The literature covered in the introductory chapter of the book addresses several points on which we can base our hypotheses. First, accountability has to be considered as a multidimensional concept (Willems and Van Dooren 2012). For the purpose of this chapter it is important to note that different mechanisms can be used to allow public organizations to account for their actions: there are 'upward' mechanisms directed towards their political principals and 'downward' mechanisms directed towards their users, interest groups and other stakeholders (see Chapter 1). In this chapter we distinguish between
- (1) individual upward accountability mechanisms, referring to the mechanisms to make individual senior managers of regulatory agencies account for their actions on the criteria of legality, administrative-financial actions, and results achieved towards the objectives set, and
- (2) organizational upward accountability mechanisms, in terms of evaluating and sanctioning the achievement of results at the organizational level.
Second, discussions of the effects of New Public Management (NPM) on administrative accountability often refer to the creation or multiplication of autonomous agencies as a centrepiece in NPM reforms in many OECD countries. Many studies of agencification point to widespread problems of upward accountability because the hierarchical control chain is broken and the minister has less direct control (for example,
OECD 2002; Flinders 2004). Basically these studies assert that the further agencies are from government, and the more formal independence these organizations have, the more problematic upward accountability becomes. This point is made even more strongly when it concerns regulatory agencies, which have to be insulated from political interference as much as possible in order to be credible for market actors and citizens (see Chapter 1 in this volume).
It remains unclear in the literature, however, to what extent new downward accountability mechanisms actually replace, complement or supplement traditional upward mechanisms of accountability (Hodge and Coghill 2007; Black 2008). One approach is to stress downward accountability in order to compensate for, or even replace, failing upward accountability in the case of formal independent agencies (see Chapter 1 in this volume; see also Wilson 1995; Black 2008).
Formal independence refers to the formal/legal distance of agencies from their government. However, as has been shown extensively in the case of agencies in general (Verhoest et al. 2004) and in the case of regulatory agencies in particular (Maggetti 2007), the formal/legal independence of an agency can diverge considerably from its de facto autonomy in terms of being more extensive or more restricted. De facto autonomy is here understood as the ability of agencies to formulate their own preferences and to translate these into authoritative actions or decisions, free from pressure from their government, minister or parent department (Maggetti 2012; Verhoest et al. 2004). More specifically we focus on managerial autonomy, referring to decision-making with regard to the use of resources by the agency management, as perceived by the agency management themselves (Verhoest et al. 2010). Following the potential divergence of formal independence and de facto autonomy, one might question whether problems of upward accountability really manifest themselves if agencies do have high formal independence but low de facto autonomy, or if in similar cases agencies do have the need and the discretion to develop downward accountability mechanisms. Formal independence and de facto autonomy might have mutually compensating features when it comes to their effects on upward and downward accountability mechanisms.
Moreover, there are several factors that might affect the level of accountability for regulatory agencies (like the policy sector, the interest groups affected, factors related to politico-administrative systems and polities; see also Chapter 1 of this volume, see e.g. Koop 2011b). We will, however, explicitly consider one of them: the political salience of the agency. Agency research has clearly shown that agency size is a proxy for political salience, as larger agencies pose greater risks to politicians in the event that something goes wrong (Pollitt et al. 2004). Additionally, large agencies are in most cases also of crucial importance for specific interest groups, which will call for sufficient ways to hold the agency management to account for their actions and their effects on the private interests of these groups (Verhoest et al. 2010). Alternatively, size also refers to the capacity of an agency to invest in mechanisms for accountability, not only to their political principals but also to their stakeholders.
In this chapter we will analyse how the levels of formal independence, de facto managerial autonomy and political salience (operationalized through agency size) of regulatory agencies interact with one another in influencing the extent of upward and downward accountability of such agencies. Which configurations of these factors bring about the highest use of upward and downward accountability mechanisms in regulatory agencies?
Considering the above-mentioned elements, we expect that agencies with low levels of formal independence, low levels of managerial autonomy or large size, or a combination of these conditions, will show a high use of individual upward accountability mechanisms (H1). A similar hypothesis can be formulated regarding a high use of organizational upward accountability mechanisms (H2). On the contrary, we expect that agencies with high levels of formal independence and high levels of managerial autonomy will show a high use of downward accountability mechanisms, whether they are large or not (H3).
Another question remains to be answered: which accountability mechanisms (upward or downward) used by a regulatory agency render the highest level of accountability towards society? There is the argument that downward mechanisms can replace failing upward accountability in the case of autonomous bodies (Wilson 1995). However, there is another more restrictive position in the literature that argues that such downward systems can never replace upward direct accountability to the minister, but can merely supplement it (Mulgan 2003). Is the level of accountability of the regulatory agency towards society, as perceived by senior management, mainly influenced by (individual/organizational) upward mechanisms, by downward mechanisms, or by a combination of both mechanisms? Based on the perspective in which upward and downward accountability mechanisms are considered to be complementary, we expect that regulatory agencies that use high levels of both upward and downward accountability mechanisms report a high level ofself-perceived accountability towards society (H4). However, this position might not be true, as Schillemans (2011) found that downward mechanisms do not deepen upward ministerial accountability.