The concept of accountability has become increasingly popular in diverse fields including development policy. This working paper from the Overseas Development Institute argues that new meanings and dimensions risk diluting its content and creating conceptual confusion – with significant implications for empirical analysis. A classic approach to concept formation is required, which suggests that accountability refers to a class of concepts under the category ‘methods of limiting power’. It is important to distinguish between accountability and responsiveness.
Commonalities in the concept of accountability from its history in political science and financial accounting include the principle of delegating authority, holding to account, and imposing sanctions. A classic approach to concept formation provides basic rules for categorisation and comparison. Accountability is subordinate to the concept of power in the classical typological sense.
- There are five main characteristics of the core concept of accountability including: (i) an agent or institution who is to give an account (A for agent) ; (ii) an area, responsibilities, or domain subject to accountability (D for domain); (iii) an agent or institution to whom A is to give account (P for principal); (iv); the right of P to require A to inform and explain/justify decisions with regard to D; and (v) the right of P to sanction A if A fails to inform and/or explain/justify decisions with regard to D.
- There are three key dimensions in distinguishing types of accountability including: different sources of the accountability relationship; degree of control which the principal exercises over the power holder; and spatial direction of the accountability relationship.
- On the basis of these three dimensions, twelve subtypes of accountability can be identified including: business, bureaucratic, audit, client-patron, patron-client, peer professional, representative, fiscal, legal, societal, political, and reputational.
Efforts are needed to resist conceptual ‘stretching’ and to overcome the growing confusion regarding the use of the concept of accountability. A classic approach to concept formation is required to protect the concept of accountability and its usefulness for empirical analysis. Some implications for empirical research include:
- The importance of distinguishing between accountability and responsiveness: Responsiveness is simply a possible desired outcome of some types of accountability, rather than integral to the concept itself. For instance, responsiveness is key to vertical-upward accountability (e.g. shareholders expect business executives to be responsive), however, vertical-downward accountability is very different (e.g. in the case of legal accountability, judges are not supposed to be responsive to politicians and citizens)
- The difficulty of aggregating findings about different subtypes of accountability to make general conclusions on causal form: Although the various subtypes of accountability may fit within a broad definition, each has its own internal logic and separate domains which require alternative empirical approaches to analysis. For this reason, causal arguments based on the study of accountability on one level of analysis using a particular type of concept do not necessarily apply to other types and levels.
