The relationship between independence and accountability: Operational definitions
As already mentioned, scholars who study regulatory independence occasionally refer to accountability, and vice versa. These authors certainly do not ignore the fact that the two concepts are interlinked. Most likely, they implicitly acknowledge that some relationship exists, but they do not spell it out. Looking at the definitions analysed in the previous section, we can attempt to establish a link between independence and accountability. In order to clearly identify the precise meaning and the implications of independence and accountability, I will use a simplified example, in which the relationship is between two generic subjects, 'regulatory agency' and 'politics'. It is worth noting that, although (for simplicity's sake) I will use the adjectives 'independent' and 'accountable' in absolute terms, these are not dichotomous concepts: no person or body is either fully independent from, or fully dependent on, other actors - and the same applies to accountability. We must rather refer to a certain degree of independence or of accountability of a particular subject.
The operational definition of independence used in this chapter is the following:
A regulatory agency is independent from politics if it can carry out its activity without interference from politics and without considering the politicians' preferences.
The operational definition of accountability used in this chapter, meanwhile, is the following:
A regulatory agency is accountable to politics if it must offer information about and explain its activity to politicians, and it may face consequences for how it operates.
Following these definitions - which are well-established in the literature, despite slightly different formulations - independence and accountability can be seen to some extent as opposites of one another. In fact, the sanctioning element is strictly related to both concepts. Features of regulatory independence normally imply insulation from politics, difficulty in dismissing agency members, long terms of office and so forth: all these elements make it difficult for politicians to steer the agency's activity, especially on a day-to-day basis. At the same time, the extent to which agencies must explain and justify their decisions has a clear impact on the possibility of sanctions: the more often information must be offered, and the more detailed this information must be, the greater the possibility for the principal to sanction the agency's behaviour.
Therefore, as long as accountability is deemed to include a sanctioning element - that is, a feature that restricts the agency's freedom to pursue its own goals and make it subject to interference from politics - the presence of independence must be negatively correlated to the presence of accountability, and vice versa. This does not mean that they are mutually exclusive and that their 'marginal rate of substitution' is fixed at one. However, some sort of trade-off between the two must certainly exist.