Query

We always assume that there is a direct link between transparency and accountability, but is this a given or is it supported by evidence?

Introduction

In general, it is assumed that the existence of transparency would result in better governance, more accountability and less corruption (Bovens 2006: Koppell 2005: Mulgan 2012). In fact, the demand for accountability is often responded by increasing the level of transparency under the assumption that better and more information would allow citizens, governments or markets to hold institutions accountable for their policies and performance (Bellver and Kaufmann 2005). However, even if in theory the link between transparency and accountability could seem obvious, in practice it is not always straightforward.

The quality of the relationship between transparency and accountability depends on their purpose, definition and implementation. There are no single conceptions of transparency and accountability. How those concepts are defined depends on why and who are demanding them, which in turn will determine how they are expected to be implemented and how they affect each other.

The concept of transparency and its demand

Attempts to define transparency emphasise different dimensions of the concept. Transparency involves the release of information and requires an open attitude about actions and decisions, indicated by the degree to which the principal (on whose behalf the agent is supposed to act) can monitor and evaluate the actions of the agent (who does the action) (Bauhr and Nasiritousi 2012).

Some definitions emphasise that the disclosure of information is not enough, and the latter has to be reliable, accessible, of good quality and on time to be effective and understandable to the principal. For instance, according to the OECD, “transparency refers to an environment in which the objectives of policy, its legal, institutional and economic framework, policy decisions and their rationale, data and information related to monetary and financial policies, and the terms of agencies’ accountability, are provided to the public in a comprehensible, accessible, and timely manner” (OECD web glossary). Others highlight the relation between the principal and the agent enhanced through transparency (Meijer 2013; Cotterrell 1999). In this sense, authors also refer to active transparency (voluntary disclosure of information) and passive transparency (release of information following a request) (Mabillard and Zumofen 2015). The U4 Anti-corruption Resource Centre defines transparency as “the quality of being open, communicative and accountable. It implies that governments and other agencies have a duty to act visibly and understandably”.

The demand for transparency has emerged in the areas of governance and the economic market mainly, and it has been targeted to governments and private companies and banks (Kosack and Fung 2014; Bellver and Kaufmann 2005). In both areas that demand has evolved over time, acquiring variations in its purpose. For example, in governance, the demand for transparency was initially justified by the citizens’ right to information (expressed in RTI legislation) and freedom of information (FOI legislation).

FOI legislation empowers citizens to request information and obliges governments to provide that information. According to Kosack and Fung (2014), the use of this kind of transparency enables citizens to better govern themselves democratically since to have access to a wide range of information and arguments empower citizens to express their preferences effectively.

In recent decades, the demand for transparency in governance has moved towards linking transparency with accountability. It is formulated by citizens as individuals, more than as a collective, in need of better public services (Kosack and Fung 2014). This change responded to civil society concerns with corruption and underperformance of public services. The ultimate purpose of transparency in this case is to improve the life of individuals by improving the quality of public services (Kosack and Fung 2014), and not so much the balancing powers in a democratic system.

In the economic market, the demand for transparency has had at least two purposes: to ensure responsible corporate behaviour and to tame undue private power (Kosack and Fung 2014; Brandeis 1913). In these cases, transparency serves customers to protect themselves from unscrupulous merchants and bankers, to make better choices in the marketplace and to negotiate from a stronger position.

The purpose of transparency plays an important role in determining how and what kind of information should be disclosed. When transparency is demanded to governments by citizens following its right to information and participation, the implementation of transparency is expected to be open and include information on policy decision-making processes, management of funding and results, among other issues. However, when transparency is demanded by citizens as costumers and is targeted to private companies, it is expected to provide information related to the product or service, and not necessarily on the business and marketing strategies of the company.

The concept of accountability and its demand

The notion of accountability refers to a relationship between the agent (who does the action) and the principal (on whose behalf the agent is supposed to act), in which the principal is able to hold the agent responsible for its actions and the proper execution of its powers. In the U4 Anti-corruption Resource Centre, accountability is defined as “the obligation of an individual or an organisation (either in the public or the private sectors) to accept responsibility for their activities, and to disclose them in a transparent manner. This includes the responsibility for decision-making processes, money or other entrusted property”.

Accountability can be expressed at two main levels: answerability (the duty of an individual or organisation to answer to their decisions and actions) and sanctions (the power to sanction misdemeanours and malpractice). Both forms involve power transmission through the obligation of the agent to inform the principal of their conduct, and the capacity of the principal to ask questions and to pass judgement (Bovens 2006; Hood 2010).

The form of accountability varies depending on the context and, in particular, on who is the principal. When the principal is a higher authority the direction of accountability is “upwards”. The direction of accountability can also be “downwards” when the principals are citizens or a community; and “horizontal” when it is part of a contract or partnership agreed for mutual benefit (Murphy et al. 2016).

In public administration, the demand for accountability has evolved following changes in institutional structure and functions. If traditional bureaucracies relied primarily on upwards accountability due to hierarchical management structures, the introduction of new public management (NPM) reforms, involving outsourcing and the privatisation of some services, has turned accountability into its horizontal form through contractual relationship with suppliers and partners (Murphy et al. 2016).

High levels of corruption and poor quality of public services, especially in developing countries, enhanced the demand for accountability from the civil society in its downward form. In recent years, bottom-up accountability mechanisms, also known as social accountability, have been highly valued as an alternative to the limited success of top-down accountability to curb corruption, especially in countries with a systematic problem of corruption and weak institutional systems (UNCAC 2004).

However, the implementation of social accountability encounters several challenges, raising questions about how effective the involvement of citizens in monitoring public management actually is.Empirical studies conclude that social accountability processes and outcomes are heavily context-dependent (O’Meally 2013).

Nevertheless, studies point to shared challenges, such as the means and quality of information provided by the government to the citizens, the capacity and education of those citizens to properly assess governmental performance, low levels of civic engagement or the co-optation of social representatives (Olken 2007; Schouten 2011; Arroyo and Sirker 2005).

Considering these differences on how transparency and accountability can be formulated, does more transparency improve accountability? The answer to this question is threefold: yes when transparency is a pre-condition of accountability; no when transparency has high costs; and it depends on the type of transparency and the type of accountability.

Transparency as pre-condition for accountability

In general, a positive view among theorists and practitioners predominates on the benefits of transparency on accountability. In extreme cases, both concepts are seen as indistinguishable and inseparable from one another, what Hood (2010) calls “Siamese twins”. This understanding might lead to the assumption that simply the implementation of transparency measures automatically means more accountability.

In other cases, transparency is seen as pre-condition of accountability because it provides the ability of the principal to observe how the agent behaves and the consequences of the agent’s behaviour (Prat 2005). Moreover, the disclosure of information enhances the prosecution and sanction of corrupt behaviour by helping to dissuade officials from acting in that way (Murphy et al. 2016; Hood 2010).

Mabillard and Zumofen (2015) argue that transparency may reinforce horizontal accountability. This has been especially relevant regarding the incorporation of NPM reforms in public administration involving increasing public-private partnerships through the outsourcing and privatisation of public services. The horizontal accountability required in this new public management style relies on data transparency and open competition (Ferry et al. 2016). According to Mabillard and Zumofen (2015), transparency also plays a supporting role in vertical accountability, in the sense that informed citizens can alert authorities and denounce dysfunctions.

Nevertheless, the positive impact of transparency on accountability is often more theoretical than empirical (Hale 2008; Meijer 2013; Mulgan 2012; Bovens 2006). Empirical studies point to common challenges, such as the quality of information and how information is managed.

It is not sufficient to make information accessible; the information needs to be relevant, released on time and communicated in a way that it can be understood by the receivers (Schouten 2011). For instance, the reforms undertook in English municipalities in the last years illustrate the potential consequences of how transparency requirements could diminish accountability because of how information is visible and to whom (Ferry et al. 2016). For long lasting hierarchical accountability, reforms in local governments involved the creation of bottom-up mechanisms, such as greater transparency and open markets for public services. That led to the establishment of horizontal accountability where municipal governments were required to publish details of all transactions over £500, the salaries of senior staff and some performance information. Citizens are invited to assess organisational performance on the basis of gigabytes of raw data without assistance from professional auditors, which implies that the data will be inaccessible and meaningless to great part of the public (Ferry et al. 2016). Thus, what was intended to increase transparency turns out to be quite opaque.

In countries in conflict and fragile states, a common challenge is the capacity of public institutions to generate and manage data (Schouten 2011). Often the data generated by different departments of the same government is inconsistent, and discrepancies across political and administrative units prevents a consistent process of consolidation and verification of the data on services such as health and education (Schouten 2011).

The costs of transparency

Contested questions regarding transparency are how much information should be revealed and what might be the costs of transparency. In some contexts, the demand for transparency finds limitations in areas in which public decision making is protected and exempt of the requirement to be disclosed by law (Prat 2005).

Arguments made against transparency are based on the following ideas:

Protection: some information could be used in a hurtful way by a third party (Prat 2005). For example, in politics, to not disclose information on national security strategy would prevent its misuse by hostile countries. In business, to keep information on new products or business plans secret would impede competitors to use that information to position themselves ahead in the market.

Separation: if the agent and the principal have different understandings of the purpose of transparency, the disclosure of information might create separation between them. For example, when a government discloses data rather than meaningful information just to tick the box of “being transparent”, not only it does not have an effect on accountability (O’Neill 2006), but might create tension and prevent further communication between the principal and the agent.

Fulfilling expectations: in some cases agents might be tempted to fulfil requirements for transparency by providing what is expected rather than honest and meaningful information. For instance, transparency on actions rather than on consequences might reduce the chances for accountability since the agent might be more interested in selecting the information according to what is expected and is seen as “normal”, while withholding challenging but important information (Pratt 2005). The European Council of Ministers, which has made efforts to be more transparent by making votes public, offers good examples of how members of the council take positions in private that differ those expressed in public (Stasavage 2005). This is facilitated by the fact that decision making in the council is still made in secrecy. Arguments in favour of closed door decision making point to the danger of using transparency to “fulfil expectations” and state that secrecy favours less public posturing and gives space to governments to deliberate freely (Heisenberg 2005). In this case, the combination of transparency with secrecy might nullify transparency measures and impede accountability.

Effective performance: some authors argue that excessive transparency and accountability might have a negative effect on organisational performance. The concern is based on the idea that excessive monitoring and rules can hinder decision making (Bovens 2006), and that the costs of operating accountability systems could be greater than the benefits they bring.

Making accountability confusing: in some cases, different requirements to be transparent might make accountability confusing and difficult to achieve. For instance, between 2010 and 2015 the UK’s coalition government introduced reforms in the police forces in England and Wales for them to publish a range of performance and financial information online and directly elected police and crime commissioners to oversee those forces (Murphy et al. 2017). These new transparency arrangements implied the multiplication of actors involved in assuming the role of both principal and agent in different upwards and downwards accountability relationships. That created confusion among the public regarding roles and responsibilities of the forces, making it difficult to hold officials to account.

What type of transparency for what type of accountability

The link between transparency and accountability is a two-way relationship. On one hand, the model of accountability defines the type of transparency necessary. On the other hand, the quality of the information released determines what kind of accountability is possible. These interdependent relationships require thinking of transparency and accountability not only as meaningfully related concepts but in terms of what type of transparency and accountability can best serve to each other.

With that question in mind, Fox (2007) analysed the impact of transparency on accountability according to three institutional capacities: dissemination of and access to information, answerability, and the power to sanction and compensate. Those institutional capacities are analysed against two types of transparency – clear and opaque – and two types of accountability – soft and hard. Clear transparency refers to information‑access policies and to the revelation of reliable information about institutional performance including official’s responsibilities and funding destinations. In an opaque transparency, the information shared does not reveal how an institution behaves in practice, how decisions are made and the consequences of its actions. Moreover, when transparency is opaque, the information might not be reliable. Soft accountability refers to the answerability of the institution, and hard accountability is the answerability with the possibility of sanctions.

Fox (2007) concludes that the dissemination and access to information on its own does not make an institution accountable. Only when there is institutional answerability is there an impact of transparency (in its “clear” form) on accountability (in its “soft” form). This conclusion refers to institutions with the power not only to reveal existing data but also to produce answers by investigating and producing information on institutional behaviour.

In turn, accountability cannot be expected from opaque transparency. Furthermore, answerability does not necessarily imply the power of sanctions; therefore, hard accountability cannot be expected from “answerability”. The power to sanction requires more than transparency, and it is influenced by the nature of the governing regime and the capacity of the civil society to mobilise.

Some empirical examples contradict Fox’s conclusion on the disconnect between transparency and hard accountability. For instance, in 2001, 25 million students in Bangladesh started the school year without textbooks because the sole supplier failed to deliver them on time. Even then, the few books that were available were full of errors (Arroyo and Sirker 2005). After an investigation by Transparency International Bangladesh (TIB) about the availability of books, the textbook errors and the quality of the books printed, the government filed a lawsuit against the corrupt institutions that had caused the shortage.

In analysing the Swiss context, Mabillard and Zumofen (2015) also find that the disclosure of information may actually lead to sanctions. What can make the difference, according to the authors, is the content of information, which can be of two types: compromising and non-compromising. These authors distinguish between active transparency and passive transparency, and within accountability they differentiate between answerability and enforcement. They argue that, when there is passive transparency on compromising information, the form of accountability might be enforcement through sanctions. The scandal in Switzerland of the State Secretariat for Economic Affairs (SECO) and its attribution of more than 40 public mandates to enterprise without going through a public bidding system supports this argument. After the Federal Data Protection and Information Commissioner granted access to the documents in response to journalists’ request, the manager of the public official involved resigned. In this example, a simple access to information provoked sanctions.

Mabillar and Zumofen (2015) also argue that when there is active disclosure of compromising information both accountabilities – answerability and enforcement – might be affected through remediation. For instance, when the information on the mismanagement and failures in the surveillance of the INSIEME project, in charge of merging the IT systems of the direct federal taxes and the value added tax in Switzerland, was revealed, the finance minister decided to suspend the person responsible for the project and initiate an investigation. That scandal enhanced the demand for more transparency within the administration, which resulted in accepting in 2014 a motion about the annual publication of all information related to public markets when a transaction is higher than CHF50,000 by the Federal Council. In this example, the accountability process took place before the transparency process and the remedy that led to the disclosure of information happened later (Mabillar and Zumofen 2015).

How worldviews can influence transparency and accountability relations

Hood (2010) offers an alternative model regarding the typology and relationship between transparency and accountability. According to Hood, the form that transparency and accountability acquire, as well as the relationship between them, depends on particular ways of looking at the world. He distinguishes four worldviews: hierarchist, egalitarian, individualist and fatalist.

From a hierarchist point of view, accountability is upwards, and answerability is owed to authorities in the higher position of the social order. Accountability will be concerned mainly with the adherence to rules and standards set by the authorities. The type of transparency required in this model is not general access to information but access to information on a “need to know” basis. The entitlement to know is instrumental, and it depends on status. In this view, both transparency and accountability are seen as separable but needed to produce good governance (“matching parts”).

In an egalitarian world, accountability is downwards and owed to the people and community at large. Examples of this type of accountability are citizens’ forums, such as the town meetings established in New England in the 17th century where the community comes together to make decisions on policies and budgets for the local government. The suitable transparency in this model is general openness and disclosure of information on the actions of organisations and individuals, without limitations set by rules or statues. Transparency is considered a right. The relationship between transparency and accountability in an egalitarian view is as inseparable concepts (“Siamese twins”).

From an individualist perspective, accountability is horizontal and takes place between contracting parties (for example, between buyer and seller). Individualism adds the notion of accountability to oneself, independently of any collective process or authority. The value of accountability rests on individual choices. The type of transparency required for this kind of accountability is one that respects individual rights to privacy, intellectual property and commercial confidentiality. The expectation will be transparency on information that helps individuals to make choices or manage negotiations.

According to a fatalist perspective, since life is unpredictable, accountability should put emphasis on processes and structures that make the accountability environment unpredictable for individual and organisations. The same view applies to transparency, where the intentions of transparency and the actual transparency is seen as problematic due to the unpredictable or reverse effects that transparency measures might have. In fatalists’ opinion, when transparency takes place it is due to coincidental circumstances. In this view, there is not necessarily a connection between transparency and accountability, and their relationship can be of any type.

Rethinking accountability and transparency

The lack of expected results from both top-down and bottom-up accountability mechanisms, together with the often absence of integration between transparency and accountability initiatives, has triggered a reflection on how to do accountability and, in turn transparency too, more efficiently. Proponents of doing accountability differently speak about ‘the vertical integration of civil society policy monitoring and advocacy’ (Fox and Aceron 2016). According to Fox, vertically integrated accountability involves (Fox and Aceron 2016):

-scaling up (not in the sense of replication but of impact) to ‘connect the dots’ looking at how different levels of decision-making at the district, provincial, national and transnational levels interact with each other;

- ‘vertical integration’, which implies coordination of monitoring and advocacy across the governance process ‘from policy debate and agenda-setting to the formulation of policy and budget decisions, as well as to their implementation throughout different agencies and levels of government’;

- combination of policy monitoring and public interest advocacy creating synergies.

The importance of the context

The potential of transparency to lead to more accountability depends in great manner on contextual considerations. Kosack and Fung (2014) highlight the following contextual factors as especially relevant: market conditions, political will and the type of relationship between agents and principals. In a market where there is competition, the effect of transparency on accountability increases. The information provided by transparency will give customers the capacity to switch to higher-performing providers, which might increase the willingness of companies to be accountable so that they do not lose customers. Having the will to be more accountable is a crucial aspect, especially in politics since the effect of competition only happens in electoral times. The lack of political will might explain the challenge of implementation of existing laws and institutional mechanisms in place. It also requires going beyond the minimum transparency requirements and ensuring the use of transparency for accountability purposes.

In the case of political will, the relationship between the agents and the principal is most relevant (Kosack and Fung 2014). On one hand, that relationship can be one of collaboration; for example, when reform-minded officials are allies of advocates to improve services. On the other hand, the relationship can be of confrontation. In a confrontational relationship, the assumption is that the agents will misbehave unless they are constantly watched. From these definitions, a way to know whether the relationship would be collaborative or confrontational can be to assess if and how proactive agents are in being transparent and taking responsibility for their behaviour.

These two types of relationship imply very different uses of transparency (and the subsequent effect on accountability). In a collaborative relationship, transparency provides the basis for joint problem solving. In the confrontational model, transparency mechanisms will be more in the form of complaint hotlines, citizen charters and social audits, to expose corruption and malfeasance (Kosack and Fung 2014).

Both types of relationship, often coexisting in the same situation, might have a positive effect on accountability. An example of a successful collaborative relationship is the local government intervention in the health sector in Uganda in 2004 (Kosack and Fung 2014). Transparency was guaranteed through “community score cards” containing detailed information about the quality of health services in each community and compared to those in other communities and to government benchmarks. This information allowed communities to identify problems in the health sector and develop plans of action to solve them. A year later, mortality in children under five fell by 33%, and immunisation rates and treatment practices rose by 20%.

An example of a successful confrontational relationship are the social audits to prevent corruption under the Mahatma Gandhi National Rural Employment Guarantee Act (MGNREGA) in Andhra Pradesh, India. Among the positive impacts of the social audits were means to detect and contribute to the containment of wage theft (although there is still significant corruption regarding the purchase of material for construction sites) and to force corrupt officials to give back stolen money during public hearings (around one‑third of the funds determined as misappropriated by the auditors has been recovered).

The difference between collaborative and confrontational systems might lie in the level of commitment of the government to be accountable. In a collaborative relationship, government commitment is potentially higher than in a confrontational relationship since there is an internal motivation. In a confrontational relationship, however, the motivation is external and, therefore, more vulnerable to be affected by external circumstances.

The type of political regime can also play a significant role in determining how transparency relates to accountability. In a democratic society it is assumed that transparency would have a positive effect on accountability due to the power of voters to change a government in an election. The risk of not being voted makes politicians more responsive to public demands, and it is understood that proper elections require informed voters, which implies the need for transparent and open governments.

In authoritarian regimes, where the power to sanction behaviour lays in elite government officials rather than on the electorate, transparency might have a different effect. In fact, transparency in authoritarian assemblies may have contradictory effects (Malesky et al 2012): on one hand it might help to reduce illicit bargains and political self-dealing by exposing questionable politics and trades in the assemblies to the public; on the other hand, transparency could reduce forms of participation. In trying to determine whether transparency interventions can improve delegate performance in authoritarian parliaments through a field experiment in Vietnam, Malesky et al (2012) find that the higher the exposure to transparency –measured in terms of internet penetration in a province and online exposure- the most likely a delegate was to behave in a conformist manner. On one hand, a closed forum of debate sometimes represents a safe place in which delegates might have more freedom to criticize regime leaders and for the latter to accept it, facilitating the debate and solution-seeking among authorities and other actors. Transparency might disrupt this balanced closed order diminishing delegates freedom to provide honest opinions. Conformist behaviour not only would eliminate a necessary debate but also would prevent actions and decisions that could yield better outcomes (Prat 2005). Malesky et al (2012) finding questions arguments on the benefits of transparency in authoritarian national legislatures for its potential to increase responsiveness to stakeholders such as NGOs and media (Kaufman and Bellver 2005).