Copyright Protection of Government-Related Material Before the Courts of the United States and Canada: Considerations for Future Reforms

Abstract

Copyright protection of government-related material lies at the intersection of private incentives, public interest, and political motivation. These interests naturally clash. Therefore, the justification and scope of copyright protection for such materials has been the subject of intense controversy ever since. Recently, the Supreme Court of the United States and the Supreme Court of Canada handed down landmark decisions on the application of the respective century-old doctrines and provisions. Moreover, courts in the U.S. and Canada have lately addressed the protectability of privately created, government-adopted industry standards. This article takes these decisions as an occasion to reflect on the copyright protection of government-related material against the background of rapid technological advancement and substantial ongoing societal and political change. Taking into account the regulatory experiences in the EU, this article questions the prevalent assumptions of trustworthy state action and undistorted functioning of markets, which considerably underlie the design of current government copyright regimes around the globe. In this light, the article aims to provide avenues for future legislative reforms that address the copyright of government-related materials. It suggests a more focused, nuanced, and holistic regulatory approach for strengthening and maintaining open, democratic societies.

1 Introduction

States create and disseminate a vast amount of material that can potentially qualify as works under the copyright laws. This is true, for instance, for laws and court decisions, maps, reports, brochures, or information videos. But also, in some jurisdictions, public registers or databases may qualify for exclusive protection. While these materials are created by public servants or employees of the respective branches of the state, the question of copyrightability also stretches to commissioned works like studies or expert opinions. The debate on whether copyright should in fact protect such material – no matter if disseminated as print or online – is distinct from the classical copyright debate. The concerned materials usually touch on the public interest, they involve the state which represents the public interest but also follows strategic considerations, and they may be tax-funded to a considerable extent.

How copyright treats such materials varies significantly across the globe. At least at first glance, the legal regimes in the U.S. and Canada appear to stand at opposite ends. While U.S. copyright law follows a public domain paradigm, Canadian copyright law – according to the tradition of the Commonwealth countries – provides Crown copyright protection for government works. The respective rules and doctrines are more than a century old. In this light, it came as a surprise that recently, and almost at the same time, both the Supreme Court of the United States (USSC) and the Supreme Court of Canada (SCC) handed down landmark decisions on the application of the relevant copyright rules. While the USSC case (Georgia v. PRO) concerned the protectability of an annotated version of a legal code in the state of Georgia, the case before the SCC (Keatley Surveying) concerned the copyright status of privately created plans of survey after they were provided to the land registry office in the province of Ontario.

The decisions of the Supreme Courts come at a time of considerable technological progress and societal transformation, which concerns the state-market interface that lies at the core of these copyright disputes. The issue of “copyright and the state” has become particularly relevant for mass digitization projects that often involve public–private partnerships. But this debate has nowadays advanced into the discourse on data-driven innovation and artificial intelligence, and in particular the smart city/state context, where public and private actors are technologically linked. At the same time, economic, political, and societal transformation causes many societies to reconsider the notions of the market and the state as such. This includes the question of reconfiguring information powers between private and public actors. Given these developments, this article takes the recent decisions of U.S. and Canadian courts as an opportunity to re-think copyright protection for government works. It proposes a more integrated regulatory approach and questions whether copyright protection for government-related materials is needed at all in open, democratic societies.

For this purpose, this article will discuss the decisions of the USSC, which stands for the public domain paradigm (at Sect. 2), and of the SCC, which stands for the Crown copyright paradigm (at Sect. 3). The examinations will focus on the context of the disputes, the reasoning of the courts as well as on the doctrinal consequences and economic effects. Also, both decisions provide implications for the long-debated issue of copyright protection for government-adopted industry standards, with which other courts have also recently dealt. On this basis, the article will then turn to a wider reflection on the future of copyright law in government-related materials (at Sect. 4). For this purpose, it considers lessons that can be drawn from the attempts to harmonize the re-use of public sector information in the EU. The article concludes by sketching considerations for future legislative reforms on the copyright of government-related materials.

2 USA: The Public Domain Paradigm (Georgia et al. v. Public.Resource.Org, Inc.)

2.1 Public Domain: The Government Edicts Doctrine

On 27 April 2020, the Supreme Court of the United States (USSC) handed down the decision in Georgia et al v. Public.Resource.Org (Georgia v. PRO),¹ which is a landmark copyright case. At its core lies the issue of copyrightability of the law, which not only evokes the challenge to adequately interpret copyright provisions, but also touches on a philosophical reflection on the nature of the law itself. Georgia v. PRO exemplifies a problem that constantly accompanied the history of legal publishing: Governments may heavily depend on private entities to edit and publish (online and offline) material of public interest. This leads to an unescapable clash of interests.²

The case centered around the Official Code of Georgia Annotated (OCGA).³ This code contains every statute of Georgia that is currently in force as well as non-binding annotations which appear beneath each statutory provision. These annotations include summaries of judicial opinions, summaries of the state attorney general’s opinions, and a list of related reference materials. The OCGA is assembled by the Code Revision Commission, a state entity predominantly consisting of legislators,⁴ funded through legislative branch appropriations, and staffed by the Office of Legislative Counsel. Furthermore, the OCGA is subject to the legislature’s approval.⁵ Yet the Code Revision Commission commissioned a private publisher (LexisNexis⁶) for the production of the OCGA under a work-for-hire agreement. According to this agreement, any rights vest exclusively in “the State of Georgia, acting through the Commission”. However, LexisNexis enjoys the exclusive rights to publish, distribute, and sell the OCGA. In return, LexisNexis agreed to limit the price it may charge (412 USD per hard copy) and to offer an unannotated online version of the statutes for free.⁷ Public.Resource.Org (PRO), an NGO, provided the OCGA online to the public. Georgia claimed copyright for the annotations and sued PRO.

Several court decisions have dealt with the copyrightability of “the law”, in particular of statutes, court decisions, model codes,⁸ and government-adopted industry standards. But Georgia v. PRO did not concern the copyrightability of the law itself, but rather of additional materials (annotations).⁹ The USSC had to deal with this constellation for the first time. It could revert to the “government edicts doctrine”, which the Supreme Court itself had stipulated and refined in the 19th century through three decisions:¹⁰Wheaton,¹¹Banks,¹² and Callaghan.¹³ However, all these cases concerned court opinions and not material of the legislature. Strictly speaking, the government edicts doctrine so far only affected works of the judicature,¹⁴ notwithstanding that the Wheaton decision of 1834 already called it “absurd, for a legislature to claim the copyright” and expressed the common notion that “statutes were never copyrighted”.¹⁵ In its Banks decision of 1888, the USSC held that judges could not assert copyright in “whatever work they perform in their capacities as judges”, which would also include the non-authoritative portions of decisions like the statement of the case and the syllabus or the headnote.¹⁶ In the same year, the Callaghan decision upheld a reporter’s copyright interest in several self-created explanatory materials. The USSC justified this by saying that this material was authored by someone who had – unlike judges – no authority to speak with the force of law.¹⁷

After 1888, only lower courts in the U.S. have further inquired into the nature and application of the government edicts doctrine.¹⁸ This explains why the theoretical foundations of the three USSC decisions are held as “generally implicit and unstated”¹⁹ until this day. In 2020, meaning more than a century later, the USSC took Georgia v. PRO as a chance to clarify both the doctrinal reasoning and its extension to materials of the legislature. The court could substantially build on the Eleventh Circuit’s examination of possible foundations of the government edicts doctrine. Both the Eleventh Circuit and the USSC refer to the Banks decision when drawing their conclusions, because Banks provides several reasons why judges cannot be considered the “authors” of their work.

2.2 The Georgia v. Public.Resource.Org Decision of the USSC

The five-justice majority opinion of the USSC held that the annotations are ineligible for copyright protection. It thereby confirmed the decision of the Eleventh Circuit,²⁰ albeit on different grounds. The USSC based its “straight forward” rule on the identity of the author (judge) with regard to the law.²¹ By referring to the Banks decision (“whatever [emphasis added] work they perform in their capacity as judges”), the USSC derives the general presumption of the judge’s empowerment to make and interpret the law.²² This view installs a doctrine that inquires into the nature of the author instead of the nature of the work. Therefore, the court did not need to address the controversial question which materials would constitute “the law” and whether the binding force of the material was a relevant factor.²³ At the same time, however, no one would doubt that judges can also be authors under the meaning of copyright law – e.g. when they write novels, poems, or their memoirs. This explains why the USSC ultimately stipulates the formula that judges could simply not be “authors” of such works that they prepare “in the discharge of their judicial duties”.²⁴

Regarding the underlying rationale, the USSC argues that “[n]o one can own the law”,²⁵ because every citizen is presumed to know it. The court therefore refers to the societal significance of wide access to particular works,²⁶ which copyright protection could put at risk. In contrast to this consequentialist reasoning, the Eleventh Circuit regarded the people as the constructive author of the law and considered judges and legislators as “merely draftsmen exercising delegated authority”.²⁷ The choice of reasoning has implications, which will be discussed below. Particularly the USSC’s emphasis on access is important to keep in mind for understanding how it extended the government edicts doctrine to material of the legislature.

On the merits, the USSC unsurprisingly confirmed the settled²⁸ view that the government edicts doctrine not only covers material of the judiciary, but also legislative materials: “[L]egislators, acting as legislators” cannot be “authors” either.²⁹ As a consequence, one must clearly attribute the creation to a judge or to the legislators. In the case at hand, the USSC argued that the Code Revision Commission supervised the work of the commissioned producer (LexisNexis) and functions as an arm of the legislator.³⁰

However, the exact scope and the criteria for determining the limits remain vague. The USSC majority opinion argues that legislators – similar to judges – are vested with authority to make the law.³¹ Therefore, legislators could not serve as authors when they produce works within their official capacity.³² But what does official capacity mean? This reference stems from the Banks decision,³³ and rather than leaving us with a clear-cut rule, the USSC only provides some loose factors. The majority opinion refers to the exercise of legislative duties and argues that the legislature has deemed the annotations relevant to understanding its laws.³⁴ Also, it resembles Banks when subsuming materials as part of the “whole work done by [legislators]”.³⁵ Moreover, the majority opinion states that the doctrine can cover explanatory and procedural materials as well (e.g. floor statements, committee reports and proposed bills of legislators), and it is indifferent to the practical significance of the material.³⁶ Finally, the majority opinion did not agree with the limitations that Justice Ginsburg highlights in her dissent in which she presumes a missing link between lawmaking and annotations and stresses their mere purpose of convenience for reference.³⁷ In sum, the majority opinion’s understanding stretches far because it identifies the exercise of legislative duties whenever legislators act within their public tasks.

2.3 Limited Scope

By extending the government edicts doctrine to works of the legislature, the USSC has enlarged the pool of public domain materials. The general formula of the USSC refers to the exercise of legislative duties and allows for an extensive reading. However, when it comes to details, it also appears challenging to apply this formula in practice.

In any case, the decision’s implications are restricted to government edicts, i.e. works of judges and legislators. The doctrine does not apply to materials created by government officials or other actors who lack the authority to make or interpret the law.³⁸ The impact of the doctrine is therefore limited, which becomes clearer when looking at the different rules on copyright in government works across the U.S. On the federal level, 17 U.S.C. § 105 maintains that copyright does not protect works of the U.S. Government.³⁹ In contrast, the states’ rules on copyright in government works vary considerably.⁴⁰ So far, any political initiatives to extend the government exemption to the state level have failed. A crucial (albeit somewhat obsolete) explanation for this resistance can be found in the states’ lack of adequate printing facilities, which makes them frequently rely on copyright protection to incentivize private publishers to print and publish such works on a commercial basis.⁴¹ Therefore, the USSC’s decision significantly impacts those states which claim copyright protection for materials of their legislature. At the same time, the USSC has confirmed that the states are free to assert copyright for works other than legislative or judicial materials, which applies to a vast majority of expressive works they produce (e.g. universities, libraries, tourism offices etc.).⁴²

2.4 Towards an Access Paradigm?

Different strands of reasoning could justify the extension of the public domain. When interpreting and making the law, courts and legislators explicitly or implicitly refer to them.⁴³ The majority opinion bases the government edicts doctrine on the function of the law and emphasizes the significance of access to it. This long-standing strand of reasoning stresses the seminal role of information about the law for the democratic process.⁴⁴ Moreover, the USSC highlights that due process concerns call for accessibility when arguing that citizens must have access to the law if they are held responsible for complying with it.⁴⁵ This accessibility claim is based on public interest reasoning which lies beyond copyright law. If one assumes that the state has to create the law anyway, the discussion revolves only around the distribution of works and does not need to consider incentives for their creation. However, one can challenge the accessibility presumption as well as the due process objection in the particular case of annotations. First, one can doubt to what extent accessibility to annotations is in fact necessary in the light of the democratic process and due process concerns.⁴⁶ Second, the “incentives to create” play a seminal role regarding the editing of annotations as opposed to the editing of the law itself. These two aspects conflate in the majority opinion’s critical remark about the divide between economy-class readers (access to the mere law) vs. first-class readers (access to the law plus annotations) and its societal implications.⁴⁷ In sum, the key message of the USSC’s reasoning lies in its emphasis on accessibility and its shift towards public policy considerations. However, due to the specific facts and the limitation to legal material, one should not over-interpret it as a general access paradigm⁴⁸ for the further application of copyright law.

In contrast, the USSC did not take up three other doctrinal approaches, which have already been extensively discussed. First, the court did not refer to intra-copyright doctrinal reasoning, as there was no occasion to apply the merger doctrine according to which the idea of the law (and annotations) would be inseparable from its expression.⁴⁹ Second, the USSC rejected the Eleventh Circuit’s “collective authorship” approach, according to which words of judges and legislators carry the force of law because they express the voice of the people as true author and therefore articulate the sovereign will.⁵⁰ This idea of collective authorship leads to the metaphorical concept of “citizens ownership”, which other courts have previously suggested.⁵¹ It would then be crucial to ask if the law is “attributable to the constructive authorship of the people”.⁵² This, however, raises challenges, as it would inevitably result in philosophical questions about the nature of the law.⁵³ Moreover, it does not appear compelling that collective ownership implies that the works fall under the public domain. One could equally well argue for state ownership when presuming that the state represents the peoples’ will in a democracy.⁵⁴ Third, the USSC has not taken up another strand of reasoning, which regards the government as “employee of the people”. The core of this analogy is that the public pays the salaries of those who draft legislation and that judges are paid out of the public treasury.⁵⁵ While this notion contains some logical flaws,⁵⁶ its application would certainly reach beyond legal materials. It could support the view that everything created by publicly financed servants should fall into the public domain. This reasoning is reflected in 17 U.S.C. § 105 and also in “open government data” (OGD) policies.⁵⁷ At the same time, however, this argument is often turned upside down when arguing – albeit in a doubtful way from an economic perspective – that exclusive rights enable the government to sell the material in order to recoup costs which would otherwise be passed on to taxpayers.⁵⁸

2.5 Consequences for Government-Adopted Industry Standards

The USSC decision Georgia v. PRO affects a practically important and long-discussed issue: the copyrightability of government-adopted model codes and industry standards. So far, courts in the U.S. appear to agree that model codes, which have been adopted by municipal or state governments through official, verbatim incorporation, are not copyrightable once they become “the law”.⁵⁹ While the USSC has so far not directly addressed this question, its reflection on and expansion of the government edicts doctrine confirms this position.

In contrast, it is still highly disputed whether privately created industry standards lose copyright protection once the legislators adopt them by mere reference.⁶⁰ When following the “access to the law” rationale, one would argue that citizens must be able to easily consult such privately authored standards in order to understand their legally imposed obligations.⁶¹ However, so far the courts have left open whether the inclusion through such reference results in the loss of copyright⁶² and to what extent the incentives for and the purpose of their creation affect their copyrightability.⁶³ The issue is delicate because thousands of technical standards have in fact been incorporated into the law by reference.⁶⁴ And even more so, a loss of copyright could also call protection for many other private reference works into question – e.g. the Bluebook or school books.⁶⁵ Recently, the D.C. Circuit deliberately sailed around this “thorny question”⁶⁶ in another case where – again – PRO was the driving force. PRO provided privately created technical standards, which had been incorporated in the Code of Federal Regulations by reference, on a public website. The court left “for another day the question of whether the Constitution permits copyright to persist in works incorporated by reference into law”.⁶⁷ Instead, it reversed and held that the district court erred in its application of the fair use doctrine.⁶⁸

Does the USSC’s advancement of the government edicts doctrine eventually affect copyright protection of referenced private industry standards? The court’s holding would require that the legislators actually “created” the material in question. Such creation does not relate to the inclusion of the reference as such into the law;⁶⁹ instead it concerns the development of the referred standards as such. In Georgia v. PRO, the USSC held as decisive that even if a private party prepared the materials, the Code Revision Commission, for which Nexis as a private publisher made a work-for-hire, “functions as an arm of it for purpose of producing the annotations” as the material in question.⁷⁰ But this arrangement was part of an ex ante scheme that had been deliberately set up for producing the annotations. In contrast, a mere ex post reference to private standards, which have already been created, does not constitute such a scheme. This distinction is important, because the USSC regards the preparation process with its clear purpose and functional context as decisive. Therefore, it appears reasonable to argue that the standards were not created in the course of legislative duty. From a strictly doctrinal view, the holding of Georgia v. PRO does not therefore directly affect cases where external standards are adopted into the law by mere reference. In contrast, providing access as the general rationale of the USSC’s decision may allow for easily extending the government edicts doctrine to government-adopted industry standards in the future.⁷¹ Such extension, however, would have broad implications for future incentives to create privately developed industry standards, considering that the legislators could basically refer to any private material and render the material uncopyrightable as a sudden consequence.

2.6 Impact on the Incentives to Create Legislative Materials

Finally, the actual impact of the Georgia v. PRO decision on the creation of materials (such as annotations) remains open and appears ambivalent. As setting up public–private partnerships for publishing revolves around money, it would seem naïve to believe that the USSC decision causes more material to fall into the public domain. Such a static view would overlook possible future dynamic effects on the creation of the material.⁷² The loss of copyright may adversely affect the incentives of the actors involved in public–private partnerships. As a consequence, all involved actors must reconsider the arrangements on the creation and publication of annotations. This is relevant for 25 jurisdictions in the U.S.⁷³ Should legislators continue to provide such annotations in the future, they have to find ways to fund their production.⁷⁴ In the Georgia v. PRO proceedings, various parties argued that a lack of copyright protection would cause such state annotated codes to simply disappear.⁷⁵ The states would be unable to induce private parties like LexisNexis to assist in preparing affordable annotated codes for widespread distribution,⁷⁶ because a lack of copyright protection would undermine incentives to contract with third-party publishers to prepare the annotations, analyses, and commentary about their codes.⁷⁷

Indeed, whether or under what conditions such annotated codes will be offered in the future relies entirely on market forces, should the states abstain from funding such material. If the material will still be produced and offered, one can at least expect a price increase. In the worst case, the societal consequences run counter to what the USSC’s majority opinion actually wanted to prevent: an even bigger divide between “first- and second-class readers”. Ultimately, it depends on the states’ policies on how to frame the public task. This is subject to democratic choice which goes beyond the scope of this article. However, when thinking about this choice, one should at least be aware of the risk that lawyers might tend to overestimate the citizen’s general desire “to searching in internet or library for annotated version of the law”.⁷⁸

What sort of public–private arrangements will be concluded in the future depends on the concrete markets, existing business models, and future opportunities to be found. These factors may significantly differ across the states. In general, one needs to pay attention to whether revising public–private cooperation (e.g. through reshaping procedures and institutions) will decrease public control over the process, in case states decide to redesign their schemes so that they can retain copyright protection by escaping from the scope of the government edicts doctrine. Moreover, a legislative amendment could potentially safeguard the involved interests. The USSC gave a broad hint when it held itself as not being responsible to decide on policy concerns and that they are more appropriately addressed by Congress, which should decide how best to pursue the copyright clause’s objectives.⁷⁹

In sum, Georgia v. PRO may appear to be a rather old-fashioned case that at first glance revolved around the mere editing and publishing of books. However, at second glance and from a wider angle, the case involved a complex scheme that included online business models because the editing and publishing of the book is deliberately linked to obligations to provide internet access. At the same time, the case has a limited impact on the public–private interface because it is limited to works of the legislature only and does not affect any other materials. This is different in the Keatley Surveying decision, which the SCC delivered in 2019. This case addresses state copyright and public–private partnerships beyond the law and therefore deserves a closer look.

3 Canada: The Crown Copyright Paradigm (Keatley Surveying v. Teranet)

3.1 Crown Copyright

Government materials in Canada – as in other countries of the Commonwealth – are protected by Crown copyright. This is in stark contrast to the public domain status of government works on the U.S. federal level. The Keatley Surveying decision of the SCC has Crown copyright protection at its core, and it is therefore necessary to briefly contextualize the concept of Crown copyright.

Crown copyright, also called government copyright,⁸⁰ is highly debated. Calls for its entire abolishment have frequently been raised across Commonwealth jurisdictions.⁸¹ Originating from the United Kingdom, where printing as such was once considered to be a matter of the state,⁸² the Crown’s prerogative power over publishing was reserved to all publishing in the sixteenth and seventeenth century.⁸³ The Crown prerogative is based on the reasoning of the monarchy, according to which Kings derive their just powers from God.⁸⁴ As a consequence “the property of all law books is in the king, because he pays the judges who pronounce the law”.⁸⁵ Effectively, the Crown prerogative was used as a censorship tool.⁸⁶ Besides its philosophical basis, the justification for the printing privileges were seen in ensuring preservation, authenticity, accuracy, and reliability of government materials,⁸⁷ but also in public revenue generation.⁸⁸ The Crown’s proprietary rights cover law, judicial opinions,⁸⁹ and religious works⁹⁰ in particular.

This common-law-based Crown prerogative persists until this day, even though it could in principle be abolished by statute.⁹¹ Moreover, the establishment of modern copyright acts usually added a second pillar: statutory Crown copyright. In Canada, the Crown copyright provision was introduced in Sec. 12 of the Copyright Act of Canada of 1921, which reads:

Where copyright belongs to Her Majesty.

Without prejudice to any rights or privileges of the Crown, where any work is, or has been, prepared or published by or under the direction or control of Her Majesty or any government department, the copyright in the work shall, subject to any agreement with the author, belong to Her Majesty and in that case shall continue for the remainder of the calendar year of the first publication of the work and for a period of fifty years following the end of that calendar year.

The beginning of the section (“Without prejudice to any rights or privileges of the Crown […]”) reflects the Crown prerogative, which grants to the Crown a monopoly in printing the above-mentioned works in perpetuity, because the time limits of the Copyright Act do not apply.⁹² This explains why the Reproduction of Federal Law Order of 1997 provides a blanket license for the reproduction of federal statutes and court decisions in Canada.⁹³ The wording of Sec. 12 implies that statutory Crown copyright stretches far and covers almost any material one could think of as work.

Section 12 is based on (and is still almost identical with) the original version of Sec. 18 U.K. Copyright Act of 1911, which was the first statutory Crown copyright provision. However, the United Kingdom expanded and specified Crown copyright legislation in 1956 and further amended the provisions in 1988, when it included a separate provision on Parliamentary copyright.⁹⁴ Also, the reform of 1988 deleted exactly the passage that is now contested in the Keatley Surveying case: instead of referring to works “by or under the direction or control” of the Crown, the law now covers works “by an officer or servant of the Crown in the course of his duties”.⁹⁵ This amendment aimed to overcome difficulties with the vague and overly broad wording.⁹⁶ The U.K. reforms affected many Commonwealth jurisdictions.⁹⁷ New Zealand, for example, also renewed the Crown copyright sections because of the ambiguous drafting of the sections,⁹⁸ and nowadays explicitly exempts bills, acts, regulations, bylaws, debates, reports, and jurisprudence from copyright protection.⁹⁹ In contrast, however, Canada – just like Australia¹⁰⁰ – has not revised Crown copyright to this day, even though this has been frequently initiated.¹⁰¹

3.2 The Keatley Surveying v. Teranet Decision of the SCC

In September 2019, the SCC handed down the Keatley Surveying decision and ruled for the first time on the century-old provision of Crown copyright. The decision falls right into a recent policy debate on Canadian copyright reform, including Crown copyright protection.¹⁰² While the Keatley Surveying decision also focused on government works and public–private partnerships, it did not deal with the copyrightability of the law itself and is therefore to be distinguished from Georgia v. PRO.

The facts of the Keatley Surveying case are as follows: Ontario provides an electronic land registry system (ELRS). This is a database about all properties in the province, which inter alia contains plans of survey. ELRS is run by Teranet, a private company that also helped in building the database. In the underlying agreement with Teranet, Ontario retains all rights, title and interests, including IP rights, to the data used in the electronic land registry system, including plans of survey.¹⁰³ To build the database, Teranet contracted directly with individual surveyors and surveying firms to provide plans of survey.¹⁰⁴ After the surveyors provide the plans of survey to the land registry office, Teranet digitizes them and makes them electronically available on Ontario’s behalf for a statutorily prescribed fee (16.30 CAD per plan). Keatley Surveying Ltd. is a land survey company and brought a class action on behalf of approx. 350 land surveyors who provided plans of survey to the land registry. They claim that the surveyors (and not the Crown) retained copyright in the plans of survey they had created. Teranet would infringe the land surveyors’ copyright by storing and copying the plans of survey. The dispute centers around the interpretation of Sec. 12 when it refers to works that are “prepared or published by or under the direction or control of Her Majesty or any government department”. Keatley argues that on this basis, the Crown would only obtain copyright in works that it created itself (or where it ordered or controlled creation by someone else). In contrast, Teranet argues that the Crown would obtain copyright in everything it published.

The four-judge majority opinion of the SCC found that Ontario owned the copyright in the plans of survey according to Sec. 12.¹⁰⁵ The court held as decisive whether the Crown possesses sufficient direction or control over the preparation or publication of the surveys. When interpreting this standard, the court finds some middle ground. As regards the preparation, the court follows a narrow interpretation, according to which the production of the work must be the principal object and not only a peripheral consequence of the government’s direction or control.¹⁰⁶ Also, the fact that government can demand changes, veto publication, or refuse to accept the work for any reason would not establish sufficient direction or control.¹⁰⁷ In the case at hand, the land surveyors were independent contractors and were neither directed nor controlled by the Crown.¹⁰⁸ As regards the publication, the SCC held that merely making available someone else’s work is insufficient; rather, the Crown “must exercise direction or control over the publication process, including both the person publishing the work and the nature, form, and content of the final, published version of a work”.¹⁰⁹ The SCC then refers to different indicia for publication by the Crown, namely the presence of a statutory scheme transferring property rights in the works; strict controls on the form and content of the works; the exclusive control to modify the work; the opt-in nature of the statutory scheme; and the necessity of the Crown to make its work accessible for the public.¹¹⁰ Given these indicia, the court evaluated the scheme in its entirety and concluded that the land registration regime would give the Crown complete control over the process of the publication of the registered and deposited plans of survey.¹¹¹

Ultimately, it did not make a difference that Teranet, a private party, actually published the works. The SCC held that the degree of direction and control exercised by the Crown over the third-party publisher would be decisive.¹¹² Again the court refers to indicia, namely that Teranet acted in accordance with the comprehensive statutory regime on registration and depositing of the plans, publication, conditions, and the power to amend the plans, which would provide the government complete control over the process of publication.¹¹³ As a consequence of Crown copyright, government could legitimately permit Teranet to publish and make copies of the registered deposited plans as Ontario’s licensee; and the licensing agreement ensures that the publication is eventually done under the Crown’s full direction or control.¹¹⁴ What can be seen is that at the end of the day, rather than constituting a bright line rule, the majority opinion sets out a test that requires a contextual, fact-specific inquiry on a case-by-case basis to evaluate the necessary “degree” of control.

The concurring opinion criticized the lack of guidance of the majority opinion’s test.¹¹⁵ While it agreed on the result of Crown copyright protection, it construes the phrase “prepared or published by or under the direction or control” of the Crown according to its literal meaning and to inquire into the sufficient direction or control over the work itself.¹¹⁶ However, it admits that this broad interpretation could lead to over-protection. To mitigate this consequence, the concurring opinion proposes to read a requirement into Sec. 12, according to which the work must be a “government work”, meaning a work that “serves the public purpose and in which vesting the copyright in the Crown furthers that purpose”, especially accuracy, integrity and dissemination.¹¹⁷ Ultimately, this suggestion would end up in a work-specific approach that reconciles the wording and the purpose of the copyright act. Unlike the proposed test of the majority opinion, the approach would not focus on the process and therefore not rely on the fact-specific circumstances of the publication.¹¹⁸

3.3 Impact of the Decision

Unlike the Georgia v. PRO decision in the U.S., the Keatley Surveying decision will most likely not cause any immediate effects in Canada. The SCC has confirmed Crown copyright protection and therefore the copyright status on which Ontario’s existing regulatory scheme already relied. However, the court only arrived there by concluding that government surpassed the necessary threshold which Sec. 12 would require for sufficient direction or control. Therefore, the result depended on specific facts and circumstances, and the SCC shares the similarity with the USSC in leaving us with uncertainties on the decisive criteria how to determine the threshold. As a consequence, the implication is similar in a sense that also Canadian governments have to carefully consider how they structure government-related services and to what extent and under which legal framework they include private parties in the fulfillment of public tasks.

The striking difference is that Canadian governments are in a stronger initial position than their U.S. counterparts, because the default position allocates copyright ownership to them. Therefore, private contractors who wish to retain copyright need to closely examine their arrangements with the government to derogate from this default position.¹¹⁹ In the Keatley Surveying case, it appears doubtful whether the surveyors are in fact able to negotiate a clause that would retain their copyright. If they fail to do so and if there is no price regulation, surveyors might compensate the loss of copyright by charging Teranet higher prices for their services. However, higher prices will most likely not prevail if the surveying market is competitive.

From a more general perspective, the Keatley Surveying decision underlines the significance of government copyright in Canada. The SCC reiterates the “principle of copyright balance” between creators’ and users’ rights,¹²⁰ which Canadian courts have repeatedly articulated.¹²¹ The Keatley Surveying decision has now explicitly included the state’s interest in holding copyrights into this equation: On the one hand, the court acknowledges the aim of ensuring accuracy and integrity of government documents. On the other hand, the SCC warns that this could “not lead to such an expansive Crown copyright regime that public interest in accessing information is harmed”.¹²² The court emphasizes that “Crown copyright cannot be so expansive in scope that it allows for the routine expropriation of creators’ copyright in their works”; because that would risk “impeding the public interest in accessing these works and could compromise the existence of a robust public domain”.¹²³ Therefore, the decision aims to reconcile the efficiency of using third-party contributions with the need to maintain copyright in works for ultimately serving the accuracy, integrity, and dissemination of such works. The SCC found a diligent approach to reconcile the wording of a century-old provision with its rationale. However, the court also questions Crown copyright’s general necessity and leaves it to the legislators to take further actions.

3.4 A Side Glance at Government-Adopted Industry Standards

The issue of copyright ownership in government-adopted industry standards has not only been before the U.S. courts in recent times. In the Knight v. Canadian Standards Association (CSA) decision of 2018,¹²⁴ the Canadian Federal Court of Appeal (FCA) ruled on the copyright protection of privately developed electrical safety codes that the legislators incorporated into Canadian law by mere reference. CSA is a not–for–profit corporation that is engaged in developing, testing, and certifying voluntary electrical standards. The Electrical Code sets out safety standards for installation and maintenance of electrical equipment in Canada. The CSA Electrical Code was registered in April 2015 in favor of the CSA. PS Knight Co., a commercial competitor of the CSA, published the CSA Electrical Code without license.

The majority opinion of the FCA held that the CSA owned valid copyright in the Electrical Code and that PS Knight Co. infringed copyright.¹²⁵ A seminal question was whether the incorporation into Canadian law through reference constituted Crown copyright protection under Sec. 12. Obviously, the Crown copyright protection would lead to the opposite result of the U.S. discussion, according to which such reference could trigger the referred part of the standard to fall into the public domain. Remarkably, the FCA emphasizes the irrelevance of the US case law and doctrine for solving the Canadian case, due to the fundamental differences in the legislative schemes and traditions.¹²⁶

On the merits, the FCA held that Sec. 12 does not apply to the CSA Electrical Code and therefore PS Knight Co. could not rely on Crown copyright to defeat the CSA’s claim. The court explicitly refers to the Keatley Surveying decision of the Court of Appeal for Ontario¹²⁷ and agrees with the Sec. 12 analysis of that court but distinguishes the facts. With regard to statutory Crown copyright, the majority opinion does not see the necessary degree of direction or control of the Crown for preparing or publishing the CSA Electrical Code.¹²⁸ CSA is a private corporation and the Standards Council, which adopts the standards, is neither an emanation of the Crown nor a government department.¹²⁹ With regard to rights under the Crown prerogative, the majority opinion held that stretching the prerogative to codes that have been incorporated only by reference would amount to an “impermissible broadening” of the Crown’s right.¹³⁰ Even though the case had the interpretation of Crown copyright under Sec. 12 at its core and exposes obvious overlaps with Keatley Surveying, the SCC eventually denied its leave to appeal.¹³¹

4 Learning from Both Paradigms: On the Possible Future of Copyright in Government-Related Material

4.1 The Call for Legislative Reforms

It appears like a remarkable coincidence that both the USSC and the SCC dealt with their century-old rules on the copyrightability of government works at the same point of time. But the advancement of information technology and the increasing significance of public–private collaborations in digitization has inevitably assigned copyright (and therefore the initial allocation of ownership) a decisive role and may explain this coincidence. At first glance both courts seem to follow their opposite traditions by progressing along paradigmatically different paths¹³² that their jurisdictions entered centuries ago: the USSC strengthens the public domain in the U.S., while the SCC confirms Crown copyright protection in Canada. But this view would be overly simplistic. A closer look reveals that the paradigms are limited to the federal level to a considerable extent; and the states/provinces can (and actually do) deviate from these paradigms.¹³³ Moreover, both Supreme Courts seem to doubt the contemporaneity and adequacy of the law they had to interpret. It is striking how bluntly both majority opinions highlighted the (self-evident) role of the legislators to amend the law and, therefore, to initiate policy reforms in case they are held to be necessary. From a perspective of international law, the Berne Convention leaves it to the jurisdictions themselves to decide on the copyright status of government works.¹³⁴

In this light, the following wider reflection on the future of copyright law in government-related material must take the rationales and conclusions of both Supreme Court decisions into account. On this basis, the reflection reconsiders the general justification for copyright in government works and highlights the significance of the underlying presumptions (at Sect. 4.2). There are good reasons to argue that copyright should focus on the “incentive to create” problem, which is mostly relevant for public–private partnerships (at Sect. 4.3). A side glance at the difficulties in the regulatory attempts of the EU (which stands before the challenge to reconcile different legal orders and traditions of copyright) to partly harmonize copyright in government works is informative (at Sect. 4.4). Finally, the proposed solution questions whether copyright law is well placed at all to solve the identified problems, and it suggests that additional or alternative instruments appear more promising (at Sect. 4.5).

4.2 Accuracy and Integrity as Questionable Justifications for Copyright

Historically, the Crown prerogative is based on a philosophical justification which follows a non-democratic conception of government. Nowadays, this notion appears obsolete in its entirety, and its flaws also become practically evident when trying – and failing – to explain Parliamentary copyright which the United Kingdom introduced in 1988.¹³⁵ Rather, the leading¹³⁶ and most convincing¹³⁷ argument for government copyright seems to lie in the goal to provide and maintain the accuracy and integrity of government works. However, one can seriously challenge this reasoning, especially in the online environment where one can quickly verify the integrity of sources. Also, it has been repeatedly argued that the goals of accuracy and integrity can be (even better) served through other means than copyright, be it technological¹³⁸ or legal.¹³⁹ This is supported by the consideration that accuracy and integrity relate to information, while the question whether this information is incorporated in a work that is protected by copyright seems arbitrary. Moreover, the argument that a lack of copyright would put accuracy and integrity at risk appears empirically questionable.¹⁴⁰ While official printers also make mistakes,¹⁴¹ economic theory may even support the claim that in competitive markets, private printing may be more accurate than state printing. A reliance on different presumptions as regards the state and the market can explain why the U.S. aims to achieve the same goals of accuracy and integrity but presumes to achieve them with an instrument (namely the public domain as expressed in the government edicts doctrine and 17 U.S.C. § 105) opposite to the Canadian Crown copyright protection.¹⁴²

Another issue challenges the accuracy and integrity assumption. Based on the negative experiences with the Crown prerogative’s censorship,¹⁴³ there is general concern that government copyright can negatively affect public opinion, as it may be used to censor materials, delay access, and chill discussion.¹⁴⁴ From a public interest point of view, it is not the actual practice, but already the potential for abuse and its consequences that matters. And while one may argue that statutory Crown copyright has in fact rarely been used in more recent times to restrain reproduction and copying of government materials,¹⁴⁵ the strategic use of government copyright for withholding information of general interest is becoming increasingly visible in other jurisdictions.¹⁴⁶

4.3 Public–Private Partnerships in the Center of the Creation-Dissemination Trade-Off

The USSC highlighted in Georgia v. PRO the significance of a broad dissemination of government works and pointed to the paramount importance of citizens’ access to legal material in a democracy under the rule of law. This main justification for the absence of copyright for legislative and judicial material appears undisputable as long as it does not touch upon the incentives to create.¹⁴⁷ But the decision has exemplified that this reasoning finds its limits where the costs of creation do play a role and therefore affect incentives. In this case, the state often involves private parties in the fulfillment of public tasks, and exactly such public–private partnerships evoke clashes of interest which eventually end up before the courts. Looking at the development in the last three decades, such public–private partnerships have moved away from merely outsourced editing, printing, and publishing, over mass digitization and online publishing schemes, to collaborations where private actors pursue multifaceted digital business models in the data-driven economy (e.g. smart city context).¹⁴⁸ Copyright (re-)determines the scope of action for the government when it comes to information policies and the entrustment of a third party with fulfilling the public task in particular.

At the end of the day, the scope of action that copyright sets for the allocation of rights in such partnerships is decisive. The U.S. and the Canadian copyright regime actually allow for both: On the one hand, copyright can end up in the hands of the government. This is the default situation under the Canadian Crown copyright regime; but it can also be reached in the U.S., because 17 U.S.C. § 105 explicitly allows a transfer of rights to the U.S. Government. The government may therefore compel government-financed private contractors, who have acquired copyrights in works they have developed, to assign these rights to the government.¹⁴⁹ On the other hand, copyright can end up in the hands of the private actor. For the U.S., this is true in the said cases where 17 U.S.C. § 105 does not preclude private copyrights in government-commissioned works. The Canadian Copyright Act leaves it to the governments whether to retain copyright in its own name or to let the contractor acquire the right. Therefore, public and private actors can agree to derogate from the default position of Crown copyright by contractual agreement to the contrary, which would assign copyright to the private contractor.¹⁵⁰ As can be seen, similar results can eventually be achieved in the U.S. and Canada. However, the opposite default allocation of initial ownership leads to different ways to get there. This matters from a transaction cost perspective. Moreover, the distribution of actual bargaining power between the involved actors predetermines how ownership rights are eventually balanced against the public interest and which arrangements will ultimately be found. This implication is vital for further policy considerations, because bargaining power is highly context-specific and depends on the respective market structure.

It is remarkable that both the USSC and the SCC explicitly highlighted the legislators’ right to amend the law.¹⁵¹ The Canadian legislators could either entirely abolish Crown copyright, clarify the necessary level of “direction or control” as has been outlined by the majority opinion, or follow the concurrence approach and define specific government works. Also, the USSC pointed to possible legislative reforms, but implied the opposite direction which would shape copyright law in a way that the extension of the public domain would effectively find its limits. Ultimately, the question remains which avenues of reform would appear advisable.

4.4 Lessons from the EU Regulation of Public Sector Information

Seeing it realistically, major policy reforms appear unlikely in this area,¹⁵² considering that the state would have to give up its own rights and privileges, at least to some extent. In fact, legislators across jurisdictions are generally hesitant when it comes to legislative reforms on removing protection of government works. This is in contrast to the actual development in the last 20 years, where the online dissemination and the global trend towards “open government” have raised strong calls for extending the public domain on government works. These calls are supported by the innovation argument, according to which a wider dissemination and a lack of copyright protection would lead to more value-added products and services.¹⁵³ But rather than tackling the “ground layer” of copyright protection, policy reforms focused on the “second layer”, which is licensing.¹⁵⁴ Indeed, some evidence may suggest that licensing practices rather than the initial copyright status seem decisive as regards the accessibility of government works and information.¹⁵⁵ However, when pointing to the empirical effect in specific jurisdictions one overlooks the general issue that relying on licensing policies implies dependency on the goodwill and practices of government policies and decisions. Considering that governments and political leadership can change quite unforeseeably and quickly, a regulatory regime on government information that prevents concentrations of information power in the hands of the state appears favorable.

Regulating only the second layer of licensing is also problematic from a market perspective. Looking at the EU policies on Public Sector Information (PSI) confirms this claim. EU jurisdictions vary considerably regarding the copyright status of government-related material. The significant differences are partly based on the different copyright philosophies of the Member States (copyright countries vs. droit d’auteur countries). Yet even the copyright tradition does not clearly indicate the regulatory approach to the copyright protection of government works.¹⁵⁶ EU legislation addressed the re-use of PSI early on. Originally enacted in 2003, the PSI Directive¹⁵⁷ aims to foster the wide re-use of PSI; but it has not touched the copyright status of government works to this day. Rather than the existence of exclusive rights, the Directive regulates the exercise of these rights, and therefore licensing.¹⁵⁸ But harmonizing only the second layer of licenses¹⁵⁹ without tackling the first layer of ownership does not address or correct the deficits related to the different modes of transaction costs and the arbitrary divergence of bargaining power. In addition, market structures and traditions regarding information creation and dissemination differ between the Member States. Finally, licensing provisions may cause new problems because they might mistakenly create “illusionary property rights” for the state.¹⁶⁰

When looking at the fierce debate on copyright protection of government-adopted industry standards that are incorporated by reference, the European jurisdictions also appear informative. While there is no EU-wide harmonization, Germany introduced a compulsory licensing provision in its Copyright Act in 2003.¹⁶¹ This provision aims to balance public access and private incentives to create such standards. It was a reaction to previous decisions of the German Federal Supreme Court, according to which statutory references to such standards could trigger their loss of copyright protection under German law.

4.5 Considerations for Future Reforms on the Copyright of Government-Related Material

Keeping in mind that copyright protection can establish control over information, the technical, economic, and political developments over the last decade may add complexity to the challenge of reforming copyright in government works. The Snowden revelations and the flourishing of the platform economy have considerably fed into a general debate on redistribution of information power between public and private actors.¹⁶² Opinions on whether strengthening the state provides an adequate counterbalance to private power concentrations appear polarized, and skepticism on excessive private power is (re)discussed in the recent competition policy debate.¹⁶³ In any case, one cannot initiate prudent reforms without having a clear and honest view on the implicit presumptions behind the claims of the integrity of state action and the undistorted functioning of markets. Given the dynamic technological advancement and societal change of the last years, it still remains an open question whether and how these presumptions need to be adapted. As for now, there are good reasons to argue that a legal regime that covers government works should require a substantial justification for both over-control by the sovereign and the excessive private propertization of information to the detriment of the general interest.

These considerations lead to a more nuanced proposal for future legislative reforms on government copyright, which is based on the following three presumptions: First, the better reasons (namely to provide broad dissemination and access to information, to foster follow-on innovation, and to prevent government censorship) speak for government-related material falling in the public domain by default. Second, the call for integrity and accuracy is legitimate but does not sufficiently justify copyright protection for government works in itself, as it is based on outdated premises and lacks empirical evidence. Third, most relevant and controversial for a future regulatory regime are situations where the initial allocation of rights provides incentives for public–private partnerships regarding the fulfillment of public tasks. It therefore touches on the creation of works or new services for the benefit of society at large.¹⁶⁴ Regulation (and copyright – if needed here at all) should therefore focus on such public–private collaborations. Yet the analysis revealed that the effect of the initial allocation of rights (state, private party, or public domain) cannot be evaluated in such constellations without accounting for the possibility of contractual derogation/circumvention, the occurrence of transaction costs, the distribution of bargaining power, and the underlying market structure. Copyright in government works does not address these decisive factors systematically. Rather, the copyright system appears ill-equipped to adequately account for the involved private and public interests.

Therefore, regulatory reforms should address clashes of individual and public interest with regards to exclusivity in such public–private partnerships systematically and holistically. This covers the question of initial ownership as well as the contractual layer and its interaction with potential property rights. Also – as seen in the legal approaches to government-adopted industry standards – compulsory licenses can provide an effective instrument to overcome the “incentive to create” problem. Yet a holistic regulatory system would need to go a step further by integrating and matching procurement and competition rules. Moreover, aspects of information re-use should be addressed systematically. This has become especially prevalent, for instance, as regards the use of text- and data-mining and artificial intelligence when it comes to government-related information.¹⁶⁵ Evidently, a holistic regulatory regime would then address information as such, so that it would reach beyond works and, therefore, go beyond copyright. Whatever concrete regulatory regimes legislators eventually choose – the regulatory design should be based on natural skepticism towards both the underlying assumptions about the integrity of state action as well as the undistorted functioning of markets. Recent times demonstrate that both can fail. In this light, transparency rules and justification requirements on public interest grounds are not only sensible tools for future regulation.¹⁶⁶ They can also considerably contribute to preventing that the inglorious early history of the Crown prerogative regarding government-related works will repeat itself.