Democracy, regulatory institutions and accountability
The typical image of the liberal democratic state as a coherent structure that relies on clearly defined upward and horizontal institutional accountability mechanisms is under significant strain. While democratic legitimacy has been traditionally entwined with accountability, the complex patterns of governance today make it increasingly difficult to determine who is responsible for what and how to hold them accountable. Accountability has thus become a far-reaching concern for institutions, especially for new institutions with a certain level of political and organizational delegation from political principals. Debates about the emergence of the regulatory state (Loughlin and Scott 1997; Majone 1997; Lodge 2001; Moran 2002) have also led to an interest in accountability. The fragmentation of the public sector associated with the emergence of the regulatory state in democracies involves the transformation of institutional structures and the appearance of new actors operating out of ministerial hierarchies (Bouckaert et al. 2010). Moreover, delegation of important policymaking powers to non-majoritarian institutions raises novel problems of democratic legitimacy (Majone 1999).
Based on autonomy, independence and expertise as fundamental elements to the improvement of policy outcomes and economic efficiency, numerous regulatory agencies have been established since the 1990s to regulate both economic and social issues. While 'the diffusion of the regulatory agency model has challenged the structure of power in many countries, often reducing centralization in the governance of many sectors' (Jordana and Sancho 2004: p. 307), questions of legitimacy and accountability have been raised (Flinders and Smith 1999; Scott
2000; Schmitter and Trechsel 2004). On the one hand, extensive delegation of policymaking competencies to agencies has been portrayed as promoting greater policy continuity and consistency and as a way of compensating for the credibility problems of democratic politicians (Majone 1999). On the other hand, the main criticism is that vital state functions are being delegated to agencies operating at arm's length from government, to the extent that 'the real work of running democracies is now carried out by the unelected' (Vibert 2007: p. 186). Moreover, this type of state bureaucracy has been seen as a contradiction of the parliamentary model of most European governments (Dohler 2011).
A main concern in regard to upward accountability is the emergence and diffusion of autonomous regulatory agencies in different policy areas, at both the national and sub-national level (Scott 2000; Flinders 2004; Levi-Faur and Jordana 2005). The regulatory agencies are one of the most prominent forms of new institutional structures characteristic of the regulatory state. Often established as independent authorities, these agencies operate outside ministerial hierarchies and, in consequence, are not subject to traditional hierarchical control and accountability mechanisms (Verhoest et al. 2004). The political isolation that these institutions enjoy, as well as the considerable autonomy they benefit from in the performance of their duties, has raised concerns regarding their accountability (Weir 1995; Flinders and Smith 1999; Scott 2000; Schmitter and Trechsel 2004). However, this is not a new problem. As early as the 1950s, critics in the United States - the regulatory state avant la lettre - first questioned the lack of political (upward) accountability of independent regulatory commissions created in prior decades (Majone 1996).
The shift to the regulatory state has created concerns about the democratic legitimacy of the new regulatory authorities. Thus, scholars have long discussed how these structures differ from the traditional accountability mechanisms of bureaucracies, and whether or not new forms of accountability have emerged or have eventually been reinforced. In fact, decision-making processes within regulatory institutions often involve politically sensitive choices, such as balancing economic efficiency with social or environmental objectives (Baldwin and Cave 1999). The decisions of regulatory organizations are not merely technocratic, but are often based on values-based choices (Lodge 2004: p. 124). In this regard the creation of autonomous agencies was expected to improve regulatory performance and efficiency, while having no negative side effects on other values, namely accountability (Self 2000; Pollitt et al. 2004). However, analysts have also underscored the accountability deficit of these new institutions (Baldwin et al. 1998). Clearly, the diagnosis of accountability problems of regulatory agencies will differ depending on the particular theoretical accountability perspective adopted.
Some authors argue that although agencies may not always be controlled by traditional bureaucratic procedures, they can be held accountable through alternative means (Kickert 2001; Koppell 2003; Christensen and L^greid 2004). Agencies are constrained by procedural and substantive rules that guide their discretion. Autonomy from direct political control does not mean immunity from other directions of accountability, in particular inter-institutional accountability (Majone 1999). Therefore, weaker upward accountability mechanisms towards the executive may be compensated with stronger downward accountability mechanisms towards consumers and citizens (Flinders 2004; Verhoest et al. 2005). In other words, responsiveness to users and clients might become a complement or a partial substitute for upward accountability of autonomous agencies (Bouckaert and Peters 2004). Accountability should be framed within a wider context that includes institutional checks and balances, consumer sovereignty and the empowerment of citizens (Lodge 2004). In this sense, scholars point out that the emergence of the regulatory state may imply a change in the quality of citizenship (Lodge 2004). When compared to the conception of citizenship in the era of the welfare state, the regulatory state is portrayed as bringing a more reduced version of citizenship, one that moves from a political conception to an economic consumer-like one, also relying on less effective (downward) accountability mechanisms. This criticism has led to claims of deterioration in the quality of citizenship standards (Haque 2001), and consequently a decline in overall accountability has been suggested (Falconer and Ross 1999).
Second, and from another perspective, the principal-agent (PA) literature has pointed to a wider discussion of accountability instruments given its concern with information asymmetries and incomplete contracting (for an overview, see Huber and Shipan 2000). Because of their isolation, agencies might have more opportunities to shirk, misuse and abuse political power than ministerial departments have (Koop 2011). To account for this, the PA perspective analyses whether the weakness of upward accountability mechanisms facilitates possibilities for drift due to limited control in a context of incomplete information (see McCubbins et al. 1987; Horn 1995; Epstein and O'Halloran 1999). However, autonomous regulatory agencies also face de facto incentives to remain accountable. At any given moment, principals might find that the discretion allowed to the agency has produced agency losses that are too significant. They may then decide to redesign the agency - through the modification of its mandate, operating procedures or delegation rules - or even dismantle it. Though effective intervention may not be necessary, we expect (credible) threats of political intervention to affect the agency's behaviour. Furthermore, multiple veto players may produce delegation to regulatory agencies that is more credible as the risk of political intervention to the agency diminishes. As the number of veto players increases, changes to the status of an agency turn out to be more complicated since such changes require the participation of a larger number of veto players, reducing in turn the credibility of threats to intervene in the agency. How this affects the political accountability of the agencies requires further scrutiny: are agencies that are less exposed to political intervention also less accountable, or is it the other way round?
Third, regulatory agencies are also accountable beyond the national state and its territorial scope. This is a particular view that emerges as national regulatory agencies become enmeshed in new modes of network governance. While the implementation of these networks in different policy domains constitutes another manifestation of the shift from a bureaucratic state to new ways of policymaking, they also entail the articulation of new structures of cooperation between national, regional - this being especially evident in the case of the EU - and international agencies. Here, hierarchical accountability mechanisms are almost non-existent, horizontal mechanisms hardly work properly, and most of the accountability requirements rely on downward mechanisms. Thus, regulatory agencies become accountable to broader audiences that interact with these agencies in the various governance networks and across multiple regional and international settings in which the agency is involved. From an EU perspective, the literature has underscored agencies as hubs in a network of national experts that develop informal resources and linkages, such as information, expertise, reputation and trust (Coen 2005; Sabel and Zeitlin 2008) and promote learning opportunities. While this can certainly apply to international policy networks, both networked structures generate their own audiences and forums. Hence, when analysing accountability within the politics of regulation, we also need to understand the ways in which regulatory agencies become articulated within these transnational 'governance webs' (Jayasuriya 2008) encompassing domestic, regional and international actors, and the social relations and accountability mechanisms through which agencies interact and dialogue with these wider audiences.
Many discussions of accountability have a strong normative basis. Accountability is usually seen as a positive quality of public officials and institutions, and is thus linked with normative understandings of citizenship and democracy. Based on these assumptions, most analyses focus on the assessment and evaluation of the actual conduct of public actors and agencies. However, we should recall that 'accountability and transparency are not just a good thing of which we should have more' (Lodge 2004: p. 142). In fact, accountability is a matter of adequacy, not of quantity, and dosage may vary in different settings and contexts. The normative understanding of accountability also emphasizes transparency, openness and responsibility (Bovens 2005), but analysing accountability as a social relation or mechanism opens the door to more nuanced analysis of the ways in which these relations and mechanisms are selected, how they operate and what effects they produce, including those cases of oversupply and undersupply. Hence, the study of accountability should go beyond the obsession with traditional upward accountability mechanisms and horizontal mechanisms of control, and capture to what extent and how these mechanisms are adequate to specific policy and national circumstances, while the introduction of a richer variety of well-adjusted downward mechanisms of accountability contributes to the promotion of better dialogues and improved political communities.