. This note series is intended to summarize good practice and key policy find- ings on PREM-related topics. The views expressed in these notes are those of the authors and do not necessarily reflect the views of the World Bank. PREMnotes are distributed widely to Bank staff and are also available on the PREM website (http://prem). If you are interested in writing a PREM- note, email your idea to Sarah Nedolast. For additional copies of this PREM- note please contact the PREM Advisory Service at x87736. PREMnotes are edited and produced by Communications Development Incorporated. Prepared for World Bank staff