Article
Perspectives
Case Law Analysis
Book Review

The Right to Access Public Information: A Legal Comparison between Sweden and the Netherlands

Toon als PDF
L. F. D. Honée - PhD candidate, Leiden Law School, the Netherlands, A. Drahmann - Associate Professor, Leiden Law School, the Netherlands*
Keywords: Access to Public Information, Netherlands, Sweden, Administrative Culture

Abstract

In both Sweden and the Netherlands, citizens have the legal right to access public information held by their respective governments, yet significant differences in administrative culture regarding openness remain. This article aims to examine how that right is implemented in law. It compares the scope of the right, who makes decisions to keep information confidential, the grounds for those decisions and how those decisions are made. Two key differences between the Netherlands and Sweden are particularly noteworthy, as they relate to fundamental theoretical debates about publicity and secrecy in democracies. These differences also shape the broader legal framework governing the right to access public information. The first difference concerns the principle of legality, and the second involves the protection of confidential deliberations. By comparing how the executive function is organised in the Netherlands and Sweden, this article demonstrates how, both in practice and principle, these two matters are closely tied to issues of responsibility and the division of public powers within the executive.

Introduction

In both Sweden and the Netherlands, citizens have the legal right to access public information held by their respective governments. At the same time, however, according to some observers, the administrative cultures of the two countries can be described as opposites. Hall, for one, notes that openness is one of the four features that together form the backbone of the Swedish administrative model.
The other features are the “dualism” (i.e. the relatively high degree of autonomy that Swedish state agencies enjoy in relation to the government), decentralisation and corporatism. P Hall, ‘The Swedish Administrative Model’, in Jon Pierre (ed), The Oxford Handbook of Swedish Politics (Oxford Academic 2015) 299-314.
 In Sweden, the principle of publicity and the fundamental right to access all official documents were already enshrined in the Constitution in 1766 and have thus acted as an underpinning for the Swedish administration model.
Tryckfrihetsförordningen (Freedom of the Press Act) (1949:105), ch 2 s 1.
 According to Andeweg, Irwin and Lauwerse, the Dutch administrative culture can be typified historically as one of confidentiality.
RB Andeweg, GA Irwin andT Louwerse, Governance and Politics of the Netherlands (5th ed, Houndmills, Basingstoke, Hampshire: Palgrave MacMillan 2020) 169-173.
 It can be characterised as one that seeks compromises (known in the Netherlands as the poldermodel), something that can benefit from the intimacy and confidentiality of closed doors.
D Stasavage, ‘Open-Door or Closed-Door? Transparency in Domestic and International Bargaining’ (2004) 58 International Organization 667.
 The Netherlands codified the right to access public information back in 1978, making it an early adopter of this right compared to most other western liberal democracies.
JM Ackerman and IE Sandoval-Ballesteros, ‘The Global Explosion of Freedom of Information Laws’ (2006) 58 Administrative Law Review 85.
 However, government reports and Dutch scholars have pointed out that the Dutch administration has not fully internalised this right to information in practice and rather sees it as a favour to its citizens.
J De Meij, ‘De WOB blijft een vergiet’ (1992) 10 Mediaforum 108; Evaluatiecommissie wet Openbaarheid, Openbaarheid tussen gunst en recht (The Hague April 1983) 33; Nota ‘Open de oester’, Parliamentary papers II (2004/05) 30214, nr. 1 1; SEO Economisch Onderzoek, invoeringstoets wet open overheid, knelpunten, best practices en neveneffecten, addendum Parliamentary Papers II (2023/2024) 33328 nr. AH ii.
Of the three remaining features of the Swedish administration that Hall identifies—dualism, corporatism and decentralisation—Sweden and the Netherlands share two: corporatism and decentralisation. Corporatism refers to the influence of stakeholders within administrative affairs, and both countries have strong traditions of involving organised interests in policymaking and administration.
JG Christensen and K Yesilkagit, ‘Delegation and specialization in regulatory administration: a comparative analysis of Denmark, Sweden and the Netherlands’ in P Lægreid and T Christensen, ‘Autonomy And Regulation: Coping With Agencies in the Modern State’ (Edward Elgar Publishing 2006); A Lijphart and ML Crepaz, ‘Corporatism and consensus Democracy in Eighteen Countries: Conceptual and Empirical Linkages’ (1991) 21 British Journal of Political Science 235.
 Regarding decentralisation, both countries have recognised autonomy for decentralised bodies and aim to organise power as much as possible at the level of government closest to citizens.
LM Raymakers, Leidende motieven bij decentralisatie: Discours, doelstelling en daad in het Huis van Thorbecke (doctoral thesis, Leiden University 2014); V Renko ea, ‘Pursuing decentralisation: regional cultural policies in Finland and Sweden’ (2021) 28 International Journal of Cultural Policy 342.
 That leaves the principle of dualism, which together with the principle of openness, sets the administrations of the Netherlands and Sweden apart from each other.
The principle of dualism refers to the separation of policymaking and administration. In a dualistic government, responsibility for the execution of these tasks is divided among different entities, each of which can be held accountable for its own tasks.
JG Christensen and K Yesilkagit, (n 7).
 The opposite of a dualistic organisation is a monocratic organisation. In a monocratic organisation of the executive, ministers can determine all matters in their portfolio.
Ibid.
 As a result, all responsibility and accountability is assigned to the ministerial position. In the existing literature, particularly by Swedish authors, access to information and the principle of dualism are discussed together.
PT Levin, ‘The Swedish Model of Public Administration: Separation of Powers – The Swedish Style’ (2009) 4 JOAAG 42; I Cameron, ‘Secrecy and Disclosure of Information in Sweden’ (2024) 2 European public law 117, 121-122; P Hall (n 1) 299-314.
 In the Dutch literature, a discussion on the right to information is often combined with a discussion on ministerial responsibility.
For example: A Drahmann, ‘De Wet open overheid is slechts een tussenstap; een preadvies over mogelijkheden tot verbetering van de openbaarheid van overheidsinformatie’ in AWGJ Buijze, CJ Wolswinkel, NN Bontje and EC Pietermaat, Transparantie en openbaarheid (VAR-reeks 167, Boom juridisch 2022) 132-136.
 There is also an intuitive link between the two concepts: the principle of openness and the right to access public information are both rooted in the need for public accountability. Without access to information, citizens are unable to hold their governments to account. The aim of this article is to gain a better understanding of how these two concepts interact.
Both Sweden and the Netherlands are parliamentary democracies and decentralised unitary states. They both recognise certain fundamental principles of administration, such as the principles of legality, impartiality and proportionality. This warrants a comparison. The aim of this article is to explore why two European democracies that both codify the right to access public information and share many other institutional similarities exhibit such differences in administrative culture regarding openness.
It is, of course, impossible to completely isolate the independent variable—dualism—and the dependent variable—openness—in this study. Despite the similarities, other institutional and cultural differences remain. One notable difference in this context is the role and functioning of the courts and the ombudsman. These institutions fall outside the scope of this article, as their primary function is to protect citizens’ rights, and they only indirectly serve as mechanisms for holding the government accountable. Their role differs from the unconditional right to access public information, which is specifically aimed at promoting public accountability. Also, the position of state-owned companies and other semi-public bodies fall outside of the scope of this article.
Moreover, it is important to recognise the relevant cultural differences between Sweden and the Netherlands. In particular, the right to access public information must be understood in relation to the protection of private information. According to Westin, one of the fundamental characteristics of democratic states is that governments are transparent, while citizens are guaranteed a private realm.
In contrast to totalitarian states which rely on secrecy for the regime and high surveillance and disclosure for citizens. AF Westin, Privacy and Freedom (New York, Atheneum 1967) 23-32.
 However, the exact boundary between the public and private spheres is shaped by culture. In the Netherlands, there is a greater emphasis on privacy, whereas Sweden places greater weight on publicity. For example, many Dutch citizens find it inconceivable that personal details such as names, addresses, phone numbers and house prices are publicly accessible, while in Sweden, this is generally accepted and considered uncontroversial. One possible explanation for this difference is that the Netherlands might be characterised as more liberal, while Sweden leans more towards a socialist model, leading to different trade-offs between classical individual rights and collective rights. Culture and law interact in complex ways, each influencing the other.
M Mautner, ‘Three Approaches to Law and Culture’ (2011) 96 Cornell Law Review first pg, 839-868.
 Therefore, it is difficult to pinpoint a clear causal link between cultural differences and legal approaches to openness and privacy. While these cultural and institutional distinctions are acknowledged, they are not the primary focus of this article.
This is not the first legal comparison to be made between different codifications of the right to access public information.
DC Dragos, P Kovac and AT Marseille (eds), The Laws of Transparency in Action (Palgrave Macmillan 2019); HJ Blanke and R Perlingeiro (eds), The Right of Access to Public Information: An International Comparative Legal Survey (Springer-Verlag 2018).
 This study contributes to existing research conducted by comparing two nations that both codified the right to access public information, but do not share the dualistic relationship between government and public administrations. To this end, it is necessary to compare the legislation of the two countries. First, Section 2 discusses some constitutional provisions regarding ministerial accountability and access to information. Sections 3 to 5 then compare the legislation on the access to public information based on three topics: the scope of the right to access public information (Section 3), secrecy clauses (Section 4) and publicity on request (Section 5). To properly understand the differences that emerge from this legal comparison, these topics will be examined within the context of how the executive function of the state is organised and the mechanisms of accountability that legitimise the use of public powers. Section 6 compares, from an overarching perspective, the two main differences identified in the comparisons of Sections 3 to 5. It also analyses how these two main differences relate to how the administration is organised. Section 7 provides conclusions.

Constitutional principles: openness and public accountability

To gain a good understanding of both countries, the constitutional arrangements of each first need to be outlined. This section analyses the institutional arrangements laid down in the constitutions of both countries that relate to the interaction between executive and legislative bodies and how the administration is organised. It will then discuss the nature and extent of the principle of publicity in Sweden and the Netherlands.

Institutional arrangements

Sweden

Sweden does not just have one constitutional document—it has four: the Regeringsformen (Instrument of Government), the Tryckfrihetsförordningen (Freedom of the Press Act), the Yttrandefrihetsgrundlagen (Fundamental Law on Freedom of Expression) and the Successionsordningen (Act of Succession). Although the roots of the Swedish parliament (Riksdag) go back to the 15th century, it was in 1809 that the forerunner of the current Instrument of Government was adopted and the separation of powers was established.
T Bull and I Cameron, ‘The Evolution and Gestalt of the Swedish Constitution’ in A von Bogdandy, P Huber and S Ragone (eds), The Max Planck Handbooks in European Public Law Volume II: Constitutional Foundations (OUP 2023) 603.
 Since then, the government (regering), which is formed by the ministers, has been accountable to the Riksdag. Any member of the Riksdag is entitled to ask questions to the ministers about the performance of their duties. Sweden does not have a ministerial government. Under the Instrument of Government, all government decisions are made by the entire government, even if a decision was drawn up by just one minister.
ibid.
The Swedish Instrument of Government guarantees the dualistic relationship between the government and public authorities. On the one hand, the independence of public authorities is guaranteed by law. The Instrument of Government grants them the right to interpret and apply the law in individual cases without interference from the regering or Riksdag.
PT Levin, (n 11) 42.
 On the other hand, public authorities must obey directives from the regering. The specific powers of the various public authorities are therefore precisely designated by law. Ministerial responsibility does not extend to the faults of public authorities. Instead, these public authorities are responsible (sometimes even criminally responsible) for their actions.
T Bull and I Cameron (n 16) 631.
 Most of the state public administration is performed by national administrative agencies.
Although many public tasks are also performed on local and regional level. T Bull and I Cameron (n 16) 631-634.
 As a result, Swedish ministries are quite small compared to most other countries, with only around 4,000 employees.
M Ribbing and J Reichel, ‘Codification of administrative law in Sweden’ in F Uhlmann (ed), Codification of administrative law: a comparative study on the sources of administrative law (Oxford: Bloomsbury Academic 2023) 252; I Cameron, (n 11) 119.
 It should be noted that the dualistic nature of the Swedish administration already existed in 1809 and was further entrenched to limit the power of the King, who still had political power when the first Instrument of Government was enacted. On an international level, the dualism of the Swedish system is rather unique.
With Finland possibly being the main exception, which can be explained by the fact that it was ruled by Sweden for 500 years. PT Levin (n 11) 42.

The Netherlands

While the modern Dutch Constitution (Grondwet) was enacted in 1848, the current arrangement of the parliamentary system also emerged from conventions.
LF Besselink, ‘The Evolution and Gestalt of the Dutch Constitution’ in A von Bogdandy, P Huber and S Ragone (eds), The Max Planck Handbooks in European Public Law Volume II: Constitutional Foundations (OUP 2023) 391-392.
 The Grondwet sets out ministerial responsibility, including responsibility for any action taken by the monarch to the parliament (Staten-Generaal) The legislative function in the Netherlands is attributed to the government (the regering) and the Staten-Generaal together. The regering (ministers and the King) also has designated independent legislative powers, though those powers are rarely utilised as their use is controversial. Instead, legislative power may be designated by law to the regering as a whole or to individual ministers.
Ministries hold a central position within the Dutch central administration, typically consisting of several directorates-general. The directorates-general have extensive tasks covering every aspect of government work: policy planning and analysis, policy support, political support and advice, as well as administrative casework and operational management.
JG Christensen and K Yesilkagit (n 7); A Lijphart and ML Crepaz (n 7) 12.
 Ministries in the Netherlands are, especially compared to Swedish ministries, rather large, with around 140.000 employees in total in 2023.
Ministerie van Binnenlandse Zaken en Koninkrijkrelaties, Jaarrapportage Bedrijfsvoering Rijk 2023 (Publisher 2024) 26.
Dutch ministers each govern their own ministry and are the direct supervisors of the public officials employed there.
Grondwet, s 44.
 On the one hand, this means that ministers are directly responsible for the actions of public officials and are accountable for their activities in relation to parliament. On the other hand, public officials themselves are not accountable to parliament for their own performance.
WJM Voermans, Onze Constitutie (Amsterdam: Prometheus 2023) 349.
 Ministers are only responsible for the tasks of which they have legislative powers. However, if those powers exist, ministerial responsibility applies regardless of the minister’s level of personal involvement or culpability.
ibid 350-351.
Looking at the central administration in the Netherlands, some public tasks are performed by independent public authorities or, in special cases, by independent public officials employed by ministries. Independent public authorities are organisations that implement legislation and, in theory, do not make political decisions.
ibid 352-353.
 Especially in the 1980s, a considerable number of independent public authorities were established.
Belinfante ea, Beginselen van het Nederlandse staatsrecht (19th ed, Deventer: Wolters Kluwer 2020) 189.
 Examples of independent officials are tax inspectors and the Prosecution Service employees.
WJM Voermans (n 27) 351.
 In both cases, ministerial responsibility does not extend to the individual decisions made by these independent authorities and officials. Nonetheless, ministers do have legislative powers over independent public authorities. They can implement binding policy for the independent authorities, appoint board members and otherwise influence their performance. Since ministers are authorised and therefore responsible for these tasks, they are also accountable to parliament for them.

Principle of publicity

Sweden

The Tryckfrihetsförordningen (Freedom of the Press Act) guarantees the Offentlighetsprincipen (principle of openness).
CF Bergström and M Ruotsi, Grundlag i gungning? En ESO-rapport om EU och den svenska offentlighets-principen (Regeringskansliet, Finansdepartementet, Rapport till Expertgruppen för studier i offentlig ekonomi 2018); S Lamble, ‘Freedom of information, a Finnish clergyman’s gift to democracy’ (2002) 97 Freedom of Information Review first page, 2–8; and A Bohlin, Offentlighetsprincipen (Norstedts Juridik AB i Stockholm 2015).
 This principle is regarded as a foundational principle because it has existed since the inception of the modern Swedish state in a period now referred to as ‘the age of freedom’. It was first adopted in 1766, and although it was shortly curtailed from 1772 as a result of the political reform of Gustav III, it was reinstated in 1809.
B Wenngren, HG Axberger, J Hirschfeldt and K Örtenhed (eds), Press Freedom 250 Years: Freedom of the Press and Public Access to Official Documents in Sweden and Finland – A Living Heritage from 1766 (Stockholm, Sveriges riksdag 2017) 5.
 Notably, the reinstatement of this principle coincides with the adoption of the Instrument of Government.
The principle holds that citizens have the right to an unimpeded view of government activities.
P Jonason, ‘The Swedish Legal Framework on the Right’ in HJ Blanke and R Perlingeiro (eds), The Right of Access to Public Information: An International Comparative Legal Survey (Springer-Verlag 2018) 236.
 The most well-known legal right under this principle is the right to access all official documents, but the principle also extends to access the meetings of decision-making assemblies, the right to access court sessions and the right to freedom of speech for public officials.
ibid 236-237.
 Furthermore, the principle protects whistleblowers and functions as a barrier for managers in government to research the identity of public officials that leak information to the press.
P Hall (n 1) 305.
 The right to access official documents is contained in the Tryckfrihetsförordningen. The right to access public information has always been connected to the right to access information, which is an integral part of freedom of the press and the right to freedom of speech.
B Wenngren, HG Axberger, J Hirschfeldt and K Örtenhed (n 31) 475.

The Netherlands

The Dutch Constitution contains four provisions that can together be said to codify the principle of publicity. First, the meetings of the Staten-Generaal are held publicly.
Grondwet, s 66.
 Second, the Staten-Generaal has the right to request information from the ministers—which they are then obliged to provide, unless this is not in the interests of the State.
ibid, s 68.
 Third, both court sessions and judgments are public.
ibid, s 121.
 Fourth, in performing its various duties, the government must ensure openness in accordance with the rules laid down by law.
ibid, s 110.
 This latter obligation was first included in the Grondwet in 1983. The rule referred to in this provision is the (per 2022) Wet open overheid (Dutch Open Government Act, Woo), which is the successor of the 1980 Wet openbaarheid van bestuur (Dutch Government Information (Public Access) Act). Notably, this law preceded the constitutional obligation to carry out public duties in public.
Compared to Sweden, the public nature of the Dutch administration is a relatively recent addition to the Dutch Constitution. The rationale and grounds for the right to access public information were found in the emancipation and democratisation of society and the decreasing control of the Staten-Generaal on the expanding public administration.
Staatscommissie heoriëntatie overheidsvoorlichting (Commissie Biesheuvel), Openheid Openbaarheid, ’s-(Gravenhage: Staatsuitgeverij 1970) 3-11.
 The recognition of a general right to access public information was therefore intended to complement parliamentary oversight, since ministerial responsibility as the basis of accountability was increasingly becoming a fictional concept.
ibid, 14-24.
 Notably, the adoption of the principle of publicity for public authorities occurred in a period in which independent public authorities were established en masse in the Netherlands.
In 1974 Scheltema first gave academic attention to the many independent public authorities. M Scheltema, Zelfstandige bestuursorganen, (Groningen, H.D. Tjeenk Willink 1974). During that time, the first access to information legislation was pending in the Staten-Generaal.

Scope of the right to access public information

From this paragraph onwards, the legislation of the Netherlands and Sweden will be compared, starting with the scope of the right to access public information.

The Netherlands

In the Netherlands, the right to access public information is codified in Section 1.1 of the Woo. ‘Public information’ means any information recorded in documents that are held by or can be requested by public authorities.
Woo, s 2.1.
 This right is therefore limited to information that is already recorded; citizens do not have a right to request the creation of a new document.
ABRvS [2013] ECLI:NL:RVS:2013:CA2102.
 ‘Documents’ are defined as any carrier of data that is received or drawn up by a public authority, which in nature relates to the public duty of that public authority.
E-mails or notes which do not relate to any public task, do not even qualify as documents according to the Woo, let alone public information.
 The term document is a technological neutral concept and interpretation by the courts have extended this concept in modern times to text messages and WhatsApp-conversations.
ABRvS [2019], ECLI:NL:RVS:2019:548.
 All preparatory work is considered to be drawn up, so drafts and memoranda fall under this definition. Furthermore, data that can easily be extracted from a database with certain prompts also fall under this definition.
ABRvS [2013] ECLI:NL:RVS:2013:CA2102.
According to the definition of public information, information in documents is public if those documents in their nature refer to the ‘public duty’ of the public authority. With this definition, the legislator wanted to adopt a broad definition of public information in line with pre-existing case law. Under the previous law, documents were public if they referred to ‘public affairs’, which was then interpreted by the courts as referring to affairs of the administration in all aspects, both internal and external.
ABRvS [2016] ECLI:NL:RVS:2016:2375.
 An example of documents that fall outside the definition is private information of public officials that is stored at the public authority.

Sweden

According to the Tryckfrihetsförordningen everyone has the right to access official documents in order to promote a free exchange of ideas.
Tryckfrihetsförordningen, ch 2 s 1.
 It is notable that the Constitution not only grants this right, but also contains substantive rules about what ‘official documents’ entail.
A ‘document’ is defined as any recording, and is a technologically neutral concept.
Administrative Supreme Court (Previously: Regeringsrättens; Currently: Hogsta forvaltningsdomstolen), 1999 ref. 18, cases n. 3148-1998 and 5556-1998 (Judgment of 19 April 1999).
 A document is an ‘official’ document if it is held by a public authority, and it has been received or drawn up.
Tryckfrihetsförordningen, ch 2 s 4.
 A document is deemed to be drawn up if one of three conditions is fulfilled. First, if the document has been dispatched outside the organisation.
ibid, ch 2 s 10
 As a result, all incoming and outgoing documents are official documents. Second, a document is drawn up if the matter is finalised by the public authority.
ibid.
 Third, a document can become an official document if the document itself has received its final form.
Tryckfrihetsförordningen ch 2 s 10; Hogsta forvaltningsdomstolen 08-06-2004, Målnummer 8324-03.
 There are some exceptions specified in the Tryckfrihetsförordningen such as diaries, which are still official documents although they are never truly completed. Furthermore, some documents only acquire official status later, such as court decisions that only become official documents when they are dispatched.
Official documents can be distinguished from internal information, which falls outside the scope of this right to access. Internal working material is not finished and is therefore usually excluded from the right to access.
I Cameron (n 11) 121-122; P Jonason (n 34) 244; PT Levin (n 11) 40.
 The Tryckfrihetsförordningen makes special mention of certain preparatory documents such as memoranda and drafts.
Tryckfrihetsförordningen ch 2 s 12.
 Memoranda that serve to present information and that do not add any new factual information are not considered official documents, even though they may be drawn up. Drafts are not public documents unless archival law prescribes their archiving.
Lastly, a special category, which serves the principle of publicity in modern times, is formed by ‘potential documents’. These are documents that do not exist in their current form but can be created using technical aids such as computers. The right to access official documents does not in principle oblige public authorities to produce new documents. However, with technological advancements, information is stored in ways in which no official document is created, for example by adding data in a database. The Tryckfrihetsförordningen stipulates that if the public authority can extract data ‘using routine measures’, citizens have a right to it as if they were finished documents.
P Jonason (n 34) 240; Tryckfrihetsförordningen ch 2 s 3.

Comparison

As can be seen clearly from above, there is a difference between the scope of the Tryckfrihetsförordningen and the Woo. The Dutch Woo has a broader scope. Although the Woo specifies that documents only become public if they are drawn up or received just as in the Tryckfrihetsförordningen, this is not interpreted as meaning that only finished documents become public documents. It is therefore not uncommon in the Netherlands that all draft versions of particular documents are requested under the Woo.
Rechtbank Midden-Nederland [2024] ECLI:NL:RBMNE:2024:3285.
 Furthermore, in the Netherlands, internal correspondence between public officials always qualifies as public information so long as it refers to a public task, while in Sweden, this is only the case if this correspondence contains new factual information or must be archived for other reasons. However, this does not yet lead to the conclusion that more information is actually made public in the Netherlands, since secrecy clauses are also a relevant factor. These clauses are therefore compared in the next Section.

Secrecy clauses

In both countries the right to access public information is laid down in law, and as a result, exceptions to that right must also be laid down in law. In this section, the system for exceptions of both countries will be analysed.

The Netherlands

In the Netherlands, a distinction must be made between exception grounds and the exemption regarding the protection of personal policy views of public officials in the internal deliberations within public authorities.
The exception grounds are laid down in section 5.1 Woo and the protection of personal policy views is regulated separately in section 5.2 Woo.
 The exception grounds apply to all public information, whereas the exemption regarding personal policy views only applies to information produced within internal deliberations in preparation of policy decisions.
First, the exception grounds are discussed. There are two types of exception clauses in the Woo: absolute exceptions and relative exceptions.
Woo, s 5.1(1) and (2).
 The difference is that if an absolute exception clause applies to information, that information is not public. If a relative exception clause applies, public authorities must weigh the interest of publicity against the interest of secrecy that the relative exception clause is intended to protect.
The Woo contains five absolute exception clauses. Absolutely excepted is information that may cause danger to the safety of the state, information that endangers the unity of the Crown, information about business secrets which are shared in a confidential manner with the State, sensitive personal information,
A connection is made with sensitive personal information as defined in the General Data Protection Regulation.
 and national identification numbers. However, in addition to the Woo, there are also sectoral laws that contain more absolute exceptions. These sectoral laws are mentioned in the appendix of the Woo. The reason for this is that due to the general nature of the Woo, specific sectoral law may deviate from the Woo provided that the sectoral law is intended to exhaustively regulate the publicity of information.
It is not required that the legislator has clearly indicated the exhaustive nature of the sectoral law. The court can also deduce the intention of the sectoral legislation from the system of the law. ABRvS [1999] ECLI:NL:RVS:1999:AE8211.
 Examples of exhaustive clauses in sectoral law are rules regarding the secrecy of tax-information, police records and medical records.
Algemene wet inzake rijksbelastingen (General law regarding State taxes), s 67; Wet politiegegevens (Police records act), s 3; and Burgerlijk wetboek (Civil code), Book 7 s 457.
The Woo contains nine relative exception clauses. These clauses do not contain qualifications of types of information but refer to interests which can be taken into account when deciding whether information is public. These interests are in summary as follows:
  • International relations;
  • Economic and financial interests of the government;
  • The investigation and prosecution of criminal offences;
  • Inspection, control and supervision by public authorities;
  • Privacy;
  • Protection of business secrets (other than those that are absolutely excepted);
  • Protection of the environment;
  • Safety and protection of property of individuals and legal persons;
  • The proper functioning of the government.
For each document—and even parts of it—public authorities must decide whether these interests outweigh the interest of publicity. This implies a weighing of interests between the interest of publicity and the interest of secrecy.
When making this assessment, public authorities must take the principle of publicity into account.
ABRvS [2023] ECLI:NL:RVS:2023:488; ABRvS [2015] ECLI:NL:RVS:2015:2263.
 The law stipulates that publicity of public information is presumed to be in the interest of democracy.
Woo, s 2.5.
 The interest of publicity therefore has a fixed weight that is not influenced by the requestor or the content of the information.
ABRvS [2015] ECLI:NL:RVS:2015:2217; ABRvS [2020] ECLI:NL:RVS:2020:2417.
 Therefore public authorities do not estimate the interest of publicity in the specific case. Instead, public authorities must decide whether the interest of secrecy outweighs this general principle in a specific case. Only those parts of the documents to which the exception grounds apply can be refused.
ABRvS [2021] ECLI:NL:RVS:2021:2064.
A special comment is made with regard to the exception ground regarding the proper functioning of the State. This is a new ground that was introduced in 2022. Public authorities have attempted to use this exception ground to generally except all drafts of documents, but courts have so far overturned these decisions since the Woo requires that drafts are assessed on a paragraph-by-paragraph basis as to why this interest justifies an exception; a general refusal to disclose certain types of documents (drafts) is not permitted under the Dutch legal system.
Rechtbank Midden-Nederland [2024] ECLI:NL:RBMNE:2024:3285.
The Netherlands also has one exception ground that does not refer to specific information or a specific interest. This non-specific clause can be applied if publicity would disproportionately harm an interest that is not covered by the other exceptions.
Woo, s 5.1(5).
The system of relative exception clauses gives public authorities considerable discretionary power in weighing the importance of publicity against the importance of secrecy. This discretionary power is limited by the supervision of the administrative courts. As part of the general administrative law doctrine that leaves a margin for public authorities to decide when the law allows for discretion, courts also apply this margin in cases about the right to access public information.
ABRvS [2008] ECLI:NL:RVS:2008:BG5356.
 Courts will generally distinguish between two questions. First, has it been correctly decided that a secrecy interest applies to the document? Second, has the public authority given a reasonable motivation as to why the secrecy interest outweighs the interest of publicity? Courts are usually stricter when it comes to the first question. If the interest of secrecy does indeed exist, courts are usually more reserved regarding the weighing of interests that public authorities have made.
EJ Daalder, Handboek openbaarheid van bestuur (Den Haag: Boom Juridisch 2023) 317.
In addition to these exception grounds, the Woo contains a specific exemption that protects the personal policy views of public officials expressed in the internal deliberations of public authorities.
Woo, s 5.2.
 These are neither absolutely nor relatively excepted. Instead, public authorities must decide whether they believe it is in the interest of democratic and good governance to publish these policy views of public officials in an anonymised manner or to keep them secret from the public. Public authorities must provide reasons for their decision, but generally this motivation is not intensively assessed by the courts.
Courts have only criticised decisions to exempt personal policy views if the information was very old while it was also information of great importance for society.
 The rationale for this exemption is that the intimacy and confidentiality of closed doors allows officials to be franker and have freer discussion, resulting in better decision-making.
Parliamentary papers II (2011–2012) 33 328, nr. 3, 44.
The application of this exception depends on two criteria. First, public authorities must decide whether the policy view is expressed within internal deliberations. Internal deliberations are not limited to deliberations between public officials; outsiders, such as lawyers or consultants, may participate in internal deliberations if they have no personal interest in the matter discussed.
ABRvS [2017] ECLI:NL:RVS:2017:3497.
 Second, the information must contain a personal view of the public official. Only personal views and not factual information are exempt. In Section 5.2 Woo this distinction is further clarified by providing examples of policy views (advice, views, arguments) and more objective information (facts, forecasts, policy alternatives). If factual information is intertwined with the policy view and cannot be separated, both the factual information and the policy view can be redacted.
ABRvS [2018] ECLI:NL:RVS:2018:314.

Sweden

Section 2 Chapter 2 of the Tryckfrihetsförordningen states that the right to access official documents may only be limited if it is required with regard to:
  • The security of the kingdom or its international relationships.
  • The kingdom’s central financial policy, monetary policy, or currency policy.
  • The activities of authorities for inspection, control, or other supervision.
  • The interest in preventing or deterring crime.
  • The public’s financial interest.
  • The protection of individuals’ personal or financial conditions.
  • The interest in preserving animal or plant species.
Furthermore, any limitation of the right to access official documents must be carefully specified in a provision of a special law or, when it is more appropriate, in another law to which the special law refers to. This special law mentioned by the Tryckfrihetsförordningen is the Offentlighets- och sekretesslag (OS). This means that only the Riksdag is competent to create secrecy provisions, although it can delegate the power to make supplementary rules to the Regering.
P Jonason (n 34) 246.
In the OS, the secrecy clauses are formulated with great precision, and it is a lengthy law of about two hundred pages long. For example, the OS contains chapters with provisions on the secrecy of tax-information, police records and medical records.
OS, Chapter 27 on the Confidentiality to protect individuals in activities related to tax, customs OS; also see Chapter 25 Secrecy to protect individuals in activities relating to health care; and Chapter 35 Secrecy to protect individuals in activities aimed at preventing or deterring crime.
 Another example is that information concerning the planning of an inspection that a public authority must carry out can be withheld if this would thwart the purpose of such an inspection.
OS, ch 14 s 1.
 The scope of the secrecy clauses is always limited. The precision of the secrecy clauses limits the discretion of individual public officials, which is Tryckfrihetsförordning’s goal for this requirement.
P Jonason (n 34) 246.
With regard to secrecy clauses, a distinction can be made between clauses that provide for absolute secrecy and clauses that contain a requirement of harm.
ibid 247.
 In cases where secrecy must be the main rule, the requirement of harm is reversed.
An example is Section 3 Chapter 32 OS which states that information regarding the individual’s personal circumstances that has been obtained through camera surveillance in the interest of monitoring, control or supervision is secret, unless it is clear that the information can be disclosed without the individual or someone close to him suffering harm.
 Although public officials have none or a very limited margin of discretion regarding the scope of the secrecy clauses, they do have some discretion in assessing whether harm will occur to a particular interest if the information becomes public.
P Jonason (n 34) 247.
 Many provisions also contain a maximum term for secrecy.
For example, according to Section 1 Chapter 15 OS, information regarding Sweden’s relationship with another state or international organisation, secrecy applies for a maximum of forty years.
 Although there has been parliamentary debate on the question whether the requirement of harm should be replaced with a weighing of interests, this proposal was not implemented because this would lead to increased discretion for public officials and parliamentarians feared that as a result more information would be marked as secret.
Prop 2007/08: 09: 150 Regeringens proposition Part 10 133-135.
Within clauses that include a harm requirement, that harm can be either skada or men.
Prop. 1979/80:2 Regeringens proposition 79.
 Skada refers to financial harm and men refers to privacy infringements. Whereas skada can be more objectively assessed, there is a higher level of subjectivity involved in men. The premise is that purely subjective harm is not sufficient to fulfil the harm requirement; the decision must refer to more objective criteria, such as the prevailing views in society on privacy infringements. When making an estimate of the risk of harm, public officials usually do not consider who the requester is.
ibid 80-82.
 An advantage of this is that the requester can remain anonymous if he wishes. However, when applying secrecy clauses with a reversed harm requirement, information about the requester may be necessary to assess whether no harm will occur.
ibid 80.

Comparison

There is a lot of overlap in the interests of secrecy that the open access laws of the Netherlands and Sweden protect. The most striking difference between the two legal frameworks is that the relative exception clauses in the Woo can be applied directly by public officials, while the interests stated in the Tryckfrihetsförordningen are further specified in the OS. As a result, the clauses in the OS are much more precise, leaving less discretionary room for public officials, which is the intention of the Tryckfrihetsförordningen. The exception clauses laid down in the OS qualify the category of information which limits the scope of the secrecy clause. In the Netherlands, there is considerable discretionary power because the relative exception grounds are formulated as interests that the public authority must consider before granting access to the information. As an example: the Swedish OS allows public authorities to withhold information about the planning of an inspection, if the purpose of which would be thwarted by granting access to the information. In the Netherlands, public authorities can withhold information if the interest of inspection, control or supervision outweighs the interest of publicity. This Dutch clause allows for more flexibility for protecting the interest of secrecy, while the Swedish clause clearly states the type of information and therefore can only be applied in those instances. As a result, Dutch public authorities have a rather large room for discretion compared to Swedish public authorities. In addition, the Dutch Woo also allows public authorities to withhold information in the interest of the proper functioning of the government. This clause has no Swedish counterpart and clearly allows for its application in a wide range of cases.
At first glance, the relative exception clauses in the Woo seem to require a different weighing of interests than the exception grounds in the OS, because the OS contains a harm requirement. On closer inspection, this difference is more nuanced than it appears. When Dutch public officials must weigh up interests, the interest of publicity does not depend on the particular circumstances of the request, such as the person of the requester or the content of the information. Instead, public officials must justify that the harm done to an interest of secrecy outweighs the principle that information should be public. As a result, the Woo also has a harm requirement disguised as a weighing of interests. That only leaves two (minor) differences. First, the Swedish system uses both negative and positive harm requirements. Therefore, some exception clauses contain stronger protections of the interest of secrecy than others. Such fine-grained legislation does not exist in the Netherlands. Second, Dutch sectoral laws contain many absolute exception grounds, while in Sweden all clauses are laid down in the OS and absolute clauses are quite rare.
Another important difference is the exemption regarding personal policy views made in the internal deliberations of the public authority which is codified in the Woo, but not in the OS. However, internal working material does normally fall outside the scope of the right to access according to the Tryckfrihetsförordningen as it does not qualify as an official document.
In this section, only the system for exceptions to the right to access public information has been compared and not the specific interpretation of these clauses. Outlining and comparing hundreds of clauses would exceed the scope of this article. However, the special interaction between the principle of publicity and the right to privacy should be pointed out. In the Woo, public authorities must weigh the interest of publicity against the interest of protecting privacy. Names, e-mail addresses and telephone numbers are normally withheld from public access on this basis.
Ministry of the Interior and Kingdom Relations (Ministerie van BZK), Rijksbrede instructie voor het behandelen van Woo-verzoeken, 2022, 26-27. Online access at rijksoverheid.nl.
 In addition, courts have ruled that the names of public officials must be redacted, unless they hold a position in public or the requester can state a reason why publicity of their names is required.
ABRvS [2018] ECLI:NL:RVS:2018:321.
 Swedish law normally allows for access to non-sensitive personal data. Names and contact details are in principle public because they do not meet the harm requirement as laid down in the OS. More personal information, for example on pensions, student support, school grades and taxable income is also public.
O Jørgensen, Access to Information in the Nordic Countries (Nordicom 2014) 22-23.
The difference in the wording of the secrecy clauses, which gives Dutch public officials much more discretionary power than in Sweden, align with differences in administrative and constitutional culture as described in Sections 1 and 2. This will be further elaborated in Section 6. First, it will be examined to what extent the procedure for making information public differs between the two countries.

Access to public information on request

Both the Dutch and Swedish legal system grant citizens the right to request for information. It must be noted that the Dutch Woo contains far-reaching obligations regarding active transparency by public authorities.
A Drahmann, ‘De Wet open overheid is slechts een tussenstap; een preadvies over mogelijkheden tot verbetering van de openbaarheid van overheidsinformatie’ in AWGJ Buijze, CJ Wolswinkel, NN Bontje and EC Pietermaat, Transparantie en openbaarheid (VAR-reeks 167, Boom juridisch 2022) 104-112.
 Because these active transparency obligations have not yet come into effect, they will not be discussed further in this article. In Sweden there are no active transparency obligations linked to the right of access to official documents. This paragraph focusses on information on request.

The Netherlands

Only a valid Woo-request will result in a corresponding obligation for public authorities to grant access to public information. For a valid request, the requester must indicate either the public affair about which he wishes to receive information or the documents he wishes to receive.
Woo, s 4.1.
 The request must also be stated with sufficient precision to allow the public authority to identify the information that relates to the request.
ABRvS [2023] ECLI:NL:RVS:2023:1451.
 If the request is too general or unspecific, public authorities must assist the requesters in order to specify their request.
Woo s 4.1(5).
 The size of the request is, in itself, not a reason that can lead to an invalid request.
ABRvS [2022] ECLI:NL:RVS:2022:1984.
 However, public authorities are not obliged to respond to abusive requests.
Woo s 4.6.
 The condition for invoking abuse of the right to request information is that the requester uses this right for a purpose other than gaining access to information; for example, a personal vendetta or to disrupt the functioning of public authorities.
ABRvS [2022] ECLI:NL:RVS:2022:1984.
If a valid request is made, the public authority has a duty to respond to the request. In cases in which the information is granted without restrictions, this response does not have to be a formal decision by the public authority.
Section 4.3 Woo. A decision to grant access can be made orally, which means it does not qualify as a formal decision according to the General Administrative Act, since those are in writing.
 However, if there is any restriction to access information, such as the application of a secrecy clause or limits to further disseminate the information, a formal decision by the public authority must be made. The previous Section has shown that the Netherlands has many exception grounds, meaning that a formal decision is almost always made.
The statutory term for responding to information requests is as soon as possible, although not longer than four weeks.
ibid s 4.4(1).
 This term can be extended by two weeks if the size or the complexity of the request justifies such an extension.
ibid s 4.4(2).
 In many cases, this term is not met by public authorities. According to the government, this is because many information requests are so extensive or complex that six weeks is too short a term.
Letter of the Minister of Internal Affairs to Parliament dated 17 July 2023, Parliamentary Papers II 32802, 73.
 In some cases, the government does make a valid point here. After the Covid-19 crisis, 241 information requests were filed at the ministry of Health relating to an estimated 1.8 million documents.
ABRvS [2022] ECLI:NL:RVS:2021:2348 and A Drahmann & L F D Honée, Algemeen Bestuursrecht 2022/12 and 2022/13 case note to ECLI:NL:RVS:2021:2346.
 That many documents cannot be assessed in four weeks. Three reasons can be given why such large requests are possible. First, the scope of the Woo extends to all preparatory work—such as drafts and internal e-mails. This can result in copious amounts of information for complex decision-making, such as the choice concerning which population groups must be vaccinated first. Second, citizens only need to mention the public affair about which they want to receive information rather than which documents they want to receive. Third, in practice, information management in governments is inadequate despite a duty of care in the Woo to keep information orderly.
Parliamentary interrogation committee, Ongekend onrecht, 2020; Parliamentary Papers 35 510 1.
 It is therefore possible that one request could lead to an obligation for public authorities to identify thousands of documents and apply exceptions. The legislator responded to this by including a clause in the Woo which stipulates that if the size or complexity of the request is such that a decision within the statutory term is not possible, public authorities will consult with the requester to make agreements about the term and the prioritisation of the documents to be assessed.
Woo, s 4.2.
In the Netherlands, two compliance mechanisms exist. One concerns the general arrangement for legal protection against decisions as laid down in the Algemene wet bestuursrecht (General Administrative Law Act). In this procedure, an objection must first be filed with the same public authority that made the decision under the Woo.
Awb, s 7:1.
 This objection can be filed by the requester or by another interested party.
ibid ss 1:2, 7:1, and 8:1.
 After this stage, it is possible to appeal to a general court and to appeal against a court decision to the Administrative Law Division of the Council of State. The other mechanism is a complaint procedure that can only be initiated by requesters who professionally request information, such as journalists and researchers.
Woo s 7.3.
 The complaint is then filed to the Adviescollege openbaarheid en informatiehuishouding (Public Access and Information Management Advisory Board), which mediates between the parties and can also give non-binding advice.
ibid s 7.1.

Sweden

In Sweden everyone has the right to access official documents (as opposed to information). The requester must be able to provide the necessary information so that the public authority can identify the requested documents.
P Jonason (n 34) 257-258.
 If a requester can name specific documents, the work required to apply exception grounds can in principle never be a reason to reject the request.
Kammarrätten Göteborg, Avgörandedatum: 14 March 2017, Mål nr 183-17. in this case the estimate was that one public official would have to work for one year to respond to the request.
 To make a valid request, requesters do not need to describe the documents precisely. A request is considered sufficiently precise if the requester can describe the contents of a document or a limited group of documents so that authorities can identify the document without much effort.
Hovrätten för Västra Sverige 26 March 2007 Mål nr. Ö1733-07.
 Authorities, with the help of registers or diaries, must assist citizens when they cannot provide sufficiently precise information. However, authorities are not obliged to conduct extensive archival research if citizens can only provide general descriptions of documents.
Regeringsrättens 1979 Ab 6; Regeringsrättens 1991 ref. 50.
 It must be assessed on a case-by-case basis whether the request is reasonable.
The decision to grant access is primarily the responsibility of the individual public official involved with the documents that are requested; for example, the registrar or the public official on the case.
P Jonason (n 34) 257.
 A formal decision by the public authority is only required in two instances. First, if the official does not fully disclose the information, or second, at the individual’s request.
OS ch 6 s 3.
 A formal decision is necessary for any appeals to the decision, a fact of which the requester must be notified.
ibid ch 6 s 3.
In Sweden, document registers play a significant role because public authorities are obliged to register official documents in order to fulfil their obligation to expeditiously grant access to official documents and to provide citizens with a good opportunity to search for documents.
OS ch 4 s 1.
 Some official documents are exempt from the obligation to register. If there is no secrecy obligation, these documents may be kept in such a way that it can be ascertained whether they have been received or drawn up.
OS ch 5 s 1.
 If the documents are of minor importance for the government’s operations, they do not need to be registered or kept in order.
OS ch 5 s 1.
 Official documents that require registration must be registered as soon as they are received or drawn up. The register must include the date, the document number, information about the sender or recipient of the information if relevant, and briefly what the document is about.
ibid. ch 5 s 2.
 If applicable, the secrecy mark is added during registration of the document.
ibid. ch 5 s 5.
In Sweden, the term for compliance with requests for official documents is immediately or as soon as possible if it concerns a request for access on site, and promptly if copies are requested.
Tryckfrihetsförordningen, ch 2 s 15-16; The question of what is ‘as soon as possible’ is regularly put to the Ombudsman, see JO dnr. 3725-2015; JO dnr. 4773-2003; JO dnr 3843-2009, JO dnr 2859-2011; JO dnr 1039-2017; JO dnr 5158-2015; JO dnr 639-2012.
 No specific term is given. Public authorities are required to prioritise information requests over their other general duties.
JO dnr. 3725-2015; JO dnr 5158-2015.
 If there are no special circumstances, such as a voluminous or legally complex request, the request must be responded to on the same or next day.
Public authorities are expected to organise their tasks in such a way that sickness or vacation of officials will not interfere with the procedures to grant access to official documents, see JO dnr 5158-2015. Usually access is given immediately. O Jørgensen (n 93) 31.
 However, there are cases where the circumstances of the case allow for a longer term.
Sweden has two compliance mechanisms: filing an appeal to a court and filing a complaint with the Parliamentary ombudsman.
P Jonason (n 34) 259-261.
 Both procedures are only accessible to the requester. Other involved third parties (for instance because the information relates to them) are not allowed access to the courts.
ibid 260.
 In order to file an appeal, the requester must first receive a formal written decision made by the public authority. If the initial decision was not a formal decision by the authority, a review by the public authority is first necessary to access the courts.
OS ch 6 s 3.
 After the formal decision, an appeal can be filed with an administrative court of appeal and a further appeal to the Hogsta forvaltningsdomstolen (Supreme Administrative Court).
OS ch 6 ss 9-10.
 In addition to the appeal procedure, citizens can file a complaint with the Parliamentary Ombudsman.
Lag (2023:499) med instruktion för Riksdagens ombudsmän (JO) (Law containing instructions for the Parliamentary Ombudsman) ss 11-12. Their task is to supervise the application of the law by public authorities, especially regarding issues of objectivity, neutrality and the protection of fundamental rights.
 The Parliamentary Ombudsman mainly deals with issues of a procedural nature, such as delays by public authorities.
P Jonason (n 34) 261.
 The Parliamentary Ombudsman provides authoritative advise that is non-binding.
ibid.

Comparison

Comparing the procedures to request information, there are five differences which are worth highlighting.
First, in Sweden, usually the public officials dealing with the information decide on information requests. Only upon subsequent request of a citizen will a public authority take a written formal decision. In the Netherlands, almost every response to an information request is a formal decision by the public authority.
Second, in the Netherlands, citizens usually request for information about a specific public affair. In Sweden, citizens usually request for official documents. However, this difference is mitigated. In Sweden, citizens are not required to describe documents precisely, for example, by naming a document number. Instead, citizens can rely on a more general description of the content of a document or a limited group of documents. Swedish authorities are not obliged to conduct extensive archival research if a request is not specific. In the Netherlands, for there to be a valid request, it is sufficient that a request mentions a public affair. The amount of work required to find all relevant information is then not relevant for triggering the obligation to grant access to all requested information. The Woo does contain a ground for refusing abusive requests, while this is not the case in the Swedish system.
The third difference is that Swedish law requires public authorities to register or otherwise organise official documents, unless it is clear that the documents are not important to the functioning of the authority. Dutch law does provide for a general duty of care to keep information in order but does not require certain important documents to be registered.
Fourthly, Sweden and the Netherlands both require that information requests are handled as soon as possible. The Woo provides for a specific term of four weeks, while the Tryckfrihetsförordningen and the OS do not. In Sweden, public authorities are required to prioritise the handling of requests for official documents over other tasks, while no such obligation exists in Dutch law.
Finally, in terms of compliance mechanisms, an important difference lies in the fact that Sweden has a Parliamentary Ombudsman, to whom everyone can file a complaint and who can give non-binding advice, for instance about delays in granting access to official documents. The Netherlands also has a complaint procedure, but this is only accessible for those that use information professionally. Following a complaint, the Dutch advisory board mediates between parties.

Comparison from an overarching perspective

Above, both legal systems have been discussed and compared on specific topics. Now, it will be examined how they are connected from an overarching perspective. In doing so, two specific overarching perspectives will be discussed, namely the principle of legality and the interest of open and free deliberations. Although both countries protect this principle and this interest, they do so in distinct ways. These two distinctions are then discussed from two perspectives. First, a connection will be made to the structure of the administrations (dualistic or monocratic) of the two states. Secondly, the importance of these two distinctions in understanding the various other legal differences that were found in Sections 3-5. This comparison rests on the assumption that Sweden and the Netherlands share their commitment to democracy and the rule of law and that this, at minimum, requires that the use of public powers must be accountable.

The Principle of legality

In summary, Section 4 showed that the exceptions to publicity in Sweden are all clearly laid down in legislation drawn up by the Riksdag. If information in an official document does not fall under one of these exception clauses, it is clear that access to the document can be granted. An assessment of the risk of harm must only be made when one of the clauses applies. This is a major difference compared to the Netherlands, where the legislation only specifies interests of secrecy that could outweigh the interest of publicity in particular cases. This gives administrative bodies much discretion to weigh up interests before deciding whether information is confidential or not.
While in both countries, ministers are accountable to parliament, the scope of ministerial responsibility differs significantly. In Sweden, most national-level tasks are not usually performed by ministers themselves but by separate public authorities, to which ministerial responsibility does not extend. By contrast, in the Netherlands most public tasks are carried out under the mandate of the minister, meaning that ministerial responsibility covers the performance of those public tasks.
Parliamentary accountability—or the lack thereof—is vital for understanding this difference in legislation between the two countries. In order to make this point clear, it is necessary to understand how the concepts of publicity and accountability relate to each other.

Relationship between access to information and accountability

The relationship between publicity and accountability is not straightforward. According to Thompson, ‘democracy requires publicity, but some democratic policies require secrecy’.
DF Thompson, ‘Democratic Secrecy’ (1999) 2 Political Science Quarterly first page, 181–193.
 Without secrecy, for instance, it is hard to imagine how the police could function effectively. Without publicity, the public cannot hold those in power to account. Democracy therefore faces a dilemma: sacrifice accountability or sacrifice the policy. This ‘all or nothing’ trade-off can be mitigated by finding a middle ground: second-order publicity. Secrecy is allowed, but the decision as to what is deemed secret is a public decision, for which governments are accountable. The decision to sacrifice accountability is then in itself accountable.
ibid.
Second-order publicity is a factual term. For the purpose of this article, this can be expressed in legal terms as the principle of legality. The principle of legality holds that government is only allowed to act if that action can be traced back to a pre-existing and specific basis in the law. Second-order publicity is the factual state that one arrives at when parliament makes legally binding decisions in advance about the cases in which governments may withhold information. Both Sweden and the Netherlands can be said to have second-order publicity. In both countries, the right to access public information is laid down in law. As a result, exceptions to this right are therefore also restricted by the secrecy clauses laid down in law. All exceptions must be adopted by the Dutch Staten-Generaal and the Swedish Riksdag. It must be noted that Thompson, along with other critics, did recognise that it is impossible in practice for lawmakers to foresee every situation in which an exception to the principle of publicity must be made.
DF Thompson (n 134); R Sagar, ‘On Combating the Abuse of State Secrecy’ (2007) 15 The Journal of Political Philosophy 404, 409-410.
 Thus, some room for discretion for public authorities is unavoidable to adequately protect the interests of secrecy.
Mokrosinska explains that the choice between secrecy and public accountability rests on a false contradiction.
D Mokrosinska, State Secrecy and Democracy (Routledge 2024) 74-78.
 Publicity may be a sufficient requirement for accountability, but it is not a necessary one. She points out that national security, for instance, is something that the general public is not informed about, but parliament and other specialised bodies exist that are tasked with holding the government to account. They must have access to information as it is a necessary component of accountability, but that does not necessarily mean that information must become publicly available. In most cases, parliamentary accountability and publicity happen simultaneously, since parliaments usually deliberate publicly. However, they are not dependent on each other and function independently in practice.
Although accountability is necessary, publicity is not always necessary to achieve accountability. If governments are accountable to parliament, publicity is not strictly necessary for public accountability, although it can contribute to achieving that end. In the absence of another mechanism for public accountability, publicity becomes vital for public accountability.
This also has implications for the argument that Thompson presents for second-order transparency. In his view, second-order transparency achieves the aim of making the decision to keep information that is hidden accountable. However, that rests on the assumption that governments are not accountable for the decision to keep information secret itself. If parliamentary accountability extends to the decision to keep information secret from the public, that assumption does not hold up. Strict adherence to the principle of legality to provide for second-order transparency is, in that case, not strictly necessary to achieve public accountability.

Organisation and publicity

To put theory into practice, the legislation of the Netherlands and Sweden will be compared in two ways. Firstly, it will be compared based on the role of publicity in relation to public accountability in general. Secondly, within the context of the legitimacy of the specific decision to grant or deny access to information.
In both countries, ministers are accountable to parliament, and linked to this is the concept that parliament has a right to information about the performance of those public tasks to which ministerial responsibility extends. The difference lies in the scope of the ministerial responsibility; this responsibility is broad in the Netherlands since most public tasks are performed monocratically through the minister and narrow in Sweden due to the principle of dualism. In Sweden, publicity is necessary for public accountability regarding the performance of all public tasks to the extent that they are not covered by ministerial responsibility. In the Netherlands, ministers are accountable to parliament for those tasks. Publicity as a principle is comparatively more important to achieve public accountability in Sweden as compared to the Netherlands. This fact may explain why, in Sweden, legal guarantees and protection of the right to access public information are much stricter.
Furthermore, in the Netherlands, a public official’s decision, on behalf of their minister, to restrict access to information is one for which the minister is accountable to the Staten-Generaal. This accountability provides democratic legitimacy for the decision, since parliamentarians can supervise the minister’s functioning. The mere existence of accountability in the Netherlands does not fully explain why restrictions to the right to access public information need not be carefully and specifically detailed in law. However, it does show that strict legality requirements are not absolutely necessary to achieve conformity with the principle of accountability. Conversely, in Sweden, decisions regarding publicity and secrecy from public authorities are not subject to parliamentary oversight. In the absence of direct democratic oversight, the legitimacy of such decisions must be rooted in a strict and precise application of the law. Through second-order transparency, the appeal to secrecy is reconciled with accountability.

Practical implications

The legal comparison of Section 5 showed that in both the Netherlands and Sweden, the decision to grant access to information does not have to be a formal decision. The laws on this point are similar. However, in the Netherlands, almost all information requests based on the Woo, in practice, result in formal decisions, whereas in Sweden this is not the case. The reason for this is twofold. First, in the Netherlands, all personal information is typically withheld, which means that almost all decisions to grant access carry a restriction. In Sweden, this is not the case, as noted in Section 4. The second reason is connected to the principle of legality. Legality is not only connected to the use of public powers, but also to legal certainty. If the law clearly stipulates in which instances information can be withheld, it is possible for public officials to decide whether the information falls within the category of information that is governed by a secrecy clause. This means that Swedish public officials normally (leaving aside more complex cases) will be able to decide whether a secrecy clause is applicable. In the Netherlands however, the public official must be certain that no interest of secrecy is relevant in a specific instance. Since the Dutch secrecy clauses can be applied more flexibly, their application in specific instances is legally more uncertain. As a result, public officials may be hesitant to informally provide requesters with information without a formal decision. This results in a formalisation of the procedure, which in turn can lead to delays and legal conflicts.

The interest of free and open deliberation

As described in Section 3, in Swedish doctrine, everyone has a right to access official documents. Preparatory documents, such as drafts and internal correspondence fall outside the scope of official documents and are therefore ‘exempt’. These documents are neither secret nor public, since the right does not extend to them. Based on the Woo, Dutch citizens have a right to public information, which includes all preparatory work. The Netherlands, however, restricts the right to access the policy views of participants in internal deliberations. The internal discussion is exempt from publicity under Dutch law and it can neither be said to be public nor secret.
These clauses also share similarities in their application in particular instances. According to the Woo, personal policy views must be distinguished from objective information, such as facts, as a determining criterion. According to the Tryckfrihetsförordingen, memoranda are not official documents if they do not add new factual information to the administrative case. Both clauses therefor have a similar area of application, namely information that does not add factual information to the case.
For this discussion, it is important to note that the objectives of this exemption also overlap. According to Cameron, the exclusion of preparatory material from the definition of ‘official documents’ grants public authorities a moment of ‘peace and quiet’, protected from constant demands of disclosure, during which it can unhindered consider different policy alternatives and have open discussions.
I Cameron (n 11) 121-122.
 The exemption in the Woo seeks to protect a similar value: the intimacy and freedom of the decision-making process.
Parliamentary papers II (2011–2012) 33 328, 3 44.

Publicity and confidentiality of deliberations

One thing that authors on both sides of the debate agrees on is that publicity changes the nature of deliberations and the pressures on those who deliberate. The exact implications of publicity are heavily debated.
This debate is well summarised by D Mokrosinska (n137).
 It must be kept in mind that the public sphere is a construct that, in reality, is made up of all spaces in which people interact, such as parliament, political debates on television, discussions on social media, town hall meetings, quality and tabloid journalism, etc.
S Chambers, ‘Behind closed doors: Publicity, secrecy, and the quality of deliberation’ (2004) 12 Journal of Political Philosophy, 398.
 The arguments stated in the previous section tend to generalise, while their particular strength may depend on the particular context.
Bentham noted that the eye of the public makes the stateman virtuous.
S Baume and Y Papadopoulos, ‘Transparency: From Bentham’s inventory of virtuous effects to contemporary evidence-based skepticism’ (2018) 21 Critical Review of International Social and Political Philosophy 2 169.
 Public actions require public scrutiny in order to induce those in power to act morally. If officeholders were monitored at all times, they would think twice about using public powers for their own private interests. Another argument closely related to this is that public reasoning discourages sloppy reasoning and encourages public (instead of private) reasoning.
D Mokrosinska (n 137) 11-12.
 Almost everyone recognises the pressures of speaking publicly: nobody wants to lose a debate, and so they come prepared. Moreover, politicians must carefully decide on their positions on particular issues in advance since there is a clear democratic incentive to be able to appeal to at least a larger part of the population than their opponents. This roots out narrow, sectarian reasoning in deliberations and improves public justification, which in turn forms a constitutive part of the legitimacy of public action.
A Gutmann and D Thompson, Democracy and Disagreement (Belknap Press 1996).
Not everyone agrees that publicity is the solution to all problems in politics. Some scholars have pointed out that publicity can also formalise deliberations in a way that hampers free, frank discussion and removes time for contemplation and the gradual accommodation of divergent views within the organisation.
AF Westin, Privacy and Freedom (Atheneum 1967) 46-47.
 Connected to this is the view that publicity will not lead to moral behaviour, but rather defensive behaviour. Officials are less likely to put controversial information in writing and may, in the most extreme cases, destroy public records.
D Mokrosinska (n 137) 8.
 Contemporary scholars have also cast doubts on the idea that the publicity of deliberations necessarily leads to strong and public reasoning. Chambers, for example, notes that public deliberations, especially in the context of asymmetrical communication through mass media and social media, can be characterised by shallow (plebiscitary) reasoning that aims to pander to its audience rather than spark genuine political discussion.
S Chambers, (n 141) 398.
In both Sweden and the Netherlands, parliament deliberates in public. It is almost impossible to imagine a democracy without such public deliberations. The laws of both countries do, however, allow space for confidential government deliberations as part of its executive function. This can be viewed as a way to balance the public nature of deliberations with the confidentiality of deliberations. The way that both countries operationalise this space differs significantly.

Organisation and publicity

In the Netherlands, administrative processes take place internally in monocratic ministries, while in Sweden these are divided among different institutions, each with its own responsibility to work together. In this context, Cameron notes that policy (and the documents forming the basis of it), whether initiated ‘bottom-up’ by administrative agencies themselves or ‘top-down’ by the government, becomes public when it ‘leaves’ the government department or administrative agency on its way from one to the other.
I Cameron (n 11) 119.
 This makes sense, as public administrations and government departments in Sweden are separate entities that share information externally. By contrast, in the Netherlands, the directorates-general within a ministry share information internally. While policy must eventually be published externally to take effect, the process of developing that policy consists of internal working material, which is not relevant for the ministerial accountability.
Only the decisions and opinions of the political leadership over the ministries (ministers and secretaries of state) are relevant for parliamentary accountability. The opinions of civil servants are not relevant and may embarrass their minister, which provides the incentive to keep them confidential.
On the one hand, this explains why, in the Netherlands, the scope of the right to access public information is relatively broad. If all internal working material were to be excluded, as in Sweden, the right to access public information would become meaningless, since the relevant information is only shared internally within the public authority. However, this broad scope also means that access to information can infringe upon internal deliberations. Therefore, the exemption of personal opinions on policy drawn up during internal deliberations is a necessary provision to protect the interest of free and open discussion among public officials. On the other hand, it also explains why, in Sweden, limiting the scope of the right to access public information does not significantly undermine the principle of publicity. Since policy decisions are based on information that is publicly shared between public administrations and government departments, policy formulation is carried out in the public domain. A specific clause that protects free and open discussion among public officials is not necessary, since working material typically falls outside the scope of the right to access. The dualistic nature of the Swedish administration fosters openness in policy preparation, as information shared between the departments and public administrations is public, barring any restrictions laid down in law.

Practical implication

These different approaches of Sweden and the Netherlands to protect internal deliberations have three legal implications. First, in Section 5.2 it was noted that in Sweden, the OS requires public authorities to register official documents or keep them in order in another way. In the Netherlands, only a general obligation to keep information in order is laid down in the Woo. This difference may be related to the difference in scope between the right to access information in Sweden and the Netherlands. Since in Sweden, the term ‘official document’ is clearly defined, obligations to register or keep those documents can be specified. In the Netherlands, all documents containing information relating to a public task, including all working documents, are treated equally for the Woo. It is therefore not obvious to lay down specific obligations to keep certain (important) documents in order in the Woo.
Secondly, this difference can be linked to the requirements for making a valid request under both laws. In Sweden, a request must be specified in order to identify the documents that the requester wishes to access. This can only properly function if interested parties can find out about the existence of documents. A public document register, which Swedish law mandates in contrast to Dutch law, is almost indispensable in this respect. A similar requirement for a valid request would raise a significant barrier in the context of the Dutch Woo. As a result, a request in the Netherlands is valid if a public affair that the information relates to is stated. Thus, also this requirement can be linked to the limited scope of the right to access information in Sweden as compared to the Netherlands.
Lastly, differences in scope and exemptions of the legislation have important legal and practical effects. In the Netherlands, all internal working documents, concepts, and internal e-mails among public officials are considered public documents and the Woo treats them the same as documents containing decisions, formalised advice, incoming and outgoing documents, contracts, etc. All these documents must be gathered and decided on if requested. This is different in Sweden since internal working documents are not included in the scope of the right to access official documents.

Conclusion

This article compares the right to access public information in the Netherlands and Sweden, focusing on the most notable differences by examining the legislation that codifies this right. These differences were further clarified by linking them to two overarching perspectives: how decisions regarding publicity and secrecy are made, and which information falls within the scope of the legislation.
In Sweden, general rules about publicity and secrecy must be laid down in law as clearly and specifically as possible, in accordance with the Tryckfrihetsförordningen. Public authorities apply secrecy provisions in specific cases, which may require an assessment of potential harm. In the Netherlands, the law outlines the interests that justify confidentiality, and public authorities must balance these interests against the principle of publicity in individual cases. This distinction is significant and helps explain differences in administrative culture regarding openness and secrecy. It also clarifies why the Dutch system tends towards formalisation, whereas the Swedish system requires formal decisions only in some instances.
An important similarity between the legal frameworks of the Netherlands and Sweden is that both protect the internal processes of public authorities, but the legal mechanisms through which this is achieved differ significantly. In Sweden, this is done by limiting the scope of the right to access public information to finalised information—either documents related to completed matters or information that has been dispatched or received. In the Netherlands, the right to access generally extends to this, but opinions formed during internal deliberations on policy are exempt from disclosure. This difference explains much of the legal variation in procedures for granting access to information.
This article compares two nations that both recognise the right to access public information but differ in terms of administrative organisation and administrative culture. Several Swedish legal scholars have highlighted the significance of the dualism between ministerial departments and public administrations in shaping the right to access official documents in Sweden. The Netherlands recognises the right to access public information but does not separate ministerial departments from public administrations. Dualism provides a strong foundation for publicity in public administration, as it becomes the primary mechanism for public accountability. However, dualism requires that the rules regarding publicity and secrecy be precisely codified in law, since there is no alternative source of democratic legitimacy for decisions to withhold information. In the Dutch system parliamentary accountability and publicity often coincide. It can be argued that this reduces the importance of publicity as a separate institution, because decisions to withhold information indirectly acquire democratic legitimacy.
According to Cameron, the principle of openness serves to underpin the dualistic nature of Swedish administration that separates government and administration.
I Cameron (n 11) 120.
 Publicity allows for accountability. The fact that administrations are accountable for their own role and tasks means that they make decisions as separate entities. In the Netherlands, the Woo aims to open government with a broad right to public information, while at the same time maintaining broad ministerial responsibility. This may explain why the Dutch government struggles with publicity of the administration. This tension may not only exist in the Netherlands, but also in other countries that rely on broad ministerial responsibility. This would yield interesting follow-up research.


Indien u een los artikel wilt bestellen, stuur een e-mail naar info@uitgeverijparis.nl