Abstract
Across the globe, ‘old’ open government is gradually transforming into ‘new’ open government. While freedom of information (FOI) legislation is usually associated with old open government, this legislation itself is also on the move. This country report discusses to what extent the Dutch Open Government Act, that has been enacted very recently, reflects this shift from old to new open government. While certainly incorporating features of modern open government, such as the emphasis on machine-readable formats of government documents, traditional elements of FOI legislation, such as the right to request for information without stating an interest, are preserved to ensure adequate access to government information.
Introduction
The Netherlands has long been considered, or at least has considered itself, a frontrunner in terms of access to government information (Ruijer & Meijer, 2016; De Graaf et al., 2019). Admittedly, the Freedom of the Press Act of Sweden (1766) and the Freedom of Information Act of the United States (1966) preceded the entry into force of the Dutch Wet openbaarheid van bestuur (Public Access Act) in 1980. Nonetheless, this piece of freedom of information (FOI) legislation was still one of the first in the world regulating access to government information.
Since 1980, ideas on openness and transparency have developed. These developments have provoked the question whether the Wet openbaarheid van bestuur is still fit to meet the demands of open government in a data- and technology-driven era. After a lengthy legislative process of roughly one decade, this debate has finally resulted in the replacement of the Wet openbaarheid van bestuur (Public Access Act) by a new Wet open overheid (Open Government Act) in 2022.
This country report sketches the developments with regard to Dutch open government legislation since the adoption of its first FOI legislation. It first discusses in general the worldwide shift from old to new government and the role of the applicable legislative framework therein (Section 2). Next, it provides for a brief history of the Wet openbaarheid van bestuur (Section 3), followed by a characterization of the new Wet open overheid (Section 4). This characterization is the basis for discussing the central question whether the Wet open overheid reflects a shift from old to new open government (Section 5).
From old to new open government
‘Old’ open government is usually associated with freedom of information (FOI) legislation, conferring individual rights to citizens to request for government information. ‘New’ open government, by contrast, is more often considered a non-legal phenomenon, which origins are usually traced in the Open Government initiatives of President Obama in 2009 (McDermott, 2010) and in the international Open Government Partnership initiated in 2011 (Moon, 2020).
Yu and Robinson (2011) signal that over several decades, the term ‘open government’ has been used primarily as a synonym for public access to previously undisclosed government information. However, with the advent of ‘open data’, the meaning of ‘open government’ has become blurred, as it has become a label for both technological innovation (with open government data) and political accountability.
According to Noveck, the rise of open data could even be considered the death of FOI legislation (Noveck, 2016). She identifies three important differences between open data and freedom of information policy. First, FOI legislation deals with disclosure at request (ex post), whereas open data thrives on ex ante, proactive publication. Second, whereas the information typically involved in FOI requests is created by the government about its workings, open data also extends to information collected by public authorities. Finally, where FOI legislation is used mainly by journalists, open data have a wider audience, as everyone can use these data for any purpose (Noveck, 2017). Ultimately, she argues for a complementary approach where FOI legislation can be strengthened by the emphasis within the open data discourse on proactive disclosure, whereas open data laws usually lack a private right of action to compel disclosure as is the case with FOI legislation (Noveck, 2016; Noveck, 2017).
Pozen (2017) also signals how practices of open government have evolved beyond applicable FOI legislation, with its “reactionary” form of transparency. Instances of affirmative disclosure have developed beyond the minimal requirements set by FOI legislation. He therefore proposes to replace FOI’s request-and-respond paradigm by a model based on affirmative disclosure with standardized disclosure methods to facilitate analysis and oversight.
In addition, Berliner et al. (2018) observe that the position of FOI as the focal point of transparency advocates has been challenged by the greater diversity and breadth of the computer-mediated open government movement, introducing a wider focus on participation, collaboration, and technological innovation. Thus, FOI no longer occupy the same position of prominence that it once did. Since FOI makes up only a small slice of the open government reforms, the eminent question is which role FOI legislation must play in the future. Berliner et al. identify both positive (complementary) and negative (subtractive) effects of open government reforms on FOI legislation. A positive effect is that FOI and open data advocates share similar goals of transparency, which might strengthen support for broadening government priorities (ranging from democratic accountability to product innovation). A negative effect might be that open data with its emphasis on proactive disclosure leads to increased discretion on the side of public officials, as they can decide freely which information to publish proactively and which not.
Where FOI legislation is considered a representative of ‘old’ open government, it is clear that some shift from ‘old’ open government to ‘modern’ open government is currently taking place. Moon (2020) has characterized this shift along four dimensions. The first dimension relates to policy where freedom of information policy (‘old’ open government) is contrasted with open government policy (‘new’ open government). The second dimension deals with the distinction between information and data: where old open government is associated with traditionally segregated information, digitalized connectable and reusable data are at the core of new open government. The third dimension relates to administrative values: where transparency and accountability are at the heart of FOI legislation, open government promotes interactivity in the forms of participation and collaboration. The fourth dimension, finally, concerns the type of citizen involved: while FOI legislation takes the informed citizen as starting point, modern open government initiatives treat citizens not only as information receivers, but also as active users and even co-producers of public services using open government data.
The common denominator in this literature is that old open government is often considered a legal phenomenon, rooted in some form of FOI legislation, whereas modern open government is a tech-driven approach, less restricted by legal constraints (Afful-Dadzie & Afful-Dadzie, 2017). However, since the applicable legal framework is a relevant factor for open government practices, if not the most important one (Zuiderwijk & Janssen, 2014), modern open government should not be isolated from the applicable legal framework. Instead, the applicable legal framework itself is not a static relict from the past but is constantly evolving between ‘old’ and ‘new’ instances of open government. This regulatory transition will be explored below by an in-depth case analysis, considering the evolution of open government legislation in the Netherlands.
‘Old’ open government: The Dutch Public Access Act
Although the origins of government transparency in the Netherlands have a much longer tradition (Meijer, 2015), the adoption and enactment of the Wet openbaarheid van bestuur (Wob) in 1980 marked a new era in open government legislation. This act was a delayed response to the governmental report Openbaarheid overheid (Openness of Government) published one decade before, which suggested to adopt legislation granting rights to citizens to have access to government information (Commissie Heroriëntatie Overheidsvoorlichting, 1970). Following an evaluation of this first piece of FOI legislation, the Wet openbaarheid van bestuur was modernized in 1992, but its contents remained rather stable over time (Ruijer & Meijer, 2016).
The main characteristics of the Wob can be characterized as follows. First, the Wob prescribed that any person could demand information related to an administrative matter if that information was contained in documents held by administrative authorities themselves or by companies carrying out work on behalf of an administrative authority (Article 3). The requester did not have to state his interest in acquiring the information requested. As a rule, the information should be provided in the way as requested, e.g. by means of a copy or a summary of the document (Article 7).
The provisions of the Wob mainly dealt with disclosure of government documents at request. However, the Wob also contained a provision on proactive disclosure. It stated in rather general terms that an administrative authority should provide, of its own accord, information on its policy, and the preparation and implementation thereof, whenever the provision of such information was in the interests of effective, democratic governance. In case of proactive disclosure, the Wob further specified that information should be provided in a comprehensible form and in such a way as to reach the interested party and as many interested members of the public as possible (Article 8). Contrary to passive disclosure, proactive disclosure was not based on an enforceable right (Ruijer & Meijer, 2016). The only way to enforce proactive dissemination was to request the document at issue.
The right to government information was not absolute under the Wob. Although the interest of openness was presented as the main concern for administrative authorities when providing information (Article 2), the Wob listed several exceptions to disclosure of government information, which applied to both proactive and passive disclosure. Some of these exceptions had an absolute character, opposing disclosure under any condition, such as state security or the protection of confidential business information. Other exemption grounds had a relative nature in the sense that administrative authorities should weigh the interest of making government documents public against the interest of protecting these other legitimate grounds, such as privacy or inspection (Article 10).
For some time (2006–2015), the Wob also contained provisions on the re-use of government information. These provisions implemented the EU PSI Directive on the re-use of public sector information (PSI) (European Parliament and Council, 2003). This directive obliged Member States to make the government information that they had disclosed, accessible in an electronic way. The Dutch legislator initially chose to implement this directive in the Wob, as re-use is closely linked with access: no re-use without access. After the revision of the PSI Directive in 2013 (European Parliament and Council, 2013), which required Member States to facilitate re-use by making government documents available in a machine-readable way, the Dutch government decided to remove the re-use provisions from the Wob, as the Wob with its focus on democratic accountability was said to differ from the re-use regime, which had the primarily economic motive of facilitating the development of new information services.
Although the Wob has been amended several times since 1980, its main features have always been preserved with its focus on disclosure at request and a rather general provision on proactive disclosure. However, the evaluation of the Wet openbaarheid van bestuur in 2002 started the debate whether the Wob should be subject to a more fundamental revision with more emphasis on proactive disclosure of government information. Since the Dutch government turned out to be hesitant in proposing a legislative proposal to transform or even replace the Wob, some Members of Parliament took the initiative and proposed a new Open Government Act in 2012. This legislative proposal met fierce resistance from administrative authorities fearing an enormous increase of administrative burden in the implementation of this legislation (De Graaf et al., 2019). After some amendments to key elements of the initial proposal, e.g. removing the obligation to have a register where all government documents are specified, the Wet open overheid (Woo) was finally adopted in October 2021 and entered into force as of May 2022, replacing the Wet openbaarheid van bestuur. Its main features will be discussed thematically below.
‘New’ open government? The Dutch Open Government Act
From disclosure at request to proactive disclosure
The most important change that the Woo has introduced, is a shift from ‘passive’ disclosure (at request) towards ‘active’ disclosure at the initiative of administrative authorities themselves. To that end, the Woo does not only reverse the order in which proactive disclosure and disclosure at request are discussed (Chapters 3 and Chapter 4 respectively), but also makes proactive disclosure less voluntary. Although proactive disclosure is still presented as an obligation administrative authorities should strive for but not necessarily achieve at any price (Article 3.1), the Woo specifies a number of categories of government documents that public authorities should publish at their own initiative. Among these documents are not only general legislation and policy rules, but also individual administrative decisions, expenditure decisions exceeding €250k, opinions of advisory committees, etcetera (Article 3.3). Because administrative authorities need time to prepare this shift from disclosure at request towards proactive disclosure, the obligation to publish these categories of documents proactively has not yet entered into force.
From open government to open semi-government?
Another shift that the initiators of the Woo proposed, was an extension of the personal scope of the Woo. This legislation should not only apply to traditional administrative authorities, but also to semi-public institutions. This proposed extension was inspired by the Council of Europe Convention on Access to Official Documents, also known as the Convention of Tromsø (Council of Europe, 2009), which explicitly allows for the opportunity to extend the definition of ‘public authorities’ to ‘natural or legal persons insofar as they perform public functions or operate with public funds, according to national law’ (Gerards et al., 2020). At the final stage of the legislative process, this proposed extension of the scope of the Woo was cancelled, as this extension would not be in line with the prevailing constitutional order.
Interestingly, the scope of the aforementioned PSI Directive, which has been succeeded by the EU Open Data Directive in 2019 (European Parliament and Council, 2019), has gradually been extended to so-called ‘public sector bodies’ and certain ‘public undertakings’. Thus, the European re-use regime applies to a broader set of documents than the Dutch access regime. However, since the Open Data Directive does not affect the access regimes of Member States (Broomfield, 2023), a restricted scope of the access regime, such as the Woo, limits the opportunities for re-use of documents of semi-governments.
From open to semi-open government
A third important shift is a more nuanced understanding of ‘openness’ of government documents. According to the traditional approach in the Wet openbaarheid van bestuur, the citizen does not have to state his interest in requesting the documents. This interest and the identity of the requester are irrelevant when deciding upon the request. Consequently, when a government document is open to someone, it is open to anyone.
Nonetheless, there may be good reasons for limiting access to government documents to certain interested parties. Where other legislation, such as the General Data Protection Regulation or a domestic Administrative Procedure Act, usually allows for such private access regimes, the Woo also establishes a limited access regime, allowing only certain persons access to government documents. In particular, the Woo allows for limited access to the requester as far as information concerning him or her is involved (Article 5.5). Limited access can also be granted to researchers or journalists when this information cannot be made public to everyone (Article 5.7).
This limited access regime resembles to some extent the limited re-use regime for certain categories of protected data held by public sector bodies under the EU Data Governance Act (European Parliament and Council, 2022). While agreements or other practices granting exclusive rights to a re-user are in principle prohibited, such an exclusive right to re-use data may be granted to the extent necessary for the provision of a service or the supply of a product in the general interest that would not otherwise be possible (Article 3). Thus, just like the access regime, the applicable re-use regime is not entirely open.
From open information to open data?
While the Woo applies to information laid down in documents, a document should be understood broadly as any written piece or other set of recorded ‘data’. This broad definition is in line with the EU Open Data Directive, considering a document as ‘any content whatever its medium (paper or electronic form or as a sound, visual or audiovisual recording)’. Thus, the opposition between ‘information’ and ‘data’ is not strict under open government legislation. Nonetheless, explicit reference to ‘open data’ is lacking in the Woo.
Interestingly, however, the Woo marks a shift towards the idea behind ‘open data’ when regulating the way in which information is provided. Government information should preferably be provided in a machine-readable open format in accordance with the requirements applicable to re-use of government information. Only where this is not possible, other ways of information provision can be considered, such as providing a paper copy of the document or a summary (Article 2.4). Consequently, even though the EU Open Data Directive does not apply to access to government documents, this new provision in the Woo ensures that proactive disclosure of government documents is directly fit for re-use purposes.
From decentralized to centralized open government
A final shift that the Woo brings about, is a shift from decentralized to centralized open government. Under the Wob, which applied to public authorities at any level of government, both central and decentral, each authority bore its own responsibility for complying with the Wob. Thus, when government documents were published online, this publication took place on the website of the authority concerned. Under the Woo, government documents that have to be published proactively, must be published on a centralized publication portal (open.overheid.nl). This portal is also available for the disclosure of other government documents (Article 3.3b). The underlying idea is that disclosure of government documents on a centralized portal will make them better available, findable, and searchable. Nonetheless, the development of such a centralized publication portal has so far been unsuccessful (Adviescollege ICT-toetsing, 2022). Thus, in the meanwhile, administrative authorities will need to continue publishing their documents in a decentralized way on their own websites.
Discussion: ‘New’ open government in the Netherlands?
Disclosure legislation in general and FOI legislation in particular is traditionally considered a re-presentative of ‘old open government’ with its rights-based approach, emphasizing the individual right of citizens to request for government information. One might therefore wonder whether this new piece of Dutch open government legislation, replacing the existing FOI legislation, should be considered a representative of ‘new open government’. This question will be discussed below following the four critical dimensions of open government identified by Moon (2020).
Policy
The Wet open overheid (Open Government Act) does not seek to replace the traditional freedom of information approach that was characteristic for its predecessor, the Wet openbaarheid van bestuur. Instead, it seeks to strengthen the regime of affirmative disclosure that already existed in the Wob but could hardly be enforced. Under the Wet open overheid, administrative authorities enjoy less discretion in determining whether they publish government documents at their own initiative. Thus, one of the presumed negative effects of open government on FOI legislation, i.e., increased discretion in determining which documents will be published, is mitigated here by a legal obligation to disclose certain categories of government documents. An unintended drawback of this prescriptive rule-based approach in the Woo on affirmative disclosure could be that administrative authorities do not feel inclined to disclose other government documents than the categories prescribed.
Information/data
New open government is said to focus on a different representation of information: where under FOI legislation information is traditionally segregated, new open government advocates structured, raw datasets that can be easily connected and reused. While the Wet openbaarheid van bestuur did already adopt a broad view on government documents, applicable to any set of recorded data (text, audio, video), this broad view has been continued under the Wet open overheid.
However, the Wet open overheid does create a shift in the way government documents have to be published. Even though the EU Open Data Directive on the re-use of public sector information does not affect the domestic regime on access to government documents, the Dutch Wet open overheid establishes voluntarily a connection between proactive disclosure of government documents and disclosure of government documents in a machine-readable open format. Consequently, even though reference to ‘open data’ is almost absent in the legislative process of the Wet open overheid, the obligation to make government documents public in a machine-readable open format amounts to an obligation to publish these documents as ‘open data’. At the same time, there remains some tension with the ‘old’ legal regime applicable to disclosure at request, as requested documents should still as a rule be provided in the way as requested.
What is more, the Wet open overheid obliges administrative authorities to publish certain government documents on a centralized government information portal. This is clearly a departure from traditional FOI legislation, which lacks a centralized “reading room” (Noveck, 2016), as it is strikingly decentralized not only on the requester side (an individual’s request determines whether a document is published or not), but also on the government side (Pozen, 2017). By publishing all disclosed government documents on a centralized portal in accordance with harmonized standards, the opportunities for connecting information and datasets will be enhanced.
Administrative value
New open government is said to broaden the scope from transparency as merely a means to promote democratic accountability to more interactive objectives of participation and collaboration. In this respect, it is interesting to note that the initiators of the Wet open overheid identified several reasons why public access to government documents is important: (i) for democratic accountability and the rule of law, (ii) for the individual citizen, (iii) for the administrative authority itself and (iv) for economic, cultural and academic development. These different reasons for public access reflect the potential of ‘open government’ to unite several proponents of government information disclosure.
In particular, the potential of public access to government documents to foster economic development was clearly rooted in the idea that disclosure of information can enable citizens and companies to create additional value with that information. During the legislative process, this motive of creating economic value with government information has lost some weight. This is due to the fact that during the legislative process, the re-use regime was removed from the Wet openbaarheid van bestuur. Consequently, any reference to re-use of government information was also removed from the legislative proposal except for the requirement that government documents should be published preferably in a machine-readable open format in accordance with the requirements on re-use of government information. Thus, although the value of transparency is still dominating the new Wet open overheid, it is not entirely neglecting the needs for participation and collaboration.
At the same time, where data from semi-public governments, such as energy network operators, are often identified as having high potential in fostering the development of new information products and services, for example to realize the energy transition, the Wet open overheid takes a traditional approach as to its personal scope: semi-public institutions are excluded from the scope of the Wet open overheid, as inclusion would not be in line with the prevailing constitutional order. This can be seen as a concrete example where the traditional administrative value of transparency to promote democratic accountability has prevailed over other values promoting collaboration in the use of (semi) government information.
Citizens
Finally, new open government is said to promote a different idea of citizenship: not just the informed citizen (with the help of media or not) should be the focal point of FOI legislation, but also the active user and producer of government information (Van Maanen, 2023). Admittedly, the Woo acknowledges the needs of active users of government information. Apart from the obligation to disclose information proactively in a machine-readable way, the Wet open overheid confers privileges to journalists and researchers affording them a right to limited access to government documents. Here, the interest of a specific subset of citizens in having access to government information prevails over the general interest of giving every citizen open access to that information.
It would go too far, however, to state that the Wet open overheid is tailored around this new type of active citizen. Instead, the ‘interested’ citizen is still the point of reference for the Wet open overheid. What is more, some provisions in the Wet open overheid even depart from the idea that government information should be provided in a machine-readable way. Instead, those provisions promote the idea of human-readable information disclosure, for example by developing a register on certain government documents that do not contain the integral text of those documents, but only its main elements (Article 3.3a).
Conclusion
The adoption and enactment of new legislation on disclosure of government information in the Netherlands could be considered a shift from old to new open government. To some extent, this Wet open overheid indeed reflects ideas underling ‘new’ open government. For example, it establishes a more detailed, rule-based regime on proactive disclosure, thereby making proactive disclosure less voluntary for administrative authorities. Furthermore, it incorporates some requirements on the publication format, which is crucial for re-use of government information.
At the same time, the Wet open overheid cannot be considered a complete departure from ‘old’ open government. Traditional mechanisms, such as the individual right to request for government documents, are preserved as an indispensable mechanism to secure access to government information. In that sense, the Wet open overheid can better be considered a legislative effort to combine the best of both worlds by providing for a legal framework within which both forms of open government, both old and new, can take shape. Whether this will actually happen, is not just a matter of drafting ‘better legislation’ but will also depend on the implementation and use of this legislation by administrative authorities and citizens (Kempeneer, Pirannejad & Wolswinkel, 2023).