Independence and accountability: Definitions
It is rare to see the concepts of independence and accountability properly distinguished in political science language. Authors who address and analyse independence occasionally refer to accountability in their works (see, for instance, Elgie 1998; Gilardi 2002, 2005; Edwards and Waverman 2006; Wonka and Rittberger 2010),3 and vice versa (see, for instance, Mulgan 2000; Bovens 2007; Philp 2009).4 Although everyone would probably agree that independence and accountability are closely related, the relationship between these two concepts has not yet been investigated to a sufficient extent. Before analysing the different applications of the two concepts in the study of RAs, I will review the most relevant literature in order to examine how they can be defined and understood.
Independence, generally speaking, is the condition of a person or body of being able to take decisions without interference from other people or bodies. Beyond this vague definition, many particular definitions have been formulated. In the literature on central banks' autonomy, 'political' independence has been defined as 'the ability [... ] to select [... ] policy objectives without influence from the government' (Alesina and Summers 1993: p. 153). A similar definition is given by Elgie (1998: p. 55). Hanretty and Koop (2012: p. 199) define 'political independence' as 'the degree to which the day-to-day decisions of regulatory agencies are formed without the interference of politicians and/or consideration of politicians' preferences'. Surprisingly, many authors who study, measure or assess independence in their works do not explicitly define the term (Cukierman et al. 1992; Gilardi 2002; Edwards and Waverman 2006; Wonka and Rittberger 2010), but rather prefer to focus on the empirical operationalization they choose.
An important distinction to be made is that between formal (or 'institutional', 'legal' or 'de jure') independence and actual (or 'de facto') independence (Cukierman et al. 1992; Gilardi and Maggetti 2011). The former indicates whatever is explicitly mentioned in laws and statutes that establish RAs, dictate their rules of procedure and confer specific tasks on them. The latter refers to how the rules are translated into practice in the agencies' day-to-day activity: frequency of 'revolving doors' phenomena, partisanship of nominations, internal organization and so on (see Maggetti 2007; Hanretty and Koop 2013). Most empirical analyses (see Gilardi 2002, 2005; Elgie and McMenamin 2005; Wonka and Rittberger 2010; Guidi 2014), and this chapter as well, focus on formal independence. The reason formal independence is particularly interesting is that it is a variable parliaments can modify by drafting new legislation, whereas features of de facto independence are often embedded in the customs and traditions of a country. Actual independence is not only more difficult to measure and compare, but also less useful when it comes to recommending legislative interventions. Especially if a study aims to indicate political decisions that can produce better regulatory performance, it is crucial to assess the causes and the impacts of legislative provisions - though it is important to control for country-specific features, including those concerning actual independence.
Accountability has been defined in a variety of ways as well. Following Koop (2011b; see also Koop 2011a), we can distinguish among three possible definitions. Basically every definition of accountability entails that there is some person or body that must offer information or explanation with regard to its activity to some other individual or body. What is not agreed amongst scholars who have dealt with this question is whether accountability also entails the possibility to sanction the accountable person or body. The first definition states that the sanctioning element is not necessary, as the obligation to inform and explain is the only constitutive element of accountability (see Philp 2009: p. 32; Flinders 2001). The second definition has been proposed by other authors, according to whom the sanctioning element may or may not be present (Schedler 1999; Stinga 2008), accountability being a 'radial concept' (Collier and Mahon 1993). Finally, most authors argue that the sanctioning element must be present, together with the information/explanation element (Mulgan 2000; Behn 2001; Fearon 1999; Strom 2000; Bovens 2007). Bovens (2007: p. 450), for instance, defines accountability as 'a relationship between an actor and a forum, in which the actor has an obligation to explain and to justify his or her conduct, the forum can pose questions and pass judgement, and the actor may face consequences'. Although in many cases accountability exists when there is a principal-agent relationship (as argued by Fearon 1999), a person or body can be accountable also vis-a-vis people or organizations other than the principal. This chapter adopts the definition of accountability based on Bovens's (2007) set out in the introduction of this book, that is, a relationship in which the sanctioning element is present together with the information/explanation one.