Defining By Andreas P. Kyriacou Associate Professor of Economics, University of Girona (Spain). Background paper prepared for Aids International (AAI) workshop on May 12-13, 2008, Stockholm. I. Introduction has been defined in the following terms: "A is accountable to B when A is obliged to inform B about A s (past or future) actions and decisions, to justify them, and to suffer punishment in the case of eventual misconduct" (Schedler, 1999, p.17). To help elucidate this definition consider the role of elections in promoting political accountability. An election gives a chance for the candidate to run for campaigns and attend forums so as to explain and inform their purposes and goals if they are elected. On the other hand, it is also a sanction for those who misbehaved or failed to represent one s interests in the past tenure by giving the vote to someone else. The concept of accountability implies that the actors being held accountable have obligations to act in ways that are consistent with accepted standards of behavior and that they will be sanctioned for failures to do so (Grant and Keohane, 2005, p.29). In this brief background paper I will begin by making what I hope will prove a helpful distinction between horizontal and vertical accountability 1. Having done so, I will present what has been called the accountability cycle pointing to some of the key elements involved in making institutions or bodies more accountable. I will then present an overview of vertical accountability across different world regions so as to give an idea of the sharp differences that exist. I will end this article with some concluding comments on the relevance of these ideas to AAI s stated mission to help improve the quality of the global response to AIDS and ensure its implementation by holding key actors accountable for their promises and performance. II. Horizontal versus Vertical A distinction has been made between horizontal and vertical accountability (O Donell, 1999). In a well functioning state, the government is subjected to accountability that is both imposed upon it from outside by citizens (vertical), and accountability that it imposes upon itself through public institutions empowered to restrain the political executive (horizontal). Vertical accountability may include citizens acting through the electoral process or indirectly via civic organizations ( civil society ) or the news media. Horizontal accountability covers the range of public entities created by the state to check its own abuses and inefficiencies (for example, the judiciary, auditors general, anti-corruption units and Ombudsmen). 1 It s important, I think, to take seriously Mulgan s (2000) warning of the danger of unnecessary academic complication of the concept. 1
Government Horizontal Judiciary, Auditors General, Anti-corruption Bodies, Ombudsmen Vertical Elections, Civil Society (NGOs), Media Figure 1. Types of accountability Horizontal accountability crucially depends on the degree of autonomy or independence of such Institutions of (IA). Moreover, governments are more likely to bind themselves through institutions of horizontal accountability under circumstances where they may be punished for failing to do so. In relation to this, horizontal accountability can be buttressed by strong vertical accountability. The effective operation of vertical accountability, through the electoral process, the news media and concerted civic action, causes governments to take seriously the perils of failing to sustain horizontal accountability. At the same time, the effectiveness of mass media and civil society in holding government accountable depends to a large extent on the actions that properly authorized state agencies may undertake in order to investigate and eventually sanction wrong doings. Contrary to the case of horizontal accountability, most expressions of vertical accountability have little possibility of applying sanctions other than public disapproval. The distinction and relationship between horizontal and vertical accountability is illustrated in figure 1 above. III. The Cycle Schacter (2000, p.2) talks of an accountability cycle which has three stages namely, information, action and response (Figure 2). At the information stage, the IA for example the Ombudsman in the case of horizontal accountability or an NGO in the case of the vertical kind is able to obtain relevant, accurate and timely information about the activities of the government. In stage 2 the IA will act on the basis of this information something which depends on its capacity both to analyze the information and to make demands on government. The final response stage refers to the degree to which government feels compelled to respond to the IA. The accountability cycle helps clarify the difference between accountability and transparency. The former refers to the government s response at the end of the cycle while the latter refers to the quality of information available in the first stage. Another concept worth distinguishing is that of checks and balances. These are mechanisms designed to prevent action that oversteps legitimate boundaries while accountability mechanisms operate after the fact. As 2
explained by Grant and Keohane (2005, p.30) the executive veto power in the U.S. Constitution is part of a system of checks and balances while the impeachment power is an accountability mechanism. Notwithstanding the fact that accountability mechanisms operate ex post, they can exert influence ex ante, since the anticipation of sanctions may deter the powerful from abusing their positions in the first place (ibid). Information Institution of Action Government Response Figure 2. The Cycle What emerges from this perspective is the importance of the quality of the information available to the IA (relevance, accuracy, timeliness, comprehensiveness) as well as its capacity to analyze this information and to transform it into practical demands on government. This said, the IA has relative control over stage 2 of the accountability cycle but less so over stage 1 to the extent that government is an important source of quality information. The IA s influence in stage 3 is constrained, among other things, by the attitude of political and bureaucratic leaders towards accountability since insufficient high-level commitment to robust accountability negatively affects the impact of an IA s work. IV. Vertical in the World The World Governance Indicators generated by the World Bank cover 213 countries and territories and measure six dimensions of governance since 1996 until end-2006 (see Kaufman et al, 2006). The indicators are based on several hundred individual variables measuring perceptions of governance, drawn from 31 separate data sources constructed by 25 different organizations. Some of these sources are commercial risk rating agencies, others governments and multilateral organizations, and yet others NGOs. Several are surveys of individuals or domestic firms with first-hand knowledge of the governance situation in the country. One governance variable which is included in these indicators is termed Voice and (VA) and defined as the extent to which a country s citizens are able to participate in selecting their government, as well as freedom of expression, freedom of association, and free media. In terms of our previous discussion therefore, this captures the notion of vertical accountability. This variable like all the others is measured on a scale from -2.5 and 2.5, with higher scores corresponding to better 3
outcomes. In figure 3 below I represent this indicator across regions and for the period 1996-2006. Source: Author s calculations based on World Bank Governance Indicators at http://www.govindicators.org Figure 3. Voice and in the World, 1996-2006. What become immediately apparent are the large differences in perceived accountability levels across regions. As expected, governments in North America and Western and Central Europe and to a lesser extent those in Oceania, are perceived to be accountable while those in South and South East Asia, Eastern Europe and Central Asia, and especially Sub-Saharan Africa and North Africa and the Middle East have consistently much lower perceived accountability levels. It s important to note that such a regional based classification masks strong differences across countries. This is clearly the case for Sub-Saharan Africa where the strong negative values for Eritrea, Congo Democratic Republic, Burundi, Sudan and Somalia are in sharp contrast to the positive values for countries like Benin, Botswana, Namibia and South Africa. It is obvious that such an accountability indicator could be used to measure the impact of greater accountability in health outcomes in different countries and across time. Is greater accountability associated with improved outcomes after controlling for other variables deemed to influence such outcomes 2? From an empirical perspective this question could be approached either by way of cross-section analysis or, ideally, a fixed effects panel data analysis that can go some way to address the problem of omitted variable bias which plagues macro-analyses of public policy outcomes. 2 Outputs ideally should be measured using indicators of the actual provision of government services, rather than their ultimate effects, outcomes, since the latter depend on many factors beyond the government s control. 4
V. Implications for AAI As reported in an advanced draft of the Pilot Study Initiated by AAI, there is both a need and a demand for independent entities [in the field of HIV AIDS] that: 1) monitor the performance of key stake holders in the response to the AIDS epidemic; 2) promote improved data collection and analysis; 3) provide tools to help bridge the gap between data collection and analysis, policy and advocacy; 4) hold leaders and institutions accountable for how their policies and programmes impact high-burden countries; and 5) reinforce leadership of the response, particularly in high-burden countries.... [There is a] need for an independent, scientifically based assessment tool that can systematically compare actors performances on a regular basis. (AAI, p. 4). The literature reviewed in this background paper brings to light several insights concerning the factors which may improve the effectiveness of Institutions of in general and AAI s efforts in particular. AAI is right to place a considerable emphasis on the need for good quality information and a scientifically rigorous rating methodology. Its impact will, to a great extent, depend on generating a consensus among the scientific community with regards to the appropriateness of its methodology and the reliability of its ratings. In relation to this, the independence or autonomy of AAI is an important factor contributing towards its legitimacy and, ultimately, towards achieving its mission. The previous discussion also suggests that the effectiveness of AAI s efforts would be buttressed by the strength of horizontal accountability since this would potentially introduce stiffer sanctions which, recall, are an important element in making organizations accountable. Thus, beyond soft sanctions such as public or peer pressure mobilized by AAI s ratings, the existence of public entities created by governments or organizations to check their inefficiencies may introduce the possibility of hard sanctions thereby enhancing accountability efforts. Notwithstanding the possibility that AAI may also contribute towards strengthening vertical accountability in the area of HIV Aids, the effectiveness of its efforts in generating a response from the organizations rated, may ultimately depend on their willingness, disposition or commitment to be held accountable in the first place. References Aids International (2008). Developing an HIV/AIDS Country Rating, Pilot Study Initiated by AAI, Draft 7 January 2008. Kaufmann, Daniel, Aart Kraay, and Massimo, Mastruzzi (2006). Governance Matters: Aggregate and Individual Governance Indicators for 1996 2005 World Bank Policy Research Working Paper 4012 Mulgan, Richard (2000). : An Ever Expanding Concept? Public Administration 78(3): 555-573. O Donnell, Guillermo (1999). Horizontal in New Democracies, in Andreas Schedler, Larry Diamond, Marc F. Plattner: The Self-Restraining State: Power and in New Democracies. London: Lynne Rienner Publishers, pp. 29-51. Schacter, Mark (2000). When Fails: A Framework for Diagnosis and Action, Institute of Governance Policy Brief 9, Ottawa, Canada. Schedler, Andreas (1999). "Conceptualizing ", in Andreas Schedler, Larry Diamond, Marc F. Plattner: The Self-Restraining State: Power and in New Democracies. London: Lynne Rienner Publishers, pp. 13-28. Grant, Ruth W.; Keohane, Robert O. (2005). " and Abuses of Power in World Politics". American Political Science Review 99 (1): 29-43. 5