VOLUME 71 2020 TOME 71
EDITORIAL BOARD—COMITÉ DE RÉDACTION 2020
Editors-in-Chief Associate Editors 2020 CanLIIDocs 3123 MITCHELL MCGOWAN SHANNON BENJAMIN ALEXANDER RIMMINGTON ALEXANDRA DEJONG Rédacteurs en chef CAITLIN GALLANT EMILY MACLEOD Rédactrices adjointes
Honourary Editor-in-Chief Volunteer Associate Editors THE HON. G. V. LA FOREST JOHN CALDWELL Rédacteur en chef honoraire DELANEY STYMIEST-LOSIER Rédactrice et rédacteur adjoints bénévoles
Faculty Advisors ANNE WARNER LA FOREST JANE THOMSON Conseillères academiques
TO BE CITED AS: (2020) 71 UNBLJ MODE DE RÉFÉRENCE : (2020) 71 RD UN-B
ISSN: 0077-8141 Copyright 2020 Droits d’auteur University of New Brunswick Law Journal © Revue de droit de l’Université du Nouveau-Brunswick Printed in Canada—Imprimé au Canada
Since 1947, the University of New Depuis 1947, la Revue de droit de Brunswick Law Journal has published l’Université du Nouveau-Brunswick a articles, comments, and reviews in wide- publié des articles, des commentaires, et ranging areas of law. The Journal is des avis juridiques dans plusieurs produced by students from the Faculty of domaines du droit. La Revue est publiée Law at the University of New Brunswick par les étudiants de la Faculté de Droit de with the advice of one or more faculty l’Université du Nouveau-Brunswick sous members. The objective of the Journal is to la direction d’un ou plusieurs membres du promote academic discussion of current corps professoral. L’objectif de la Revue legal issues, problems, and philosophies. est de promouvoir les discussions académiques concernant les The Journal expresses its gratitude to all préoccupations, problèmes et philosophies past and present supporters including the juridiques actuels. Law Society of New Brunswick, the New Brunswick Law Foundation, the Faculty of La Revue désire exprimer sa Law and Librarians at the University of reconnaissance envers ses sympathisants New Brunswick, and patrons for their passés et présents qui incluent le Barreau ongoing encouragement and assistance. du Nouveau-Brunswick, la Fondation pour l’avancement du droit du Nouveau- The opinions expressed in the Brunswick, la faculté de droit et les
contributions do not necessarily represent bibliothécaires de l’Université du 2020 CanLIIDocs 3123 the views of the Journal. Nouveau-Brunswick, ainsi que ses donateurs pour leur encouragement et leur soutien continu.
Les opinions exprimées dans les contributions ne reflètent pas nécessairement les opinions de la Revue.
Copyright Droits d’auteur Copyright © 2020 University of New Droits d’auteur © 2020 Revue de droit de Brunswick Law Journal. All rights l’Université du Nouveau-Brunswick. Tous reserved. Unless otherwise stated, no part droits réservés. Sauf dispositions of the Journal may be reproduced in any contraires, aucune partie de la Revue ne form or by any electronic or mechanical peut être reproduite sous aucune forme que means, including information storage and ce soit ou par aucun moyen mécanique ou retrieval systems, without permission from informatique, y compris le stockage the Editors-in-Chief. Reproduction of this d’information et systèmes de recherche de material without authorization by any données, sans la permission des method of duplication whatsoever is a Rédacteurs ou Rédactrices en chef. Toute violation of copyright. reproduction de ce matériel par n’importe quelle méthode duplicative sans avoir Except as otherwise provided, the author obtenu l’autorisation est une violation des of each article in this volume has granted droits d’auteur. to the Journal the non-exclusive licence to publish, reproduce, distribute, and use the Sauf dispositions contraires, l’auteur de article in print and electronic form. The chaque article de ce volume a donné un Journal authorizes the reproduction of droit non-exclusif à la Revue de publier, articles to be made for classroom use, reproduire, distribuer et utiliser les articles provided that the author and the Journal are en format papier et électronique. La Revue identified and proper notice of copyright is autorise la reproduction des articles à des affixed to each copy. fins académiques, tant quel’auteur de l’article et la Revue soient identifiés et que l’avis relatif aux droits d’auteur soit inscrit sur chaque copie.
Subscriptions and Back Issues: Abonnements et anciens numéros :
Subscription Price: $15.00 CDN per volume Abonnement : 15,00 $ CA par volume Back Issues: $12.00 CDN per volume Anciens numéros : 12,00 $ CA par volume Index, Vols. 1-36: $10.00 CDN Index, Volumes 1-36 : 10,00 $ CA
Please address your request to: Veuillez communiquer avec :
Editors-in-Chief Rédacteurs en chef University of New Brunswick Law Journal Revue de droit de l’UN-B Ludlow Hall, PO Box 4400 Ludlow Hall, Case postale 4400 University of New Brunswick Université of New Brunswick Fredericton, NB E3B 5A3 Fredericton (N-B) E3B 5A3 Email: [email protected] Courriel : [email protected]
Correspondence and Submissions: Correspondance et soumissions :
The Journal welcomes the submission of La Revue accepte des manuscrits non unsolicited manuscripts. Manuscripts must sollicités. Les manuscrits doivent être be written in either English or French and écrits en français ou en anglais et doivent must be submitted electronically in être soumis sur support électronique en 2020 CanLIIDocs 3123 Microsoft Word format. Citations should format Microsoft Word. Les références conform to the Canadian Guide to doivent être conformes au Manuel Uniform Legal Citation, 9th ed (Toronto: canadien de la référence juridique, 9e éd, Thomson Carswell, 2018). Preference is Toronto, Thomson Carswell, 2018. La given to submissions that are fewer than 60 préférence sera accordée aux manuscrits pages double-spaced. Manuscripts de mois de 60 pages, rédigés à double submitted simultaneously to other journals interligne. Les manuscrits présentés à are not accepted. d’autres revues de façon simultanée ne seront pas acceptés.
All correspondence and submissions La correspondance et les soumissions should be addressed to: doivent être adressés à :
Editors-in-Chief Rédacteurs en chef University of New Brunswick Law Journal Revue de droit de l’UN-B Ludlow Hall, PO Box 4400 Ludlow Hall, Case postale 4400 University of New Brunswick Université du Nouveau-Brunswick Fredericton, NB E3B 5A3 Fredericton (N-B) E3B 5A3 Email: [email protected] Courriel : [email protected]
To be cited: (2020) 71 UNBLJ Reference : (2020) 71 RD UN-B
ISSN: 0077-8141 ISSN : 0077-8141
Printed and bound by KKP in Fredericton, Imprimé et relié par KKP à Fredericton, New Brunswick, Canada. Nouveau-Brunswick, Canada.
2020 CanLIIDocs 3123
UNIVERSITY OF NEW BRUNSWICK LAW JOURNAL
REVUE DE DROIT DE L’UNIVERSITÉ DU NOUVEAU-BRUNSWICK
EDITORS’ PREFACE Mitchell McGowan & Alexander Rimmington ...... 1
PART I: IVAN C. RAND MEMORIAL LECTURE
ENVIRONMENTAL PROTECTION AND THE ABJECT FAILURES OF THE COMMON LAW BRUCE ZIFF ...... 3
PART II: VISCOUNT BENNETT MEMORIAL LECTURE
BREAKING THE CODE: CRYPTOCURRENCY AND PROGRAMMING PROPOSAL 2020 CanLIIDocs 3123 ALLAN C. HUTCHINSON ...... 15
PART III: FORUM—LEGAL ISSUES IN THE MODERN ECONOMY
UNIQUELY TREACHEROUS WATERS: MORE ON THE “SELLER IN POSSESSION” STATUTORY REGIMES OF NEW BRUNSWICK, NEWFOUNDLAND & LABRADOR, NOVA SCOTIA AND PRINCE EDWARD ISLAND Clayton Bangsund ...... 30
PLATFORM WORKERS AND COLLECTIVE LABOUR ACTION IN THE MODERN ECONOMY Bethany Hastie ...... 40
THE UNEASY CASE OF PROGRAMMED OBSOLESCENCE Pierre-Emmanuel Moyse ...... 61
ALGORITHMS AS LEGAL DECISIONS: GENDER GAPS AND CANADIAN EMPLOYMENT LAW IN THE 21ST CENTURY Anthony Niblett ...... 112
A STUDY ON HOW STOCK INDEX FUTURES CONTRACTS CAN BE LEGITIMIZED IN IRAN’S CAPITAL MARKET: A COMPARISON WITH SINGLE STOCK FUTURES CONTRACTS Mohammad Soltani & Nafise Shooshinasab ...... 132
THE EQUITY INCENTIVE CANADIAN STARTUPS NEED (HINT: IT IS NOT STOCK OPTIONS) Bryce C. Tingle, QC……………………………………………………………. 156
PART IV: COMMENT
REVIEW OF ALAN M. SINCLAIR & MARGARET E. MCCALLUM, AN INTRODUCTION TO REAL PROPERTY LAW, 7TH ED. (TORONTO: LEXISNEXIS CANADA, 2017) Paul T. Babie ...... 185
THE FEDERAL COURT OF APPEAL OF CANADA: A VIEW FROM THE INSIDE OF A BIJURAL AND BILINGUAL COURT The Honourable Richard Boivin ...... 190
LOCAL REPRESENTATION AND LITIGANT VOICE: THE STORY OF CLASS ACTIONS IN NEWFOUNDLAND & LABRADOR Suzanne Chiodo ...... 204
THE ABOLITION OF ADVERSE POSSESSION OF CROWN LANDS IN NEWFOUNDLAND AND LABRADOR Gregory French ...... 227
LEGAL ETHICS AND JUDICIAL LAW CLERKS: A NEW DOCTRINAL ACCOUNT 2020 CanLIIDocs 3123 Andrew Flavelle Martin ...... 248
TO PARTICIPATE OR NOT TO PARTICIPATE: JUDICIAL INVOLVEMENT IN THE COMMUNITY The Honourable Mr. Justice Malcolm Rowe, QC, & Dahlia Shuhaibar...275
PART V: STUDENT SUBMISSION
“RECOVERING” FROM COVID-19: INSURING THE PANDEMIC Ryan Chute ...... 295
EDITORS’ PREFACE
Forum: “Legal Issues in the Modern Economy”
The University of New Brunswick Law Journal Editorial Board is pleased to present the forum topic for Volume 71: Legal Issues in the Modern Economy. The forum is inspired by the numerous legal issues surrounding work, inequality, the gig economy, and other topics related to business law and democracy that have come to the fore in the decade following the financial crisis of 2007–08. The significance of these issues has recently been underscored by the COVID-19 pandemic, which has brought about the worst economic downturn since the Great Depression.
Volume 71 begins with two important events in the University of New Brunswick Faculty of Law calendar: the 25th Ivan C. Rand Memorial Lecture, and the 41st Viscount Bennett Memorial Lecture. In October 2019, Professor Bruce Ziff visited the University of New Brunswick Faculty of Law and delivered the Rand Lecture on the abject failures of the common law to protect the environment. Professor Ziff’s lecture was originally scheduled for the previous year, in conjunction with Volume 70 2020 CanLIIDocs 3123 the University of New Brunswick Law Journal. We are delighted nonetheless to include Professor Ziff’s article in Volume 71 one year later.
In January 2020, Professor Allan C. Hutchinson visited the University of New Brunswick Faculty of Law and delivered the Viscount Bennett Lecture on the prospects and challenges of regulating cryptocurrencies. Professor Hutchinson is a distinguished research professor at Osgoode Hall Law School. His article “Breaking the Code: Cryptocurrency and Programming Proposal” is the anchoring submission of Volume 71.
In addition to the Viscount Bennett and Rand lectures, the Editorial Board is honoured to publish two articles from the judiciary. The first is an article on Canada’s Federal Court system, authored by Mr. Justice Richard Boivin of the Federal Court of Appeal. The second is an article on judicial involvement in the community, coauthored by Mr. Justice Malcolm Rowe of the Supreme Court of Canada.
Volume 71 features articles from contributors across Canada and beyond that explore a variety of legal issues in the modern economy. These issues include organizing platform workers, gender gaps in employment law, programmed obsolescence, equity incentives in Canadian startups, “seller in possession” statutory regimes in Atlantic Canada, and legitimizing stock index futures contracts in Iran’s capital market. Volume 71 also includes several articles on subjects beyond the scope of our forum topic. We are delighted to publish two Newfoundland legal history articles—one on class action lawsuits and the other on the abolition of adverse possession of Crown lands—as well as an article on legal ethics and judicial law clerks. The Editorial Board is also pleased to present a book review of former University of New Brunswick Faculty of Law professor Margaret McCallum’s 7th edition of An Introduction to Real Property Law. Finally, Volume 71 concludes with a submission by third-year University of New Brunswick Faculty of Law student Ryan Chute (J.D., ‘20) on insurance law and the COVID-19 pandemic.
The University of New Brunswick Law Journal is the product of many people working together in the pursuit of developing legal discourse. We are extremely grateful to our supporters in this endeavour: the University of New Brunswick, its Faculty of Law, its professors and staff, the librarians of the Gérard V La Forest Law Library, our contributors, our peer reviewers, our subscribers, the Law Society of New Brunswick, and the New Brunswick Law Foundation. We also extend our sincerest thanks and gratitude to our Associate Editors and volunteers for their hard work, dedication, patience, and meticulous editing. Without the labour of Shannon Benjamin, Alexandra Dejong, Caitlin Gallant, Emily MacLeod, John Caldwell, and Delaney Stymiest-Losier, Volume 71 could not have come into being.
Lastly, we are grateful to you—our readers. We hope that you enjoy the contents of Volume 71 and learn something new from the pages that follow.
Mitchell McGowan & Alexander Rimmington Editors-in-Chief, Volume 71 2020 CanLIIDocs 3123
ENVIRONMENTAL PROTECTION AND THE ABJECT FAILURES OF THE COMMON LAW
Bruce Ziff*
A. Introduction
I am honoured to be presenting the Ivan C. Rand Memorial lecture on the twenty-fifth anniversary of this important public event in the UNB law school calendar.
In preparing this lecture, I sought to find intersections between Justice Rand's illustrious career and the present occasion—a talk about environmental law delivered by a visitor from Alberta. As to the latter, it is likely well known that Ivan Cleveland Rand began his professional life as a lawyer in Medicine Hat, Alberta. It might seem like a curious first destination for a freshly minted Harvard law grad from New 2020 CanLIIDocs 3123 Brunswick, but at the time, circa 1913, Medicine Hat was a town with a promising future. Natural gas had been discovered nearby, and there was plenty of legal work available. All in all, it was predicted that Medicine Hat, situated close to the American border, was destined to become a leading urban centre in the recently created province of Alberta. That was not to be. Among other things, the onset of WW1 had a chilling effect of the region's growth. By 1920, Rand decided to return home, where he set up practice in partnership with Clifford Robinson, a seasoned lawyer.
William Kaplan’s biography of Rand’s time ably chronicles his time in Medicine Hat.1 But Kaplan does not mention Rand’s other connection with Alberta. Shortly after hanging his shingle in Moncton, Rand received a telegram from Henry Marshall Tory, the founding president of the University of Alberta. Tory’s message concerned the position of dean of the faculty of law at the university. “Would you consider” Tory wrote in the clipped style of telegrams of the day, “appointment as Dean Law Faculty University Alberta.” The starting salary was stated to be $4,400. Tory was impressed by Rand’s Harvard pedigree. In the last line of the telegram, he advised that “Pound recommends you strongly.”2 Pound was none other than the famous jurist, Roscoe Pound, then the dean of the Harvard Law School. Rand declined, though the extant record does not reveal much else.
* Professor Emeritus, Faculty of Law, University of Alberta. Presented at the Ivan C. Rand Memorial Lecture, Faculty of Law, University of New Brunswick, October 23rd, 2019 1 William Kaplan, Canadian Maverick: The Life and Times of Ivan C. Rand (Toronto: Osgoode Society for Canadian Legal History & University of Toronto Press, 2009) at 18–28. 2 Canadian Pacific Telegraph, Henry Marshall Tory to Ivan C Rand, (nd), copy on file with the author. See figure 1. 4 UNBLJ RD UN-B [VOL/TOME 71
2020 CanLIIDocs 3123
Figure 1. Canadian Pacific Telegraph, Henry Marshall Tory to Ivan C. Rand, (nd).
The second point of intersection concerns Justice Rand’s connection to Supreme Court of Canada jurisprudence on environmental issues. Rand served on the Supreme Court from 1943 to 1959. During that time, he presided in 697 reported decisions. If those cases are cross-referenced in the Westlaw database under the heading “Environmental Law”, only one case is cited. This was a unanimous decision of Court (discussed below3), penned not by Rand, but rather by Justice Patrick Kirwin.
This single reference may be misleading. During Ivan Rand’s tenure on the Supreme Court of Canada, the category of environmental law was not in common usage; it is a cognate legal category of more recent origin. Environmental law in Canada as that term is now understood emerges in the late 1960s, and early 1970s. As Jamie Benedickson noted in the preface of his text on Canadian environmental law, “[w]hen I graduated from law school in 1972, nowhere in the curriculum could one find a course called Environmental Law.”4
As the 1960s were drawing to a close, a number of public interest advocacy groups were created, including preeminently, Pollution Probe (1969) and Greenpeace, which was founded in Vancouver in 1971. In that same year, the Canadian Environmental Law Association was formed. David Estrin, a recent graduate of the Faculty of Law at the University of Alberta, was the driving force behind its creation. The following year, Estrin pecked out the first edition of the Canadian Environmental
3 McKie v KVP Co, infra note 26. 4 Jamie Benidickson, Environmental Law, 4th ed (Toronto: Irwin Law, 2013) at xv. See also David R Boyd, The Right to a Healthy Environment (Vancouver: UBC Press, 2012) where an account of the history of Canadian environmental law (chapter 3, passim) begins in 1969. 2020] ENVIRONMENTAL PROTECTION AND ABJECT FAILURES 5
Newsletter on an IBM electric typewriter. A few years later it became the Environmental Law Reports. In 1976, Estrin opened the first law office in Canada devoted exclusively to environmental law matters. He was, in essence, the founding father of Canadian environmental law.
The relevant ‘law’ governing the environment was viewed fundamentally as being statute-based. A large array of statutory instruments was introduced in and around the same period.5 This alone signals a central theme of my talk: the common law offered very little potent legal protection for the natural and built environment. Legislation filled a huge chasm.
A brief exercise in discourse analysis highlights this point. Benidickson’s work, referred to above, is a student textbook of more than 400 pages. He devotes 14 pages to what he sees as the pertinent common law causes of action.6 A similar scope of coverage can be found in the published casebooks.7 Likewise, a detailed practitioner’s reference book devotes virtually no space to the common law.8 The same is true of the leading resource on Brownfields remediation.9
2020 CanLIIDocs 3123 Benedickson’s analysis is somewhat typical. The causes of action discussed are: nuisance, negligence, strict liability under the rule in Rylands v Fletcher, breach of statutory duty, trespass, and riparian rights.10 The author identifies the shortcomings of these doctrines. In short, except for public nuisance, these causes of action rely on the willingness of private actors to pursue legal recourse; there are issues of standing that can arise; the costs of litigation can be prohibitive; and plaintiff’s may need to contend with formidable issues of proof, especially where scientific questions arise. Limitation periods can also pose a barrier if latent harms take years to gestate into discernable damage. And in the end, the available remedies may be inadequate. In particular, seeking an injunction is pointless in instances in which irreparable environmental destruction has already occurred.11
These concerns are cogently presented. Still, I am surprised by the confined ambit of the analysis. With one exception (riparian rights), the common law is viewed through the doctrinal lens of the law of torts. However, there is far more to the interplay
5 The keystone provision in New Brunswick is the Clean Environment Act, RSNB 1973, c C-6. See also more recent measures: Clean Water Act, SNB 1989, c C-6.1; Clean Air Act, SNB 1997, c C-5.2; Climate Change Act, SNB 2018, c 11; Beverage Containers Act, RSNB 2011, c 121. 6 Benidickson, supra note 4, at 101–14. 7 See Elaine Hughes, Environmental Law and Policy, 3rd ed (Toronto: Emond Montgomery, 2009), 45 of 685 pages; Meinhard Doelle & Chris Tollefson, eds, Environmental Law: Cases and Materials, 3rd ed (Toronto: Thomson Reuters, 2019), 120 of 1,000 pages. 8 Allister R Lucas et al, Canadian Environmental Law, 3rd ed (Toronto: LexisNexis Canada, 2017). 9 See Ahab Abdel-Aziz & Nathalie J Chalifour, eds, The Canadian Brownfields Manual (Markham, ON: LexisNexis Canada, 2004). 10 Benidickson, supra note 4, at 102–14. 11 Ibid at 114–16. 6 UNBLJ RD UN-B [VOL/TOME 71 between environmental protection and the common law than just this. To be more precise, I am thinking of core principles of property law. This omission is ironic, since a good deal of the environment is property. Correlatively, most of the law of property has direct implications for the environment. Accordingly, the remainder of this presentation is devoted to some significant ways in which the common law of property has proven inadequate to the challenge of environmental stewardship. As I hope will be apparent, this is not merely an exercise in semantics and pedantic legal categorization. A large number of ownership principles come into focus when the lens is property law not tort, broadening the analysis in critical ways.
B. Property Law and the Environment
(a) Fundamental principles
Private property has been justified on the basis of a range of values.12 One claim, of major significance in the framing of Canadian property law, is that private property 2020 CanLIIDocs 3123 enhances material well-being. It is trite to say that private property is the mainstay of all market economies. Moreover, this has implications that extend beyond economic efficiency. It has been argued that the financial incentives connected to property rights will typically prompt owners to use their holdings in a way that reduces the imprudent destruction of land and goods. To put the matter starkly, it can be argued that the most important concept in the law of property pertaining to environmental protection is very idea of private ownership itself.
A well-known parable known as the tragedy of the commons is sometimes used to illustrate that assertion.13 The story involves a common pasture. Ranchers and farmers in the region are able to use this pasture for grazing at no cost. It looks ideal: those using the pasture reap the full benefit of the supply of feed, but bear none of the costs of doing so. A rational wealth-maximizer would therefore wish to graze as many cattle as possible. However, so would all others owning livestock. At some point, overgrazing is likely to lead to the destruction of the pasture.
One way in which to avoid this tragedy is to privatize the commons. At that point, an owner using the land for grazing will now absorb the full costs of so doing. A rational wealth-maximizer will need to take steps to ensure that the pasture is not destroyed by overuse. Put another way, both the benefits and burdens of grazing are internalized, that is, reposed in the owner of the land.
The American scholar Robert Ellickson has succinctly described this link between private property (especially fee simple title) resource management, and conservation:
12 See further Bruce Ziff, Principles of Property Law, 7th ed (Toronto: Thomson Reuters, 2018) at 11ff. 13 Garrett Hardin, “The Tragedy of the Commons” (1968) 162:3859 Science 1243. 2020] ENVIRONMENTAL PROTECTION AND ABJECT FAILURES 7
[T]he preeminent advantage of an infinite land interest is that it is a low- transaction cost device for inducing a mortal landowner to conserve natural resources for future generations. … A rational and self-interested fee owner therefore adopts an infinite planning horizon when considering how to the use [a] parcel, and is spurred to install cost-justified permanent improvements and to avoid premature exploitation of resources. The fee simple in land cleverly harnesses human selfishness to the cause of altruism toward the unborn, a group not noted for its political clout or bargaining power.14
This is an attractive argument when examined from a distance. The premise is that self-interest is the motivation for conservation. However, on closer inspection, some critical flaws are revealed. Notice first that in the parable, the cattle are privately owned. The inducement to exploit the pasture results from that allocation. In addition, the commons that Hardin imagined is one of full open-access: there are no built-in restrictions. A well-managed commons can be regulated so as to avoid the tragedy.
The use of private property as a mechanism for environmental stewardship has other limitations. The private owner is incentivized to use the property efficiently and for maximum benefit, which is why he or she will be drawn to adopt appropriate 2020 CanLIIDocs 3123 husbandry measures. However, the actual goal of the rational wealth-maximizer is material (economic) and not environmental well-being. All else being equal, nothing would prevent a private owner of pastureland from converting it into a shopping centre, if that is thought to be the best use of the parcel. Moreover, while each owner may make sensible decisions as to their individuals holdings, there is no assurance whatsoever that these will be co-ordinated with neighbouring owners. A complex web of ecosystems transcends any one property holding. To take a simple example, a decision to clear land for agriculture may destroy wildlife habitats. That may be of little consequence to an eco-system as a whole if only one parcel is affected. But every landowner may see the value in similar action, thereby transforming the habitat beyond recognition. Nothing in the common law compels co-operative use allocations that might prevent that result.15
In recent years, the label of brownfields has been coined to describe lands that have been contaminated by commercial activity. Often these properties contain the toxic by-products of some manufacturing process. The law now requires that owners remediate these properties. Lengthy legal manuals describe the myriad rules governing the mandated restoration processes.16 The tragedy of the commons reasoning suggests that owners would not wish to create brownfields; there is an economic incentive to do otherwise. Yet, the need for legislative responses to the brownfields problem demonstrates the failure of the private-property-as- environmental stewardship model.
14 Robert C Ellickson, “Property in Land” (1993) 102:6 Yale LJ 1315 at 1368–69. 15 Eric T Freyfogle, The Land We Share (Washington, DC: Shearwater Books, 2003) at 157ff. See further Amy Sinden, “The Tragedy of the Commons and the Myth of the Private Property Solution” (2007) 78 U Colo L Rev 533. 16 See e.g. Abdel-Aziz & Chalifour, supra note 9. 8 UNBLJ RD UN-B [VOL/TOME 71
(b) Key relevant doctrines
Among the many property law principles that can have an impact on the environment, there are several that have a particularly strong nexus with conservation and stewardship. Some of these, such as the law governing restrictive covenants, the support of land, and riparian rights, are geared towards conservation. Others are antithetical, or at best agnostic, about that goal. Of particular importance here are the right of destruction and the rule of capture. All of these discrete areas are reviewed below.
(i) Restrictive covenants
The law of covenants running with land can be traced to the 1848 English case of Tulk v Moxhay.17 The dispute there involved Leicester Square, located in the heart of London in an area now known as the theatre district. Some 40 years before the dispute in Tulk arose, the owner of the property had covenanted to preserve and maintain the 2020 CanLIIDocs 3123 lands as a garden. The Court of Chancery held that this promise was enforceable in equity against a purchaser of the land who had notice of the covenant. Leicester Square remains a park, in some measure due to the enduring efficacy of the original covenant.
It did not take long for the principle to take root in Canada. About a year after Tulk v Moxhay was decided, it was cited with approval by a court in Upper Canada.18 The first reported New Brunswick judgment to invoke that authority was decided in 1872. The case involved the enforcement of a positive or affirmative covenant, that is, not one that restricted use, but required action by a subsequent owner (e.g., one acquiring title from the original covenantor).19 Over the next several decades, prerequisites to the application of the concept developed. In addition to the requirement of notice (which was central to the Tulk ruling), it became established that only negative covenants could run with the land. In other words, one could restrict use of the land via a covenant, but one could not compel action by a subsequent owner. Moreover, there had to be some other parcel of land that truly benefited from the promise.20
The deployment of covenants as a mode of conservation—the function of the covenant in Tulk—is still possible. However, the prerequisites grafted onto that ruling—the need for benefitted lands and the prohibition on the enforceability of positive duties—have proven to be limiting. In consequence, legislative improvement has been undertaken in many common law jurisdictions. The New Brunswick
17 [1848] EWHC Ch J34. For an account of the backstory, see James C Smith, “Tulk v Moxhay: The Fight to Develop Leicester Square”, in Gerald Korngold & Andrew P Morriss, eds, Property Stories, 2nd ed (New York: Foundation Press, 2009) 171. 18 Province of Canada (AG) v McLaughlin (1849), 1 Gr 34, 1849 CarswellOnt 2 (UC Ch). 19 Ryan v Lockhart (1872), 14 NBR 127, 1872 CarswellNB 13 (CA), en banc. 20 See further Ziff, supra note 12 at 454 et seq. 2020] ENVIRONMENTAL PROTECTION AND ABJECT FAILURES 9
Conservation Easement Act21 is typical of these initiatives. Under the Act, positive covenants are capable of running with the land.22 And the covenant can exist in gross, provided that the right to enforce the promise is reposed in a government authority or a designated conservation agency.23
(ii) Support of land
The law of support protects against the physical impairment of land by nearby landowners. All else being equal, a landowner is entitled to the physical support of land by all neighbouring properties. Accordingly, excavation on Blackacre that results in subsidence of the surface on Whiteacre is actionable. Indeed, liability is strict, for it does not matter if all reasonable precautions have been taken. In one sense, this protection is potentially far reaching: it applies not only to directly contiguous lands, but any land within the vicinity that happen to undergo subsidence as a result of excavation activities taking place somewhere else. In addition, these rules can apply when otherwise lawful mineral extraction affects the surface owner situated directly above. 2020 CanLIIDocs 3123
Even so, there are some practical limitations on the effectiveness of this protection. One’s land is entitled to support, but not the buildings on that land. Hence, if subsidence results in damage to a structure, recovery for loss of support can only be maintained if it can be shown that subsidence would have resulted even absent the presence of the building. The weight of the building may exacerbate the subsidence, but it must nevertheless be shown that subsidence would have occurred had there been no structure. Only then would damage (and consequential damage to the building) be compensable.24
(iii) Riparian rights
Riparian rights are those enjoyed by landowners in relation to an adjacent body of water. For example, at common law the owner of a riparian tenement is entitled to draw water from a river for certain designated purposes. There are no restrictions on the right of appropriation for ordinary domestic uses. However, the law imposes general limits on more extensive uses, such as manufacturing and large-scale irrigation. When such uses are undertaken by owners of upstream tenements, the downstream owners remain entitled to the natural flow of water “without sensible diminution or increase and without sensible alteration in its character or quality”25. If
21 RSNB 2011, c 130. 22 Ibid, s 11(1). 23 Ibid, s 5. 24 See further Ziff, supra note 12 at 134–35, and the references cited there. 25 30 SLR 964, [1893] UKHL 964 at 965, Macnaughten LJ. 10 UNBLJ RD UN-B [VOL/TOME 71 we focus on the ‘quality’ element, it can be seen that the law confers a protection against water pollution caused by those who own land upstream.
The Supreme Court of Canada decision in McKie v KVP26 exemplifies of this principle in action. The case concerned the Kalamazoo Vegetable and Parchment Co (KVP), which operated a pulp and paper mill on the Spanish River in Ontario. The by- products of the manufacturing process were dumped into the river, resulting in extensive pollution. In 1947, the downstream owners, having exhausted other avenues of recourse, brought suit, seeking an injunction against the company. The action, the basis of which was a breach of the owners' riparian rights, succeeded at trial, and an injunction was ordered.27 That ruling was affirmed with minor variation by both the Court of Appeal28 and the Supreme Court of Canada.
In this case, the riparian rules were shown to have considerable potency, perhaps excessively so. The injunction threatened the survival of a major industry in the community. In response, the province of Ontario decided, controversially, that the court order was an imperfect mediation of local interests, and following the Supreme 29 Court decision, legislation was passed to dissolve the injunction. Even so, the KVP 2020 CanLIIDocs 3123 case and others that followed in its wake signaled the need for a new legislative framework to control pollution, and balance commercial and environmental needs. As a result, modern legislative frameworks impose highly detailed water use frameworks.30 As with the law of covenants, legislation has been required to shore up the deficiencies of the common law.
(iv) The right of destruction
As alluded to above, not only do property doctrines offer little to promote conservation, there is at least one core entitlement that is patently at odds with that goal. The common law offers nothing to prevent an owner, while alive, from destroying his or her holdings.31 Ancient oak trees can be felled; architectural treasures can be razed; historically significant papers can be reduced to ashes; precious jewellery can be transformed into bullion, and so forth. As Robert Sax graphically put, an art lover may play darts with the Rembrandt in a private collection if that is desired.32
26 KVP Co Ltd v McKie et al, [1949] SCR 698, 4 DLR 497. See further Jamie Benidickson, “KVP: Riparian Resurrection in 20th Century Ontario”, in Eric Tucker et al, eds, Property on Trial: Canadian Cases in Context (Toronto: Osgoode Society for Canadian Legal History & Irwin Law, 2012) 71. 27 McKie et al v KVP Co Ltd, [1948] 3 DLR 201, OR 398 (HC). 28 KVP Co Ltd v McKie et al, [1948] OWN 812, [1949] 1 DLR 39 (CA). 29 The KVP Company Limited Act, SO 1950, c 33. 30 Clean Water Act, SNB 1989, c C-6.1. 31 See generally Lior J Strahilevitz, “The Right to Destroy”, 114:4 Yale L J 781 (2005). 32 Joseph Sax, Playing Darts with a Rembrandt: Public and Private Rights in Cultural Treasures (Ann Arbor: University of Michigan Press, 2001). 2020] ENVIRONMENTAL PROTECTION AND ABJECT FAILURES 11
Absent legislation, the only protection against such action is the self-interest of the owner to do otherwise.
There are a few limited exceptions to the owner's right to destroy at common law. One is offered by the law of waste. Where one has a limited proprietary interest in land, such as a life estate, that interest-holder is prevented from undertaking most acts of destruction, and even the extensive improvement, of the land. This measure is designed to preserve the property for those ultimately entitled to the fee simple in the lands.33
It would also appear that the right to destroy may be circumscribed when contained in a testamentary disposition. There are only few authorities on point. The leading authority on point in Canada is the New Brunswick decision in Re Wishart.34 That case involved a will in which the testator directed that his four horses be put down. The validity of the clause was called into question; indeed the matter came to national attention. In the end, the Court gave a rather generous (and dubious) interpretation of the language in the will. It was concluded that the testator had been motivated by deeply held concern that the horses would not be properly treated after 2020 CanLIIDocs 3123 his passing. Their future having been assured by arrangement made by his estate, it was held that the testator’s worries had been allayed, rendering the clause unnecessary. Importantly, it was held in the alternative that, if that reading was not tenable, the clause was void as being contrary to public policy:
In my opinion, the destruction of four healthy animals for no useful purpose should not be upheld and should not be approved. To destroy the horses would benefit no one and would be a waste of resources and estate assets even if carried out humanely.35
But, to return to the main rule, had Wishart decided that the horses should be put down prior to his own demise, the common law posed no impediment. Moreover, as long as the animals are killed in a humane manner, there does not appear to be criminal liability, as matters now stand.36
33 Still, the right to commit waste may be waived in the granting document, or permitted by those entitled to the remainder. See further Ziff, supra note 12 at 208–10. Waste makes a cameo appearance in the law governing Aboriginal rights, where it is invoked as a means of defining the permissible uses of lands held under Aboriginal title. The restrictions were said to resemble the limits defined by the concept of equitable waste: Delgamuukw v British Columbia, [1997] 3 SCR 1010 at para 130, 153 DLR (4th) 193. I do not think the analogy is apt: see Bruce Ziff, “The Supreme Court, Fundamental Principles of Property Law, and the Shaping of Aboriginal Title”, in Paul Daly, ed, Apex Courts and the Common Law (Toronto: University of Toronto Press, 2019) 385 at 396–98. 34 Re Wishart Estate (No 2) (1990), 129 NBR (2d) 397, 1992 CarswellNB 69 (QB) [Re Wishart cited to NBR]. 35 Ibid at para 23, Riordon J. 36 Morris Manning & Peter Sankoff, Criminal Law, 5th ed (Toronto: NexisLexis, 2015) at para 24.59. 12 UNBLJ RD UN-B [VOL/TOME 71
(v) First occupancy
Some forms of unowned property can be acquired by the first party taking possession. The classic instance of this doctrine involves the capture of wild animals. A party first acquiring possession of such an animal obtains title.37 The title is qualified in that it may be lost if the animal escapes. At first glance, one might take this idea to be sensible, fair, and environmentally neutral. However, these assessments are contestable. As James Krier has warned, capture “has been shown in a number of instances to result in relatively rapid depletion rather than long term conservation because it induces people who seek to exploit common property resources to gear up, to get more, and to get it faster.”38 Two examples illustrate Krier’s concern.
The case law regulating the Newfoundland seal hunt bears this out. For centuries the annual seal hunt occurs in a span of just a few weeks, usually in early Spring. In the 19th century, the practise was for sealing crews to stockpile pelts on ice pans (floes), mark them, and set off on the hunt again, returning to retrieve the catch a later time. Occasionally, the initial ship (call it Ship #1) would not be capable of returning to recover the pelts; those might later be collected and stowed by a 2020 CanLIIDocs 3123 competitor (Ship #2).
During the latter half of the 19th century, a handful of cases confronted the question of the entitlement, if any, of Ship #2. Of concern to the courts and the sealing industry was the avoidance of waste. In one case it was noted that some 3,000 sculped pelts had been left behind, their fate being unknown.39 In the 1870 decision of the Full Court of the Newfoundland Supreme Court in Clift v Kane,40 a majority held that Ship #1 retained title to pelts left on a pan, but that Ship #2 would entitled to levy a salvage fee in circumstances in which the pelts might otherwise have been lost. An alternative approach, which also garnered judicial support, holds that where Ship #1 is unable to recover the pelts, they should be treated as having returned to the commons, and can therefore be fully claimed by Ship #2.41
The rule in which the ownership of Ship #1 is preserved, subject to a salvage claim by Ship #2, seems fair enough. It allows the contributions and interests of both crews to be recognised. But consider how this rule might affect incentives. If Ship #1 has collected enough pelts to fill is hull, it need not return to port. Rather, it could continue the hunt, for any pelts it then collects on the pans remain its property (subject only to the payment of a salvage fee). To be blunt, there is no reason for Ship #1 to stop killing. Moreover, the sealers on Ship #2 may elect not to salvage those pelts.
37 The locus classicus is Pierson v Post, 3 Cai 175 (NY Sup Ct 1805). 38 James E Krier, “Capture and Counteraction: Self Help by Environmental Zealots” (1996) 30:4 U Rich L Rev 1039 at 1052. 39 Doyle v Bartlett (1872), 5 Nfld LR 445 at 454 (SC), en banc. 40 Clift v Kane (1870), 5 Nfld LR 327 (SC), en banc. 41 See further the analysis in Bruce Ziff, “The Law of Property in Animals, Newfoundland-Style”, in Tucker et al, eds, supra note 26 at 9. 2020] ENVIRONMENTAL PROTECTION AND ABJECT FAILURES 13
Rather, it may prefer to allow those pelts to be lost in favour of its own catch, for which it can receive one hundred cents on the dollar. The salvage rule is, therefore, potentially quite wasteful.
The rule of capture for wild animals has been applied by analogical extension to other fugacious substances, such as oil and gas deposits. A natural reservoir of these substances may be found to extend below any number of surface parcels. Extraction on Lot #1 can lead to a depletion of the minerals under the others. As long as the owner of Lot #1 does not commit trespass, say, by undertaking drilling operations on or under nearby lands, the entire reservoir can be lawfully drained from a well on Lot #1. That principle was affirmed in the Privy Council decision in Borys v Canadian Pacific Railway:
If any substance is withdrawn from [its] property, thereby causing any fugacious matter to enter [its] land, the surrounding owners have no remedy …. The only safeguard is to be the first to get to work, in which case those who make the recovery become owners of the material which they withdraw from any well which is situated on their property or from which they have authority to draw.42 2020 CanLIIDocs 3123
The moral of this case is that, as the quote reveals, the law actually encourages the rapid depletion of the oil and gas reserves, just as Krier has warned.
C. Conclusion
In this presentation I have tried to demonstrate that the common law principles governing property—and not just those pertaining to tort—have largely failed to advance the interests of environmental protection. Every book you pick up on environmental law, layered as it invariably is with statutory fixes of all kinds, reminds us of the inadequacies of the common law.
The main reasons for the law’s failings are not difficult to discern. One involves structural or institutional issues. The same difficulties that impair the use of tort law (costs, issues of proof, standing, and so forth43) apply to civil actions brought to enforce property rights. Moreover, the common law is a blunt tool. The case-by- case development of the law can produce broad guiding principles, but will often be unable to provide a nuanced and calibrated conservation regime that is responsive to growing needs and emerging scientific discoveries. Of greater importance, while one can detect a conservationist strain immanent in certain common law doctrines (such as riparian rights), these principles are often in tension with other values that inform private property rights in liberal-democratic politics. The law promotes development
42 Borys v Canadian Pacific Railway, 1953 CarswellAlta 25 at para 6, [1953] 2 DLR 65 (PC), Porter LJ [emphasis added]. 43 See Part A, above. 14 UNBLJ RD UN-B [VOL/TOME 71 and the pursuit of material well-being. As mentioned above,44 while that end can be congruent with stewardship of the natural environment, that may not always be so.
In the last 30 years, a supposedly new way of thinking about legal frameworks for environmental protection has emerged. The label often used to describe this reform is “free market environmentalism” (FME).45 Terry Anderson, a leading proponent of FME, has described its fundamental nature:
… the first premise of FME is that “wealthier is healthier,” meaning that markets generate the wealth that gives us the wherewithal to solve environmental problems. Although many people mistakenly think that markets can only generate consumerism[,] … in reality it is markets that produce wealth and thus help the environment.
The second major premise of FME is that “incentives matter.” Positive incentives can turn the environment from a liability into an asset for a resource owner. If we own the water and land, we have the incentive to manage and conserve them properly.46
2020 CanLIIDocs 3123 In 2015, Anderson reflected back on his earlier writing on the subject, offering that in the early 1990s, “property rights economics was still in its infancy, our application of property rights to environmental issues was only a bit beyond the gestation stage”.47 I find this characterization of FME revealing, for it ignores, or at least downplays considerably, the historical antecedents discussed in this talk. At root, free market environmentalism is not new; it's as old as the right to private property itself. The underlying premise of FME draws us right back to the tragedy of the commons, and in fact FME has been promoted by reference to that parable.48 The framing of free market environmentalism as a fresh approach exposes the problems that can surface by seeing the common law's contribution to the field of environmental law as primarily a matter of tort law. Property doctrines have a sad story to tell as well. For centuries, the main protective device has been private property. That must be seen as a critical reason why we now find ourselves in environmental peril.
44 See Part B(a), above. 45 See Terry L Anderson & Donald R Leal, eds, Free Market Environmentalism for the Next Generation (New York: Palgrave Macmillan, 2015). 46 Terry Anderson & Candace Jackson Mayhugh, “Free Market Environmentalism Explained”, Hoover Digest, 1998, vol 2, online:
BREAKING THE CODE: CRYPTOCURRENCY AND PROGRAMMING PROPOSAL
Allan C. Hutchinson*
The problem to be faced in regulating cryptocurrency is the general thrust of the ‘governance paradox’1—how do you regulate an innovative scheme that demands some regulation in the public interest, when you know that any regulation will transform the very features of that scheme that not only makes it what it is, but also makes it especially useful and attractive to that same public? More specifically, how do you regulate an off-the-grid, decentralized and distributed scheme without making it into an on-the-grid, centralized and undistributed scheme? This is the challenge to
be met in devising any kind of proposal to create a tailored and efficacious regulatory 2020 CanLIIDocs 3123 regime for cryptocurrency. Consequently, in making this effort, it will be important to remember that regulation is not a technical end in itself, but a means to a larger and more substantive end. As regards cryptocurrency, this means that regulation must serve to advance and protect broader social and democratic goals—the shared notion of putting ordinary people and their interests at the heart of any regulated society, not those of many civic or state-controlled institutions that tend to put their own interests ahead of others. Accordingly, any proposals to regulate cryptocurrency must be guided by that broader and more encompassing ambition.
In this paper, I explore how to go about that exciting and, some might say, daunting task of designing and implementing such a regulation scheme. In the first section, I examine the present self-regulatory arrangements that underpin cryptocurrency; it is important to appreciate where things presently stand if there is any chance of making progress forward. In the second section, I pull back the institutional curtain and reveal the people and processes that maintain the blockchain technology that drives cryptocurrency; the nature of its operation and alleged consensus-based character are examined. The third section looks at some of the objections that are made to any effort to monitor the work of the code-makers and programmers; an important distinction is made between what is possible and what is desirable. In the fourth section, I draw comparisons and contrast between programmers and corporate directors with an eye to adopting some disciplinary strategies from corporate governance. The fifth section examines the difficulties to applying a
* Distinguished Research Professor, Osgoode Hall Law School, York University, Toronto, Canada. I am grateful to Taylor Trottier, Eyoel Negaye, Murad Wancho and, especially, Eliza Coogan for critical assistance and intellectual support. Thanks as always to Jennifer Leitch. I am grateful to the faculty and students at UNB Law School for the generosity and hospitality around the delivery of the 2020 Viscount Bennett Memorial Lecture. 1 Kevin Werbach, The Blockchain and the New Architecture of Trust (Cambridge, MA: The MIT Press, 2018) at 133–38. 16 UNBLJ RD UN-B [VOL/TOME 71 modified fiduciary duty to the work of programmers. Throughout the chapter, the goal is to lay bare the inner workings of cryptocurrency so that a better and more effective job can be made of regulating it in a sensible and sensitive manner. Taking seriously Lessig’s notion that ‘code is law’,2 I strive to bring the code-makers into the disciplinary fold.
In Those We Trust
Cryptocurrency is not so much an unregulated domain, but a self-regulated sphere of activity. While this kind of regulation is not the kind of central or governmental intervention that engages and enrages critics and supporters, it is a foundational aspect of cryptocurrency. In so many ways, cryptocurrency is entirely dependent on blockchain technology and, therefore, those who create, organise and maintain it. There can be no cryptocurrency without a very structured and sustained set of programs, codes and protocols that combine to form the underlying blockchain- technology of cryptocurrency. As such, one of the primary and neglected entry-points for possible regulation is the small, but influential group of programmers who have so 2020 CanLIIDocs 3123 far largely flown under the regulatory radar. With the power to maintain and change the codes and protocols of the blockchains, they are the heirs to Satoshi Nakamoto’s originating vision and have enormous responsibility and power, especially within permissionless systems. Even if the credo of crypto-programmers is to “reject kings, presidents and voting,”3 they will, if only by default, assume those royal and almost autocratic powers.
Although cryptocurrency users do not need to trust any central intermediary or other transacting party, they have no other choice than to trust the technology itself. Indeed, as a way to alleviate the need to trust others, the blockchain demands that you trust the system of algorithmic and cryptographic proofs and the software that enables the blockchain platform to underpin cryptocurrency transactions. This can be termed the lex cryptographica: it is an informal governance tool that both enables and puts limitations on what can and cannot be done. Indeed, it is the code and protocols that comprise the network itself that must be trusted. Of necessity, it determines the nature of people’s interaction within the network by channeling and constraining those interactions; there is no network without a code to realise it and there is no cryptocurrency without a network. As such, the defenders of cryptocurrency as a trust- free mode of interaction and financing must stake their claims on the controversial assertion that “technology is much more trustworthy than people.”4 There are obvious
2 Lawrence Lessig, Code: And other Laws of Cyberspace, Version 2.0 (New York: Basic Books, 2006) at 1. 3 Niels ten Oever & Kathleen Moriarty, “A Novice’s Guide to the Internet Engineering Task Force” (8 November 2018), online: Internet Engineering Task Force
First, it seems axiomatic that there are all modes of technology are vulnerable to error or manipulation in one way or another. Indeed, the short history of cryptocurrency itself demonstrates that distinct possibility. The Mt Gox fiasco in 2014 is perhaps the most well-known. Hackers were able to infiltrate this busy Tokyo-based Bitcoin exchange (with about 75% of all Bitcoin transactions); not only were millions in Bitcoin lost, but it destabilised the entire global cryptocurrency market. Although on a lesser scale, the antics of Quadriga resulted in investors losing C$200M; the death of the sole founder and operator left investors with no way to access the various e- wallets in which the company’s passwords were held. Secondly, unquestioning faith in the trustworthiness of cryptocurrency’s technology drives home the crucial point from a regulatory point of view that technology is only as good as those who design, run and maintain it; the programmers and code-makers must be trusted to fulfil their responsibilities and exercise their powers in a professional, competent and scrupulous manner. However (and again), experience teaches that, no matter how professional, competent or scrupulous those people are, there will be mistakes and errors. Of course, 2020 CanLIIDocs 3123 if the code-makers are not professional, competent and scrupulous (as in the Mt. Gox and Quadriga situations), trouble and turmoil will lie ahead. In such circumstances, the introduction of some regulation of programmers seems to be not only wise, but close to essential; the legitimacy of the cryptocurrency demands nothing less.
Accordingly, it is with programmers that regulatory efforts might be able to intervene at ground-zero in the cryptocurrency world. By addressing the work and world of the code-makers, a more innovative mind-set might be able to intervene in ways that are both effective and consensual. The regulatory impulse might be able to influence the lex cryptographica and engineer the kind of changes, like a scaling-back of the system’s pseudonymous characteristics, that might be demanded. Indeed, whatever crypto-purists might demand and defend, the maintenance of such a characteristic is both unwarranted and indefensible.5 In short, any agency or institution entrusted with regulatory responsibilities agency might seek to nudge and chivy the software guardians of the blockchain to design and build code that instantiates and reflects the kind of values and incentives that would be thought to best advance the goals of a more fairly and lightly regulated cryptocurrency world. Indeed, by so acting, these latter-day heirs to Nakamoto might begin to instantiate in the overall and animating benevolent spirit of that originating genius. This holistic approach to regulation would allow a blend of the lex cryptographica with what might be termed the lex traditionis for the mutual benefit of each.
Importantly, any intervention that targeted the programmers and core developers would also have a very significant and attractive effect—it would permit the blockchain to remain its own regulator by continuing and developing an internal
5 It is extremely difficult to understand why cryptocurrency should be relieved of criminal and tax obligations. However, a degree of protection might be appropriate and possible by the imaginative use of judicial screens and similar institutional devices. See Allan C Hutchinson, Paying The Price: Cryptocurrency and the Regulatory Challenge [forthcoming 2020]. 18 UNBLJ RD UN-B [VOL/TOME 71 mode of algorithmic governance.6 This is a tantalising prospect. Despite the arguments put forward by the good faith defenders of the status quo (as opposed to its less savoury and ill-intended ones who seem keen to protect illicit activity), the present structure and operation of permissionless systems of cryptocurrency can be enhanced by the right kind of regulation. The main arguments against directing regulatory initiatives— ‘free speech’ and impracticality—do not hold water. They are diversionary and last- ditch tactics more than they are genuine and serious obstacles to regulation. As such, the effort to bring the code-makers within the regulatory field of consideration are worth pursuing; they hold the potential to both unlock and boost the beneficial possibilities of targeting the work and influence of programmers.
Before recommending how to go about this challenging task of regulating the programmers and protocol-makers, it is important to explain who these people are and how they operate. For a process that is touted for its transparency and ‘trust-free’ qualities, there is a definite amount of mystery that swirls around how the technological integrity and maintenance of the system is achieved. Indeed, the identities of the core developers are far from simple to discern and their modus operandi is also far from transparent to the uninitiated eye. Nevertheless, as they are 2020 CanLIIDocs 3123 located at the dynamic heart of the cryptocurrency enterprise, it would be foolish not to look more deeply into the code-makers’ mandate and canvass the possibilities for bringing them more squarely into the regulatory fold.
At The Core
The first core developer (and, therefore, first regulator), of course, was Satoshi Nakamoto. He/she/or they created and implemented the blockchain technology in January 2009. However, Nakamoto’s involvement in this development role only lasted for about 22 months until December 2010. At that time, Nakamoto had over 1 million Bitcoins in their name; these remain inactivated and are now worth close to US$1 billion. Their activation would have a serious and negative effect on the price of Bitcoin. During Nakamoto’s short tenure as the chief code-maker, Nakamoto made several small alterations to the blockchain software that maintained its efficiency and security, but did not affect its basic operating process and workings. Nakamoto’s last e-mail message was in April 2011 and they have not been heard from (or identified) since then. In effect, Bitcoin and blockchain technology has its own ironic and myth- making genesis-story; Nakamoto plays the role of benevolent creator.
When Nakamoto bowed out in 2010, the responsibility for maintaining and taking care of the source-code and operating protocol was handed over to five people. This process of the existing developer or developers deciding who gets to be part of the core development team has become part of the Bitcoin tradition. Programmers are invited into the core group as a result of having built up a strong reputation as highly competent and solidly reliable within the existing group of core developers: it seems
6 See generally Primavera De Filippi & Aaron Wright, Blockchain and the Law: The Rule of Code, (Cambridge, MA: Harvard University Press, 2018) at 193–204; Werbach, supra note 1 at 157–60. 2020] BREAKING THE CODE 19 very much an inside or elite practice for an overall enterprise that is supposed to be open and distributed. Since 2010, there have been only 14 core developers who have access to the software to maintain, modify and update the blockchain programs. There are presently six core developers—Wladimir van Der Laan, Pieter Wuille, Jonas Schnelli, Marco Falke, Samuel Dobson and Michael Ford. Among this group, one of them is entrusted with the considerable power and responsibility to lead the overall project and coordinate any modifications or developments. After Nakamoto, Gavin Andresen took on this role until April 2014. It is now occupied by Wladimir van der Laan; he is funded by the Digital Currency Initiative at the Massachusetts Institute of Technology (MIT).
However, as powerful as this group are, it would be misleading to give the impression that this elite band of core developers amount to a technocracy that wield absolute power and exercise untrammelled authority. The convention is that these programmers are only supposed to act and make changes to the basic blockchain technology when there is sufficient consensus among the larger group of contributors (of which there are presently around 600 or so) and the larger community of Bitcoin users. As such, in order to assess the power of the core developers and possibilities for 2020 CanLIIDocs 3123 regulating them, it is essential to have a better grasp of when and how changes to the basic source-code can be implemented.
There exists what is termed a Bitcoin Improvement Proposals (BIPs) process. All of this occurs through a communications platform, titled the GitHub repository where BIPs can be proposed and discussed. This is a relatively elaborate and involved process that can be generally understood as comprising five distinct and important phases:
• Submission of proposals—anyone is free to make a proposal for changes to the Bitcoin network’s code and protocols provided that they follow the standardized process as outlined on GitHub. There are presently about 600 or so contributors to Bitcoin Core. Since Bitcoin’s creation, there have been 10s of thousands of proposals. Only a relatively small number of them are ultimately accepted for implementation. However, it is estimated that more than 75% of Nakamoto’s original source-code has been changed or discarded in its 10 years of existence;
• Discussion and editing—once a BIP is made, there is considerable discussion and commentary about the proposed changes on GitHub. Again, anyone can contribute and suggest revisions or alterations to the proposal to make it more feasible or likely to obtain approval and implementation. By relying on the participation and insights of other contributors, there is intended to be a greater degree of accountability and cooperation. However, not surprisingly, more weight is given to the proposals and comments of those contributors who have a strong track-record and lengthier involvement than others. The ambition of this discussion is to garner sufficient support for particular proposals 20 UNBLJ RD UN-B [VOL/TOME 71
so that a momentum is generated to improve the chances of community consensus;
• Community consensus—in order to have a chance at success, any BIP must be able to generate a rough consensus. Exactly what this amounts to is not entirely clear. In generating this level of collaboration and agreement, the views and interest of miners are given special consideration; there are about 20 mining pools with a total of over 2,000 miners involved. Their participation and approval is vital because they are the ones who validate transactions and, thereby, create more Bitcoins. Without them, the whole process would stagnate. Consequently, it appears that there needs to be a very high threshold of miners who are on-board with a particular BIP—as much as 95%—before it can receive the necessary degree of community consensus;
• Implementation—once the code has been reviewed and generated the
required rough consensus, the core developers will take a sterner 2020 CanLIIDocs 3123 look at the BIP to ensure that it is compatible with the general principles of the overall Bitcoin project and that it will enhance the performance and technical integrity of the Bitcoin protocol. This confers considerable power on the core developers because not all BIPs with a rough community consensus make it through to implementation.7 Those BIPS that are accepted by the core developers are then implemented and made available for public adoption; and
• Community upgrade—even if a change is implemented to the governing code, it will not automatically become part of the users’ network by way of a centralized decree. Each node operator must take steps to update and upgrade the code that they run and are operating with. Consequently, even after discussion and review, the users have the final say on the acceptability of any new change or alteration. For those who champion the truly decentralized, consensual and bottom-up nature of Bitcoin technology, this user- oversight is a vital feature of the system’s set of checks and balances.
The rationale for proceeding in this way is that, as an open-source and distributed network, the fate of Bitcoin depends on its continuing security, enhanced technological integrity, broad participation, and effective decentralization. The challenge is to ensure that no person or group can hijack Bitcoin and make changes that do not serve or adversely affect the interests of the Bitcoin community. However, the reality is not so comforting. Despite the general and genuine sense of community spirit, Bitcoin (and other related blockchain-based entities) are more of a faux- democracy. They are long on rhetoric, but short on action. For instance, although the
7 See Forking Off below. 2020] BREAKING THE CODE 21 leading core developer, van der Laan, insists that changes are made by the core developers in a “janitorial” or housekeeping way, he also stated that the “GitHub repositories are not democratic… [and] difficult technical issues are not solved by popular voting.”8 This is far from a democratic commitment or even the appearance of one.
Nevertheless, there is much in this system of upgrading and modifying that warrants serious respect and attention. In particular, it makes the regulatory task even more challenging and difficult; this is especially so with the practice of giving users the ultimate ability to accept or reject any implemented changes. In light of this, some throw up their hands and hold that regulation is unfeasible—the only choice is between an outright ban on permissionless cryptocurrency or a complete hands-off stance to it. Accordingly, my favoured approach of regulation-lite becomes even more impractical and even impossible in some people’s eyes, whether they are supporters or critics of cryptocurrency. However, I remain chastened, but undaunted. Behind the front of impregnability and distributed power, I maintain that there is still a viable and pragmatic possibility for regulation. Moreover, I also insist that such regulatory interventions can actually improve cryptocurrency’s operation and facilitate its 2020 CanLIIDocs 3123 broader acceptance. This is no easy task, but is an achievable one.
Forking Off
The first and most serious objection to targeting programmers or code-makers is that, whatever its claimed benefits and advantages, such regulation is simply not possible. They are constrained by and held hostage to the wishes and control of the broader cryptocurrency community. In other words, they are simply amanuenses or janitors who respond to and do the bidding of their communal bosses and leaders. Moreover, whatever they decide to do, they are open to correction and admonition by those very same community members; users can simply refuse or fail to adopt any changes made by the core developers to the operating platform and protocols. In short, the select group of core developers follow, not lead; they are not kings, but king-makers. Consequently, it is argued that it would be pointless and a waste of organisational resources to take steps to reorganise or monitor their programming behaviour. This is no small hurdle to surmount for any attempt at regulation. However, it is important to remember that there is a vital difference between what can be done and what programmers (and users) think can be done or would want to be done.
The past decade offers several examples of both what programmers or core developers can do and how those efforts are received by users. There have been two general kinds of change that have been made—soft forks and hard forks. As regards soft forks, these are small and incremental changes to the code that help with the overall functionality of the existing system (e.g., readability and memory usage). These changes are backwards compatible in that the blockchain itself is not changed
8 Danny Bradbury, “Why Bitcoin’s Core Developers Want Multiple Versions” (19 October 2014), online: Coindesk
The other and more controversial kind of changes are hard forks. These have more significant effects on Bitcoin’s operation because they generally change the consensus rules (i.e., those that demand almost universal). The consensus rules are the technical rules that all Bitcoin clients must adhere if the network is to continue working properly. A hard fork, therefore, is one that makes changes or up-grades that are not backwards compatible with the previous version and, therefore, demand a break with or fork from the existing blockchain. There are two well-known examples of this. In August 2017, there was a deep-seated disagreement over how to handle increasing congestion on the blockchain. A contentious change was implemented that obliged users to decide whether they wanted to stay with the original blockchain (Bitcoin) or to go with the new blockchain (Bitcoin Cash). Because one only recognised 1MB 2020 CanLIIDocs 3123 blocks as valid and the other recognised 8MB blocks, the two were incompatible. In 2018, Bitcoin Cash itself forked into two cryptocurrencies—Bitcoin Cash, and Bitcoin SV. The core developers, of course, played a major role in all this and, depending whose side you were on, were either the heroes or villains of the piece.
Some draw upon these examples to support their argument that the core developers are simply not a viable target or entry-point for regulatory intervention. Because their power and impact are conditional at best, it would be futile to hinge regulation on their programming efforts. Whether the core developers do or do not implement changes to the basic protocols and operations of the blockchain by way of a change to the consensus rules, the users get to decide whether they adopt them, ignore them or implement their own changes. The ultimate option for users is to fork and establish their own version of a blockchain-enabled cryptocurrency that is incompatible with other versions and does not have a shred history with earlier versions: the original core developers will not have access to or involvement in the new operating software. Examples of this possibility are Litecoin and Dogecoin; they did not split the blockchain itself, but altered the source-code so that new blockchains were created entirely. All of this demands that users, if they are to exercise this ultimate control, must be aware of any changes, appreciate its significance, be prepared to take a stand, and persuade others to get on board with them. Importantly, none of this was antithetical to Nakamoto’s original vision. Indeed, it can be argued that it fits aptly within the original vision of a truly decentralised, distributed and user-empowering system.
So where does this leave any recommendations to focus on the programmers as a major entry-point for regulation efforts? How is it possible and effective, let alone desirable, to bring programmers and their programming products into a regulatory framework that can be circumvented by users? Many might say that this leaves my project nowhere: programmers and particularly users will end-run any efforts at regulation. They will either set up new versions of cryptocurrency or refuse to migrate 2020] BREAKING THE CODE 23 to new versions of the operating program. This response is all well and good in theory: it is true that users can do much to evade the regulatory reach of any more centralised or state-connected organisation. However, this response begins to lose much of its traction and force when placed in a more realistic and practical context. When this is done, a sensitive and light-handed approach to putting programmers in the regulatory mix becomes more realisable and possible.
There are only so many forkings that can occur without undermining the whole utility of the cryptocurrency market. As the number of cryptocurrencies on offer grows further, they will become less attractive to both transactional users and investors. For the transactional users, the attraction of a widely-used digital coin is that they will be less restricted in using it for commercial purposes among a larger community of users; niche-cryptocurrencies have much less appeal for trading and transacting. Consequently, although the option of constantly shifting loyalty between different crypto-brands is available, it will sooner or later become less attractive and viable from a user’s own cost-benefit perspective: the avoidance of regulatory intervention at some point becomes self-defeating. Of course, that might not be the case for those who want to engage in money-laundering or related illicit activity; they 2020 CanLIIDocs 3123 will want to be beyond regulatory surveillance at almost all costs. However, this can hardly count as an argument against regulation; it is exactly those kind of activities that need to be rooted out and prevented.
For those who use cryptocurrencies for investment purposes, the constant fragmentation into more and more alternative digital products is likely not only to increase the existing price-volatility of cryptocurrencies, but also to drive many existing and would-be investors from the crypto-market entirely. Indeed, the brief history of cryptocurrency to date suggests that, despite the hard forks and new alt- coins that result, Bitcoin will continue to remain the primary and leading cryptocurrency: it is more likely to consolidate its dominant position than lose it. As such, it is fair to say that users seem to have calculated that the costs of moving away from Bitcoin are greater than the costs of remaining with it. In short, the greater technological sophistication or functionality of other cryptocurrencies that might elude regulatory reach is insufficient to offset the relative stability and reliability of remaining with Bitcoin. Consequently, it is far from clear, even from the users’ standpoint, that regulatory engagement and focus on programmers will be the bane that many crypto-insiders predict or fear, especially if such regulation is built more around innovative incentivisation than interventionist commands. At bottom, therefore, a more realistic dilemma for users, whether of a transactional of investor-oriented kind, is not a choice between remaining with regulated and unregulated cryptocurrencies. Instead, it is a choice between staying with cryptocurrencies, as litely-regulated, or shifting back to the traditional system of banking and financing. Understood in this way, some will contend that such a regulated crypto-sphere will be tantamount to making it a permissioned network. This is not the case. Bitcoin will still remain entirely open; users will be free to join at any time and without anyone’s approval or permission. For some, of course, this will be a rank betrayal of Nakamoto’s vision; the ideological (and also criminally-facilitating) appeal of cryptocurrency is its private and unregulated nature. There is little to say to such purists or ideologues. However, looked at from a more balanced and pragmatic 24 UNBLJ RD UN-B [VOL/TOME 71 perspective, a subtle and supportive scheme of regulation might actually be viewed as a way to strengthen, not subvert cryptocurrencies like Bitcoin. As with private property generally, the value and benefit of a person’s resources are enhanced and even made possible by the involvement of the state and its regulatory arm.
Finally, the other objection to the regulation of programmers is more principled in nature. The basic claim is that any effort to regulate the programmers or code-makers will be an infringement of their rights to free speech. This is a stretch. The argument is that they simply design different technological platforms and protocols; it is the users of these processes who, for good or bad purposes, use the software and, therefore, should be responsible for their activities. The defensive analogy relied on is that between knife-makers and knife-users—it would be unfair and inappropriate to make a knife-maker liable for the knife-wielding actions of a criminal. Although the analysis of that knife-using situation seems persuasive, it misses the main point of the analogy with crypto-code. First, there is no suggestion that the code-maker will be liable for the illicit use of code for criminal purposes: their liability will be by way of administrative fine or discipline.9 The basis for a regulatory imposition of responsibility is that programmers are aware that such activities occur 2020 CanLIIDocs 3123 and cannot deny that they are facilitating this in a reasonably direct way. Those who act criminally through the cryptocurrency platforms can and should be dealt with directly. Secondly, there are ample and entirely legitimate regulations imposed upon knife-makers; they are required to manufacture knives to a high standard of quality (and can be sued for bad craftsmanship) and they are prohibited from producing certain kinds of dangerous knives. Accordingly, while it is important to respect the bona fides and skills of many code-makers, there is no plausible or compelling argument to insulate them from appropriate regulatory oversight as a result of their alleged speech rights.
Code Duties
Code-developers and miners are at the heart of the blockchain process and operation. Although there already exist several checks upon their authority and possible abuses of power (e.g., decentralisation, consensus requirements, and the like), it seems sensible to consider whether there are sufficient safeguards in play to protect the interests of crypto-users. After all, there is considerable disparity in power between code-developers (and, to a lesser extent, miners) and crypto-users. In actuality, this places the crypto-users in a position where they have to place trust in the code- developers that they will act in good faith and for the benefit of all sectors of the
9 The UK is considering applying anti-money laundering and counter terrorist financing regulations “to impose data collection and reporting requirements on not only cryptocurrency developers, but all open- source software developers and those who facilitate the peer-to-peer exchange of crypto-assets.” See James Foust, “Hot Takes” (10 June 2019), online: Coincenter
The obvious, although far from identical, comparison is with the situation of corporate directors. There is an established and sophisticated jurisprudence that addresses the dynamic relation between directors and shareholders; it explores the details of that relation, lays out standards of behaviour that are expected from directors, and examines the remedies available to shareholders if those are breached. The basic notion is that directors owe a duty to act in the best interest of the corporation. As such, they stand in a fiduciary relation to the company’s shareholders; they must place the interests of those they represent and at whose behest they hold power ahead of their own interests. This can be unpacked into a variety of sub-obligations, including the avoidance of conflicts of interest and the effecting of statutory compliance. The board of directors, therefore, is be held, as the inimitable Cardozo put it, “to something 2020 CanLIIDocs 3123 stricter than the morals of the marketplace; ... only thus has the level of conduct for fiduciaries been kept at a level higher than that trodden by the crowd.”10 Suitably modified to their technological circumstances, this is far from being an onerous or inappropriate duty that might be placed on code-developers.
Code-developers, of course, are not corporate directors: there is a vast difference between their respective roles and responsibilities. The most important are that there is no centralized organisation that is in any way analogous to the corporation in the decentralized world of cryptocurrency, that code-controllers have no direct control over the crypto-users’ property (i.e., the crypto-asset) and that they do not act as the agents of individual crypto-users or the group as a whole. Nevertheless, allowing for these important differences, it is still useful to canvass whether a similar regime to that in corporate governance, suitably adapted and tweaked, might be imported into the technological world of cryptocurrencies. That said, if there is to be a transplant of legal doctrines and rules from the corporate to the crypto-world, at least two caveats will be necessary—such fiduciary duties ought not to be enforceable by way of private litigation, and there should be some kind of cap on the liability of code-developers to crypto-users individually or collectively.
The idea that code-developers are in a fiduciary relationship with crypto- users and have a duty to act in the best interests of crypto-users, not their own, is not difficult to sustain. In brief, they provide services (in contrast to products) that demand a certain reliance by crypto-users on the code-developers’ superior knowledge and influence whose exercise carries significant risks to crypto-users if they are not performed properly. Because of this imbalance, the beneficiary/crypto-user is vulnerable and needs protection from the possible untrustworthy and self-serving
10 Meinhard v Salmon, 164 NE 545 at 464 (NY 1928), Cardozo J. For my own take on corporate governance generally, see Allan C Hutchinson, The Companies We Keep: Corporate Governance for a Democratic Society, (Toronto: Irwin Law, 2005). 26 UNBLJ RD UN-B [VOL/TOME 71 actions of the fiduciary/code-developers Traditional relationships recognized as giving rise to a fiduciary duty include lawyer/client, physician/patient, and trustee/beneficiary. In an important sense, this fiduciary approach treats code- developers as being in the same position as other skilled professionals and asks that they be judged by the same standards as them.11 Consequently, by being obliged to turn their focus to more ‘public’ and less private ends, the managers of the cryptocurrency universe will be required to bring themselves more in line with the regulatory goals of improved governance, systemic stability, user protection and financial legitimacy.
However, as with most regulatory or legal regimes, the devil is in the details. It will be important to ensure that a reasonable standard is imposed upon code- developers in fulfilling this responsibility. As in the corporate world, the expectation is that they will owe uncompromised loyalty to the corporation and its shareholders, take professional care in fulfilling their responsibilities, and make decisions in good faith. This is not an absolute standard of behaviour and action, but it does demand a close attention by code-developers to the interest of all those involved in and affected by their decisions and actions. As one Canadian court has nicely phrased it in regard 2020 CanLIIDocs 3123 to corporate officials,
[the law] looks to see that the directors made a reasonable decision not a perfect decision. Provided the decision taken is within a range of reasonableness, the court ought not to substitute its opinion for that of the board even though subsequent events may have cast doubt on the board’s determination. As long as the directors have selected one of several reasonable alternatives, deference is accorded to the board’s decision.12
Consequently, mindful of the influence and power that they have, code- developers should be expected to have their conduct scrutinised and to be held responsible to other members of the crypto-community, especially users, for their actions. As with all other large and complex institutions, the crypto-challenge for a crypto-regulatory agency will be to introduce structures and measures which will contribute to ‘closing the gap’ between those relative few entrusted with authority to make decisions and those relative many affected by those decisions. To do this, it will be essential to the well-being of the crypto-community as a whole that code-developers appreciate and act upon the imperative to advance as far as practicable all the combined and often competing interests of the different stakeholders. This is no easy task, but it is one that must be assumed or imposed. Whether said by Voltaire or
11 See generally Tamar Frankle, Fiduciary Law (Oxford: Oxford University Press, 2010). For arguments in favour of treating Code-developers as fiduciaries, see Angela Walch, “In Code(Rs) We Trust: Software Developers As Fiduciaries In Public Blockchains” in Georgios Dimitropoulos et al, eds, The Blockchain Revolution: Legal and Policy Challenges (Oxford: Oxford University Press, 2018). See also Philipp Hacker, “Corporate Governance for Complex Cryptocurrencies? A Framework for Stability and Decision Making in Blockchain-Based Organizations” in Georgios Dimitropoulos et al, eds, Regulating Blockchain: Techno- Social and Legal Challenges (Oxford: Oxford University Press, 2019). 12 Maple Leaf Foods Inc v Schneider Corp (1998), 42 OR (3d) 177 at 192, 44 BLR (2d) 115 (CA). There is likely less of a case for miners being held to the same standards, although they should have some responsibility to act in the general interest of users as much as their own interests. 2020] BREAKING THE CODE 27
Spiderman’s uncle Ben, it ought to be axiomatic that ‘with great power comes great responsibility’.
An Unfair Burden?
The argument to treat code-developers as fiduciaries will, of course, not persuade everyone. There are those who maintain that the realms of corporate directors and code-developers are so different that the rules in the former should not be used in the latter; the differences far outweigh the similarities. Apart from the general disposition against public intervention of any kind, the primary thrust of this resistance to introducing a fiduciary duty is two-pronged—that skilled programmers will be discouraged from becoming code-developers (and that will work to everyone’s disadvantage); and that there are sufficient safeguards in place to prevent the typical abuse of authority that fiduciary duties are intended to control against.13 Both objections are genuine and need to be taken seriously, but they are over-stated. The first concern is true for anyone who assumes a degree or power and authority. The code-developers’ assumption of responsibility will come with costs and liabilities as 2020 CanLIIDocs 3123 it does for doctors, lawyers and trustees: there is no evidence that this has dried up the supply of those professional ranks. The challenge will be to ensure that code- developers receive adequate recognition and benefits (acknowledgements, prestige, rewards and otherwise) to off-set such legal liability.
The second concern about existing safeguards is also wide of the mark. The argument made is that code-developers have no real capacity to bind and, therefore, harm the interests of crypto-users because any changes to the blockchain and its accompanying software must be approved by those users. As such, any changes that are proposed and are assessed by users to be for the benefit of code-developers and against their own will be rejected. Again, while there is some force to this objection (i.e., users are free to run any forked version of the software that they choose), it fails to confront the realities of both the corporate and crypto worlds. Like disgruntled shareholders who can sell their holdings, crypto-users can opt out and deploy their funds elsewhere. But, as in the corporate world, this is not in itself a stand-alone reason for code-developers to forego the imposition of a fiduciary duty; ‘exit’ is not the only or optimal solution. Mindful of the information and technical asymmetry between code-developers and users, the addition of a ‘voice’ option by way of a fiduciary mandate seems to be both necessary and desirable.14 If crafted and implemented in a sensitive way, placing a fiduciary duty on code-developers will not only protect users from any self-serving or anti-communal behaviour, but it will also fill the trust-gap that is created by the present programming arrangements and changes.
13 See e.g. Raina S Haque et al, “Blockchain Development and Fiduciary Duty” (2019) 2:2 Stan J Blockchain L & Pol’y 139. 14 See Albert O Hirschman, Exit, Voice, and Loyalty: Responses to Decline in Firms, Organizations, and States (Cambridge, MA: Harvard University Press, 1970). 28 UNBLJ RD UN-B [VOL/TOME 71
Accordingly, as part of the regulatory quid pro quo, the code-developers must be given rights and discretion that will allow them to plot the best course for the crypto- community in light of its multiple constituencies. As in the corporate context, the code- developers would not be the agents of the users: their lodestar would be the best interests of the crypto-community as a whole, not only those of the largest coin-holders or users. At the same time, they should have strong responsibilities placed upon them which will ensure that they appreciate that their power is provisional and dependent on those they serve and those to whom they ought to be accountable. In short, the introduction of a fiduciary duty will give substance to the oft-claimed notion by code- developers that “we reject kings, presidents and voting.” In a manner of speaking, this will go some of the way to requiring code-developers to put their money where their technological mouths are—establishing “rough consensus and running code”.15 As such, the main challenge is to encourage the code-developers to remain innovative and creative, but to do so in a responsible and reasonable manner for the crypto-community at large.
Of course, achieving this will demand some difficult trade-offs and delicate compromises; there is no simple metric for measuring such a feat. However, two prominent 2020 CanLIIDocs 3123 recommendations come to mind that will go most of the way to assuaging some of the more obvious concerns:
• No private right of action—it would seem sensible, at least as a first step, to enforce a fiduciary duty by way of administrative process and compliance. A crypto-regulatory agency would monitor the code- developers’ behaviour; the infractions of any stipulated standard would be dealt with by means of fines, suspensions de-certifications or other administrative steps. Such a restriction on private litigation and judicial intervention would contain and hopefully reduce the scope for expensive and lengthy disputation; and
• No open-ended compensatory remedy—this follows from the first. The penalties for infringing the code of practice would be broad and inventive—reprimands, supervision, fines, appointment bans, and the like. In taking such a line, the paralyzing challenge of quantifying the nature and extent of users’ losses brought about by code- developers’ malfeasance would be obviated; the reliance on litigation caps and other such devices would not be needed. This would work as the beneficial quo to the liability quid.
As always (and as even critics contend), a more sophisticated appreciation of the range and character of regulation suggests that the most telling contrast is not between heavy-handed government intervention and an entirely hands-off laissez- faire approach. While the state possesses a wide range of repressive powers to discipline its subjects, there is a much wider and more subtle set of institutions, practices and, as cryptocurrency reminds us, technologies that shape and influence, as
15 Oever & Moriarty, supra note 3. 2020] BREAKING THE CODE 29
Foucault put it, ‘the order of things’.16 Within such a modern world, the choice is between how to mix and match these various and often competing protocols for achieving an optimal state of affairs. This suggest that, when it comes to regulating cryptocurrency and blockchain technology, the critical decision will be how to balance public processes and procedures with the various private tools of regulation. My recommendations are intended as a first-cut at such a balance.
Conclusion
As George Orwell might have put it in Animal Farm (his justly famous parable about totalitarian government), “all … are equal, but some are more equal than others” in the technological farmyard of cryptocurrencies.17 This would seem to be the case with code-makers and programmers. Although they claim to be neither kings nor presidents in their work and approach (and, therefore, are no more equal than others), their commitment to voting and consensus is not entirely genuine. Like most politicians, they say one thing, but on important occasions do another; the posture of selfless servants of the public interest is unconvincing. Consequently, any serious effort to 2020 CanLIIDocs 3123 regulate cryptocurrency must entail some genuine willingness to focus on the work and influence of the code-makers. Without bringing them into the regulatory equation, the chances of advancing and attaining a democratic and progressive agenda for the regulation of cryptocurrency will be thwarted.
16 Michel Foucault, “Truth and Power” (1979) 4:13/14 Critique of Anthropology 131. 17 George Orwell, Animal Farm, (London, UK: Penguin Books, 1989) at 90.
UNIQUELY TREACHEROUS WATERS: MORE ON THE “SELLER IN POSSESSION” STATUTORY REGIMES OF NEW BRUNSWICK, NEWFOUNDLAND & LABRADOR, NOVA SCOTIA AND PRINCE EDWARD ISLAND
Clayton Bangsund*
Aucupia verborum sunt judice indigna.1
I. Introduction
In a trio of recent articles published in the Saskatchewan Law Review,2 Dalhousie Law 2020 CanLIIDocs 3123 Journal3 and Supreme Court Law Review,4 I have criticized numerous aspects of the diverse provincial and territorial statutory regimes that govern “seller in possession” title disputes in common law Canada. In this fourth edition of the pentalogy, the theme continues as I sharpen my critical focus on the quartet of regimes in Atlantic Canada.
A “seller in possession” title dispute occurs when an initial buyer leaves bought goods in the possession of the seller who then transfers them to a subsequent transferee.5 Presently, there are four distinct statutory models in force, across common law Canada, that apply to resolve such conflicts.6 Except for Manitoba, Ontario and Yukon (the “Model 1” jurisdictions), all provinces and territories have incorporated electronic personal property registry infrastructure into their statutory priority regimes. However, the manner of integration sharply diverges between western and northern jurisdictions on one hand (Alberta, British Columbia, Northwest Territories, Nunavut, Saskatchewan, the “Model 2” jurisdictions), and eastern jurisdictions on the other. Only in New Brunswick, Newfoundland and Labrador, Nova Scotia (the “Model 3”
* Associate Professor, University of Saskatchewan, College of Law; Founder & Editor, JuliusErwin.com. I dedicate this article to Marlon (2007-2019). 1 Translation: quibbling over words is unworthy of a judge. 2 Clayton Bangsund, “ABCD Remoteness Problems: Nemo Dat & Its Exceptions Under Subsection 26(1.2) of Saskatchewan’s The Sale of Goods Act” (2018) 81:2 Sask L Rev 133 [Bangsund, “ABCD Remoteness Problems”]. 3 Clayton Bangsund, “A Survey and Critique of the “Seller in Possession” Statutory Regimes of Common Law Canada: An ABC Prequel” (2019) 42:2 Dal LJ 243 [Bangsund, “ABC Prequel”]. 4 Clayton Bangsund, “Eliminating Redundancy in Legislation Governing the Sale of Goods: A Threequel” (2019) 93 SCLR (2nd) 367. 5 MG Bridge, Sale of Goods (Toronto: Butterworths, 1988) at 633. 6 See Bangsund, “ABC Prequel”, supra note 3. 2020] UNIQUELY TREACHEROUS WATERS 31 jurisdictions) and Prince Edward Island (the sole “Model 4” jurisdiction) does the Personal Property Security Act7 (PPSA) potentially apply to resolve the title dispute between the initial buyer and subsequent transferee.
In the Dalhousie Law Journal, I identified problems with statutory Models 3 and 4 that deserve attention from eastern Canadian lawmakers.8 In short, by defining a “sale of goods without a change of possession”9 as a transaction that creates a deemed security interest under the PPSA, legislators in Atlantic Canada have exposed buyers, who postpone possession of bought goods, to intolerable levels of risk of loss to a variety of competitors in an array of arguably unwarranted circumstances. For this reason, I think New Brunswick, Newfoundland and Labrador, Nova Scotia and Prince Edward Island should uniformize their “seller in possession” statutory regimes with Model 2 provinces and territories. Here, in my penultimate piece on this subject, I briefly highlight two additional technical difficulties with Models 3 and 4 that strengthen the case for statutory and regulatory reform in Atlantic Canada even if my principal advice is rejected. The first problem concerns attachment, the second concerns perfection.
2020 CanLIIDocs 3123
II. Attachment
Creation
To ensure a sound conceptual framework for the governance of security interests, personal property security legislation is replete with technical language and concepts. “Attachment” concerns the creation or establishment of a security interest in personal property. The term is not statutorily defined but the elements of attachment are clearly laid out in Part II of the PPSA. A secured party acquires a non-possessory security interest in goods only when the conditions for attachment have been satisfied, namely: (1) the secured party gives value;10 (2) the debtor signs a security agreement describing the goods;11 and (3) the debtor has rights, or the power to transfer rights, in the goods.12 Until these conditions of attachment have been met, the secured party has, at most, an
7 Personal Property Security Act, SNB 1993, c P-7.1 [NBPPSA]; Personal Property Security Act, SNS 1995- 96, c 13 [NSPPSA]; Personal Property Security Act, SNL 1998, c P-7.1 [NLPPSA]; Personal Property Security Act, RSPEI 1988, c P-3.1 [PEIPPSA]. 8 Bangsund, “ABC Prequel”, supra note 3 at 256–259. 9 NBPPSA, supra note 7, s 1(1); NLPPSA, supra note 7, s 2(1)(ll); NSPPSA, supra note 7, s 2(1)(an); PEIPPSA, supra note 7, s 1(1)(nn). 10 NBPPSA, supra note 7, s 12(1)(a); NLPPSA, supra note 7, s 13(1)(a); NSPPSA, supra note 7, s 13(1)(a); PEIPPSA, supra note 7, s 12(1)(a). 11 NBPPSA, supra note 7, ss 10(1)(b), 12(1)(c); NLPPSA, supra note 7, ss 11(1)(b), 13(1)(c); NSPPSA, supra note 7, ss 11(1)(b), 13(1)(c); PEIPPSA, supra note 7, ss 10(1)(b), 12(1)(c). 12 NBPPSA, supra note 7, s 12(1)(b); NLPPSA, supra note 7, s 13(1)(b); NSPPSA, supra note 7, s 13(1)(b); PEIPPSA, supra note 7, s 12(1)(b). 32 UNBLJ RD UN-B [VOL/TOME 71 inchoate interest.13 Upon satisfaction of the conditions, the secured party acquires a statutory charge that may be asserted against third parties.
Rights or Powers
For many secured transactions involving goods, the third element of attachment—rights, or the power to transfer rights, in the collateral—is easily satisfied. Where, for example, the debtor owns the goods that are subject to the security interest, the secured party encounters no conceptual or practical difficulty attaching its security interest. But this is not so for the deemed security interests.14 Consider the “lease for a term of more than one year”15 and the “commercial consignment”,16 the other two types of deemed security interests involving goods. In respect of these transactions, it is not clear that the putative debtor (the lessee or consignee, as the case may be) has rights in the goods capable of supporting attachment in favour of either the original deemed secured party (the lessor or consignor) or a future true secured party. Indeed, nemo dat17 suggests the very opposite. It is for this reason that the PPSA explicitly 18 provides, ex abundanti cautela, that a lessee under a lease for a term of more than 2020 CanLIIDocs 3123 one year, or a consignee under a commercial consignment, has rights in the goods sufficient to support attachment of a security interest.19 Subsection 12(3) of the NBPPSA (in this article, New Brunswick law will serve as proxy for the law of all Model 3 and 4 provinces) is reproduced below:
For the purposes of paragraph (1)(b) and without limiting other rights, if any, which the debtor has in the goods, a lessee under a lease for a term of more than one year or a consignee under a commercial consignment has rights in the goods when the lessee or consignee obtains possession of them under the lease or consignment.20
13 Royal Bank of Canada v Radius Credit Union Ltd, 2010 SCC 48 at para 31. 14 NBPPSA, supra note 7, s 1(1); NLPPSA, supra note 7, s 2(1)(pp)(ii); NSPPSA, supra note 7, s 2(1)(ar)(ii); PEIPPSA, supra note 7, s 1(1)(rr)(ii). 15 NBPPSA, supra note 7, s 1(1); NLPPSA, supra note 7, s 2(1)(y); NSPPSA, supra note 7, s 2(1)(y); PEIPPSA, supra note 7, s 1(1)(y). 16 NBPPSA, supra note 7, s 1(1); NLPPSA, supra note 7, s 2(1)(h); NSPPSA, supra note 7, s 2(1)(h); PEIPPSA, supra note 7, s 1(1)(h). 17 Nemo dat quod non habet—translation: no one can give what he does not have. 18 Translation: out of abundant caution. 19 See Jacob S Ziegel & Ronald CC Cuming, “The Modernization of Canadian Personal Property Security Law” (1981) 31:3 UTLJ 249 at 262; RCC Cuming, “True Leases and Security Leases Under Canadian Personal Property Security Acts” (1983) 7:3 Can Bus LJ 251 at 259. One peer reviewer disagrees with this interpretation of the provision’s purpose and effect, suggesting instead that it is solely focused on clarifying the time at which a lessee or consignee acquires rights or powers sufficient to support attachment. On one hand, this narrow interpretation is appealing because it furnishes a plausible explanation as to why the other deemed security interests are not referenced in the provision. On the other hand, it is plainly inconsistent with the stated intentions of the chief architects of Canadian personal property security legislation. 20 NBPPSA, supra note 7, s 12(3); NLPPSA, supra note 7, s 13(3); NSPPSA, supra note 7, s 13(3); PEIPPSA, supra note 7, s 12(3). 2020] UNIQUELY TREACHEROUS WATERS 33
Argument from Silence
It is peculiar that there is no express reference to a “sale of goods without a change of possession” in the above provision in light of such transaction’s characterization as a deemed PPSA security interest. Does a seller under a sale of goods without a change of possession have rights in the goods, or the power to transfer rights, sufficient to support attachment of a security interest? Recall that the seller no longer owns the goods, instead occupying the role of bailee in possession of the buyer’s property. Accordingly, there is foundation for the view that the absence of express statutory language deeming a seller in possession to have rights or powers sufficient to support attachment, coupled with the presence of such language respecting the other deemed security interests involving goods, implies that a seller in possession lacks these rights or powers: argumentum ex silentio.21
Reduction to the Absurd 2020 CanLIIDocs 3123
Then again, to adopt the above interpretation would defeat the purpose of including a “sale of goods without a change of possession” within the scope of the PPSA. Indeed, the more sensible interpretation is that a seller under a sale of goods without a change of possession, as bailee in possession, does have rights or powers sufficient to support attachment of a security interest.22 Supporting this purposive interpretation is the operative language “and without limiting other rights, if any, which the debtor has in the goods”. To interpret the PPSA otherwise would be to render unattachable any security interest given by the seller in possession: reductio ad absurdum.23
Out of Abundant Caution
The absence of express statutory language reinforcing a seller in possession’s ability to confer a valid security interest, in this sui generis24 context, appears to be a legislative oversight. It can be remedied with the addition of appropriate language to the existing statutory provision. Specifically, subsection 12(3) of the NBPPSA should be amended as follows:
For the purposes of paragraph (1)(b) and without limiting other rights, if any, which the debtor has in the goods, a lessee under a lease for a term of more than one year, or a consignee under a commercial consignment, or a
21 Translation: argument from silence. 22 See Ronald CC Cuming, Catherine Walsh & Roderick J Wood, Personal Property Security Law, 2nd ed (Toronto: Irwin Law, 2012) at 258. 23 Translation: reduction to the absurd. 24 Translation: of its own kind or class; unique or peculiar. 34 UNBLJ RD UN-B [VOL/TOME 71
seller under a sale of goods without a change of possession, has rights in the goods when the lessee, or consignee or seller obtains or retains, as the case may be, possession of them under the lease, or consignment or sale.25 (additional language underlined, deleted language struck out)
To date, this anomaly has not caused any interpretive difficulties for courts. However, it is a point worth clarifying for the legislative record in case the issue does arise in the future. For those Atlantic provinces whose legislators consider implementing the reforms detailed in the Report to the Canadian Conference on Personal Property Security Law on Proposals for Changes to the Personal Property Security Acts,26 this additional housekeeping item is also worthy of consideration, again, as a second-best alternative to my principal recommendation.
III. Perfection
Status 2020 CanLIIDocs 3123
“Perfection” concerns the status of a PPSA security interest; it occurs upon the coincidence of attachment and a validly taken perfection step.27 A secured party perfects its security interest in collateral in order to optimize its priority position vis- à-vis third parties. For a secured party to establish priority to any given collateral, it is generally necessary, though not always sufficient, that the security interest be perfected. Registration is the most common method used by secured parties to perfect security interests in goods.28
Serial Numbered Goods
The second problem with the “seller in possession” statutory regimes of the Atlantic provinces concerns the regulatory provisions governing registration against a special class of goods, namely, “serial numbered goods.” Before pinpointing the issue, I must first describe the regulatory rules and their connection to the PPSA priority
25 Legislators may wish to take the additional step of including confirmatory language for the deemed security interests involving accounts and chattel paper. To date, no province or territory has introduced such statutory language, but the case for its inclusion is equally compelling. 26 Ronald CC Cuming et al, “Report to the Canadian Conference on Personal Property Security Law on Proposals for Changes to the Personal Property Security Acts” (prepared by a working group of the Canadian Conference on Personal Property Security Law and ratified at the CCPPSL Annual Meeting in Edmonton, Alberta, 21-23 June 2017) [Cuming et al, “CCPPSL Report”]. 27 NBPPSA, supra note 7, s 19; NLPPSA, supra note 7, s 20; NSPPSA, supra note 7, s 20; PEIPPSA, supra note 7, s 19. 28 Cuming, Walsh & Wood, supra note 22 at 318. 2020] UNIQUELY TREACHEROUS WATERS 35 scheme. I begin by reproducing the definition of “serial numbered goods”, set out in section 2 of New Brunswick’s Regulations:29
“serial numbered goods” means a motor vehicle, trailer, mobile home, aircraft, boat or an outboard motor for a boat;30
Serial numbered goods are characterized by their mobility and availability for resale in secondary markets. Each such good is inscribed, by its manufacturer, with a unique identification number that distinguishes it from all others.31 Registration by serial number facilitates enhanced ex ante32 discovery by subsequent searching parties who, due to their remoteness from the secured transaction, may not be in a position to discover the security interest by conducting a debtor name search. A remote party with knowledge of an item’s serial number can discover a prior interest via serial number search and thereby avoid or resolve any potential conflict from the outset of the acquisition-transaction. The personal property registry’s serial number functions enable registering parties to perfect their interests in serial numbered goods, and searching parties, whether immediate or remote, to discover such interests.33
2020 CanLIIDocs 3123
Variable Requirements
The PPSA expressly provides that a person may register a financing statement in the personal property registry in accordance with the Regulations.34 The Regulations, in turn, impose detailed registration requirements35 that vary depending
29 General Regulation, NB Reg 95-57 [NB Regulations]; Personal Property Security Regulations, NLR 103/99 [NL Regulations]; Personal Property Security Act General Regulations, NS Reg 129/97 as amended by NS Reg 143/2015 [NS Regulations]; Personal Property Security Act Regulations, PEI Reg EC1998-270 [PEI Regulations]. 30 NB Regulations, supra note 29, s 2, “serial numbered goods”; NL Regulations, supra note 29, s 2(p); NS Regulations, supra note 29, s 2(1)(t); PEI Regulations, supra note 29, s 1(t). 31 Not all high-value goods bearing unique manufacturer-issued serial numbers constitute serial numbered goods (e.g. designer watches, electronic equipment, artwork, etc.). Goods of this nature are subject to the general collateral registration rules, and are only discoverable via debtor name search. 32 Translation: from before; before the event. 33 Bangsund, “ABCD Remoteness Problems”, supra note 2. 34 NBPPSA, supra note 7, s 43(1); NLPPSA, supra note 7, s 44(1); NSPPSA, supra note 7, s 44(1); PEIPPSA, supra note 7, s 43(1). 35 NB Regulations, supra note 29, ss 23–25; NL Regulations, supra note 29, ss 23–25; NS Regulations, supra note 29, ss 23–25; PEI Regulations, supra note 29, ss 22–24. 36 UNBLJ RD UN-B [VOL/TOME 71 on whether serial numbered goods are held by the debtor, at the moment of attachment,36 as “consumer goods,”37 “equipment”38 or “inventory.”39
Registration by serial number is currently40 mandatory for consumer goods.41 Failure to register by serial number results in non-perfection of the security interest and loss of priority to an array of third party competitors including, most notably, the debtor’s trustee in bankruptcy.42
Serial number registration is not a de jure43 requirement for serial numbered goods held as equipment. Technically, it is optional.44 However, failure to register by serial number leaves an equipment financier vulnerable to other secured parties, buyers, and lessees.45 Since equipment financiers are heavily incentivized to register by serial number, this method of registration is a de facto46 requirement for any secured party intent on optimizing its priority position.
Since commercial inventory is in constant flux, inventory financiers are not required to describe each item of inventory by serial number as it comes into the hands of the debtor. Such a system would be cost prohibitive and exceedingly labour 2020 CanLIIDocs 3123 intensive, requiring continual updating of the associated financing statement. Accordingly, for the sake of efficiency, the registration rules are relaxed for inventory
36 NBPPSA, supra note 7, s 2(2); NLPPSA, supra note 7, s 3(2); NSPPSA, supra note 7, s 3(2); PEIPPSA, supra note 7, s 2(2). 37 NBPPSA, supra note 7, s 1(1), “consumer goods”; NLPPSA, supra note 7, s 2(1)(i); NSPPSA, supra note 7, s 2(1)(i); PEIPPSA, supra note 7, s 1(1)(i). 38 NBPPSA, supra note 7, s 1(1), “equipment”; NLPPSA, supra note 7, s 2(1)(p); NSPPSA, supra note 7, s 2(1)(p); PEIPPSA, supra note 7, s 1(1)(p). 39 NBPPSA, supra note 7, s 1(1), “inventory”; NLPPSA, supra note 7, s 2(1)(x); NSPPSA, supra note 7, s 2(1)(x); PEIPPSA, supra note 7, s 1(1)(x). 40 The Canadian Conference on Personal Property Security Law proposes alignment of the registration and priority rules governing consumer goods and equipment, which is to be attained via relaxation of the registration requirements for serial numbered goods held as consumer goods. See Cuming et al, “CCPPSL Report”, supra note 26 at 122–25 (Item III.1). 41 NB Regulations, supra note 29, s 23(1)(a); NL Regulations, supra note 29, s 23(1)(a); NS Regulations, supra note 29, s 23(1)(a); PEI Regulations, supra note 29, s 22(1)(a). 42 NBPPSA, supra note 7, s 20(2)(a); NLPPSA, supra note 7, s 21(1)(a); NSPPSA, supra note 7, s 21(2)(a); PEIPPSA, supra note 7, s 20(2)(a). 43 Translation: as a matter of law. 44 NB Regulations, supra note 29, s 23(1)(c); NL Regulations, supra note 29, s 23(1)(c); NS Regulations, supra note 29, s 23(1)(c); PEI Regulations, supra note 29, s 22(1)(c). 45 NBPPSA, supra note 7, ss 30(6)–(7), 34(1), 35(4); NLPPSA, supra note 7, ss 31(6)–(7), 35(1), 36(4); NSPPSA, supra note 7, ss 31(6)–(7), 35(1), 36(4); PEIPPSA, supra note 7, ss 30(6)–(7), 34(1), 35(4). 46 Translation: in point of fact. 2020] UNIQUELY TREACHEROUS WATERS 37 financiers, who need only describe their serial numbered collateral in the general collateral field to gain optimal protection against third parties.47
Potential Confusion
The above rules for registration in relation to serial numbered goods sensibly apply to conventional secured transactions and the other deemed secured transactions, but it is not altogether clear how they apply to a sale of goods without a change of possession. The problem is akin to that of the square peg and round hole. Here, our putative “debtor”48—the seller in possession—has sold the goods to the initial buyer and, at the critical moment of attachment of the buyer’s deemed security interest (or the granting of any future true security interest),49 holds the goods as bailed property, not as inventory, consumer goods or equipment per se.50 A buyer under a sale of goods without a change of possession may be confused about the manner in which serial numbered goods are to be described in the financing statement—by serial number in the serial number field, or by item or kind in the general collateral field? In Canada’s sole Model 4 province, Prince Edward Island, the problem is exacerbated because the 2020 CanLIIDocs 3123 registration rules set out in the PPSA Regulations apply, mutatis mutandis,51 to the parallel “seller in possession” provisions of the Factors Act.52
Plausible Interpretation in a Unique Context
In the unique context of a sale of goods without a change of possession, in which the seller occupies the bare role of bailee with no accompanying right to use or alienate the goods, one might embrace a modified interpretation of the existing registration requirements by focusing on the manner in which the seller/debtor held the goods immediately prior to the sale to the buyer out of possession. This interpretation, though probably mistaken, is likely to be adopted by all but the sharpest of legal minds. If adopted, it has the unfortunate effect of imposing more onerous registration requirements on buyers of consumer goods (who must register by serial number to gain perfected status) than on buyers of equipment and inventory (who gain
47 NB Regulations, supra note 29, s 23(1)(e); NL Regulations, supra note 29, s 23(1)(e); NS Regulations, supra note 29, s 23(1)(e); PEI Regulations, supra note 29, s 22(1)(e). 48 NBPPSA, supra note 7, s 1(1), “debtor”; NLPPSA, supra note 7, s 2(1)(m)(v); NSPPSA, supra note 7, s 2(1)(m)(v); PEIPPSA, supra note 7, s 1(1)(m)(v). 49 NBPPSA, supra note 7, s 2(2); NLPPSA, supra note 7, s 3(2); NSPPSA, supra note 7, s 3(2); PEIPPSA, supra note 7, s 2(2). 50 Translation: by or in itself; intrinsically. 51 Translation: with necessary changes. 52 Factors Act, RSPEI 1988, c F-1, s 9(2). 38 UNBLJ RD UN-B [VOL/TOME 71 perfected status through mere general collateral description).53 In any case, this alternate interpretation is inconsistent with express statutory language, which provides that “the determination as to whether goods are ‘consumer goods’, ‘inventory’ or ‘equipment’ shall be made as of the time the security interest attaches.”54
Technical Interpretation: Equipment as the Residual Category
The more persuasive view is that since the goods are not held (or are no longer held) by the seller as consumer goods or inventory at the time of attachment, they must fall under the residual category of equipment.55 While technically sound, this interpretation of the law assumes an inordinately high level of sophistication from the first buyer. Assuming he understands that the transaction is governed by the PPSA (a stretch in itself), this buyer must also appreciate that goods, held by the seller as either consumer goods or inventory, undergo instantaneous metamorphosis at the time of the sale, thereby necessitating different registration protocol. Regrettably, there is no reason to believe that many, let alone most or all, buyers possess this level of sophistication. Unlike the other deemed secured parties (i.e. commercial consignors, 2020 CanLIIDocs 3123 leasing companies, accounts factors and chattel paper financiers), who are specialized, highly sophisticated entities that regularly engage with the PPSA and its regulatory trappings, buyers represent a far more generalized class.56 Suppose that the first buyer is a member of the general public (e.g. a librarian) who buys goods from a seller’s commercial inventory outside the ordinary course of business, leaving said goods in the seller’s possession. Will such buyer be astute enough to recognize that the associated registration in the personal property registry must be effected in accordance with the technical rules for equipment rather than inventory? His failure to appreciate this nuance may very well prove fatal to the proprietary claim. Commercial legislation is useful only if it is accessible to those whom it targets and affects.
Simplicity is a Friend to the Laws
Since PPSA priorities commonly hinge on fact and time of valid registration in the personal property registry, it is imperative that buyers, under sales of goods without a change of possession, be furnished with clear and coherent rules for registering notice of their deemed security interests. For the sake of simplicity, and to better enable such buyers to transact with confidence, I recommend that lawmakers in
53 This anomaly will be addressed by any eastern province that relaxes the mandatory serial number registration rule for consumer goods as recommended by the Canadian Conference on Personal Property Security Law. See note 40. 54 NBPPSA, supra note 7, s 2(2); NLPPSA, supra note 7, s 3(2); NSPPSA, supra note 7, s 3(2); PEIPPSA, supra note 7, s 2(2). 55 NBPPSA, supra note 7, s 1(1): “equipment” means goods that are held by a debtor other than as inventory or consumer goods; NLPPSA, supra note 7, s 2(1)(p); NSPPSA, supra note 7, s 2(1)(p); PEIPPSA, supra note 7, s 1(1)(p). 56 Bangsund, “ABC Prequel”, supra note 3 at 258. 2020] UNIQUELY TREACHEROUS WATERS 39
New Brunswick, Newfoundland and Labrador, Nova Scotia and Prince Edward Island adopt a regulatory provision mandating that, in respect of a sale of goods without a change of possession, serial numbered goods be described in the serial number collateral field of the financing statement: simplicitas est legibus amica.57 Section 18 of Saskatchewan’s Regulations,58 reproduced below, is a model of such a provision:
When a registration authorized pursuant to The Sale of Goods Act, The Factors Act, The Commercial Liens Act or Part V.1 of The Summary Offences Procedure Act, 1990 is to be effected:
(a) goods that are serial numbered goods are to be described pursuant to section 13; and (b) goods other than serial numbered goods are to be described by item or kind.
Best Practice
Fortunately, for any buyer under a sale of serial numbered goods without a 2020 CanLIIDocs 3123 change of possession (and for a buyer without possession governed by Prince Edward Island’s Factors Act), the regulatory obstacles highlighted above are easily surmounted. To avoid registration invalidity, a diligent buyer under a sale of serial numbered goods without a change of possession should, ex abundanti cautela, describe her collateral in both the serial number description field and general collateral description field of the financing statement. By adopting this best practice, she jettisons any argument that a third party, like a trustee in bankruptcy, could possibly mount on grounds of technical non-compliance with the Regulations.
IV. Conclusion
I reemphasize that the proposals set forth in this article are contingent in nature, representing a mere second-best solution to the problems I have identified. For the reasons given in my recent article in the Dalhousie Law Journal,59 I think that legislators in in the Atlantic provinces should consider adopting a statutory system similar to that of western and northern Canada, incorporating personal property registry infrastructure into the priority governance of seller in possession title disputes without bringing such disputes under the strict purview of the PPSA. If, however, lawmakers in New Brunswick, Newfoundland and Labrador, Nova Scotia and Prince Edward Island reject my principal recommendation, then the solutions proposed in this article should be contemplated in the alternative. Implementation of these modest proposals would clarify and simplify the law, and better enable buyers to navigate the uniquely treacherous waters of delayed-possession sales in Atlantic Canada.
57 Translation: simplicity is a friend to the laws. 58 The Personal Property Security Regulations, RRS c P-6.2, Reg 1. 59 Bangsund, “ABC Prequel”, supra note 3.
PLATFORM WORKERS AND COLLECTIVE LABOUR ACTION IN THE MODERN ECONOMY
Bethany Hastie*
Introduction
Work in the digital platform economy, such as for Uber, Lyft, Foodora, Door Dash, and other similar services, has given rise to substantial legal and scholarly attention in recent years. Like many other forms of work and employment, platform workers are often characterized as precarious. These workers face significant obstacles, both formal and practical, in accessing legal rights and protections related to their work. Scholars in general, and legal scholars in particular, have largely been preoccupied with the question of status or taxonomy for platform workers to date, unpacking and debating the question of whether platform workers are properly characterized as 2020 CanLIIDocs 3123 employees or independent contractors. While the question of taxonomy is important, as will be discussed in this article, it has largely deflected attention away from the multitude of strategies and avenues that platform workers can, and do, use to advance their labour interests regardless of their employment status. The current focus on taxonomy reveals a deeper concern for the challenges facing gig workers as workers, regardless of the status ascribed to them, and of the possibilities for innovative solutions to improve advance their interests beyond or outside of existing legislative regimes. While a variety of legislative and policy responses to regulating the platform economy have been explored,1 this article examines how platform workers are
* Assistant Professor, Peter A. Allard School of Law, University of British Columbia. This research was supported by the Social Sciences and Humanities Research Council of Canada. I wish to thank Mitchell Horkoff for his research assistance on this article, as well as the editorial board of the University of New Brunswick Law Journal, and the anonymous reviewers for this article, for their helpful feedback and guidance. 1 For a broader discussion of various regulatory responses that fall both within and outside of the debate on classification, see e.g. Maria Mexi, “Social Dialogue and the Governance of the Digital Platform Economy: Understanding Challenges, Shaping Opportunities”, (Background paper for the ILO-AICESIS-CES Romania International Conference, 10-11 October 2019) at 7, online (pdf): International Labour Organization
This article surveys existing efforts by platform workers to collectively organize and advance their labour interests, with a view to improving their working rights and conditions. After reviewing the status of platform workers, the challenges and contours of their work, and the needs and interests that may be served through collective labour action in Section I, this article describes and comments on identified forms of collective labour action undertaken by platform workers across a number of jurisdictions in Section II. As this article discusses, collective labour action, in its many modalities, both formal and informal, creates a context in which the traditional legal debates regarding the status of the worker become less important, focusing instead on the actual needs, interests, rights and conditions of work at issue. Collective labour action, as a tool for improving workplace rights and conditions, as well as a political strategy, also creates greater space for the participation and voice of workers. The rich and diverse forms of collective labour action undertaken by platform workers provide both illustrations and lessons that can be drawn for workers in other precarious 2020 CanLIIDocs 3123 industries and jobs, and more broadly in considering the future of labour law in a modern economy increasingly characterized by work outside of traditional direct employment, a discussion taken up in Section III. This article thus sets a descriptive foundation for further dialogue on the future of labour law in the modern economy, both for platform workers and the many other, and growing, populations of workers falling outside of traditional labour and employment protections.
I. Mapping the Landscape of Platform Work
Platform work has given rise to a substantial body of literature, litigation, and legislation, concerned with classifying platform workers for the purposes of labour and employment law. This debate has largely focused on determining whether workers are ‘employees’, and thus entitled to existing rights and protections afforded in domestic labour and employment law, or ‘independent contractors’ who fall outside the purview of legal regulation of employment.2 The implications of this question of status are clear; workers who are employees have the benefit of access to rights, such as minimum wage, and protections, such as in relation to health and safety, and against unjust dismissal. Workers who are not employees, but independent contractors,
“Regulating Work in the Gig Economy: What Are the Options?” (2017) 28:3 Economic & Labour Relations Rev 420. 2 See e.g. Jeremias Prassl & Martin Risak, “Uber, Taskrabbit, & Co.: Platforms as Employers? Rethinking the Legal Analysis of Crowdwork” (2016) 37:3 Comp Lab L & Pol’y J 619; Guy Davidov, “The Status of Uber Drivers: A Purposive Approach” (2017) 6:1/2 Spanish Labour L & Employment Relations J 6; Robert Sprague, “Worker (Mis)Classification in the Sharing Economy: Trying to Fit Square Pegs into Round Holes” (2015) 31:1 ABA J Labor & Employment L 53; Abi Adams, Judith Freedman & Jeremias Prassl, “Rethinking Legal Taxonomies for the Gig Economy” (2018) 34:3 Oxford Rev Economic Policy 475; Emily C Atmore, “Killing the Goose That Laid the Golden Egg: Outdated Employment Laws Are Destroying the Gig Economy” (2017) 102:2 Minn L Rev 887. 42 UNBLJ RD UN-B [VOL/TOME 71 receive no rights or protections under employment law and are thus required to negotiate their working conditions directly with their ‘clients’. The status question also relates directly to efforts to unionize amongst platform workers.
Most jurisdictions, including Canada, historically developed legal tests to determine whether a worker is an employee or independent contractor. These tests often look to a variety of factors with a view to ascertaining the extent of control exerted over the worker and their working conditions. Factors typically include: who owns the tools of the trade; whether the worker has a uniform; whether the worker can set their own schedule or hours of work; and, others.3 Platform work poses obvious challenges in attempting to use such factors as a basis for categorization. While platform workers typically “own the tools of the trade” (their vehicles or bicycles),4 the extent of control that the companies may exert over working conditions, hours, future work prospects, and wages, is significant.5 As such, a nuanced application of historical employment tests under law produces a less-than-clear, and contestable, result for platform workers.
Recent recognition of the growing number of workers in various industries 2020 CanLIIDocs 3123 who are neither clearly employees nor independent contractors has given way to new categories of workers, such as “dependent contractor”, in some jurisdictions. For example, Ontario recognizes the category of “dependent contractor”, which it defines as: “non-employment work relationships that exhibit a certain minimum economic
3 In Canada, the leading authority is 671122 Ontario Ltd v Sagaz Industries Canada Inc, 2001 SCC 59, which summarizes the various tests and criteria historically adopted in Canada at paras 35–47. In California, the recent decision in Dynamex Operations West, Inc v Superior Court of Los Angeles, 4 Cal (5th) 903 (2018) rejected the historical Borello test used to determine whether workers were employees or independent contractors, which had, similar to the Canadian approach, emphasized the extent of control over working conditions, as well as other factors such as such as ownership of equipment, opportunity for profit and loss, and the belief of the parties. The California Supreme Court in Dynamex adopted a new “ABC test” that begins from a rebuttable presumption that the worker is an employee. 4 But see Canadian Union of Postal Workers v Foodora Inc dba Foodora (2020), CanLII 16750 at paras 92–99, 2020 CLLC 220-032 (OLRB) [Foodora], which characterizes the app (technology) as the essential tool of the trade, and which is owned and maintained by the enterprise. 5 A number of jurisdictions are currently considering, or have already made judicial pronouncements, on this question. See Jeremias Prassl, Humans as a Service: The Promise and Perils of Work in the Gig Economy (London, UK: Oxford University Press, 2018) at 11; “Employment status of platform workers (national courts decisions overview—Argentina, Australia, Belgium, Brazil, Canada, Chile, France, Germany, Italy, Nederland, Panama, Spain, Switzerland, United Kingdom, United States & Uruguay)” (8 December 2018), online (blog): Una Mirada Crítica a Las Relaciones Laborales
As with many other forms of precarious work, workers in the platform economy are said to face varying levels of exploitation associated with their work. Concerns regarding wages, health and safety, and access to legal remedies are commonly documented.8 Scholars engaged in the classification debate largely see the solution to exploitation as expanding current legal definitions of employment to encompass platform workers. Some scholars have focused attention on the underlying normative criteria that employment tests might focus more substantially on, such as 2020 CanLIIDocs 3123 subordination,9 while other approaches have argued for the creation of intermediary categories between employee and independent contractor.10 Yet others have argued for a radical shift away from classification under employment law, advocating for the extension of a set of core rights and protections for all forms and modes of work.11
The debate about status has likely been a focal point in existing scholarship precisely because of its assumed consequences in extending labour rights, as mentioned above. If platform workers are employees, they are subsumed under
6 McKee v Reid’s Heritage Homes Ltd, 2009 ONCA 916 at para 30. 7 Foodora, supra note 4 at paras 77–79. 8 See e.g. Global Commission on the Future of Work, Work for a Brighter Future, ILO, (2019) at 44, cited in Mexi, supra note 1 at 1. See also Miriam A Cherry, “Beyond Misclassification: The Digital Transformation of Work” (2016) 37:3 Comp Lab L & Pol’y J 577 [Cherry, “Digital Transformation”]; Miriam A Cherry, “People Analytics and Invisible Labor” (2016) 61:1 Saint Louis ULJ 1, cited in Mexi, supra note 1 at 9. 9 See e.g. Emmanuel Dockès, “New trade union strategies for new forms of employment” (2019) 10:3 European Labour LJ 219 at 221. 10 See e.g. Seth Harris & Alan Krueger, “A Proposal for Modernizing Labor Laws for Twenty-First Century Work: The ‘Independent Worker’” (December 2015), online (pdf): The Hamilton Project
This evidences an underlying core concern about the material working conditions of platform workers, regardless of their legally defined status.13 Indeed, the precariousness of working conditions for platform workers is increasingly documented, as are some of the negative consequences that can be associated with such work. While the debate concerning status has obvious merit and urgency, it has, in some way, deflected attention from a deeper discussion of the labour needs and interests of workers that might be served through alternative forms of collective labour action.
Many of the features of platform work are simply a new instantiation of enduring labour precariousness: the casual or on-call nature of the working hours, 2020 CanLIIDocs 3123 minimal wages, lack of clear safety protections, and others.14 These reflect a general trend associated with neoliberalism away from standard employment, characterized by full-time permanent work, towards labour fragmentation and piece-work, facilitated through short-term contracts.15 This shift in the construction of labour markets is evidenced by the shedding of legal liability for labour and employment laws by enterprises, the transfer of risk from employer to worker, and often, consequential
12 Though workers may still be covered by, for example, occupational health and safety regulations, anti- discrimination law, and other related areas of law that regulate work and workplaces. 13 Although status may be significant in determining, for example, formal access to unionization under domestic laws. See e.g. Mexi, supra note 1 at 6: “The right to collective bargaining for self-employed workers is the object of legal discussion, as it is often considered in breach of competition law by national antitrust authorities, given that this is considered "price fixing'' harming consumer welfare”, citing also Antonio Aloisi, “Negotiating the Digital Transformation of Work: Non-Standard Workers’ Voice, Collective Rights and Mobilisation Practices in the Platform Economy” (2019) European University Institute MWP Working Paper No 2019/03; Hannah Johnston & Christopher Land-Kazlauskas, “Organizing On-Demand: Representation, Voice, and Collective Bargaining in the Gig Economy” (2019) International Labour Organization Working Paper Conditions of Work and Employment Series No 94. 14 See Stewart & Stanford, supra note 1 at 428–30; Mexi, supra note 1 at 1. Similar issues are noted for a growing number of workers falling outside of traditional direct-employment relationships; See Judy Fudge, Eric Tucker & Leah Vosko, “Employee or Independent Contractor? Charting the Legal Significance of the Distinction in Canada” (2003) 10 CLELJ 193; Guy Davidov, “The Three Axes of Employment Relationships: A Characterization of Workers in Need of Protection” (2002) 52:4 UTLJ 357 [Davidov, “Characterization of Workers”]; Bethany Hastie, “Human Rights and Precarious Workplaces: A Comment on British Columbia Human Rights Tribunal v Schrenk” (2019) 52:1 UBC L Rev 169 [Hastie, “Precarious Workplaces”]. 15 Austin Zwick, “Welcome to the Gig Economy: Neoliberal Industrial Relations in the Case of Uber” (2018) 83:4 GeoJournal 679. See also Jim Stanford, “The Resurgence of Gig Work: Historical and Theoretical Perspectives” (2017) 28:3 Economic & Labour Relations Rev 382. 2020] PLATFORM WORKERS AND COLLECTIVE LABOUR ACTION 45 economic benefits passed onto the consumer.16 These are each hallmarks of platform work, though they are not unique to this kind of work.
Yet, there are features of platform work—or, at least, its representation—that are distinct in understanding the labour issues surrounding platform work, and which, in turn, produce distinct consequences for platform workers. The emphasis placed on technology—both in affirming and disruptive discourses about platform work—is of particular significance. Specifically, the role of the intermediary platform (or ‘employer’) has been especially obfuscated. This obfuscation is an essential component to the financial success of the platform economy and, relatedly, to the illusion of its operation outside of the bounds of labour law.17 The technological core of platform work is presented as merely a passive instrument through which workers can connect directly with their clients, or purchasers of goods.18 As a result, the technological core of platform work operates in a distinct way to obscure the very existence of the work as regulated labour, to the benefit of the enterprise (or ‘employer’) and, arguably, consumer.
As existing scholarship has demonstrated, technology in this arena further 2020 CanLIIDocs 3123 functions to solidify aspects of precariousness of platform work.19 In particular, technology both enables greater surveillance of workers, while also isolating them, each of which entrenches precariousness in specific ways. The level of flexibility and freedom offered for work available through digital platforms is often cited as a draw for platform workers. However, “technology, in fact, acts more as an enabler of management than an emancipator” in these contexts.20 For example, the extent to which platform-based work enables surveillance and tracking activities, such as in relation to acceptance rates, has been noted as a cause of distinct stress for platform workers as compared with more traditional forms of employment or work.21 This entrenches the precariousness of the work both through the minute tracking of worker activities and through the consequential stress and negative impact on well-being to workers.
Technology further functions in the context of platform work to create an acutely isolated and autonomous labour pool. The nature of platform work is such that
16 Zwick, supra note 15 at 681–82. 17 Mark Freedland, “New Trade Union Strategies for New Forms of Employment - A Brief Analytical and Normative Foreword” (2019) 10:3 European Labour LJ 179 at 181–82. See also Ian Fitzgerald, Jane Hardy & Miguel Martinez Lucio, “The Internet, Employment and Polish Migrant Workers: Communication, Activism and Competition in the New Oganisational Spaces” (2012) 27:2 New Technology, Work & Employment 93, cited in Mexi, supra note 1 at 9. 18 Stanford, supra note 15. 19 See e.g. Prassl, supra note 5. 20 Nicola Countouris, Valerio De Stefano & Mark Freedland, “Preface to the ELLJ Special Issue: ‘Testing the “Person Work” Relation: New Trade Union Strategies for New Forms of Employment” (2019) 10:3 European Labour LJ 175 at 176. 21 Uttam Bajwa et al, “The Health of Workers in the Global Gig Economy” (2018) 14:124 Globalization & Health 1. 46 UNBLJ RD UN-B [VOL/TOME 71 the work takes place in unfixed locations, during inconsistent times, with little or no physical workspace or opportunity to interact with other workers or with the company they work for.22 While the independence and flexibility associated with these characteristics of platform work is often cited as a benefit or advantage of this form of work,23 the resulting social isolation can negatively impact a worker’s sense of well- being. Moreover, the extent of isolation and fragmentation of this workforce affects access to important avenues through which they might mediate their working relationship, interests and concerns. Without the opportunity for interaction with similarly situated individuals (other workers), platform workers do not benefit from clear or direct access points to seek advice on problems associated with their tasks or working conditions, or to gather information about their legal rights and protections.24 Further, without clear communication channels to other workers, platform workers are inhibited from collectivization and collaboration.25 This poses a unique challenge for platform workers, as collective labour action has historically been seen as an important tool through which workers can advance their interests.
Collective action, in particular, can provide platform workers, as it does with other workers, an avenue through which they can work towards materially improving 2020 CanLIIDocs 3123 their working conditions and interests. This may include items such as working hours, wages, surveillance and tracking systems, expectations concerning acceptance rates, health and safety conditions, and others. Importantly, collective action provides a means through which to raise awareness of and directly bargain about such conditions, potentially regardless of employment status, allowing platform workers to move past the status debate in order to advance their labour interests directly. The next section takes up a direct examination of forms of collective labour action that platform workers have engaged in, with a view to understanding the promise and potential of collective labour action in furthering the interests and needs of platform workers outside of the confines of existing labour and employment law regimes.
II. Collective Action and Labour Organizing Among Platform Workers
Despite the noted difficulties and barriers that platform workers face, as discussed in the previous section, a growing number of case studies illustrate the innovative approaches workers are using to collectively organize and advance their labour
22 See e.g. Mexi, supra note 1 at 9. 23 See e.g. Elise Taylor, Isolated with Friends: Online Communities in the Gig Economy (Masters Dissertation, Northeastern Illinois University, 2017) [unpublished] at 15. 24 Juliet Webster, “Microworkers of the Gig Economy: Separate and Precarious” (2016) 25:3 New Labor Forum 56. 25 See e.g. Alex Wood, Vili Lehdonvirta & Mark Graham, “Workers of the Internet Unite? Online Freelancer Organization Among Remote Gig Economy Workers in Six Asian and African Countries” (2018) 33:2 New Technology, Work & Employment 95 at 97–98. 2020] PLATFORM WORKERS AND COLLECTIVE LABOUR ACTION 47 interests outside of formal unionization,26 and the promise that such approaches may hold in that regard. This section outlines and describes various approaches to collective labour action that platform workers have used, and how these may provide effective avenues through which to respond to the issues, needs and challenges they face. These include informal collectivization through the creation and proliferation of communication channels, union affiliation, union creation, cooperative ownership, and localized legislative responses.
In its most informal instantiation, collective labour action by platform workers has included the creation and use of communication channels as a means to connect with other workers, as well as public protests as a vehicle through which to convey a shared message to a broader audience. Public protests and demonstrations can function as a platform for workers to advance their labour interests, and increase the visibility of the workforce and the ongoing issues it faces. For example, numerous large-scale protests by Indonesian app-based drivers during 2016-2017 resulted in widespread media attention and invitation to talks with both legislators and companies.27 Similarly, app-based couriers in Italy engaged in large-scale protests 28 against Foodora in an effort to improve the wage structure associated with their work. 2020 CanLIIDocs 3123 Workers have further organized strikes and boycotts in the delivery sector, such as through “logging out en masse from apps that allocate work shifts”.29
Platform workers have also created and utilized communication channels as a vehicle for internal discussion, dialogue and collaboration. These channels, also called “mass self-communication networks”,30 provide workers with an important avenue through which to connect with other workers, discuss dissatisfaction with their employer and working conditions, and brainstorm solutions for the future.31 Communication channels can also be used to externally advocate for better conditions directly with consumers. In the United States, “Turkopticon, an online community of
26 Access to which will be shaped directly by whether such workers are understood as employees or independent contractors, and by the statutory language defining application of the relevant legislation, as discussed in section I. 27 Michele Ford & Vivian Honan, “The Limits of Mutual Aid: Emerging Forms of Collectivity Among App- based Transport Workers in Indonesia” (2019) 61:4 J Industrial Relations 528 at 541. 28 Arianna Tassinari & Vincenzo Maccarrone, “The Mobilization of Gig Economy Couriers in Italy: Some Lessons for the Trade Union Movement” (2017) 23:3 Transfer 353. 29 Mexi, supra note 1 at 11, citing Anthony Forsyth, “'Prova di Solidarietà': How Effectively are Unions and Emerging Collective Worker Representatives Responding to New Business Models in Australia and Italy?” (Paper for the 17th International Conference in Commemoration of Prof Marco Biagi, Modena, 18-20 March 2019). 30Alex Wood, “Networks of Injustice and Worker Mobilisation at Walmart” (2015) 46:4 Industrial Relations J 259. 31 Ibid. See also Kurt Vandaele, “Will trade unions survive in the platform economy? Emerging patterns of platform workers’ collective voice and representation in Europe” (2018) European Trade Union Institute Working Paper No 2018/05; Mexi, supra note 1 at 11, citing Michele Forlivesi, “Alla ricerca di tutele collettive per i lavoratori digitali: per i lavoratori digitali: organizzazione, rappresentanza, contrattazione [Looking for Collective Protection for Digital Workers: Organization, Representation, Bargaining]” (2018) 4:1 Labour & L Issues 35. 48 UNBLJ RD UN-B [VOL/TOME 71
Mechanical Turk platform digiworkers, created a web platform called “Dynamo” that focuses specifically on building collective action”.32 Dynamo has engaged in a sustained campaign involving the publication of guidelines on fair pay, for example.33
These forms of collective labour action can do some work to break down the barriers associated with isolation and fragmentation of this workforce, as discussed in the previous section. In particular, public protests and demonstrations function to connect workers with a broader audience or set of stakeholders, making visible a workforce that is largely rendered invisible through the discourse of consumerism that dominates the public sphere. The increased visibility, itself, is said to be a benefit regardless of whether such demonstrations effect material change in the short-term.34 Relatedly, the creation and use of communication channels amongst workers provides the opportunity for interaction, problem-solving and advice-seeking in relation to their working conditions. Of particular importance with regards to the “self-communication networks”, these provide a means through which workers can “[boost] their associational power” and undertake further or more formal labour organizing with a view to advancing their labour interests.35
2020 CanLIIDocs 3123 Building on the foundation that may be created through communication networks and public demonstrations, platform workers in some cases have also directly engaged with trade unions in order to advance their labour interests. In some cases, workers are engaged with unions directly with the end-goal of formal unionization in mind. Unions are providing legal support for platform workers challenging their working conditions, employment status, and ability to formally organize a union, through litigation. For example, in Canada, ongoing legal disputes with Foodora and Uber are being supported by the Ontario Federation of Labour, as well as Canadian Union of Postal Workers.36 However, beyond this core role, unions are engaging with platform workers in a variety of informal ways. For example, in some cases, workers are engaged in “union-affiliated” relationships where established
32 Mexi, supra note 1 at 11, citing Birgitta Bergvall-Kåreborn & Debra Howcroft, “Amazon Mechanical Turk and the Commodification of Labour” (2014) 29:3 New Technology, Work & Employment 213. 33 Mexi, supra note 1 at 11, citing Niloufar Salehi et al, “We Are Dynamo: Overcoming Stalling and Friction in Collective Action for Crowd Workers” (Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems, Seoul, Republic of Korea, 18-23 April 2015), online (pdf): Stanford University
In Europe, existing unions in several countries have made effort to adapt and extend their member-base to include platform workers.38 These have included inclusion of platform workers within existing collective agreements,39 partnerships with platform workers,40 as well as supporting the creation of websites for workers to “rate the working conditions of different platforms”, advocate for working conditions and wages, and connect workers with existing unions.41 For example, in Denmark, a first collective agreement on platform work was signed in 2018 between a platform for cleaning services, Hilfr, and Danish trade union 3F.42 That agreement introduces a new category of worker, “Super Hilfrs”, which attracts increased minimum hourly wages and allows workers to accrue rights to pensions, holiday, and sick pay.43 Workers are eligible to become a “Super Hilfr” after 100 hours of work, and the regime is opt-out, meaning that workers will automatically be included unless they specifically object.44
Established unions, with existing expertise and resources, offer an important 2020 CanLIIDocs 3123 support and avenue through which platform workers can collectively organize to advance their labour interests through both formal and informal channels. Such strategies are not new for unions, who have been increasingly present in similar ways in a number of precarious industries and with various populations of precarious workers. For example, a number of unions in Canada and internationally have been engaged, both formally and informally, with migrant workers in agriculture and other industries, supporting unionization as well as broader advocacy and support efforts for these workers.45 As formal unionization has declined across a number of labour sectors
37 “Gig Workers’ Fight for Employee Rights”, supra note 36; “Unionizing the Gig Economy”, supra note 36. 38 See Mexi, supra note 1 at 13, citing Maarten Keune, “Trade Unions, Precarious Work and Dualisation in Europe” in Werner Eichhorst & Paul Marx, eds, Non-Standard Employment in Post-Industrial Labour Markets: An Occupational Perspective (Cheltenham: Edward Elgar, 2015) 378; Organisation for Economic Co-operation and Development, In It Together: Why Less Inequality Benefits All (Paris: OECD, 2015). 39 In the Italian logistics sector, for example. See Mexi, supra note 1 at 15. 40 See e.g. Osborne, supra note 5. 41 Mexi, supra note 1 at 13. 42 Ibid at 15. 43 Ibid. 44 Ibid. 45 Regarding agricultural workers, see Bethany Hastie, “Renewing Labour’s Engagement with Old Forms of Precarity: A Case Study of Unionization of Migrant Agricultural Workers in British Columbia” (2019) 49:1 Labour, Capital & Society 28; Tanya Basok & Ana Lopez-Sala, “Rights and Restrictions: Temporary Agricultural Migrants and Trade Unions’ Activism in Canada and Spain” (2016) 17:4 J Intl Migration & Integration 1271. For broader discussions concerning migrant workers and union revitalization, see e.g. Maite Tapia & Gabriella Alberti, “Social Movement Unionism: A Toolkit of Tactics or a Strategic Orientation? A Critical Assessment in the Field of Migrant Workers Campaigns” in Jurgen Grote & Claudius Wagemann, eds, Social Movements and Organized Labour: Passions and Interests (London, UK: 50 UNBLJ RD UN-B [VOL/TOME 71 in past decades, these kinds of activities by trade union organizations are being documented and analyzed as a form of “union revitalisation” across a number of jurisdictions.46
New unions and worker organizations are also being created across a variety of jurisdictions specifically for platform workers. For example, in Austria, Foodora workers created Betriebsrat with the support of an existing union, while in the United States, both Seattle and California have App-Based Drivers Associations.47 New guilds are further emerging in Europe, including “the Collectif Livreurs Autonomes de Paris, the German Deliverunion, the Italian Deliverance Milano, and the Dutch Riders Union with the aim to mobilize food couriers and Uber drivers, while also seeking to establish collective bargaining and social dialogue in the platform economy”.48 The creation of targeted associations and worker organizations can allow platform workers to connect with other workers, and provide a basis for targeted and concerted labour action, such as public awareness and advocacy campaigns, and direct bargaining for working conditions, applicability of legislation, and other significant topics.
2020 CanLIIDocs 3123 Some platform worker collectives are taking steps to compete directly with existing companies like Uber and Foodora through the creation of co-operative ownership models. Worker-owned enterprises allow platform workers to “share risks and benefits and negotiate better contracts, while being in a position to impact decision-making on how the platform is organized and managed”.49 The extent of control and direction that worker-owners may have under this model provides a clear avenue for improving labour conditions and interests. ‘Worker-owned’ platforms have surfaced in a variety of jurisdictions and contexts, including: Up & Go, a New York City-based home cleaning app designed and owned by immigrant Latin American workers; 50 Mensakas, a Barcelona-based delivery app started by two former Deliveroo drivers; 51 Eva, a Montreal-based Uber-like app that provides drivers with voting rights and shares in corporate profits;52 Fairmondo, a competition with Amazon and Ebay,
Routledge, 2018); Richard Hyman & Rebecca Gumbrell-McCormick, “Resisting Labour Market Insecurity: Old and New Actors, Rivals or Allies?” (2017) 59:4 J Industrial Relations 538. 46 Christian Lyhne Ibsen & Maite Tapia, “Trade Union Revitalisation: Where Are We Now? Where to Next?” (2017) 59:2 J Industrial Relations 170. 47 Mexi, supra note 1 at 11–12. 48 Ibid at 12. 49 Ibid at 14 citing Trebor Scholz, “Platform Cooperativism vs. the Sharing Economy” (5 December 2014) online: Medium
These models, termed “platform cooperativism”,57 provide an approach where workers’ rights may be better recognized and realized given the direct control workers have over the enterprise, and which may work towards fostering more sustainable and responsible commercialism.58 The worker-owned nature of such enterprises may provide a promising response to the rise of the gig economy monoliths,59 and importantly does so in a way that refocuses attention on the needs of workers. This may, in turn, do some work to mitigate the tendency to pit consumer interests against the needs of workers, a tactic commonly used to justify the suppressed working conditions of platform workers. It may further mitigate the obfuscation issue and its resulting consequences for workers, discussed in the previous section. Finally, the cooperative model may facilitate a greater sense of solidarity among platform workers.60
Finally, localized legislative responses that address the root issues at play 2020 CanLIIDocs 3123 may provide concrete and material improvements for workers within that jurisdiction. The municipality of Bolonga, Italy, recently passed a “Charter of fundamental digital workers’ rights within an urban setting”.61 This Charter was the product of efforts by Riders Union Bologna, a group of platform delivery workers who formed this union. The Charter was signed by the city’s mayor, the Riders Union Bologna, two prominent Italian labour unions, and two food delivery platforms that, combined, employ approximately one third of food delivery riders in the city.62 The Charter prescribes a fixed rate for services meeting or exceeding the applicable minimum wage, as well as compensation for overtime, public holidays, bad weather, and insurance for accidents and illness.63 At a broader level, both California and the European Union have engaged in similar legislative responses, prescribing particular work conditions for platform
53 Mexi, supra note 1 at 14. 54 Ibid. 55 Ibid. 56 Ibid. 57 Trebor Scholz, “Platform Cooperativism: Challenging the Corporate Sharing Economy” (2016), online (pdf): Rosa Luxemburg Stiftung
Creative legislative approaches to addressing the labour needs, interests and challenges of platform workers outside of the confines of existing labour and employment law demonstrate the continued relevance and significance that extension of legal rights and protections on paper may have, and how these can be effected in a way that circumvents the lingering debate over employment status. Further, these legislative responses appear to include similar conditions and terms to those of negotiated collective agreements, such as the Danish example provided earlier in this section. The Bologna Charter is particularly interesting given the highly localized nature of the intervention, and potential for replication across municipalities in various jurisdictions. 2020 CanLIIDocs 3123
III. Platform Workers and Collective Labour Action: Lessons for the Modern Economy
The previous section documented numerous examples of collective labour action by platform workers grouped into the following general categories: communication channels; union affiliation; union creation; platform cooperatives; and, legislative responses. These forms of collective labour action contain important implications for both platform workers and for workers in the modern economy more generally, which is increasingly characterized by a shift away from traditional employment relationships, and towards stratification and fissuring of the labour market.65 Most directly, each of the forms outlined in the previous section illustrates ways in which worker representation and voice can be advanced, and is being advanced, in a modern economy where formal employers and formal unionization are both in apparent decline. These forms demonstrate how collective labour action can advance core interests and needs of and for workers regardless of enduring issues surrounding status, which is posited to be of increasing concern for workers in a multitude of labour sectors.66
64 EC, Directive (EU) 2019/1152 of the European Parliament and of the Council of 20 June 2019 on transparent and predictable working conditions in the European Union, [2019] OJ, L186/105; US, AB 5, An act to amend Section 3351 of, and to add Section 2750.3 to, the Labor Code, and to amend Sections 606.5 and 621 of the Unemployment Insurance Code, relating to employment, and making an appropriation therefor, 2019-20, Reg Sess, Cal, 2019. 65 See e.g. David Weil, The Fissured Workplace: Why Work Became So Bad for So Many and What Can Be Done to Improve It (Cambridge, MA: Harvard University Press, 2014); Hastie, “Precarious Workplaces”, supra note 14. 66 See e.g. Weil, supra note 65; Davidov, “Characterization of Workers”, supra note 14; Fudge, Tucker & Vosko, supra note 14; Hastie, “Precarious Workplaces”, supra note 14. 2020] PLATFORM WORKERS AND COLLECTIVE LABOUR ACTION 53
This section takes up an evaluation of the forms of collective labour action discussed in the previous section, with a view to understanding how they may function and be effective for both platform workers and other precarious workers in the modern economy. As discussed earlier in this article, these forms of collective labour action evidence innovative approaches to enduring labour law problems, both in overcoming obstacles associated with employment status, and in overcoming obstacles associated with practical access to rights regardless of status, as discussed in section I. As such, the forms of collective labour action being undertaken in relation to platform work have broader potential benefits for labour and the legal regulation of work in the modern economy. This section first contextualizes the broader contemporary labour landscape to which the benefits of the identified forms of collective labour action may attach. It then goes on to evaluate and discuss the identified forms of collective labour action from the previous section in relation to three key factors: representation of members and power vis-à-vis external actors; channels of participation; and, service alignment with member needs.67
The modern economy is informed by many shifting paradigms, including 2020 CanLIIDocs 3123 globalization, automation, and the “fissured workplace”,68 which describes the abandonment of historical “direct employment” relationships and trend towards contracting-out discreet aspects of a business to external entities.69 This enables enterprises to maximize profits by shedding legal responsibility for workers, who are constructed as independent contractors, and also by transferring risk to those workers. Just as the rise of the standard employment relationship in the mid-20th century was motivated by the interests of capital-intensive enterprises, so too is the current shift away from standard employment motivated by the interests of enterprise, though in the opposite direction.70
The current labour landscape is increasingly characterized by “non-standard” work.71 This label captures myriad forms of work that fall outside of the historical full- time, permanent, direct-employment model. As such, “non-standard work” can include part-time and seasonal workers, casual and contract workers, employees and contractors. Alongside the growth in non-standard work is an increasing trend in the precariousness of such work.72 Precarious work is variably defined in existing scholarship, and often includes characteristics such as: instability and insecurity of employment; low wages; lack of benefits and entitlements; and, lack of control or
67 Maite Tapia et al, “Responding to the Incongruences: Shifting Forms of Worker Representation” (Paper delivered at the CRIMT Conference “What Kind of Work for the Future”, Montreal, 25 October 2018) [unpublished]. 68 Weil, supra note 65. 69 Ibid at 4. 70 Stanford, supra note 15 at 390-91. 71 See e.g. Davidov, “Characterization of Workers”, supra note 14. 72 Hastie, “Precarious Workplaces”, supra note 14. 54 UNBLJ RD UN-B [VOL/TOME 71 autonomy in the labour process.73 The consequences of precarious work often include an erosion of effective rights and protections under law, and in practice in the workplace.74 For a growing number of precarious workers, access to formal unionization, which could assist in ameliorating these issues, is also increasingly out of reach, either formally under law or in practice. This has, in turn, contributed to a general decline in unionization in the modern economy.75
The increasing challenges of accessing formal and traditional models of unionization has resulted in a shift towards “embrac[ing] alternative forms of voice as a way to reach a broader set of worker identities and interests”76 pursued through informal agreements and extra-legal mechanisms with employers and related actors.77 As existing scholarship has noted, access to formal unionization is increasingly out of reach for a growing number of workers, including platform workers.78 Thus, alternative forms of collective labour action, like those discussed in the previous section, “typically focus on service, advocacy, or organizing to improve the working and living conditions for employees rather than winning bargaining rights through bargaining unit elections”.79
2020 CanLIIDocs 3123 The previous section briefly commented on how each identified form of collective labour action works towards advancing platform workers’ labour interests. This section builds on those comments to examine in greater depth the function and effectiveness of each form of collective labour action, not only for platform workers, but more generally, given the noted shifts in the labour market within the modern economy. Specifically, this section develops and discusses the potential benefits of the identified forms of collective labour action in relation to three key criteria: representation of members and power vis-à-vis external actors; channels for participation; and, service alignment with member needs, drawing on the framework developed by Tapia et al.80
73 Leah Vosko, ed, Precarious Employment: Understanding Labour Market Insecurity in Canada (Montreal: McGill-Queens University Press, 2006). See also Stephanie Procyk, Wayne Lewchuk & John Shields, eds, Precarious Employment: Causes, Consequences and Remedies (Halifax: Fernwood, 2017); Arne L Kalleberg & Steven P Vallas, “Probing Precarious Work: Theory, Research and Politics” in Arne L Kalleberg & Steven P Vallas, eds, Precarious Work (Bingley: Emerald, 2018) 1. 74 Hastie, “Precarious Workplaces”, supra note 14; Weil, supra note 65 at 4; Davidov, “Characterization of Workers”, supra note 14. 75 Tapia et al, supra note 67; Tapia & Ibsen, supra note 46; Hyman & Gumbrell-McCormick, supra note 45. 76 Tapia et al, supra note 67 at 1. 77 Ibid. 78 Tapia & Ibsen, supra note 46; Hyman & Gumbrell-McCormick, supra note 45. 79 Tapia et al, supra note 67 at 2, citing Annette Bernhardt & Paul Osterman, “Organizing for Good Jobs: Recent Developments and New Challenges” (2017) 44:1 Work & Occupations 89. 80 Tapia et al, supra note 67. 2020] PLATFORM WORKERS AND COLLECTIVE LABOUR ACTION 55
Tapia et al describe how scholarly inquiry has focused disparately on union revitalization or alternative worker representation forms, without sufficient attention to the overarching logics and considerations that explain why certain forms of worker representation exist and how they are situated within the economic and labour market landscape. The framework they develop moves forward from the siloed treatment of alternative representation forms to provide a “common framework with testable propositions about how incongruences are likely to lead to union restructuring or alternative forms of worker representation via the logic of membership and influence.”81 These logics, explained below, provide key indicators that can be used to assess the function and effectiveness of forms of collective labour action for workers in the modern economy.
The logic of influence focuses attention on how powerful an intermediary organization or representation model is in its relationship with relevant external actors, such as an employer. In other words, it suggests that effective representation requires an intermediary, such as a union, that is “able to both represent and control workers credibly and effectively vis-à-vis external actors, such as employers and the 82 government.” This logic examines the channels of representation, and mode and 2020 CanLIIDocs 3123 extent of control over workers, in order to determine the level of influence the intermediary has in relation to relevant external actors. This logic further examines these two factors in relation to the external environment in which the intermediary is operating. The “environment” is “a broad contextual factor that consists of the main elements outside the direct control” of the intermediary, such as “labor laws, political pressure, industry/organizational structures, and employer behavior”.83 Increasing congruency between representation and the environment, and control and the environment, work to increase the influence of the intermediary.84 The logic of membership focuses attention on the internal dynamic between the intermediary organization or model and the workers or “membership base”. Under this logic, congruence between member identities and the factors of services and participation are key. The factor of services examines what “goods” the intermediary is able to provide to the membership base, such as collective agreements, legal advice or insurance, and how provided goods serve the interests of the constituency, having regard to their identities. “Greater congruence is achieved when services maximize the interest of the constituency”.85 Relatedly, channels of participation examines how the internal structure of the intermediary reflects the constituency’s identities and
81 Ibid at 4. 82 Ibid at 5. 83 Ibid at 8. 84 Ibid. See also Claus Offe & Helmut Wiesenthal, “Two Logics of Collective Action: Theoretical Notes on Social Class and Organizational Form” (1980) 1 Political Power & Social Theory 67; Virginia Doellgast, Nathan Lillie & Valeria Pulignano, “From Dualization to Solidarity: Halting the Cycle of Precarity” in Virginia Doellgast, Nathan Lillie & Valeria Pulignano, eds, Reconstructing Solidarity: Labour Unions, Precarious Work, and the Politics of Institutional Change in Europe (Oxford: Oxford University Press, 2018) 1. 85 Tapia et al, supra note 67 at 6. 56 UNBLJ RD UN-B [VOL/TOME 71 interests. Greater congruence here is achieved when “channels of participation maximize members’ feeling of empowerment”.86
As with a growing number of works across disparate industries and occupations, platform workers face significant difficulty accessing traditional models of collective labour action unionization. The previous section discussed many other, and novel, strategies that are being employed in the context of platform work in order to facilitate collective labour action, including: the creation and proliferation of communication channels; union affiliation; union creation; cooperative ownership; and, localized legislative responses. Each of these forms of collective labour action are differently situated along the axes of the two logics described above: influence, and membership. The remainder of this section examines how these forms of collective labour action map onto these axes, having regard to the key factors of: representation and control (for the logic of influence), and services and participation (for the logic of membership).
Representation as a criterion for evaluating new forms of collective labour action has particular import in the digital platform work context, and perhaps more 2020 CanLIIDocs 3123 broadly in articulating and explaining why an increasing number of workers find formal unionization inaccessible and inapplicable. The formal and narrow requirements attending legal regimes for formal unionization often operate to the practical (if not formal) exclusion of non-standard workers, geographically dispersed workers, and autonomous workers. Platform workers can be described as encompassing all of these characteristics (non-standard, geographically dispersed, autonomous/isolated work). Moreover, the increasing population of precarious workers may be variably characterized by one or more of these factors, particularly as concerns non-standard work. This makes representation of particular importance in considering and evaluating forms of collective labour action. The forms of collective labour action discussed in section II create channels of representation that do not depend on classification or status under employment law, a noted inhibitor to accessing formal unionization under existing labour law, and thus work towards greater congruency between representation and the environment in the context of platform work, and potentially for other precarious work contexts.
The creation of new unions specifically for platform workers, particularly in jurisdictions that allow for more flexible models of unionization, such as in Europe, may be seen as promising avenues for increasing channels of representation and increasing the congruence between representation and the environment. This form of collective labour action, closely aligned to the dominant historical model of formal unionization, shows the potential for adapting existing labour law regimes in ways that ameliorate the limitations and constraints noted in respect of these channels of representation for platform workers and an increasing number of precarious workers more generally.87 Further, the bargaining power that this form of collective labour action may hold, as evidenced by the instrumental role of the Riders Union in creating
86 Ibid at 6. 87 See e.g. Tapia et al, supra note 67. 2020] PLATFORM WORKERS AND COLLECTIVE LABOUR ACTION 57 the Bologna Charter, for example, illustrates well the significance that congruence between representation and the environment can have in enabling influence vis-à-vis external actors and in effecting substantive change for worker-members.
Beyond the adaptation or expansion of unions, co-operative ownership may best illustrate an alternative form of collective labour action that creates strong congruence between representation and the environment. Co-operative ownership models that enable broad-based participation by platform workers within the defined geographic area and labour market sector provide an interesting pathway forward in considering innovative solutions to an increasingly acute and potentially widespread problem. In addition, the relatively low cost of entry for these enterprises, facilitated in many ways by the technology that in other circumstances may function coercively, makes co-operative ownership models a viable option for platform workers uniquely as compared to other populations of workers. Further, given the direct-ownership model of co-ops, congruence between control and the environment may be more readily achieved, at least where this is mediated by a small number of worker-owners in a geographically defined space. More broadly, as a model for collective labour action, co-operative ownership may function vis-à-vis external governmental actors, 2020 CanLIIDocs 3123 and can work towards advancing labour interests for platform workers in the arena of law and policy. The extent to which co-operative ownership holds promise for broader populations of precarious workers, however, may remain uncertain. The low cost of entry to the market associated with online platforms is a significant advantage of this model in the context of platform work. However, as noted earlier, some enduring forms of precarious work, such as cleaning and domestic services, may transition to online platform models, creating options for worker-owned co-operative enterprises to transform such labour sectors.
Public-facing communication channels, such as protests, boycotts and public awareness campaigns, may operate in weak ways to support channels of representation and create influence vis-à-vis external actors, particularly those outside of the historical employment relationship. These forms of collective labour action do some work to make visible the hidden and often obscured centrality of the worker and their labour in the platform economy. As such, these forms of collective labour action may assist workers in advancing their interests and effecting change through direct communication with external stakeholders, such as the consumer, enterprise, and regulatory actors. These channels can, in addition to identifying and communicating the interests to be advanced, further open channels for representation, as there are few, if any, barriers to inclusion and participation for workers. These communication channels are, however, informal and as such, the contributions provide a foundation to support representation and influence, and should largely be seen as supportive or secondary measures. This form of collective labour action is not unique to platform workers, and the previous section discussed ways in which union-affiliated activities, particularly, have engaged in similar public-facing communication channels and campaigns for agricultural workers, as an example, which is another population of precarious workers often rendered invisible and hidden. Such communication channels, again, can be important supportive measures for advocacy around workers’ rights, particularly where workers’ interests require ‘buy-in’ from external 58 UNBLJ RD UN-B [VOL/TOME 71 stakeholders outside of the direct-employment relationship, such as from consumers and regulatory actors.
Several forms of collective labour action discussed in the previous section illustrate well the ways in which service alignment and member participation positively influence internal dynamics, providing significant benefit or meaning for worker-members, separately from a consideration of influence vis-à-vis external actors. Services—what a form of collective labour action offers for its constituency or member base—is expanded in important ways for platform workers in the forms of collective labour action discussed in section II. Further, several of the identified forms of collective labour action discussed in the previous section provide meaningful channels for member participation. These attributes align with recent labour studies literature that examines how collective labour action, both including and beyond unionization, can hold benefits for workers beyond formal bargaining power and the negotiation of working conditions under a collective agreement.88
Services in the form of resources and information can be an important tool for platform and other precarious workers who may lack clarity on their existing legal 2020 CanLIIDocs 3123 rights and status. Further, as discussed in section II, resources and information-sharing amongst workers, as facilitated through some forms of collective labour action, can assist workers in identifying common challenges and trends in their working relationships. This can both create a sense of community and belonging in its own right, combatting effects of social and geographic isolation inherent in platform work, and can also be used to form a foundation for further collective labour action that looks to influence external actors, such as the enterprise ‘employer’, government, or the general public, as discussed above. Various forms of collective action described earlier provide information and resource-sharing services, which may align with basic informational needs of workers, as well as providing a secondary benefit of networking and connection amongst workers, thus combating isolation and mitigating against the noted barriers to collective organizing amongst geographically dispersed and autonomous workforces. These benefits are readily extendable beyond platform workers. Similar channels for workers in various precarious and non-standard labour contexts may work to both enhance a sense of well-being and community, while also providing important information and resources on topics of interest, such as legal rights and entitlements.
The union affiliation form of collective labour action discussed in section II, while not unique to engagement with platform workers, illustrates well how the trade union—as the historically dominant labour relations form—is further adapting its function and purpose to create better congruence between its membership and services. Beyond formally representing union members, engagement in broader advocacy and resource provision to non-union sectors and workers such as platform workers, and agricultural workers, can both assist a union in building a potential membership base for formal representation, and also reimagines the purpose of the trade union as advancing worker voice and interests more broadly. Like with the creation of unions
88 See e.g. Tapia & Ibsen, supra note 46; Hyman & Gumbrell-McCormick, supra note 45. 2020] PLATFORM WORKERS AND COLLECTIVE LABOUR ACTION 59 for platform workers, this is a modest shift in form of collective labour action from the historical dominant model of formal unionization. Yet, it similarly evidences the potential of this existing core to adapt to a modern economy, including re-envisioning its function and purpose to advocate for decent work on behalf of all workers, and perhaps especially the growing number of precarious workers for whom formal representation under a trade union is inaccessible, as discussed earlier.
In addition to service provision, evaluating forms of collective labour action asks about the internal dynamic fostered through participation of members. This criterion is clearly exemplified in many of the identified forms of collective labour action taken up by platform workers, as discussed in section II. Channels of participation should aim to “maximize members’ feeling of empowerment” and, possibly, sense of belonging within the collective.89 Given the fragmentation, geographical dispersion, and autonomous nature of platform work, as with many other forms of precarious labour, channels for member participation may be a particularly significant benefit of alternative forms of collective labour action. As discussed above, some of the identified forms, while focused on service provision, provide additional opportunity for member participation through networking as well as grassroots 2020 CanLIIDocs 3123 information sharing. This may foster a sense of empowerment and belonging, as well as a sense of community, amongst workers who are otherwise largely isolated.
New union creation and co-operative ownership models, while creating more open channels of representation, are also grounded by worker participation at their core, enhancing the sense of empowerment and ownership workers will perceive, and hold, over their labour, as well as in relation to collective advocacy or bargaining efforts. Indeed, these forms of collective labour action may hold particular promise for addressing noted gaps for precarious workers in the modern economy given the extent to which they increase congruence between representation and environment, on the one hand, and between membership and participation on the other hand. However, as mentioned earlier, new union creation will be a viable option particularly in jurisdictions that enable flexible unionization models under law, notably in Europe. This makes this form of collective labour action weaker in jurisdictions like Canada, where difficulties in opening up the channels for representation under current law would remain. However, as discussed earlier, associational models outside of formal unionization may still hold relative influence in advocating and advancing the labour interests of workers to a broader set of stakeholders, such as consumers and regulatory actors. Relatedly, the promise of co-operative ownership models is constrained by the higher cost of entry to the market outside of online platform-based work, as discussed earlier. Nonetheless, where a sector of precarious labour may find ways to transition to platform-based work, such as with respect to a few examples concerning cleaning and domestic services set out earlier in this article, such a model may prove a viable form of collective labour action for workers.
89 Tapia et al, supra note 67 at 6. 60 UNBLJ RD UN-B [VOL/TOME 71
Conclusion
Digital platform work has captured much attention from legal and other scholars, policy makers, and others, in recent years. While this form of labour presents distinct challenges for regulation in the modern economy, it also reveals enduring challenges of and for labour law. Particularly, while distinct in some ways, platform work in many other ways mirrors and aligns with noted historical trends in relation to broader conceptions of precarious work that exist across a range of labour sectors and occupations. Much of the dialogue surrounding platform work to date has focused on resolving the question of employment status. However, as this article demonstrated, this focus has distracted attention from the deeper underlying questions about the function and effectiveness of labour and employment law more broadly and, again, for both platform workers and a wider population of precarious workers.
The technology associated with platform work, coupled with the heightened attention on this form of precarious labour, has in some ways produced novel approaches to advancing labour interests for workers outside of the confines of existing labour and employment law regimes and taxonomies. These forms of 2020 CanLIIDocs 3123 collective labour action, as identified in this article, present interesting case studies from which to consider how alternative forms of collective labour action may hold some promise for labour, and its legal regulation, in a modern economy that is increasingly characterized by non-standard work and workers that fall outside of historical regulatory regimes. This article canvassed various forms of collective labour action identified in relation to platform workers, and commented on the various benefits and limitations associated with each form, having regard to the ways in which these forms open channels of representation, hold influence vis-à-vis external actors, create service alignment with member needs, and open channels for participation by members. As stated at the outset, the aim of this article is to provide a rich descriptive foundation of the identified forms of collective labour action, their benefits and limitations, in order to set the stage for further dialogue and examination of the future of labour (law) in the modern economy.
THE UNEASY CASE OF PROGRAMMED OBSOLESCENCE
Pierre-Emmanuel Moyse*
I—Planned Obsolescence ...... 66 A—London’s Law of Obsolescence ...... 66 B—Social Engineering: The Influence of Thorstein Veblen ...... 75 C—Technocrats ...... 80 II—Stories of Programmed Obsolescence ...... 85 A—The Light Bulb Affair ...... 86 2020 CanLIIDocs 3123 1—General Electric and the Phoebus Cartel ...... 86 2—Evidence of Programmed Obsolescence ...... 90 3—State Authorities and Product Lifespan ...... 93 B—Y2K Period ...... 97 III—Epilogue: The Apple of Discord ...... 105 A—The Laws of Obsolescence ...... 105 B—The Bitten Apple ...... 108 IV—Conclusions ...... 113
* The author is an Associate Professor at the McGill Faculty of Law and the Director of the Center for Intellectual Property Policy (www.cippmcgill.ca). This article is part of a research project on obsolescence funded by the Canadian Social Sciences and Humanities Research Council (2018). Pierre-Luc Purtell, François Le Moine, Georgina Hartono, Renaude Morin, Christopher Little, Arthur Scalabrini, Angèle Poupard, ZiChen Gao and Nicole Thompson helped considerably in researching, editing and commenting on previous versions of the article. Early versions of this article were presented at the Hebrew University of Jerusalem, Roma Tre University and at the University of Geneva in 2018 and 2019. The author would like to thank the Canadian Social Sciences and Humanities Research Council for the generous grant, without which this project would not have been possible. 62 UNBLJ RD UN-B [VOL/TOME 71
This article is an attempt to understand the context in which the concept of programmed obsolescence has emerged and evolved to slowly make its appearance in today’s legal landscape. It concludes the first phase of our four-year research project. The text aims at describing the genealogy of the concept of programmed obsolescence, not its legal treatment. The ambition here is to revisit common places of obsolescence, from its acceptance in Bernard London’s famous pamphlet to the popular tale of the reduced life of the light bulbs. This essay focuses on facts surrounding obsolescence narratives. It describes the grammar of obsolescence, its epistemology. It is only in the second phase of our project that we will look at the legal and technical grounds, both in terms of specific remedies and statutory initiatives, to redress some of the negative impacts attributed to the phenomenon. The findings of this subsequent phase of our project constitute the object of an article in preparation.
The history and early manifestation of programmed obsolescence tells a compelling story about consumption and the contradictions of capitalism. To keep the wheels of the economy turning and workers active, more goods must be purchased. Innovation and competitive consumption are thought to be conditions of progress.
Jean-Jacques Rousseau, in his 1754 work Discourse on Inequality, predicted such: 2020 CanLIIDocs 3123
Insatiable ambition, the thirst of raising their respective fortunes, not so much from real want as from the desire to surpass others, inspired all men with a vile propensity to injure one another, and with a secret jealousy, which is the more dangerous, as it puts on the mask of benevolence, to carry its point with greater security. In a word, there arose rivalry and competition on the one hand, and conflicting interests on the other, together with a secret desire on both of profiting at the expense of others. All these evils were the first effects of property, and the inseparable attendants of growing inequality.1
Upon walking through the early days of programmed obsolescence, we realized that law is perhaps ready to address programmed obsolescence for the reason that, in the Western world, to a large extent, both obsolescence and law are products of the market economy. Obsolescence might have been accepted as an unabridged matter of business, but not of law. Law, in our Western civilization overall embraces and supports the structure of the market. In this perspective, the law of programmed obsolescence seems to be a contradiction in terms, an oxymoron. If this assumption stands, it means that the recent legislative initiatives tackling programmed obsolescence signal an important shift and are explainable perhaps by social concerns strong enough to check the forces of the liberal economy. Such concerns involve the themes of overconsumption and environment.
Let’s start with a generic definition of programmed obsolescence. Programmed obsolescence is generally understood as the engineered, premature breakdown of a product to trigger its replacement by its own maker. One can speak of an unvirtuous circle. This proposed definition is not scientific but captures the
1 Jean-Jacques Rousseau, Discourse on the Origin of Inequality, translated by GDH Cole (Penguin, 1984) at 29, online (pdf): Egalitarianism
However, programmed obsolescence seems to bring the question of durability to another level which goes beyond the jurisdictions of consumer law and its policy objective of informed choice and of commercial law. By being indissociable from obsolescence, the production of waste becomes a matter of environmental law and brings to the fore social concerns. Until obsolescence became associated with 2020 CanLIIDocs 3123 environmental concerns, the law did not develop the antibodies for a phenomenon it generally perceived as legal. In an effort to map the movements of ideas, we noticed that the topic of programmed obsolescence moved from the discreet sphere of private law, where it remained largely unnoticed, to the realm of public law. Signs of this shift appeared in other fields first. The topic of programmed obsolescence is covered in managerial economics and marketing.3 It has also made some headway in engineering and design schools to address sustainability issues at the design and production stages of manufacturing.4 This renewed interest is due, as already noted, to the social and environmental concerns raised by the volume of electronic waste as a result of the accelerated life cycle of products and limited recycling capacity. France’s recent reform is on point. It pioneered the movement against programmed obsolescence in 2015 by enacting the Loi relative à la transition énergétique pour la croissance verte,5 which criminalizes “techniques by which a manufacturer aims to deliberately reduce the life of a product to increase the replacement rate”.6 Québec too followed suit. Inspired by France’s initiative, it tabled its Bill 197, entitled An Act to amend the Consumer Protection Act to Fight Planned Obsolescence and Assert the Right to Repair Goods.7 Similar initiatives mushroomed elsewhere. But nowhere has the very
2 Consumer Protection Act, CQLR c P-40.1 at s 38 [CPA]. 3 Claude Déméné & Anne Marchand, “Exploring Users’ Practices through the Purchase, Use and Disposal Phases to Reduce the Environmental Impact of Electronic Products: A Case Study on Televisions” (2016) 29 J Research for Consumers 4. 4 Claude Déméné & Anne Marchand, “Exploring Barriers and Drivers Related to the Repair of Electronic Products” (2016) 12:1 Intl J Sustainability Policy & Practice 1. 5 Loi n° 2015-992 du 17 août 2015 relative à la transition énergétique pour la croissance verte, JO, 18 August 2015, 14263, n°0189 [Loi relative à la transition énergétique pour la croissance verte]. 6 Ibid at s L.441-2. 7 Bill 197, Loi modifiant la Loi sur la protection du consommateur afin de lutter contre l'obsolescence programmée et de faire valoir le droit à la réparation des biens, 1st Sess, 42nd Leg, Québec, 2019 [Bill 64 UNBLJ RD UN-B [VOL/TOME 71 phenomenon of programmed obsolescence been seriously studied. Its very existence has been largely presumed. There are a plethora of data and statistics with respect to the durability of various classes of products. Consumers’ literature and social media fuel the suspicion that the premature death of objects is part of a greater industrial plan or conspiracy. We remain skeptical. Legal documentation contains very little evidence of the engineered practice of obsolescence forcing the acquisition of a replacement product offered by the same maker. Is programmed obsolescence a specific and provable practice to be regulated or is it a vague, cathartic expression which diffuses our ambiguous sentiments about consumption and capitalism?
This article is an attempt to understand the grammar of obsolescence: how the word came into being and what narratives have carried it over to the present day. The exercise is to a great extent historical and epistemological. It questions the inherent conditions of modern economy. In fact, we suspect obsolescence might have been a natural companion to the market economy before it acquired a negative ring. In theory, the market owes its dynamism to the large volume of transactions explaining why the replacement rate of goods is taken as a positive indicator. After all, the creative destruction process implies a constant substitution of solutions and products for the 2020 CanLIIDocs 3123 benefit of all.8 The modern views on innovation suggests that obsolescence is consubstantial to technological progress. Obsolescence, in the evolutionary approach of modern economy, is the natural consequence of novelty and improvement. Yet, in recent years, it has caught the eye of the regulator and lawmakers. Programmed obsolescence invites us to research the reasons why manufacturers would jeopardize the goodwill built in their brands by sabotaging their own products or rendering them purposely less effective.
Thus, our journey begins at the turn of the 20th century in the United States with an original idea, that of “planned obsolescence” (I). The name of Bernard London is indissociable to an expression of which he is regarded as the author. Any proper research on the origin of obsolescence irremediably starts with him. In 1932, he published his work Ending the Depression Through Planned Obsolescence9. His concept of planned obsolescence consists in assigning a legal term on the use of products in order to force their replacement, a solution to increase production and employment. London’s proposal was an institutional response to overproduction and stalled consumption. A point of interest already emerges: the idea of planned obsolescence developed in London’s work has little to do with programmed obsolescence as it is understood nowadays. Yet, the two expressions are used interchangeably. Our inquiry into the evolution of the meaning of obsolescence compels us, however, to distinguish London’s planned obsolescence from today’s programmed obsolescence. The former is a theoretical policy plan designed to accelerate the retirement of goods under state supervision to boost domestic economy.
197]. The English version erroneously uses the expression “planned obsolescence” instead of “programmed obsolescence”. 8 Joseph Schumpeter, Capitalism, Socialism, and Democracy, 3rd ed (New York: Harper, 1950) at 81–86. 9 Bernard London, Ending the Depression Through Planned Obsolescence (New York: 1932) [London, Ending The Depression]. 2020] PROGRAMMED OBSOLESCENCE 65
The latter refers to actual strategies put in place by manufacturers, to the same end, but for their own benefit. This clarification, which is perhaps the first contribution of this article, brings to the fore the first movement of a dialectic between the two variations of obsolescence. They are related, but distinct. This text, therefore, maintains the distinction between planned and programmed obsolescence. A second point is worth mentioning. They both arise in time of crisis and are tales about modernity. Planned obsolescence is a child of the Great Depression; conversely, programmed obsolescence is the belated realization that consumption runs amok and threatens our very existence on earth. The crisis here is ecological.
The scope of the article expands beyond London’s works. Looking for an answer as to why “planned obsolescence” appeared at this exact moment under the pen of a self-taught business man, we delved into the grey matter left by intellectuals of the time in hopes of finding the origins of this notion. We found that obsolescence belongs to a complex network of political ideas and themes already in place capturing the social dimensions of rapid industrialization. It was, in a sense, an archeological dig. New York in the 1920’s was the site to excavate. London was living in New York at the time. Our research led us to the work of the unorthodox economist Thorstein 2020 CanLIIDocs 3123 Veblen, a New Yorker as well, who published some of his most important works at the time London published his pamphlets. It also revealed an unexpected filiation between Veblen and the little known Technocracy movement and its enigmatic leader Howard Scott. The Technocracy movement grew after WWI and expanded to the United States and Canada afterwards. The history of the movement and its ideological foundations help in restoring the epistemological dimension of obsolescence.
The second section presents two stories of obsolescence that took place in the past century (II). They are selected moments. Here, we enter surreptitiously into the domain of the law, but only inasmuch as it gives us the opportunity to observe documented facts that might match our definition of programmed obsolescence. Obsolescence stories, as they emerge from our case studies, are stories of disruptions under irrepressible technological changes. They expose the uneasy relationships between law and science. They show that the part played by law is essentially to mitigate the costs and social impacts of technological changes on society. These stories offer concrete examples of historical instances where the liability of the manufacturers was considered. The first case is another inescapable station on our journey to the origins of obsolescence. It is the light bulb affair. In the 1920s, under the leadership of General Electric, the largest light bulb manufacturers established an international alliance that controlled the quality and properties of light bulbs. The second case is the Y2K conundrum, during which fear arose due to computers and other machines’ inability to record the presence of the new millennium. One can retrospectively question the liability of programmers who were well aware of the limits of the software.
The last section of the article draws on our historical and philosophical excursions (III). It builds on the two previous sections and the observations we have made along the way to better understand the recent legislative initiatives tackling programmed obsolescence. While these new legislative mechanisms have yet to be tested in court, some cases have drawn the attention of the media and consumer 66 UNBLJ RD UN-B [VOL/TOME 71 organizations, which have painted them as instances of obsolescence. The Apple Batterygate, which arose following Apple’s 2017 admission that it had throttled the performance of iPhones to avoid unexpected shutdowns, reminds us that assessing the legality or illegality of obsolescence is a highly complex task. Obsolescence, and in particular the obsolescence of software-enabled devices, continues to fall outside the purview of the law. It seems, once again, to defy the law.
I—Planned Obsolescence
This section situates us in the United States at the turn of the 20th century when the term “planned obsolescence” was first coined by philanthropist and dilettante politician Bernard London. In a series of short texts published between 1932 and 1935, London outlined possible policy solutions to end the Great Depression. He proposed to set a legal term for the use of products to force their replacement and boost production, consumption and employment. His solution was meant to resolve what was then perceived as a cruel contradiction: technology and machines optimized production and allowed for a greater quality and quantity of products - the materialistic 2020 CanLIIDocs 3123 definition of progress. However, American households, crippled by debt and plagued by unemployment, were unable to buy these products. The increasing output capacity did not meet the fluctuating spending power. The Great Depression magnified the causal or consequential links between the cogs in the machine of the modern market economy. The need to understand the economic operations and their complexities drove the development of new theories and provoked intense debate. Unsurprisingly, this period saw the emergence of a literature examining patterns of consumption and an increasing interest for the role of obsolescence.
London’s papers will be presented first (A). His views will be analyzed in the broader intellectual context in order to grasp the social and economic changes brought by modernity. To better situate London’s ideas, the second section will draw on the work of one of the most influential socio-economist of his time, Thorstein Veblen (B). Politicians and intellectuals alike, represented here by London and Veblen, sought ways to gain better control and understanding over the economy at a time where innovation in technology placed greater importance on technical expertise, thus shifting the relevant knowledge from the realm of human sciences to the schools of engineers. The Technocracy movement, born in New York in the 1930s and rooted in the fertile ground of Veblen’s ideas, exemplifies the rendezvous set by the Great Depression, one that called for a reflection on the articulation between modern economy, science and society. The Technocracy movement will be presented in the third part (C). The theme of obsolescence is the common thread that holds the three parts of this story together.
A—London’s Law of Obsolescence
Little is known of the life of Bernard London. He is reported to be a Russian-born successful real estate broker who became active in New York financial, political and 2020] PROGRAMMED OBSOLESCENCE 67 academic circles in the 1920s and 30s.10 His essays were self-published and contained little to no references, suggesting that he was a self-taught author with no formal economic training. In 1932, he introduced the concept of planned obsolescence in his pamphlet Ending the Depression Through Planned Obsolescence.11 Briefly stated, planned obsolescence consists in assigning a lease of life to products so as to limit their duration. The essay became the first chapter of a small book, The New Prosperity: Permanent Employment, Wise Taxation and Equitable Distribution of Wealth, published in 1933.12 In this piece, he further details his proposed solution for economic recovery and the creation of a governmental emergency taskforce in charge of preventing “any breakdown in the nation’s economic structure.”13 In 1935, London published his last work, Rebuilding a Prosperous Nation through Planned Obsolescence, which presents planned obsolescence as a solution to the real estate crisis.14
While London is known for coining the term “planned obsolescence,” the concept was brewing in the 1920s. In 1928, a few months before the crash, Justus George Frederik, business executive and founder of the Advertising Men’s League of
New York, suggested that “progressive obsolescence” was the path toward increased 2020 CanLIIDocs 3123 consumption:
We must induce people who can afford it to buy a greater variety of goods on the same principle that they now buy automobiles, radios and clothes, namely: buying goods not to wear out, but to trade in or discard after a short time, when new and more attractive goods or models come out… Therefore the one salvation of American industry, which has the capacity for producing 80 to 100 percent more goods than are now consumed, is still to foster more progressive obsolescence principle, which means buying for up-to-datedness, efficiency, and style, buying for…the sense of modernness rather than simply for the last ounce of use.15
Similarly, the emergence of the concept of “consumer engineering,” the increasing interest for the value of design and the consolidation of marketing as a discipline,
10 Giles Slade, Made to Break: Technology and Obsolescence in America (Cambridge, MA: Harvard University Press, 2006) at 72; Vinay Gidwani, “For a Marxist Theory of Waste: Seven Remarks” in Jini Kim Watson & Gary Wilder, eds, The Postcolonial Contemporary: Political Imaginaries for the Global Present (New York: Fordham University Press, 2018) at 187. 11 London, Ending The Depression, supra note 9. 12 Bernard London, The New Prosperity: Permanent Employment, Wise Taxation and Equitable Distribution of Wealth (New York: 1933) [London, New Prosperity]. 13 Ibid at 30. 14 Bernard London, Rebuilding a Prosperous Nation Through Planned Obsolescence (New York: 1935). Unlike his previous work, this last piece appears to be a non-solicited report intended primarily for government officials, not the general public. 15 J George Frederick, “Is Progressive Obsolescence the Path Toward Increased Consumption?”, Advertising and Selling 11:10 (5 September 1928) 19 at 19–20, 44. His wife, economist Christine Frederick, made a similar point in 1929 in Christine Frederick, Selling Mrs. Consumer (New York: Business Bourse, 1929) (arguing that “America’s triumphs and rapidity of progress are based on progressive obsolescence [emphasis in original]” at 246). 68 UNBLJ RD UN-B [VOL/TOME 71 further emphasized and articulated the role of obsolescence as a tool to stimulate the consumption of goods. In 1932, Roy Sheldon and Egmont Arens, advertising gurus, co-authored the book Consumer Engineering: A New Technique for Prosperity, in which they presented the concept of “obsoletism”:
Obsoletism is another device for stimulating consumption. This element of style is a consideration in buying many things. Clothes go out of style and are replaced long before they are worn out. That principle extends to other products—motorcars, bathrooms, radios, foods, refrigerators, furniture. People are persuaded to abandon the old and buy the new in order to be up- to-date, to have the right and correct thing. Does there seem to be a sad waste in the process? Not at all. Wearing things out does not produce prosperity, but buying things does.16
Progressive obsolescence and obsoletism prefigured the modern field of cognitive studies on consumer behaviour. The concepts also signaled the birth of industrial design and an appetite for new forms of intellectual property protections for distinguishing guise. These early variations of obsolescence claim that controlling, or at least stimulating the replacement of goods before they are out of use, is key to 2020 CanLIIDocs 3123 avoiding the saturation point of markets. The role of consumption in driving the economic engine led Joseph Mazur, a New York City banker, to conclude that “obsolescence has been a vital ingredient in American business prosperity”. He acknowledged the role of obsolescence by wear, but further highlighted obsolescence due to consumers’ changing tastes. 17 The literature on consumer behaviour and its emphasis on advertising techniques aimed at diminishing the desirability of older goods reflects a growing realization in the first half of the 20th century that physical deterioration over time is not the sole agent motivating the replacement of products. Businesses could start engineering early obsolescence to stimulate consumption., Alternatively, they could introduce, at an accelerated pace, similar but slightly improved or redesigned products into the market. Today, research is being conducted on aesthetic, psychological18 or cosmetic obsolescence to better understand how consumers perceive the utility and value of their goods.19 London may well have encountered the terminology and ideas emerging in the field of advertising during his time. His work indeed reflects the same underlying objective of supporting consumption within a linear economy rhythm by the replacement of goods and investments.
London’s concept of planned obsolescence, however, relied on regulation
16 Roy Sheldon & Egmont Arens, Consumer Engineering: A New Technique for Prosperity (New York: Harper and Brothers, 1932) at 7; Slade, supra note 10 at 66–67. 17 Joseph Mazur, American Prosperity: Its Causes and Consequences (New York: The Viking Press, 1928) at 97. 18 Claudia Déméné & Anne Marchand, “L’obsolescence des produits électroniques : des responsabilités partagées” (2015) 10:1 Ethics Forum 4 [Déméné & Marchand, “L’obsolescence des produits électroniques”]. 19 Debra Lilley et al, “Cosmetic Obsolescence? User Perceptions of New and Artificially Aged Materials” (2016) 101 Materials & Design 355. 2020] PROGRAMMED OBSOLESCENCE 69 and not on marketing strategies or other private instruments. Planned obsolescence is intended to curb the natural tendency of the impoverished mass of consumers to use their belongings as much and as long as possible. Interestingly, his emphasis on the role of government and consumption brings his work closer, although not comparable in quality or depth, to that of economists such as John Maynard Keynes who, starting in the 1920s, explored similar themes. In his 1936 book, Keynes argued that the fall in expenditure was a root cause of the Great Depression.20 In an attempt to understand the factors influencing the propensity to consume, Keynes posited that the amount a community spends partly depends on “subjective needs and the psychological propensities and habits of the individuals,” which, in his opinion, are unlikely to be altered except in “abnormal or revolutionary circumstances.”21 While the technological revolution at the time greatly increased production capacity, the cause of unemployment, he argued, was the lack of expenditures within an economy, which decreases aggregate demand:
When involuntary unemployment exists, the marginal disutility of labour is necessarily less than the utility of the marginal product. Indeed it may be much less. For a man who has been long unemployed some measure of 2020 CanLIIDocs 3123 labour, instead of involving disutility, may have a positive utility. If this is accepted, the above reasoning shows how “wasteful” loan expenditure may nevertheless enrich the community on balance. Pyramid-building, earthquakes, even wars may serve to increase wealth, if the education of our statesmen on the principles of the classical economics stands in the way of anything better.22
In his view, times of crisis require structural intervention by the state to generate wealth throughout the economy. Public spending policies, such as the New Deal, which were implemented later in the century and helped redress the economy, gave unquestionable credit to his theories.23
Naturally, writing in 1932, London may not have had access to Keynes’ insights. He did not have the ambition to reinvent modern economy either. Unlike Keynes who supports government intervention only in times of crisis, London’s proposal does not seem to have the same temporality.24 Both, however, are
20 John Maynard Keynes, The General Theory of Employment, Interest and Money (London, UK: Palgrave Macmillan, 1936). See also Peter Temin, Did Monetary Forces Cause the Great Depression? (New York: WW Norton and Company, 1976). 21 Keynes, supra note 20 at 81. See the factors in ibid at 95. 22 Ibid (Keynes used the term loan expenditure “to include both public investment financed by borrowing from individuals and also any other current public expenditure which is so financed” at 114). 23 See Milton Friedman & Anna Jacobson Schwartz, The Great Contraction 1929–1933 (Princeton: Princeton University Press, 1965). See also Karl Brunner, ed, The Great Depression Revisited (Boston: Martinus Nijhoff Publishers, 1981). In the 1960s and 1970s, however, the Monetarist, including the likes of Friedman and Schwartz, started questioning Keynesian supremacy and offered their own explanation for the Great Depression, emphasizing the “Great Contraction” of the money stock as its main cause. 24 London, Ending The Depression, supra note 9 (“[t]he present deadlock is the inevitable result of traveling along blind alleys. Chaos must unavoidably flow from an unplanned economic existence. In the future, we must not only plan what we shall do, but we should also apply management and planning to undoing the 70 UNBLJ RD UN-B [VOL/TOME 71 sympathetic to the idea of a planned economy in some circumstances and, to a certain degree, to increasing expenditures, ensuring employment and improving the living conditions of the mass.25 Their work similarly highlights the role of the government in fostering the conditions necessary to enable and provoke consumption. In this regard, London stages, with undeniable political wit, the same constitutive elements of what Keynes called “involuntary unemployment.” The problem of underconsumption and the decline of buying power is in fact a recurring theme in Ending the Depression Through Planned Obsolescence: “Factories, warehouses, and fields are still intact and are ready to produce in unlimited quantities but the urge to go ahead has been paralyzed by a decline in buying power.”26 Many at the time shared London’s belief that underconsumption was the prime driver of the Great Depression, rather than the symptom of more complex conjunctures that experts would take decades to unfold.27 The logic seems impeccable: underconsumption leads to overproduction which, in turn, and absent sufficient demand, results in unemployment.
Overproduction or underconsumption were often presented, alternatively, as self-explanatory causes of unemployment when they were symptomatic of far more 28 complex causes. In Ending the Depression, London suggests to boost the economy 2020 CanLIIDocs 3123 by legally reducing the life of consumable goods, from kitchen appliances to buildings and wheat.29 Planned obsolescence is, for London, a state-led initiative: obsolete jobs of the past” at 5). London explains that “[his] suggested remedy would provide a permanent source of income for the Federal Government and would relieve it for all time of the difficulties of balancing its budget” (ibid at 6). See also London, New Prosperity, supra note 12 (noting that emergency relief is not enough and has to be combined with solutions aimed at “permanent improvement”: “[t]he old method is temporary cure; but mine is permanent intervention, which will obviate costly cure” at 61). 25 Utilitarian economic thinking had also gained traction in the decades preceding the publications of London and Keynes. For instance, Cambridge economist Arthur Cecil Pigou elaborated his concepts of “marginal private product of an activity” (the benefits obtained by a private party undertaking an activity) and “marginal social product of an activity” (the benefits obtained by society from this activity) to understand some forms of market failures such as monopolies. See Arthur Cecil Pigou, The Economics of Welfare (London, UK: Macmillan, 1920); Roger Backhouse, The Ordinary Business of Life: A History of Economics from the Ancient World to the Twenty-First Century (Princeton: Princeton University Press, 2004) (noting that, with these concepts, “Pigou offered a detailed programme for economic policy, virtually providing a blueprint for the welfare state” at 274). In General Theory, Keynes criticized Pigou’s Theory of (Un)Employment: see Keynes, supra note 20 at 240–48. However, Pigou’s work shares some similarities to London’s proposal, which calls upon the government to implement an adequate economic organization of society to ensure that the welfare of society is not “left to pure chance and accident” and “the unpredictable whims and caprices of the consumer” thus suggesting an arrangement of individual benefits and societal benefits to avoid market failures: see London, Ending The Depression, supra note 9 at 4. 26 London, Ending The Depression, supra note 9 at 4; Gidwani, supra note 10 at 187; London, New Prosperity, supra note 12 (arguing “the real cause of unemployment is not overproduction but underconsumption” at 28). 27 See Gregory R Woirol, “Plans to End the Great Depression from the American Public” (2012) 53:4 Labor History 571. In 2012, the economist Gregory R Woirol reviewed thousands of letters sent to the Roosevelt administration by Americans during the Great Depression and found that a majority of the correspondences proposed measures to increase consumer purchasing power and consumer spending to solve the issue of underconsumption (ibid at 573). 28 See Lionel Robbins, The Great Depression (New York: Routledge, 2009). 29 London, Ending the Depression, supra note 9 at 13, 15. 2020] PROGRAMMED OBSOLESCENCE 71
I would have the Government assign a lease of life to shoes and homes and machines, to all products of manufacture, mining and agriculture, when they are first created, and they would be sold and used within the term of their existence definitely known by the consumer. After the allotted time had expired, these things would be legally “dead” and would be controlled by the duly appointed governmental agency and destroyed if there is widespread unemployment. New products would constantly be pouring forth from the factories and marketplaces, to take the place of the obsolete, and the wheels of industry would be kept going and employment regularized and assured for the masses.30
This peculiar plan was a necessity since, according to London, American consumers had blatantly ignored the law of obsolescence: “[t]hey are using their old cars, their old tires, their old radios and their old clothing much longer than statisticians had expected on the basis of earlier experience.”31 Based on these observations, London focuses on the acts of expenditures of consumers, which did not increase in line with the increased productivity and technological innovation of the market. London believes that the solution lies in the intervention of statesmen, not in the natural 2020 CanLIIDocs 3123 dispositions of the economy, stating that “[t]he existing troubles are man-made, and the remedies must be man-conceived and man-executed.”32 The market, as well as traditional economic models, had failed to redress the situation.
As we have seen, London’s plan builds on the themes discussed by academics, the advertising world, and the media: market failures, prices, overproduction, underconsumption, unemployment, obsolescence.33 London’s work thus distills a certain impression of déjà vu, using for its own political agenda themes that were already well articulated by scholars before him. He, too, sees the economy through its social actors—the government, the entrepreneurs, the capitalists, the engineers and the consumers—and not just through the narrow prism of the price mechanism.34 London’s proposal also reflects a desire, in dire times, to explain society
30 London, Ending the Depression, supra note 9 at 6. London adds, “I propose that when a person continues to possess and use old clothing, automobiles and buildings, after they have passed their obsolescence date, as determined at the time they were created, he should be taxed for such continued use of what is legally “dead’” (ibid at 8). 31 Ibid (noting, “[p]eople generally, in a frightened and hysterical mood, are using everything that they own longer than was their custom before the depression” at 4). 32 Ibid. 33 London, New Prosperity, supra note 12 at 28. 34 London’s prose shares affinities with the writings of Karl Marx. See Gidwani, supra note 10 (explaining “London criticizes as detrimental to society the structural asymmetry between the owner of the means of production (capitalist) and those who must sell for a wage the only means of production at their disposal, a power that can be neither stored nor accumulated (laborer)” at 188). Borrowing the same rhetoric, London writes in Ending the Depression, supra note 9 at 10 that “the product of the worker’s toil continues to benefit and produce income for its owner long after the one whose sweat create it.” The central role of the state in London’s plan—not only of statesmen but also of engineers—recalls scientific socialism and materialism. These are ideologies that led to economic planning—with planned obsolescence being a peculiar manifestation of it. 72 UNBLJ RD UN-B [VOL/TOME 71 as a piece of social engineering whose design can be improved:
Under the direction of the Secretary of Emergency the Government’s engineers, economists, mathematicians and actuaries would work formulating the plans and setting up the machinery for putting into operation the program of planned obsolescence.35
Hence, London adopts a pseudo-scientific approach to the economy by entrusting experts, as government agents, with the task of calculating the replacement rate of products:
Furniture and clothing and other commodities should have a span of life, just as human have… The original span of life of a commodity would be determined by competent engineers, economists and mathematicians, specialists in their fields, on behalf of the Government.36
His faith in engineering expertise and technology was not unorthodox and must be viewed in the context of the rise of the Technocracy movement in America. The technocracy ideology, influenced to some extent by the work of renowned intellectuals 2020 CanLIIDocs 3123 such as Thorstein Veblen, was actively promoted by a group of converted engineers and bohemian intellectuals based in New York and led by Howard Scott. The movement gained some momentum in Western Canada as well. The following part examines their probable influence on London’s work.
B—Social Engineering: The Influence of Thorstein Veblen
London may have been well aware of the emerging movement in the 1920s calling for the planning of the economy and the creation of a “Soviet”37 of engineers. The idea had been brilliantly exposed in the work of Thorstein Veblen in 1929 and thereafter loosely reused by the Technocracy movement. The Technocrats, as they became known, were a group of men who purported to empower scientists and engineers to replace speculators and businessmen deemed responsible for the Great Depression. London, a businessman himself, is careful not to engage in a diatribe against the behaviours of his kind. He is, nevertheless, fully aware of the accusations. In his 1932 pamphlet, he wrote the following:
If this plan were in operation, speculators would not acquire fortune simply by manipulating and creating false values or synthetic wealth. If it were decreed that the life of wheat were to be no more than two years, for example, no one would buy the grain solely for speculation.38
35 Ibid at 30. 36 London, Ending the Depression, supra note 9 at 12. 37 Thorstein Veblen, The Engineers and the Price System (Kitchener: Batoche Books, 2001), online (pdf): McMaster University
The work of London here intersects with themes developed by Veblen, starting with his attraction for a planned economy. Veblen was one of the first social philosophers to point out that money follows the path of profit regardless of people or productivity, thus creating a real disconnect between finance and the real economy.39 Veblen supports the idea of enrolling technicians and industrial experts in the government. He was convinced that their technical knowledge would make them better leaders because they are able to understand the impacts of technology on society. Entrepreneurs, and capitalists in general, he observed, had lost touch with reality. This distortion, in his own words, led to the “uneconomical use of material resources, and an incredibly wasteful organization of equipment and manpower in those great industries where the technological advance has been most marked.”40
By the time Ending Depression was published in 1932, Veblen had already outlined the idea of a “Soviet of Technicians” to conduct the economy at the time of mechanization.41 Veblen was already a well-established scholar of the left, albeit controversial, and a prominent figure in American economics. However, he would 42 always remain “an outsider to the mainstream of American society.” He first 2020 CanLIIDocs 3123 garnered attention in 1899 with The Theory of the Leisure Class, in which he raised the issue of conspicuous consumption and developed his theory of consumption by emulation as a trigger of competitive spending.43 In 1914, he published The Instinct of
39 See John Patrick Diggins, “Thorstein Veblen and the Literature of the Theory Class” (1993) 6:4 Intl J Politics Culture & Society 481 (arguing that “Veblen’s greatest contribution to economic thought [was] the demonstration that industry and business operated for different principles and thus maximization of output can jeopardize the maximization of profit” at 482). 40 See Veblen, Engineers and the Price System, supra note 37 at 40:
[T]hey have been unable to rely on the hired-man’s-loyalty of technologists whom they do not understand. The result has been a somewhat distrustful blindfold choice of processes and personnel and a consequent enforced incompetence in the management of industry, a curtailment of output below the needs of the community, below the productive capacity of the industrial system, and below what an intelligent control of production would have made commercially profitable. 41 Interestingly, despite giving engineers a key role in his plan, London avoids tackling the topic of mechanization. Numerous writings, at that time, accused machines and their makers of stealing the jobs of millions. A case in point is the essay published in Fortune entitled the “Obsolete Men” written by three- time Pulitzer prize recipient Archibald MacLeish “condemning the ‘technological unemployment’ that had led to ‘a serious decline in the number of wage earners in basic industries.” See Archibald MacLeish, “Obsolete Men”, Fortune 6 (December 1932) at 25–26, 91–92, 94. See also Henri Elsner, Messianic Scientism: Technocracy: 1919-1960 (PhD Thesis, University of Michigan, 1962), online:
Workmanship and The State of Industrial Arts in which he elaborated on the impact of the process of mechanization on class divisions and knowledge distribution. While businessmen are well versed in finance, accounting and management, they know little about the technology from which they derive profits. The theme runs through Veblen’s work: the modern businessman is out of touch with technology, unfit and incompetent to exercise surveillance over the processes of industry.44 For Veblen, technological changes impact social structures and new knowledge is necessary to govern the industrial society. In his ideal government, the role of the engineer is predominant. They are the central figures of The Engineers and the Price System published in 1921.45 In this book, Veblen called upon technicians to take command, replacing those whose “services, proximate or remote, to society are often of quite a problematical character” and “are within the law and within the pale of popular morals.”46 In the final features of the social structure which this institution of ownership touches. The possession of wealth confers honour; it is an invidious distinction” at 13–14). Some of his thoughts are visionary descriptions of modern consumption. See e.g. ibid at 48–49:
This suggests that the standard of expenditure which commonly guides our efforts is not the average, ordinary expenditure already achieved; it is an ideal of consumption that lies just beyond 2020 CanLIIDocs 3123 our reach, or to reach which requires some strain. The motive is emulation—the stimulus of an invidious comparison which prompts us to outdo those with whom we are in the habit of classing ourselves. Substantially the same proposition is expressed in the commonplace remark that each class envies and emulates the class next above it in the social scale, while it rarely compares itself with those below or with those who are considerably in advance. Standards of repute and canons of conduct, not necessity or satisfaction of reasonable needs, are the vectors of consumption. Reification is not so much a question of production than identification to a class and style. Commodities send signs of social affiliation and success. Veblen’s provocative style remains inimitable and is still very effective. See e.g. ibid at 40:
From the foregoing survey of the growth of conspicuous leisure and consumption, it appears that the utility of both alike for the purposes of reputability lies in the element of waste that is common to both. In the one case it is a waste of time and effort, in the other it is a waste of goods. Both are methods of demonstrating the possession of wealth, and the two are conventionally accepted as equivalents. The choice between them is a question of advertising expediency simply, except so far as it may be affected by other standards of propriety, springing from a different source. 44 Thorstein Veblen, The Instinct of Workmanship and the State of the Industrial Arts (New York: Macmillan, 1914) at 222 online (pdf): Internet Archive
chapter, “A Memorandum on a Practical Soviet of Technicians,” he proposes to dislodge the “captains of the industry.”47 They are the “vested interests and their absentees owners.”48 Veblen’s belief that businessmen are ill-suited to manage production stems from his theory of value that distinguishes pecuniary value (or market value) and industrial value (or “material serviceability”).49 The former relies essentially on the price system which presents itself as a teleological mode of valuation whose prime purpose is to determine profit. Ultimately, however, this business or commercial value, as reflected in the price system, has no or very little connection to the industrial value.50 Industrial value is about assessing the state of progress and the overall evolution of a given civilization. As such, the idea of industrial value espouses a sense of collective responsibility in the sharing of technological knowledge by virtue of which the industrial life should not be “organized on a pecuniary basis and managed from the pecuniary side.”51 The distinction made between industry and business values or, to put it more bluntly, between social progress and profit, foreshadows the current demands to prevent and criminalize programmed obsolescence. Growing concerns for the environmental and social costs of the premature replacement of goods and waste fuels the criticisms against the corporate focus on pecuniary value and private interests
to the detriment of the common good, that is, the safeguard of the environment and 2020 CanLIIDocs 3123 humanity.52
Veblen’s contention that the common stock of technical or pragmatic knowledge must be preserved, developed, and shared also remains particularly relevant to the current discourse on programmed obsolescence. Interestingly, Veblen does not favourably view the use and expansion of intellectual property. For him, the joint stock of knowledge cannot be reserved for private interests:
47 Veblen, Engineers and the Price System, supra note 37 at 101. 48 Ibid. 49 See Olivier Brette, “Thorstein Veblen on Value, Market and Socioeconomic Progress” (2014) 26:1/2 European J Economic & Social Systems 55. 50 Veblen, “Industrial and Pecuniary Employments”, supra note 46 at 228. 51 Ibid at 227. 52 While programmed obsolescence is in today’s context inseparable from environmental concerns and has become central to many scholars’ work, little attention was given to this issue by their forefathers. The notion of waste finds its way into the writings of Veblen and London but in a very different meaning: one that is not even remotely related to the environment. For them, waste is an economic term describing inefficiency and is not charged with any ecological undertone. In fact, London provides no detail as to waste administration and the necessary resources to collect, store or recycle obsolete goods: see London, Ending the Depression, supra note 9 at 7–8. As for Veblen, neither the concept of obsolescence nor that of waste— both of which are sporadically used in his writing—relates to the impact of waste on the environment. Veblen’s reflection on capitalism, however, raises the issue of waste, more specifically “conspicuous waste,” which he also refers to as “wasteful consumption”. In The Theory of the Leisure Class, wasteful consumption is described as the practice whereby bourgeois elites consume goods, often expensive ones, to parade their wealth and display their financial power in society. He writes the following in Theory of the Leisure Class, supra note 43 at 40: “[f]rom the foregoing survey of the growth of conspicuous leisure and consumption, it appears that the utility of both alike for the purposes of reputability lies in the element of waste that is common to both. In the one case it is a waste of time and effort, in the other it is a waste of goods. Both are methods of demonstrating the possession of wealth, and the two are conventionally accepted as equivalents”. 76 UNBLJ RD UN-B [VOL/TOME 71
The state of the industrial art is a joint stock of knowledge derived from past experience, and is held and passed on as an indivisible possession of the community at large. It is the indispensable foundation of all productive industry, of course, but except for certain minute fragments covered by patent rights or trade secrets, this joint stock is no man's individual property.53
As we will see, Veblen’s call for knowledge sharing and transparency in the industrial process challenges some of the key ingredients required to implement programmed obsolescence: the lack of information regarding product design and durability and the control of knowledge and technology through private property regimes, such as intellectual property rights. In Veblen’s opinion, only a policy focused on industrial knowledge could improve the overall living conditions in modern society:
[T]he mechanical technology is impersonal and dispassionate, and its end is very simply to serve human needs, without fear or favor or respect of persons, prerogatives, or politics. It makes up an industrial system of an
unexampled character—a mechanically balanced and interlocking system 2020 CanLIIDocs 3123 of work to be done, the prime requisite of whose working is a painstaking and intelligent coordination of the processes at work, and an equally painstaking allocation of mechanical power and materials. The foundation and driving force of it all is a massive body by technological knowledge, of a highly impersonal and altogether unbusinesslike nature, running in close contact with the material sciences, on which it draws freely at every turn. 54
It follows that engineers, who are “gifted, trained, and experienced technicians who now are in possession of the requisite technological information and experience,” must actively participate in the administration and governance of modern society.55 They make up “the General Staff of the industrial system, in fact; whatever law and custom may formally say in protest” and while “the ‘captains of industry’ may still vaingloriously claim that distinction, and law and custom still countenance their claim… the captains have no technological value, in fact.”56 Following the “abdication or dispossession” of the captains of industry, a directorate of technicians would take over the industrial system.57 The passage from the old to the new system would unfold
53 Veblen, Engineers and the Price System, supra note 37 at 19, 34. 54 Ibid at 81–82. 55 Ibid at 82. 56 Ibid. 57 Veblen notes the following in ibid at 44:
These expert men, technologists, engineers, or whatever name may best suit them, make up the indispensable General Staff of the industrial system; and without their immediate and unremitting guidance and correction the industrial system will not work. It is a mechanically organized structure of technical processes designed, installed, and conducted by these production engineers. Without them and their constant attention the industrial equipment, the mechanical appliances of industry, will foot up to just so much junk. The material welfare of the community is unreservedly bound up with the due working of this industrial system, and therefore with its unreserved control by the engineers, who alone are competent to manage it. To do their work as it should be done these men of the industrial general staff must have a free hand, unhampered by commercial 2020] PROGRAMMED OBSOLESCENCE 77 as follows:
The incoming industrial order is designed to correct the shortcomings of the old. The duties and powers of the incoming directorate will accordingly converge on those points in the administration of industry where the old order has most signally fallen short; that is to say, on the due allocation of resources and a consequent full and reasonably proportioned employment of the available equipment and man power; on the avoidance of waste and duplication of work; and on an equitable and sufficient supply of goods and services to consumers. Evidently the most immediate and most urgent work to be taken over by the incoming directorate is that for want of which under the old order the industrial system has been working slack and at cross purposes; that is to say the due allocation of available resources, in power, equipment, and materials, among the greater primary industries. For this necessary work of allocation there has been substantially no provision under the old order.58
Engineers and technical knowledge are the beating heart of London and Veblen’s vision of a planned economy. While Veblen calls on experts to design the whole 2020 CanLIIDocs 3123 industrial system—its equipment, processes and outcomes, the role that they play in London’s proposal is limited to the design of consumable goods. Although their plans do not have the same breadth, they both seek to take control away from agents who pursue their own private interests and who, by doing so, disregard the welfare of society. Engineers appear as guardians of the common good against what London saw as the “unpredictable whims and caprices of the consumer” and the “haphazard, fickle attitudes of owners,”59 or, in Veblen’s works, the profit-seeking-at-any-cost “absentee owners” and “conspicuous consumers.”60
C—Technocrats
The inquiry into the social and intellectual scene where London staged his proposal leads us to the Technocracy movement that emerged in the United States during the
considerations and reservations; for the production of the goods and services needed by the community they neither need nor are they in any degree benefited by any supervision or interference from the side of the owners. Yet the absentee owners, now represented, in effect, by the syndicated investment bankers, continue to control the industrial experts and limit their discretion, arbitrarily, for their own commercial gain, regardless of the needs of the community. 58 Drawing from Marx, Veblen thought that engineers, rather than workers, would be the ones to overthrow the industrial system. See Lewis Corey, “Veblen and Marxism” (1937) 1 Marxist Q 62; Abram L Harris, “Economic Evolution: Dialectical and Darwinian” (1934) 42:1 J Political Economy 34; Forest G Hill, “Veblen and Marx” in Douglas Dowd, ed, Thorstein Veblen: A Critical Reappraisal (Ithaca: Cornell University Press, 1958) at 129–49; Bernard Rosenberg, “Veblen and Marx” (1948) 15:1 Social Research 99; Paul M Sweezy, “The Influence of Marxism on Thorstein Veblen” in DD Egbert and Stow Parsons, eds, Socialism and American Life, vol 1 (Princeton: Princeton University Press, 1950) at 473–77. 59 London, Ending the Depression, supra note 9 at 4–5. 60 Veblen, Engineers and the Price System, supra note 37 at 101; Veblen, Theory of the Leisure Class, supra note 43 at 40. 78 UNBLJ RD UN-B [VOL/TOME 71 interwar period, then spread west and across North America.61 Chapters and associations affiliated with the movement appeared in Western Canada during the Great Depression. In Manitoba and Alberta, especially, the consequences of the crisis were harshly felt. At the University of Alberta, a dedicated fund allowed for the creation of a unique and vast collection of articles, correspondence of members, pamphlets, and other writings related to the Edmonton and Calgary chapters of Technocracy Inc. 62 Completed in 2007, the archives represent a valuable and underexploited source of documentation for whoever is interested in the social archeology of the industrial age. Technocracy Inc will be our last and third tableau in an effort to contextualize planned obsolescence.
Technocracy grew out of the post-war social, political, and economic turmoil. It presented itself as an apolitical movement to promote a certain idea of modernism and progress. “Growing public recognition of the importance of the scientific method in the operation of a well ordered society has caused the Continental Committee on Technocracy to be formed,”63 is the incipit of Introduction to Technocracy, a 1933 publication to introduce the public to the Technocrats’ program. Presenting itself as a platform to share ideas, the movement invited scientific experts and intellectuals to 2020 CanLIIDocs 3123 apply their expertise to find solutions to the economic crisis. They too blamed financiers and businessmen for the Great Depression, who had used technology to further their own interests and, often, in an intentionally suboptimal manner. Their accusations echo Veblen’s notion of “sabotage,” a term that he famously coined to describe business strategies geared towards upholding profitable prices:
Financial business has not only exercised complete control over this field and dictated what should be produced regardless of the resources available, but has also failed in the distribution of the ever-increasing volume of goods and services released by the accelerating rates of energy conversion.64
Technocrats believed in a science-based approach to governance—an idea in vogue at the time.65 Technocracy started under the leadership of Howard Scott as a movement
61 Allen Raymond, What Is Technocracy? (New York: Whittlesey House, 1933) at 6–7, online: Internet Archive
The decade just completed has witnessed a very important change… brought about almost entirely by the new problems which have sprung into existence with the rapid development of 2020] PROGRAMMED OBSOLESCENCE 79
of engineers committed to “study the physical operation of society”66 and to answer the call for “science and technology to extend the frontiers of their domain.”67 The technocratic ideology is first fully articulated in 1933 in a booklet entitled What is Technocracy? which reads as a manifesto. It proposes to translate social movements and industrial operations into mass, energy, force and motion.68 For Technocrats, the physical status of any social system can be measured quantitatively.69 Adopting a materialistic theory of history, they posit that socio-economic changes are governed by energy, that is, the capacity to do work. In times when the energy available is constant, no social change occurs. However, a few periods were marked by energy increases, which led to social changes referred to as “conversion changes in the rates of energy.”70
Although the cross-influence is uncertain, the economic plans envisioned by London and the Technocrats share many commonalities. London is sympathetic to the idea of empowering men of science. Planned obsolescence, as he conceived it, would be implemented by a committee of engineers and accountants responsible for setting the duration of the leases for consumable goods as well as complementary tax credits,
thus ensuring that old goods are replaced and new goods purchased. London believed, 2020 CanLIIDocs 3123 as Technocrats did, that the application of physical sciences to the problems of society could fix the economy; scientific determinism prevails.71 Technical expertise is the
our public utilities. The engineer is now called upon to aid in the solution of the broad question of regulating our public service corporations, and there he finds himself in contact not only with the law but that more interesting field of political science or economics, and almost of necessity he must absorb a good deal of both. 66 Frank Arkright, The A B C of Technocracy (New York: Harper, 1933) at 1. 67 Raymond, supra note 61 at 32ff. See also Scott, Introduction to Technocracy, supra note 63 at 38. The organization was based in New York City and was initially registered at Columbia University. Nothing justifies this affiliation but for the obvious benefit of bringing some legitimacy to the young organization. 68 Ibid at 7. Although Technocracy was formed in 1920, one can trace its origins to a group of reform minded mechanical engineers called the “New Machine,” some of whom have founded the “Technical Alliance” in 1919, of which Veblen was a member during his time at the New School for Social Research. The Technical Alliance had four primary objectives, which reflect their association to Veblen and later became prevalent themes of Technocracy: (1) uncover waste and leakage in the present industrial system; (2) render estimates of the raw material and human effort necessary to insure the various members of society a given standard of comfort; (3) show graphically how the present system of production and distribution operates; and (4) work out a tentative design of production and distribution completely coordinated. See Elsner, The Technocrats, supra note 41 at 23. 69 Scott, Introduction to Technocracy, supra note 63 at 19. 70 See ibid at 15–16:
From the view point of the technologist, man has experienced but few sweeping social changes, that is, few conversion changes in the rates of energy; and these are widely separated in point of time. The domestication of the crop plants and the development of them in a dim, historic past thrust man into a larger control of his environment—that is, to use a technological term, into a new energy state. In the same way, the domestication of animals gave him new powers to command and carried him a little further along the way of control. The introduction of these factors, each in its turn, wrought revolutionary changes in social scheme under which he lived. 71 See Howard Scott, The Evolution of Statesmanship: Science and Society (New York: Technocracy Inc, 1939) [Scott, Evolution of Statesmanship]. Many themes are recurring and redundant in the literature produced by the Technocrats along the years. Its quality is uneven and often mediocre. 80 UNBLJ RD UN-B [VOL/TOME 71 adequate response to the Great Depression, especially in the context where the crisis is attributed to mismanagement and a poor use of technology leading to goods being produced in excess. Whether London’s idea of establishing a committee of experts is inspired by technocratic thoughts remains a supposition.72
Nevertheless, some cross-pollination may have occurred and it can be reasonably assumed that London had been acquainted with Technocratic thoughts.73 After all, London was active in the intellectual and business circles of New York City in the 1930s when the Technocracy movement's popularity was at its peak. His pamphlet Ending the Depression through Planned Obsolescence was registered with the Library of Congress on September 21, 1932, one month after Howard Scott first came to widespread national publicity following an interview published in the New York Times on August, 21, 1932.74 The timeline suggests that London was working on his pamphlet at the same time and place where the ideas of Technocrats were gaining ground.75 From them, London appears to have derived the notion that the problem was a fundamental mismatch between production and consumption, or, in London’s words, “disturbed human relationships.”76 Factories were able to mass produce, but they did not have a corresponding demand for their goods. Mass 2020 CanLIIDocs 3123 production required mass consumption. Therefore, there was a need to synchronize the tempo of these two facets of the economy. For London and the Technocrats, balancing this equation is a task for engineers who, unlike the owners of the production means, are best placed to examine both variables and design a calculated, well-
72 Noteworthy, the Technocrats published virtually nothing of their work until 1933. The only publications available on Technocracy were published by and under the charismatic figurehead of Howard Scott. He or his followers coauthored and edited the Industrial Workers of the World monthly magazine. There is no evidence that London met Scott or any of his disciples, who were, however, numerous at the time of the publication of Ending the Depression, supra note 9 in 1932. Much of the information about Technocracy between 1921–1932 spread through meetings and presentations organized by devoted individuals in different chapters, making it particularly difficult for historians to trace. See William E Akin & William Ernest Akin, Technocracy and the American Dream: The Technocrat Movement, 1900–1941 (Berkeley: University of California Press, 1977). 73 This point is also made by others, see e.g. Slade, supra note 10 at 72–73. 74 “Declares Machines Add to Unemployment”, New York Times (21 August 1932) at f9. Elsner, supra note 41 reports on the event: “[t]echnocracy swept across the country in almost every available form. The high point was reached in January, 1933. The New York Times alone had no less than sixty articles on Technocracy that month” at 7). 75 The attention that the movement had gained in the 1930s make these inferences very plausible. On January 13, 1933, the Technocrats organized an important event at the Hotel St-Pierre in New York City. The event was much anticipated and Scott, the emblematic leader of the Technocrats delivered a speech on how engineers and technocracy could save America and assure for all the promised state of abundance and self- sufficiency. The event was broadcasted nationally. Of all accounts, it was a disaster and somewhat the death knell for the movement which at this very moment lost most of its capital. See Allen Gordon, “Scott, the Technocrats, Is Sold Out!”, Macfadden Weekly (24 November 1934) describing the event at 4:
The beginning of the act that night was tense; there was an expectant hush as the leading figure in the greatest economic drama of modern times took the stage. He began to speak haltingly; he groped for words; he sneered at times; he appeared absolutely inarticulate… Scott spoke of ergs and energy certificates and capitalistic economics—all that came over to the heaters was a jumble of unfinished and half-baked sentences. 76 London, Ending the Depression, supra note 9 at 4. 2020] PROGRAMMED OBSOLESCENCE 81
balanced solution. Technocrats argued however that, despite having played a role in bringing into existence the vast amount of America’s wealth by designing the means of production, engineers had ceded their role in the distribution of wealth to business persons and politicians.77 Technocrats believed that engineers must regain political power and maintain some control over the technology they design to balance the equation of production and consumption. If not, technology would be left in the hands of businesspeople and owners of the means of production.
Although the Technocrats adopted much of Veblen’s lexicon, they minimized his influence, probably in an attempt to preserve the intellectual originality of the movement’s leader, Howard Scott.78 The Soviet of Engineers became, under the Technocrats, the Technate79 and the “price system” was to be replaced by a new system based on energy conversion.80 The technocratic movement rapidly lost traction. The favour it initially gained quickly vanished. It was criticized for lacking direction and offering few solutions to the problems that they denounced.81
The affinity found in the works of Veblen, in the Technocrats’ platform and
in London’s writings is certainly not coincidental. Cross-influence seems highly 2020 CanLIIDocs 3123 plausible. Even without direct contributions, their works provide the necessary background to understand planned obsolescence as it was initially understood. Taken in its intellectual context, London’s plan stems from a belief that an unplanned economy is suboptimal and wasteful, creating surplus supplies, and leaving workers unemployed. Unlike his contemporaries, however, London considers that “[t]he sound remedy lies in rehabilitating the consumer, rather than in curtailing the producer.”82
77 Scott, Introduction to Technocracy, supra note 63 at 11; Scott, Evolution of Statesmanship, supra note 71 at 16. 78 In a note on the work of Thorstein Veblen, Howard Scott intended to respond to some questions regarding the credit owed to the famous author: “[t]here has been much discussion concerning the origin of the body of ideas for which the term Technocracy now stands. Speculation concerning this point has focused attention upon the work of Thorstein Veblen as the source of inspiration, with particular reference to the “Engineers and the Price System” as the animating force. Such conclusions are quite contrary to the facts.” See Scott, Introduction to Technocracy, supra note 63 at 59. According to Scott, the two men were simply caught in the same drift of “modern common sense” (ibid at 60). See Howard P Segal, Technological Utopianism in American Culture (Chicago: University of Chicago Press, 1985) at 122. 79 Klint Finley, “Techies Have Been Trying to Replace Politicians for Decades,” Wired (6 May 2015), online:
While London envisioned a state-led planned obsolescence, the current debate concerns corporate-controlled and engineered obsolescence. Manufacturers are accused of implementing techniques to deliberately reduce the life of a product to increase the replacement rate. While such strategies follow the dictate of stimulating consumption to meet production levels and sustain the economy, many of the problematic dynamics identified by London, Veblen and the Technocrats remain: the lifecycle, use, and replacement of consumable goods are pieces of the economy left in the hands of the owners of the production means, driven by profits and who, arguably, do so to the detriment of collective interests whether they are workers’ rights, consumers’ rights or the protection of the environment. While London did not openly attack profit-seeking business strategies, his view of a new economy intrinsically draws on corporate social responsibility:
Such a socially responsible system, which is anxious for the wellbeing of all of its citizens, is on a vastly sounder and more permanent basis than one which allows business merely to take out profits without improving the organization with new methods and without renewing the equipment. I maintain that with wealth should go responsibility. Too many nowadays regard wealth as license to freedom and immunity from obligation to the 2020 CanLIIDocs 3123 people. Such irresponsible possessors of wealth are shirkers, who tend to make all of us poorer.83
It is difficult to speculate on what London would have thought of obsolescence as a commercial strategy to maximize profit. Today, planned obsolescence remains a historical curiosity when its quasi homonym, programmed obsolescence, is on trial. The next section draws our attention to the issue of manufacturers and designers’ liability. Stories about manipulation of consumers and abusive control over the product are told, but the culprits are yet to be found.
II—Stories of Programmed Obsolescence
Stories of programmed obsolescence are sites to observe its manifestations. They are an invitation to reflect on the need and limits of legal interventions. This section presents two cases that have grasped the attention of the law. The first one is often presented, wrongly as it happens, as the poster child of programmed obsolescence. It concerns the Phoebus Cartel, an alliance of manufacturers that controlled the output and distribution of light bulbs and, incidentally, tampered with the durability of light bulbs to reduce their lifespan (A). The second one is not known as an instance of obsolescence. Its inclusion is perhaps less obvious. However, it imposed itself as our research progressed. It concerns the Y2K episode, the programming code limitations that made the year 2000 a daunting event for consumers (B). Instances of programmed obsolescence tell stories about science, progress, and the disruptive effects of technological changes. Strategies aimed at forcing the replacement of goods would be,
83 London, Ending the Depression, supra note 9 at 19. See London, New Prosperity, supra note 12 (noting “I outline the fundamentals of a new economic policy which would translate into terms of improved welfare for the average citizen. In the new prosperity, good business will become identical with good sociology” at 6). 2020] PROGRAMMED OBSOLESCENCE 83 in Veblenian terms, perfect examples of “sabotage,” a term he broadly defines as any measures of restraint or incitement deployed to consolidate control over production and maintain profitable prices.84 But as these two cases demonstrate, in Veblen’s words, “sabotage commonly works within the law, although it may often be within the letter rather than the spirit of the law”.85
A—The Light Bulb Affair
1—General Electric and the Phoebus Cartel
In the 1920s, an international alliance of powerful light bulb manufacturers was formed. Aimed at controlling the light bulb industry, it became an organized standardization model that imposed technical limitations to the life of the incandescent lamps produced by its members in America and in Europe. This large-scale strategy became with time the very fact and face of programmed obsolescence in documentaries, organization reports, and parliamentary debates.86 Above all, the light 2020 CanLIIDocs 3123 bulb affair was a series of anti-trust cases in the United States directed at General Electric (GE) questioning the dominant position it held for decades as a result of its pioneer patents. These actions were a governmental response to giant corporations and strategies used by them to control the market87. All the cases relate in some way to the use- or misuse - of patents held by GE and contractual techniques developed by it in order to control the light bulb industry and prices. Here, in the background of an anti- trust case, programmed obsolescence met with intellectual property.
The light bulb affair is unique in several respects. The anti-trust cases had an unprecedented and extensive evidentiary record that included reports and letters
84 Veblen, The Engineers and the Price System, supra note 37 (more specifically, Veblen addresses the withdrawal of efficiency to avoid overproduction and the risk of price deflation at 16). For Veblen, sabotage techniques are, in general, within the law and correspond to a form of legal astuteness. See ibid:
All that can be said here is that many of these wise measures of restraint and incitement are in the nature of sabotage, and that in effect they habitually, though not invariably, inure to the benefit of certain vested interests—ordinarily vested interests which bulk large in the ownership and control of the nation's resources. That these measures are quite legitimate and presumably salutary, therefore, goes without saying. In effect they are measures for hindering traffic and industry at one point or another, which may often be a wise business precaution. 85 Ibid at 6. Veblen reckons that many of the tactics that fit his definition of sabotage “are deliberately sanctioned by statute and common law and by the public conscience” (ibid at 7). 86 See Markus Krajewski, “The Great Lightbulb Conspiracy” (2014) 51:10 IEEE Spectrum 56 [Krajewski, “Lightbulb Conspiracy”]; Cosima Dannoritzer et al, The Lightbulb Conspiracy: The Untold Story of Planned Obsolescence, 2010, DVD (San Francisco: Arte France, 2010); George W Stocking & Myron W Watkins, Cartels in Action: Case Studies in International Business Diplomacy (New York: The Twentieth Century Fund, 1946); Arthur A Bright Jr, The Electric-Lamp Industry: Technological Change and Economic Development from 1800 to 1947 (New York: MacMillan Company, 1949). 87 United States v General Electric Co, 272 US 476 (1926) [GE 1926]; United States v General Electric Co, Civil Action 1364, 82 F Supp 753 (DNJ 1949) [GE 1949], supplemented by 115 F Supp 835 (DNJ 1953) [GE 1953]; United States v General Electric Company, 303 F Supp 1121 (SDNY 1969). 84 UNBLJ RD UN-B [VOL/TOME 71 between high officials and engineers containing clear instructions as to product design and profit estimates. The nature of the litigation, an anti-trust investigation, is also essential to understand these cases. The investigation enabled the discovery of such evidence and, more importantly, reflects the willingness of state authorities to inquire into business strategies and assume a role in the planning of economic operations. Two decisions are of particular interest for this paper. The first case, United States v General Electric Co,88 was commenced by a complaint lodged by the American government for price fixing. The judgment, rendered in 1926, rejected the accusations and has been widely interpreted as confirming the right of a patentholder to extract competitive advantages from the exercise of its intellectual property rights to the full extent allowed by law. According to the Court, “comprehensiveness of his control of the business of selling is not necessarily an evidence of illegality in method.”89 The case points to the role of patents in the modernization of distribution techniques and emphasizes, moreover, the functional limits and properties of a good as determined by the intellectual property owner. The owner is the “supreme agenda setter for the resource.” 90 The 1926 case, however, did not raise the issue of programmed obsolescence.
2020 CanLIIDocs 3123 The second case, United States v General Electric Co,91 was decided in 1949. It contains important background information and evidence showing how an entity, in a dominant position, can unilaterally assign properties it deems profitable to a line of products, including its durability. This 1949 anti-trust case was about collusion, not price fixing.92 It shows how effective the arrangements and international licenses put in place by GE were. They prevented foreign competition and secured GE domination over the lamp industry in the United States. Evidence filed by the litigants show that GE’s average percentage of the total of all lamps sold during the period when the arrangements were in force was 55%, which amounted to an average of 82.2% when combined with Westinghouse, GE’s cross-licensee, and other licensees.93 GE and its licensees thus formed, in the words of Judge Forman, “an almost impregnable front to those manufacturers not part of their combination.”94 The Court found GE in violation
88 GE 1926. 89 Ibid at 485. 90 Larissa Katz, “Exclusion and Exclusivity in Property Law” (2008) 58 UTLJ 275 at 278. 91 GE 1949, supra note 87 supplemented by GE 1953, supra note 87. 92 The Great Depression of the 1930s, however, saw drastic changes in public perception regarding such patent-based cartels due, in part, to the failure of the cartel-friendly National Recovery Administration to restore prosperity. Electric lamp cartels were not spared by a wave of anti-trust investigations. Already under surveillance, the US government took action in 1941 and filed a complaint against GE under allegations that the company had used the Convention of 1924 and domestic licenses to further its lamp monopoly and restrain trade and competition in the United State. See FM Scherer, “The Role of Patents in Two US Monopolization Cases” (2005) 12:3 Intl J Economics Business 297 at 299. 93 GE 1949, supra note 87 at 893 (Exhibits 21-G, GE-237 and GE-238). 94 Ibid at 893. 2020] PROGRAMMED OBSOLESCENCE 85 of section 2 of the Sherman Anti-Trust Act and ruled that its licensees conspired to further the monopoly.95 It concluded,
[b]y virtue of General Electric’s dominating position in the industry and relative lack of competition it had the power to set the standard of efficiency of incandescent electric lamps for the entire industry and in so doing to determine what should be their length of life, and this constitutes an attribute of monopoly.96
The story begins in the early 20th century. The electric lamp industry was booming due to the rapid spread of electrification and the introduction of new forms of lighting.97 Weaving a web of cross-licensing agreements, leading lamp manufacturers in industrial countries stabilized and monopolized domestic markets by eliminating competition.98 In the United States, GE consolidated control over the industry by first enjoying the benefits of pioneer patents it had acquired, then by investing heavily in machinery and expanding its portfolio of related innovations and patents. It implemented licensing and distribution schemes to share commercial routes and markets, forging alliances between light bulb makers.99 After World War I, leading lamp manufacturers around the world increasingly entered international licensing 2020 CanLIIDocs 3123 agreements which led to the formation of the Phoebus Cartel.100 In 1924, representatives from the leading light bulb manufacturers in Europe and America met in Zurich, Switzerland, to sign the Convention for the Development and Progress of the International Incandescent Electric-Lamp Industry.101 The stated objectives were to “ensure and maintain a uniformly high quality [of lamps], increase the effectiveness of electric lights, and … increase light use to the advantage of the consumer.”102 The Cartel’s leading members included Osram (Germany), Philips (Netherlands), Tunsgram (Hungary), Associated Electrical Industries (United Kingdom) and Compagnie des lampes (France).103 Together, they incorporated and held shares in a Swiss-based company named Phoebus Inc Industrial Company for the Development of Lighting. The American company GE never joined the international cartel as an
95 Ibid at 902. 96 Ibid at 899. 97 Krajewski, “Lightbulb Conspiracy”, supra note 86 at 58. 98 Stocking & Watkins, supra note 86 at 313–23. 99 Ibid at 306-309. 100 Ibid at 304 and 321. 101 Ibid at 331–32; Krajewski, “Lightbulb Conspiracy”, supra note 86 at 57. A copy of the agreement is reproduced in part at Appendix 8 of UK, The Monopolies and Restrictive Practices Commission, Report on the Supply of Electric Lamps (London: Her Majesty’s Stationery Office, 1953) at 126–51 [UK Report on the Supply of Electric Lamps]. 102 Ibid at 353 (citing the Convention for the Development and Progress of the International Incandescent Electric Lamp). 103 Ibid at 332; Markus Krajewski, “Vom Krieg des Lichtes zur Geschichte von Glühlampenkartellen” in Peter Berz, Helmut Höge & Markus Krajewski, eds, Das Glühbirnenbuch (Vienna: Edition Selene, 2001) at 184 [Krajewski, “Krieg des Lichtes”]. 86 UNBLJ RD UN-B [VOL/TOME 71 official member, but nevertheless played a central role in its formation and operations.104 While the Convention was expected to end in 1955, it became void with the outbreak of World War II.105
Programmed obsolescence was not the immediate cause of action in the 1949 litigation case. Collusion was. However, the court’s examination of anti-competitive practices brought to light the conditions under which programmed obsolescence can be implemented. The case also provides unprecedented evidence to the effect that engineers had been instructed to reduce the durability of certain products. The injunction order, granted in the 1953 supplementary decision from the same court, enjoins GE from combining and conspiring, among other things, to “reduce, restrict, or limit, in any manner, the kinds, quantities, sizes, styles, or qualities of lamps, lamp parts, or lamp machinery which may be manufactured by any person.”106
The cases thus exemplify how programmed obsolescence thrives in a monopolistic environment in which consumers, who cannot readily access more competitive alternatives, are forced to accept the durability standards set by those who control the market. Indeed, the 1924 Convention provided for the implementation of a 2020 CanLIIDocs 3123 standardization program under which the parties agreed to a “formulae for arriving at the economic life of lamps.”107 Accordingly, all members had to send samples to a testing laboratory in Switzerland where the life of the light bulbs was meticulously recorded.108 Members that deviated from the optimal range determined by the cartel were fined.109 In the United States, the Electrical Testing Laboratories (ETL) was key in implementing a parallel standardization program. Although GE claimed the ETL was independent, the court found that ETL was not only substantially financed by the defendant company, it also had enforced its own standards and obtained information, including of a confidential nature, on its competitors.110
Absent effective competition, GE was able to dictate the norms and conditions governing the sale and use of its products, even if it did so at the consumers’ expense. It controlled the science, the market, and the message through aggressive marketing. As a result, the American consumers were left with little choice and
104 Krajewski, “Lightbulb Conspiracy”, supra note 86 at 59; Stocking & Watkins, supra note 86 at 321–22. By December 31, 1940, General Electric had the following percentages of ownership in its licensees: 21.45% in Osram; 11.85% in Philips; 37.03% of ordinary shares and 33.47% of founding shares in Compagnie des lampes; and 40.66% of ordinary shares and 20.72% of preferred shares in Associated Electrical Industries. 105 Krajewski, “Lightbulb Conspiracy”, supra note 86 at 61. 106 GE 1953, supra note 87 at 860. 107 GE 1949, supra note 87 at 835 (documentary evidence, reply from Mr. Woodward, a European representative of International GE, a fully owned subsidiary of GE); Stocking & Watkins, supra note 86 at 354. 108 Stocking & Watkins, supra note 86 at 353. 109 Ibid at 353–54. 110 GE 1949, supra note 87 at 854. 2020] PROGRAMMED OBSOLESCENCE 87 purchasing power to offset corporate strategies aimed at maximizing profit and stock turnover. However, the court’s assessment of anti-trust violations in the 1949 case opened a space to discuss the “deterioration of products” as an attribute of a monopoly. In light of GE’s experiments to limit the lifespan of light bulbs, the Court concluded that “the total proof leads to the conclusion that by virtue of General Electric's dominating position in the industry and relative lack of competition it had the power to set the standard of efficiency of incandescent electric lamps for the entire industry and, in doing so, to determine what should be their length of life, and this constitutes an attribute of monopoly.”111 These conclusions are compelling even if evidence documenting engineering efforts to set standards of efficiency, or shorten the lifespan of a product, were only accessory to the anti-trust action. It undoubtedly paved the way to the current narrative on obsolescence.
2—Evidence of Programmed Obsolescence
If programmed obsolescence was only peripheral to the conspiracy accusations, it was nevertheless a central piece to the US government’s argument that the monopoly 2020 CanLIIDocs 3123 created a situation whereby American consumers were clearly on the losing side, hence adding gravitas to the complaint. A series of letters, reports and memoranda between engineers of the Lamp Department and officials holding key positions in GE left a detailed record of the engineering efforts spent on altering the quality and lifespan of light bulbs, the intended objectives, and the actual achievements.112 As for the issue of the deterioration of products, the Court took a piecemeal approach to the technical and economic considerations involved in design and manufacturing decisions. The reasoning is punctuated by short references to the expertise submitted and a dose of value judgement. While the final determination pertained to GE’s power to set the standards of the light bulb industry, the Court noted that
[d]espite assertions of good faith, sound business discretion, great technical research, and consideration for the interest of the consuming public, there was manifested in the two situations [flashlight bulbs and Christmas tree lamps] a paramount concern in what would afford the maximum return in profits to the manufacturer and that General Electric had the power in the instance of the flashlight lamp, to shorten the life of the lamp, and in the instance of the Christmas tree lamp, at least, to promote a product with a short life for a specific use over one that had four times its life.113
While it acknowledged that GE could lawfully seek to maximize its profits, the Court put a negative ring to the GE’s efforts to shorten the lifespan of bulbs and to do so with
111 Ibid at 899. 112 Ibid. In an attempt to minimize the effect of these documents, GE claimed, unsuccessfully, that the correspondence amounted to “nothing more than the personal views of the particular individuals concerned” (ibid at 844). 113 Ibid at 899. 88 UNBLJ RD UN-B [VOL/TOME 71 the primary aim of maximizing its profits to the detriment of the consumers’ interests.114
The US government brought deterioration claims for three products: flashlight bulbs, miniature lamps, and tantalum lamps. While the claim for tantalum lamps was barely discussed and dismissed, the strategies implemented in the case of flashlight bulbs and miniature lamps came under scrutiny. In the case of flashlight bulbs, the evidence revealed that GE had successfully reduced their longevity from a life coextensive with three batteries in the 1920s to one of two batteries by the 1930s. In 1932, the engineering department studied the possibility of further reducing the life of flashlight light bulbs to only outlast one battery. Their report was the basis of an inter-departmental letter to the executives of GE, stating
[t]wo or three years ago we proposed a reduction in the life of flashlight lamps from the old basis on which one lamp was supposed to outlast three batteries, to a point where the life of the lamp and the life of the battery under service conditions would be approximately equal. Sometime ago, the battery manufacturers went part way with us on this and accepted lamps of two battery lives instead of three. We have been continuing our studies and 2020 CanLIIDocs 3123 efforts to bring about the use of one battery life lamps… If this were done, we estimate that it would result in increasing our flashlight business approximately 60 per cent.115
According to the US government, GE “was primarily interested not in giving the consumer more efficient light but basically in increasing its sales of flashlight lamps.”116 GE’s first line of defence was efficiency. Efficiency, which it defined as “the amount of light that the lamp will give for the amount of electricity it consumes,”117 was a “more important consideration” in measuring the quality of a lamp than its lifespan.118 Relying on the efficiency gains of shorter-lived bulbs, GE argued
[t]he main purpose of the change in the life of the lamp was to lower the total costs of light to a flashlight user by making the design life of lamps then in use shorter so as to cause them to give more light, and that, incidentally, it was estimated that its sales of flashlight lamps would be increased about 60%.119
114 Ibid at 899. 115 Ibid at 897. 116 Ibid at 897. 117 Measured in lumens per watt. 118 GE 1949, supra note 87 at 897 (R 2162, testimony of Mr. Harrison). 119 Ibid at 897. A closer look at how an incandescent electric light operates helps understand the efficiency claims raised by GE. When an electric current flows through the filament of the bulb, the filament offers resistance. The electrical energy absorbed by it is then dissipated by the emission of heat and light. The efficiency of a filament depends on its diameter and length. A filament shorter in diameter and length can operate at a higher temperature and will produce more light output (units of luminous flux or lumens) per unit of current used (watt). Running at higher temperature, however, causes the metal of which the filament is made of, tungsten in this case, to evaporate faster. The result is a more efficient bulb with a shorter life. 2020] PROGRAMMED OBSOLESCENCE 89
Another claim concerned miniature lamps. In 1936 and 1937, GE manufactured a C- 7 night lamp and a C-7-½ Xmas Tree lamp designed for a life of 2000 hours and 500 hours respectively. In a memorandum to the managers and salesmen issued in 1937, the Incandescent Lamp Department opposed a plan by the agents in charge of the wholesaling operations (“B agents”) to order the longer lived C-7 light lamps instead of shorter lived C-7-½ Xmas Tree lamps to be used in Christmas tree sets.120 The Government contended that GE opposed the substitution because it was concerned with loss of sales of the Xmas Tree lamps and feared the opposition of the manufacturers of that string of lamps.121 GE argued that it emphasized the sale of the C-7-½ Xmas Tree lamps for Christmas tree sets for “technical reasons” that were in the consumers’ interests. Both models were of similar size and shape and the C-7 light lamp could fit into the socket of the Xmas Tree lamp. However, GE claimed that the Xmas Tree lamp had a shorter design life to provide a more brilliant light. It argued that the compromise between efficiency and longevity was determined in accordance with the purpose of the product, that is, provide “the sparkle and brilliance of particular interest to Christmas tree decorations” and a “rugged construction to meet the rougher 122 handling to which the Christmas tree lamp was exposed.” In contrast, the C-7 night 2020 CanLIIDocs 3123 lamp was designed to “throw a spot of dim light for use in hallways, nurseries” and would give “unsatisfactory service if used for Christmas tree purpose.”123 The Court did not assess these claims, but emphasized that GE had the power “at least, to promote a product with a short life for a specific use over one that had four times its life.”124
On the whole, the Court considered the strategy to reduce the lifespan of products or promote shorter-lived options as part of a larger pattern of conduct in determining anti-trust violations. The two lines of defence brought forward by GE were not fully assessed nor rebutted. On the one hand, the Court recognized the trade- off between the longevity and efficiency of a bulb but seemed to imply that the
Assuming the wattage remained the same, a reduction in the lifespan of flashlight bulbs would bring an increase in lumens per watt, which means that the lamp is brighter and produces more light for the electricity consumed. See the Subcommittee’s Preliminary Study, cited in US, Government Activities Subcommittee of the Committee on Government Operations, 29th Cong, The Short Life of the Electric Light Bulb (Washington, DC: United States Government Printing Office, 1966) [Subcommittee Report]. The second part of the argument, that a shorter-lived bulb would lower the total costs of light to a flashlight user, has a reasonable basis although it is not discussed in the court decision. The argument is that from the power of one battery, a consumer would get more light output (lumens) from a flashlight that uses a shorter-lived, more efficient bulb than from a longer-lived, less efficient bulb. Simply put, the consumer gets more light per battery even if he needs to replace the bulb more frequently. Since it was demonstrated that GE had the power, and did, shorten the lifespan of the flashlight bulbs, the court did not assess these claims and, instead, pointed to evidence that lamps were burning out before the battery was fully consumed and that there was no price difference between the two models (GE 1949, supra note 87 at 879). 120 GE 1949, supra note 87 (exhibit 1867-G, a memorandum dated 30 July 1937, from the Incandescent Lamp Department to managers and salesmen). 121 Ibid at 897. 122 Ibid at 898. 123 Ibid at 898. 124 Ibid at 899. 90 UNBLJ RD UN-B [VOL/TOME 71 compromise reached was inadequate. On the other hand, references to the consumers’ interests remain abstract without evidence of the actual power, light output and lifespan of the products, making it impossible to know, for instance, whether a shorter- lived flashlight battery produced more light for the electricity consumed or whether rugged lights were actually more resistant to shocks. It will take the intervention of state authorities, acting outside of the litigation context, to elucidate the issue of the lifespan of light bulbs.
3—State Authorities and Product Lifespan
In the years following the injunction, incandescent light bulbs continued to last between 750 to 1,000 hours according to a preliminary study prepared by the Government Activities Subcommittee of the Committee on Government Operations (hereinafter, “Subcommittee”) in 1964.125 In the context of a rulemaking proceeding in 1966, the Subcommittee decided to investigate the feasibility of extending the lives of standard bulbs.126 While the Court concluded in 1949 that it was unreasonable to infer, in light of the evidence, that GE could have delivered 60 watt gas-filled tungsten 2020 CanLIIDocs 3123 lamps with a better efficiency rating in lumens per watt, the Subcommittee was not limited to assessing violations of anti-trust laws. Therefore, it evaluated the technical considerations involved in design decisions and their effects on the consumers’ interests. Its findings severely challenged light bulb manufacturers’ design decisions.
Regarding GE’s efficiency defence, the Subcommittee recognized the “inevitable compromise” between light output and the life expectancy of an incandescent electric bulb.127 In light of the preliminary report and manufacturers’ responses, the Subcommittee concluded, however, that the compromise reached was inadequate because losses in terms of lifespan would only bring small gains in lumen output. At a same wattage, the life expectancy of a bulb is inversely proportional to about a seventh of the power of the light output.128 This means, roughly, that a 10 percent decrease in light output could double the life of a bulb.129 Compared to white- finished 100-watt bulbs manufactured by Sylvania that had life ratings of 1,100 hours and an initial light output of 1,580 lumens, the standard 100-watt bulb manufactured by GE was designed to last 750 hours and had an initial light output of 1,710 lumens.130 For 350 hours less, a consumer gained 130 lumens, an amount not visible to the naked eye.131 The Subcommittee concluded that light output should not be emphasized as the
125 Subcommittee Report, supra note 119 at 27. 126 Ibid. 127 Ibid at 28. 128 Ibid at 6. 129 Ibid at 7, 49 and 69. 130 Ibid. 131 Ibid at 4. 2020] PROGRAMMED OBSOLESCENCE 91 true measure of the bulb’s efficiency to the detriment of life expectancy.132 The gains in “brightness” that GE had put forward to justify shortening the lifespan of the flashlight bulbs in the 1949 anti-trust case were probably only minimal. Hypothetically, a reduction in the lifespan of flashlight bulbs from three to a two battery life, which represents a 1/3 decrease, would have an increase in light output of approximately only 5.9%.133
In addition to the trade-off between efficiency and longevity, efficient light bulb design needs to strike a balance between light output, life expectancy, and electric power consumption. To increase the light output and life expectancy of a bulb without compromising one or the other, the bulb needs a higher wattage, which also means, consequently, that it will use more electric current.134 The resulting increase in electric costs, according to GE, would mostly affect low-income customers.135 The underlying idea of GE’s argument is that the company calculates the optimum bulb life with the objective of providing lighting, as distinct from lamps, at lower costs.136 This line of defence highlights how two markets, light bulb manufacturing and electric production, intersect within one product. Since the bulb is a device through which a service is delivered, GE commented that “[i]t is the responsibility of the lamp design engineer 2020 CanLIIDocs 3123 to build into the product the best over-all value for the average conditions of use.”137 Based on its estimates of the costs of the bulb and electric power costs, the Subcommittee, however, found that light bulb manufacturers failed to fulfil their responsibility. The Subcommittee concluded that doubling the lives of the higher wattage bulbs would cost the consumers an additional amount “so small that it can stand as no justification for the extreme limitation the bulb manufacturers have placed upon light bulb life over the past two decades.”138
Finally, the Subcommittee did not determine whether the reduction of the lifespan of bulbs was reasonable at the time of the cartel but reasoned that the 1,000 hour shelf life was no longer justified at the time when the Subcommittee’s report was issued in 1966.139 While shorter-lived bubs have been more economical for the consumer in the 1910s, when the costs of electric current were high and lamps needed to withstand voltage fluctuations, they did not remain optimum as the electric rates
132 Ibid at 2. 133 Assuming a b exponent of 7. 134 Subcommittee Report, supra note 119 at 1, 28. 135 Ibid at 52. In its Commentary submitted to the Subcommittee, GE argued that it had not increased the shelf life of its light bulbs beyond the 1,000-hour standard, because any substantial change in design to increase life expectancy while conserving present level of light output would add $100 million to the American consumers’ electric bill (ibid at 53). 136 Ibid at 57. 137 Ibid at 59. 138 Ibid at 25 (noting that “the life of a 100-watt standard bulb could be doubled—that is, extended to 1,500 hours—for an increase in light costs to the consumer of 2.5 percent [per year]” at 15). 139 Ibid at 1. 92 UNBLJ RD UN-B [VOL/TOME 71 declined.140 Changes in the ecosystem of a product, whether it is the costs of services or the development of new technologies, make it difficult to determine when the end of life of a product is no longer reasonable and becomes premature. Moreover, it remains unclear whether the failure was programmed or whether it was a state of affairs that remained unchanged. As the Subcommittee conceded,
[t]he reasonableness of extending bulb life at a minimal and even unmeasurable annual cost is a matter of judgement… It is the considered conclusion of this subcommittee, however, that the double-life bulbs would be well worth this minimal cost. And, it is our opinion that the overwhelming majority of consumers agree.141
As highlighted in the course of the anti-trust investigation, by monopolizing the market, GE was in a position to limit the useful life of their products and control the information disclosed to the consumers, thus limiting of their freedom of choice.142
Beyond the inconvenience of frequent replacement and consumer dissatisfaction, the Subcommittee highlighted a lack of information on the product’s characteristics. It found that some statements made by the manufacturers were 2020 CanLIIDocs 3123 misleading since they inferred that efficiency of the rate at which a bulb generates light, as in lumens per watt, was the sole factor of the bulb’s economy to the consumer.143 In its statements before the Federal Trade Commission, which was presented to the Subcommittee, GE responded that they had, in the past, experienced “great difficulty in explaining [the] technical story” of the balance between life and light and that “more complex statements than these would only serve to confuse [the consumers] rather than clarify.”144 It was, GE argued, in the consumers’ best interests to disclose on the packaging of light bulbs the wattage only, and not the lumens and hours of useful life as the Subcommittee recommended in its preliminary study. While the report reflects extensive investigation, cost calculations, and comparisons of light bulbs from different manufacturers, the Subcommittee did not purport to determine the optimal lifespan and focused instead on demonstrating that light bulbs could be
140 Bright, supra note 86 at 332–33, 361, cited in Subcommittee Report, supra note 119 at 25. Electric rates have declined by more than 60 percent between 1910 and 1946 due, in part, to improvement in generating and transmission equipment with the collaboration of General Electric, Westinghouse, and other large producers of heavy electrical apparatus (Bright, supra note 86 at 361). 141 Subcommittee Report, supra note 119 at 17. 142 See ibid at 25, citing Bright, supra note 86 at 333, explaining that the problem lies in the fact that
it ha[d] been impossible for consumers to prevail on General Electric to add long-lived lamps to its standard line so that the purchaser may select for himself the type he wants. This insistence by the industry leader that it knew best what its customers needed has in individual instances proved costly to them, for what General Electric made standards was virtually forced on 90 percent of the domestic market. Consumers have not had the freedom of choice with respect to burning hours to which they would seem to be entitled). 143 Subcommittee Report, supra note 119 at 2–5. 144 Ibid at 65 (Statement Before Federal Trade Commission by Donald D Scarft, General Manager, Lamp Division, General Electric Co). 2020] PROGRAMMED OBSOLESCENCE 93 longer-lasting. Ultimately, they relied on consumer knowledge. In 1970, their recommendation became a trade regulation rule providing that
it constitutes an unfair method of competition and an unfair and deceptive act or practice to: (a) Fail to disclose clearly and conspicuously the following information for such lamps on the sleeves or paper containers in which they are packaged: (1) The electrical power consumed-expressed in average initial wattage; (2) The light output expressed in average initial lumens; (3) The average laboratory life expressed in hours.145
Information would ensure that consumers had the freedom to make well-informed purchases in light of their needs, means and competing products. As we will see, current initiatives to prevent and sanction programmed obsolescence tend to rely on obligations of information as the main weapon.
Yet, even prior to legislative action, such control over product information and life span had increasingly proven difficult to maintain as GE’s monopoly was weakening. GE faced increasing competition during the Great Depression from lamp producers, within and outside the Cartel, that started the manufacturing of longer 2020 CanLIIDocs 3123 lasting lamps.146 An official of GE suggested reconsidering their strategy, explaining, “[competitors’] lamps are at somewhat lower efficiency than ours and inherently have a longer life. It is difficult to convince the typical consumer that efficiency of the lamp is the important thing. He is prone to judge quality by life alone.”147 Competition, it seems, may be a cure to programmed obsolescence after all.
The Phoebus Cartel affair stands today as an example of the type of sabotage techniques leaders can deploy. In many respects, however, it is an exceptional and singular case from which it is difficult to extrapolate. First, it did not address the legality of programmed obsolescence. It must be recalled that the creation and implementation of the GE-Phoebus cartel’s standardization program was only put forward in support of the anti-trust case. Secondly, the scope of the evidence was such that it caught in its net the elements relevant to our story. It serves our purpose only inasmuch as it is a reconstructed narrative. It gives us a unique opportunity to explore the constitutive elements of programmed obsolescence, namely: the practice, which is the deliberate shortening of a product lifespan; the goal, which is the increase of repetitive consumption of the product of a same entity or group to generate profits; and a lack of transparency in the information disclosed to the consumers.148 Together, these three elements emerged out of the corporate records to haunt GE in the courtroom where competition laws provided space to examine a strategy to reduce the
145 35 Fed Reg 11767 (1970) (to be codified at CFR § 409.1) at 11784–91. 146 Stocking & Watkins, supra note 86 at 355. 147 Stocking & Watkins, supra note 86 at 356 (Ex 1862-G, letter from RG Morison and AL Powell to HB Myrtle, 19 May 1933). 148 Slade, supra note 10 at 80–81. 94 UNBLJ RD UN-B [VOL/TOME 71 durability of its products.149 Absent such prevailing and intrusive proceedings, which singularize anti-trust prosecutions in the United States, evidence of the same nature and scope would be particularly difficult to find. In fact, a British Report of 1951 examining the supply of electric lamps and inquiring into the same matters, concluded that
[a]s regards life standards, before the Phoebus Agreement and to this day the general service filament lamp was and is designed to have, on average, a minimum life of 1,000 hours. It has often been alleged—though not in evidence to us—that the Phoebus organisation artificially made the life of a lamp short with the object of increasing the number of lamps sold. As we have explained in Chapter 9, there can be no absolute right life for many varying circumstances to be found among consumers in any given country, so that standard life must always represent a compromise between conflicting factors. 150
Nowadays, the GE case continues to be discussed and presented as a programmed obsolescence case which, as we have shown, is not entirely accurate. Nevertheless, the light bulb affair remains a key chapter in the story of programmed obsolescence as it 2020 CanLIIDocs 3123 exposes, on one hand, the level of sophistication of manufacturers when it comes to product design and, on the other hand, the information asymmetry consubstantial to the technological dimension of such design. This implies that unless programmed obsolescence is admitted or proven in the course of a litigation case in a country where extensive discovery is possible or when it is simply presented as a legitimate business method—that is, the manufacturing of disposable products offered at lower prices— only reverse engineering can have the object speak: res ipsa loquitur. The GE case’s main contribution was to bring within the realm of provable possibility the deliberate attempts by manufacturers to reduce product durability, attempts which otherwise would remain in the world of conspiracy theories. In doing so, it called for a more inquisitive approach to the physical limitations of mass market products. Even so, while extensive evidence was uncovered revealing a clear intent to shorten the life of light bulbs, technical complexities and technological developments left evidence of programmed obsolescence under a dim light.
149 Similarly, the case of Dupont’s nylons stockings is often cited as an example of planned obsolescence, but was, like the GE case, an anti-trust investigation. The company allegedly reduced the durability of their nylon stockings by amending the original formula. In United States v Imperial Chemical Industries, 100 F Supp 504 (SDNY 1951), Imperial Chemical Industries Ltd, E I du Pont de Nemours and Company Inc and Remington Arms Company, Inc were prosecuted under s 4 of the Sherman Act. Like General Electric, Dupont and affiliated textile manufacturers monopolized the nylon industry using contract and intellectual property devices to fix durability standards. 150 UK Report on the Supply of Electric Lamps, supra note 101 at 98. 2020] PROGRAMMED OBSOLESCENCE 95
B—Y2K Period
Moving ahead, the Year 2000 problem, also known as Y2K or the Millennium bug, tells perhaps another story of programmed obsolescence.151 In the realm of intangibles, coded objects and the Internet of things, programmed obsolescence may well be ubiquitous. The integration of chips and electronic components ties the use and utility of physical objects to the performance and functions enabled by the built-in software intelligence. The devil is in the code as much as it is in the matter. The control over product durability is thus expanded: the use of less durable material, as in the case of light bulbs, remains an option, but the embedded software constitutes another effective tool to make the object subservient and the consumer obedient.152 In this environment, compatibility is another facet of obsolescence. This phenomenon is sometimes referred to as “technological obsolescence,” an expression used to describe the effects of rapid technological developments on the functionality and performance of a product.153 Ultimately, both the manufacturers and the consumers may bear some responsibility for the replacement of a software-enabled product, which quickly becomes obsolete in a market flooded by new product generations with enhanced capability. The product also becomes obsolete in the eyes of its owner who decides to 2020 CanLIIDocs 3123 replace a device that is still functional.154 In this case, the manufacturer can hardly be blamed for what is known as “aesthetic obsolescence”. We are far here from the situation depicted in our proposed definition of programmed obsolescence. To be sure, as the economy shifted toward an information society and products became the proxies of a service economy, obsolescence entered a new phase in which it “began to take on increasing abstract meanings.”155 The obsolescence of electronic or smart objects becomes the banner of a crusade of a new genre, one that challenges the control of intellectual property owners over the physical world. This issue can be summarized as follows: since software are “works” within the meaning of copyright laws,156 any
151 American Home Assurance Company v Canadian Pacific Railway Company, 2004 ABQB 758 [American Home Assurance Company]. 152 Slade, supra note 10 at 187–88. 153 Ibid. See also Joseph Guiltinan, “Creative Destruction and Destructive Creations: Environmental Ethics and Planned Obsolescence” (2009) 89 J Business Ethics 19 at 19–20; Tim Cooper, “Inadequate Life? Evidence of Consumer Attitudes to Product Obsolescence” (2004) 27:4 J Consumer Policy 421 at 424; Brian Burns, “Re-evaluating Obsolescence and Planning for It” in Tim Cooper, ed, Long Lasting Products: Alternatives to the Throwaway Society (Surrey: Gower Publishing Limited, 2010) 39. 154 For a discussion on the shared responsibility for the obsolescence of electronic goods, see Déméné & Marchand, “L’obsolescence des produits électroniques”, supra note 18 at 9–18. See also France, Rapport du Gouvernement au Parlement sur l’obsolescence programmée, sa définition juridique et ses enjeux économiques: En application de l’article 8 de la loi du 17 mars 2014 relative à la consommation (April 2017) (noting: “[a] business will renew more frequently its generations of products in order to propose the latest technological innovations to consumers in a competitive environment which is constantly evolving. [In the sector of IT products,] where technological developments are rapid, products are becoming more and more complex and are quickly seen as obsolete and replaced while they are still in working order” at 16 [translated by author]). 155 Slade, supra note 10 at 187. 156 This in accordance to international copyright law. See Marrakesh Agreement Establishing the World Trade Organization, 15 April 1994, 1867 UNTS 154, Annex 1C (Trade-Related Aspects of Intellectual Property Rights), s 10 (Computer Programs and Compilations of Data). See also, the Canadian Copyright 96 UNBLJ RD UN-B [VOL/TOME 71 attempt to tamper with or to modify the code in-board any smart object in order to upgrade or render compatible and serviceable said object could constitute an infringement. Specific exemptions dealing with compatibility and interoperability exist but are very limited in their scope.157 In other words, property, here intellectual property, gives an additional tool to control the life and use of the object, here legally, in a way that otherwise would be achieved illegally as in the GE case. Programmed obsolescence, in the literal sense, is henceforth not only possible but legitimate.
In retrospect, the Y2K episode prefigures coded obsolescence although it has never been presented in this light. On the heels of the new millennium, society found itself under an unprecedented situation: the fear that computers would be unable to record the presence of a new millennium. Several computer programs developed in the late 20th century represented four-digit years with only the final two digits. Year 1999, for instance, was entered as “99”. This two-digit date recognition code, however, also meant that year 2000 would be indistinguishable from 1900, potentially causing various errors, such as an erroneous handling of date information to the miscalculation of date-dependent data, and the crash of computers and electronic systems using this 158 format. Yet, as the clocks rolled over into 2000, computer failures, for those that did 2020 CanLIIDocs 3123 materialize, did not plunge society into the anticipated crisis.
Nonetheless, Y2K caught the public’s imagination and the media’s attention.159 As consumers, organisations and companies sought to remedy the anticipated problem, computer programming firms, upstart computer service vendors and developers marketed remediation packages.160 Fearing that “a barrage of potential litigation” would hamper efforts to fix Y2K and would dissuade individuals and business from effectively engaging in remediation efforts,161 President Bill Clinton signed, in 1999, Bill HR 775, an Act to establish certain procedures for civil actions brought for damages relating to the failure of any device or system to process or
Act, RSC 1985, c C-4 (defining “literary work” as “literary work includes tables, computer programs, and compilations of literary works”, s 2). 157 See e.g. ibid, s 30.6 (Permitted acts) and s 30.61 (Interoperability of computer programs). Copyright laws also contain provisions dealing with what is so-called “digital locks” which aim at preventing third parties to circumvent protection measures controlling access and use of copyrighted material or software. 158 US, House of Representatives, 106th Cong, Y2K Myths and Realities: Joint Hearing before the Subcommittee on Technology of the Committee on Science and the Subcommittee on Government Management, Information, and Technology of the Committee on Government Reform (Washington, DC: United States Government Printing Office, 1999) at 2 [Y2K Myths and Realities]; American Home Assurance Company, supra note 151 at para 6. 159 Robert A Martin, “Dealing with Dates: Solutions for the Year 2000” (1997) 30:3 Computer 44 at 44; Eric Andrew-Gee, “Y2K: The Strange, True History of how Canada Prepared for an Apocalypse that Never Happened, but Changed Us All”, The Globe and Mail (28 December 2019), online:
Y2K put programmers under scrutiny as it questioned how it was possible that programmers could ignore such a disruptive and expansive problem for more than three decades. A report prepared by the Committee on the Judiciary, to whom was referred Bill HR 775, defined the two-digit technology as a “innocuous short term solution to the oppressively high cost of computer memory in the 1950's and 1960's,”164 acknowledging that “although programmers and managers knew in the 1950's and 1960's that they had built software with latent defects in it, no one thought that software written then would survive to the year 2000.”165 For its part, the Senate
Special Committee On The Year 2000 Technology Problem (Senate Special 2020 CanLIIDocs 3123 Committee) recognized the two-digit date recognition code as a “tradition” in the programming field.166 Since the defect was a formatting convention that was widely followed, the legislators seemed willing to accept that programmers were not responsible for the problems that ensued. The Senate Special Committee reported that programmers simply believed that the “two-digit method of date storage would solve themselves as companies, governments and other computer-owners updated their hardware and software.”167
Across the border, in Canada, Y2K only gave rise to a few cases. As Pelletier JA summarized in the 2018 decision Canada (AG) v Access Information Agency Inc, which involved a dispute over supply contracts to remedy the anticipated consequences of the 1999 to 2000 transition, “the famous Y2K problem … at the end of the day, turned out to be a non-event. After January 1, 2000 [it] had come and gone without incident …”168 Nevertheless, the Millennium bug led to the first civil action case in Canada to explicitly refer to the planned obsolescence of a device.
162 Y2K Act, Pub L No 106-37, 113 Stat 185 (1999). 163 Ibid (for actions brought by a government entity, s 4(g)(3); for suspension of penalties for small businesses, s 18(d)(1)). 164 US, Senate Special Committee on the Year 2000 Technology Problem, 106th Cong, Investigating the Impact of the Year 2000 Problem (S Rep No 106–10) (Washington, DC: 1999) at 7 [Special Committee, Investigating the Impact of the Year 2000 Problem]. 165 Year 2000 Readiness And Responsibility Act Report, supra note 161 at 12–13. 166 See Martin, supra note 159 (describing the practice as the “two-digit convention” at 44). 167 Special Committee, Investigating the Impact of the Year 2000 Problem, supra note 164 at 7. 168 Canada (AG) v Access Information Agency Inc, 2018 FCA 18 at para 30. 98 UNBLJ RD UN-B [VOL/TOME 71
In the 2004 decision, American Home Assurance Company v Canadian Pacific Railway Company, the Canadian Pacific Railway Company (CPR) brought an insurance claim for $51,488,000 to cover the incurred costs of the Y2K remediation efforts, which included “the wholesale modification, retirement and replacement of computer systems that were identified as likely to experience problems.”169 Confronted with the first Canadian property insurance claim arising out of the Y2K problem, Hawco J sought guidance from American jurisprudence. In Port of Seattle v Lexington Ins Co,170 the insured party, like CPR, claimed indemnity against its insurers for expenses incurred upgrading its computer systems to avoid the Y2K problem. The Court of Appeals of Washington reasoned that the two-digit date recognition code was an “inherent vice,” defined as “any existing defects, diseases, decay or the inherent nature of the commodity which will cause it to deteriorate with the lapse of time.”171 Two years later, the United States Federal Court of Appeals followed Port of Seattle and found against the plaintiff, explaining the following:
We agree. As the District Court explained, “[h]ere ... the insured property, GTE’s computer systems, do contain their own ‘seeds of destruction’—that is, the two-digit date limitation.” 258 F. Supp. 2d 2020 CanLIIDocs 3123 at 377. Furthermore, “GTE is not threatened by any external force; the threat is entirely internal.”172
Applying a similar reasoning, the Queen’s Bench of Alberta concluded that the Y2K problem was not a “faulty design or design defect,” nor “the result of an act of negligence,” but rather an “inherent vice” excluded from coverage as provided for in the insurance contract.173 The Court ruled as follows:
The two-digit date recognition code was a planned and conscious design decision by the original programmers that worked well for over 25 years. The fact that the design was perpetuated due to an unwillingness to replace the system despite its obsolescence does not mean that it was a faulty design. The two-digit date recognition code was an element of programming that was initially necessary because of storage restrictions, and then became perpetuated because of the form in which data was presented and generated. The understanding of inherent vice and design defect ought not be conflated in the instant case because they are clearly two different things. Rather, the Y2K problem is more accurately understood as a planned obsolescence without taking into consideration the consequences of the decision, and is an inherent vice rather than a design defect.174
169 American Home Assurance Company, supra note 151 at paras 8, 16. 170 Port of Seattle v Lexington Ins, 111 Wn App 901 (Wash CA 2002). 171 Ibid at 909. 172 GTE Corp v Allendale Mut Ins Co, 372 F 3d 598 (3d Fed Cir 2004) at 35. 173 American Home Assurance Company, supra note 151 at paras 40, 46. 174 Ibid at para 46. 2020] PROGRAMMED OBSOLESCENCE 99
As in the case of the light bulb affair, efficiency concerns and technical constraints came into play, blurring the line between willful designs and technical limitations. The two-digit date recognition code was adopted taking into account storage limitations. Indeed, the Queen’s Bench was prompted to recognize that programmers knew that two-digit codes could render the software-embedded products obsolete with the beginning of the new millennium and took the deliberate decision to use this format regardless. The Court relied on a 1996 report, commissioned by CPR, which stated: “[t]he year 2000 problem is a limitation in computer systems [which rely on] applications built using commonly accepted standards [that] will not handle dates beyond December 31, 1999, correctly.”175 Just like the US Committee on the Judiciary accepted the two-digit code as a “tradition” of the computer industry, the Queen’s Bench held that the code “became embedded in the worldwide computer culture and remained the standard method for date processing, manipulation and storage.”176 The Court welcomed several reasons why the two-digit recognition code continued to be an industry standard, noting that, despite technological developments that eliminated internal memory and storage limitations, programmers “continued to reuse the two- digit date recognition code … in order for new programs to remain compatible with the vast amount of existing two-digit date data that had accumulated over the previous 2020 CanLIIDocs 3123 20 years.”177 With these factors in mind, the Court concluded that the two-digit date recognition code was not more than “a choice of the application programmer not to provide a century recognition routine.”178
Programmers could have written codes to mitigate the risk of obsolescence, but they did not have to. They made a conscious, not faulty, design decision that would potentially cause malfunctions or end the useful life of their products at the turn of the millennium. The case focused on the insurance contractual claim and did not raise the responsibility of programmers, but it certainly gave some indications as to the creative freedom in designing software. Programmers simply assumed the products would be replaced by an innovation through a “creative destruction” process.179 While they might not have planned for the obsolescence of their products per se, they did not ensure their reliable long-term usefulness.180 Obsolescence was preventable, but they had no duty to plan accordingly. The secrecy of software conception, concealed in the layers of integrated circuits, developed in secured laboratories and implemented by employees under strict confidentiality agreements, make the determination of the life expectancy of software all the more complex. In addition to its sophistication, software is a property unlike any other tangible object. Its polymorphic, evolving nature draws on the knowledge and continuous inputs of software engineers and developers and thus reveals itself to be a service for users more than their property. This passage from
175 Ibid at para 40 citing Keane Canada, Inc, “Year 2000 Risk Assessment Project: Final Report” (20 December 1996) [“Keane report”]. 176 American Home Assurance Company, supra note 151 at para 5. 177 Ibid. 178 Ibid at para 39. 179 Guiltinan, supra note 153 at 19. 180 Ibid at 24. 100 UNBLJ RD UN-B [VOL/TOME 71 property to service represents a paradigm shift. For software-enabled products, the question is perhaps less about durability than serviceability.
The lifecycle, performance or functionality of software or smart objects can be enhanced or diminished in various ways over time as the software systems are configurated, customized and modified (e.g., a security update of the operating system). In the digital universe, the useful life of a physical object may depend on these updates, upgrades and specifications, as well as the interdependence and interoperability, or lack thereof, of systems. The obsolescence of software-enabled products is then, in a sense, truly “programmed.” The control of the programming language, as once was the press,181 directly affects the digital economy. The fact that Application Programming Interfaces (API) are now the site of a fierce copyright war in the US.182 is indeed an indication that language and code are the new control commands. The Y2K problem can be seen as a first demonstration that code is power. In the case of smart objects, the cause of their obsolescence could be, as in the past, attributed to the programmer—should such impairment be proven— but what about issues of interoperability whereby, for instance, an application is blocked from operating, or the sudden withdrawal of authorization to use key codes? Both may affect 2020 CanLIIDocs 3123 durability. Can it then be regarded as programmed obsolescence even if the cause is external? Should law look into provoked obsolescence as a special cause of action? In the affirmative, it would be a Dantean task, especially when codes essential or instrumental to interoperability can be reserved and claimed as intellectual property, a conclusion recently reached by the United States Court of Appeals for the Federal Circuit in the case Oracle America Inc v Google LLC.183 Thirty seven packages of API were found copyrightable and infringed by Google who copied the material verbatim, without alteration, in the Android operating system of its smartphones. The Court of Appeals rejected the fair use defense and reversed the 2012 District Court holding that the functional considerations predominated in their design and therefore were not eligible to copyright protection.184 The case invigorates intellectual property and poses serious questions about code as a language to allow compatibility and interoperability. In addition, when a product combines hardware and software, there is a division of ownership and rules along the lines of tangible and intangible property. This distribution of powers over things, tangible or intangible, can be problematic: the owner of the physical object can be subjected to the software and its intellectual property owner. Presented in these terms, programmed obsolescence can be an act of
181 Marshall McLuhan, The Gutenberg Galaxy: The Making of Typographic Man, revised ed (Toronto: University of Toronto Press, 2011) at 203, 267–70. 182 Oracle America, Inc v Google, Inc, Appeal No 17-1118 (Fed Cir 2018) [Oracle]. The case deals with Java API packages, which are collections of “pre-written Java source code programs for common and more advanced computer functions” owned by Oracle (ibid at 8). Java programming language is generally free. However, Oracle imposes certain terms and conditions for use of the Java APIs in a competing platform. The case is expected to be heard by the Supreme Court of the United States in 2020. 183 Ibid. 184 Oracle America, Inc v Google Inc, No C 10-03561 WHA (ND Cal 2012), online (pdf): Electronic Frontier Foundation
The Y2K problem flags the dematerialisation of the economy, our dependence to code and the radical transformation of the equation of time, utility and property. Broadly speaking, the value of an object is determined, in part, according to the properties of the composition of matter, matter which is subject to deterioration over time, such as tungsten in the case of GE’s light bulbs. A such, matter is a determinant of durability and value. In contrast, the increased use of licences reduces the importance of the physical disposition of things. Value has morphed into a continuous current of interpersonal rights and obligations extending beyond the sale of the matter. Taking into consideration the product-service value of software-enabled goods, contract termination and chargeback guarantees offer a potential avenue to remedy provoked or premature obsolescence.
While the Senate Special Committee did not blame programmers for Y2K, it brought to the attention of the House of Representatives the lesson to be learnt from that episode: 2020 CanLIIDocs 3123
At the heart of the problem lies a serious disconnect between those who use technology and those who create it. On a worldwide scale, leaders of corporations and countries are struggling to understand the Y2K problem. In the process, they are receiving a crash course in the fragile mechanics of information technology. The Committee feels strongly that Y2K, as the first widespread challenge of the information age, must leave a legacy of increased awareness and appreciation of information technology’s role in social and economic advancement.185
Can an object speak for itself? Can it point ex post to the liability of its makers for not ensuring that the object continues to fulfil the functions it was acquired for? A tale of modernity, progress and science lies in the background of the stories of obsolescence. As objects deteriorate over time, a technology and its magics wear off, replaced by new alternatives. How a product will evolve is not immediately observable and challenges the status of the owner of the object who has only limited control over the technology. Law, because it is not written by scientists, attempts to safeguard human dignity by ensuring the enjoyment of property, thus perpetuating a long-lasting attitude of suspicion, defiance but also admiration of lawyers towards science. Obsolescence is the no-man’s land between entrenched positions, that of science and law, a terrain of dialogue and coproduction.186
185 Special Committee, Investigating the Impact of the Year 2000 Problem, supra note 164 at 7–8. 186 Pierre-Emmanuel Moyse, “Innovation: In the Shadow of Law”, WIPO/IPL/GE/16/T8 (2016) online: World Intellectual Property Organization
III—Epilogue: The Apple of Discord
Through the years, programmed obsolescence has evolved to a point where its early chapters are often quickly skimmed through. The concept, as it stands today, may seem far from London’s state-administered leases on consumer goods, yet recent legislative and jurisprudential developments bear the mark of the stories of the past. This section seeks to reconnect the current discourse on programmed obsolescence to its origins. Some recent developments, namely (a) the legislative attempts to sanction programmed obsolescence in France and Québec and (b) the wave of cases and settlements in response to the slowdown of Apple’s iPhones following the installation of an update, reveal the influence and relevance of prior academic work and cases on the obsolescence of products. Though it is non-exhaustive, this overview reveals that programmed obsolescence, as it attracts more attention, is taking on new proportions, but it still escapes the grip of the law.
A—The Laws of Obsolescence
2020 CanLIIDocs 3123 A number of legislative actions, predominantly in the area of consumer law, has been initiated in recent years to address programmed obsolescence.187 The reasons for the adoption of specific legislation are, in our opinion, twofold. First, the narrative of programmed obsolescence fueled the idea of a vast industrial conspiracy at the expense of the consumer. The GE anti-trust case became a caricatural example of unscrupulous and abusive business practices. It also showed that anti-trust law is ill-suited to address programmed obsolescence. Programmed obsolescence only constitutes a corroborative element of the conduct it targets, collusion. Therefore, from a policy standpoint, the passing of a bill rendering this practice illegal clearly seeks to respond to the growing public scrutiny and criticism of the act of programming the premature obsolescence of goods in light of the economic and social costs of the replacement of goods and waste. It shows governmental initiative even if the key issue, the very definition and reality of programmed obsolescence, remains unclear. Second, the new legislative initiatives are carried by a greater objective which also receives a growing adhesion, that of preserving the environment. In Québec, the current government recently tabled Bill 197 An Act to amend the Consumer Protection Act to fight planned obsolescence and assert the right to repair goods.188 This Bill, drafted by students and endorsed by a Member of the National Assembly, borrows from different foreign legislations, notably the 2015 French law, Loi relative à la transition énergétique pour
187 See Thierry Libaert, “Pour une consommation plus durable: en phase avec les enjeux européens” (2018), online (pdf): Actu Environnement
189 Loi relative à la transition énergétique pour la croissance verte, supra note 5. 190 C cons, s L.441-2. 191 Indeed, the French preparatory work, reports and debates, contain several references to Bernard London and stories of obsolescence, in particular, the Phoebus Cartel. See e.g. France, Assemblée nationale, Commission du développement durable et de l’aménagement du territoire, Rapport d'information déposé par la mission d'information sur la gestion durable des matières premières minérales, presented by Christophe Bouillon & Michel Havard, Report No 3880, (26 October 2011) at 14, online (pdf):
While these type of laws may be dissuasive, their normative effect are yet to be demonstrated. To this date, the French legislation has not led to a conviction directly related to programmed obsolescence. Several reasons may explain this apparent inertia. In most jurisdictions, consumers are already entitled to goods free from defects affecting their use or durability. When goods are defective, consumers can typically rely upon private law for redress. Québec’s Consumer Protection Act, for instance, provides legal warranties whereby “[g]oods forming the object of a contract must be durable in normal use for a reasonable length of time, having regard to their price, the terms of the contract and the conditions of their use.”198 The burden of proof regarding manufacturer liability is also lightened: once the purchaser has proven that the good has indeed perished or defaulted prematurely, the defect is presumed to exist.199 Secondly, while presumptions can facilitate the process of proving the existence of a defect, proving the existence of obsolescence, especially if it is provoked or programmed, remains onerous. As the General Electric anti-trust cases demonstrate, extensive investigative powers enforced by public authorities might be required to obtain evidence of the design decisions that were made and the objectives that were sought. Furthermore, technical constraints and compromises, as in the case of the light 2020 CanLIIDocs 3123 bulbs, as well as industry knowledge and standards at the time the good was produced, as in the case of the Y2K bug, may offer sufficient grounds to reject allegations of programmed obsolescence. Finally, laws to sanction programmed obsolescence, because they are by-products of property and product liability regimes, have not fully grasped the hybrid character and particularities of electronic devices and smart objects and, in particular, the risk of software obsolescence. In a context of a service economy and licences where value resides in how the user experiences the intangibles, the replacement of a good may be prompted not by the deterioration of the physical product, but rather by a decline in the uses and the experience that it provides to the user. In this new chapter of programmed obsolescence, the evolutionary law of software—governing the access to codes and platforms, interoperability and the freedom to accept updates and upgrades, to name a few—may require a new approach to durability. Unlike a burned out light bulb, the replacement of complex products may simultaneously be the result of different types of obsolescence, from design decisions to aesthetics, for which various actors may be responsible.
As countries join the cortege against programmed obsolescence, the driving force of the movement, that is, the collective aspirations for sustainable development and the need for immediate actions, is forcing a new reflection in the design of responsible modes of production, consumption and disposal of the goods. If the replacement rate was once the measure of prosperity, it has become the pathology of a dysfunctional system. Here, Veblen and his prescient and insightful comments
198 CPA, supra note 2, s 38. See also ibid ss 37, 53. 199 Ibid; 1729 CCQ, which provides that “[i]n a sale by a professional seller, a defect is presumed to have existed at the time of the sale if the property malfunctions or deteriorates prematurely in comparison with identical property or property of the same type; such a presumption is rebutted if the defect is due to improper use of the property by the buyer.” See Office de la protection du consommateur, “Examples of Judgments Concerning Legal Warranties” (2020), online: Government of Québec
B—The Bitten Apple
Across the world, a series of legal actions against Apple were launched after iPhone 6, 6s, SE and 7 slowed down following the installation of software updates of the operating system (iOS). The case has been described as the Apple Batterygate. The media and consumer organizations quickly denounced the throttling controversy to be an instance of programmed obsolescence to incite consumers to purchase new iPhones.200 A brief overview of how the cases unfolded shows the role of public authorities, the asymmetry of information between the producers and the consumers and the fight over the control of the tangible and intangible elements of complex products.
200 See e.g. “Apple et Samsung sanctionnées en Italie pour obsolescence programmée”, Le Monde (24 October 2018), online:
In France, the competition agency, the Direction générale de la concurrence, de la consommation et de la répression des fraudes (DGCCRF) opened an investigation in 2018, following a complaint by French advocacy group Halte à l’obsolescence programmée (HOP) alleging that Apple committed the new offence of programmed obsolescence. In February 2020, the DGCCRF concluded that Apple did not adequately inform iPhone users that updates of the exploitation systems iOS (10.2.1 and 11.2) could slow down the processor speed. The DGCCRF news release gives little information as to the legal characterization of the impugned activities. In a laconic statement, the DGCCRF indicates that
[t]hese updates … included a dynamic power management feature that could, under certain conditions and especially when the batteries were old, slow down the operation of iPhone 6, SE and 7 models. Failure to revert to the previous version of the operating system would have forced many consumers to change the battery or even buy a new phone.201
The DGCCRF concluded that the failure to adequately inform the consumers amounted to a “deceptive commercial practice by omission.”202 Following an 2020 CanLIIDocs 3123 agreement between the French prosecutor and Apple, the company agreed to pay €25 million and to publish a communiqué on their website for a month.203 In a statement, the decision was celebrated by HOP, which nevertheless expressed regret that the decision deprived the public of the opportunity to put planned obsolescence on trial.204 While the DGCCRF carefully pointed out that the update only slowed down degraded batteries and noted that the reduction in performance could force the replacement of the phones, it was not the technique in itself that was to blame. The DGCCRF acknowledged the impossibility for owners of the affected phones to cancel the update and recognized that their freedom to control how they use and experience their goods was infringed.
In this sense, the current debates on obsolescence become a way to address the issue of the asymmetry of power between producers and consumers and the need for more transparency with respect to a product’s functionality, durability and complexity. In 2018, the Italian Autorita’ Garante della Concorrenza e del Mercato (Italian Competition Authority, ICA) fined, under two separate decisions, both
201 France, Direction générale de la concurrence, de la consommation et de la représsion des fraudes, “Transaction avec le groupe APPLE pour pratique commerciale trompeuse” Ministère de l’économie des finances et de la relance (7 February 2020), online:
Apple205 and Samsung206 for unfair commercial practices concerning software updates affecting the performance of their products. The two firms were fined €10 million and €5 million respectively. The media hastily reported the case of Apple’s iPhones as a programmed obsolescence case although it was not filed as such. The allegation of programmed obsolescence only made its way in the decision through the arguments made by the consumers’ representatives, seemingly with the objective of introducing proof of bad character and taking advantage of the public outcry and attention that the concept had gained.207 However, the allegation was not determinative of the statutory violations for which Apple was found liable. Indeed, Apple was condemned under specific provisions of the Italian Consumer Code208 pertaining to unfair commercial practices, for, inter alia, having misled and omitted to properly inform the consumer and for insistently asking for updates without customers being adequately informed in advance about the inconvenience that the installation of these updates might cause, and giving only limited and belated advice about how to remedy these shortcomings.209 Again, the ICA’s decision highlighted the lack of clear alternatives or remedies to the problem even if Apple had implemented its worldwide Reduced Price Battery Replacement Program.210
2020 CanLIIDocs 3123 In the United States, Apple had to answer questions from the Senate Commerce Committee regarding its throttling practices.211 Apple publicly released their response in February 2018.212 In its statement, Apple insisted that its throttling practices were dictated by technological necessity rather than programmed obsolescence:
We have never—and would never—do anything to intentionally shorten the life of any Apple product or degrade the user experience to drive customer upgrades. Our goal has always been to create products that our customers love, and making iPhones last as long as possible is an important part of that.”213
205“PS11039 [Apple]” (25 September 2018), online (pdf): Italian Competition Authority
Apple provides in their letter a timetable of the process that led to this software update.214 They appear to credibly lay out the reports that they have received from consumers whose phones were unexpectedly shutting down around 30% battery charge. With the iOS 10.2, released on December 2016, Apple included a telemetry diagnostic tool that revealed issues with the degradation of the lithium batteries: when iPhones were hitting peaks of processor power, batteries that were cold, old or low on charge were unable to give a sufficient power supply and the phones were shutting off. The iOS 10.2.1, released in January 2017, was sent to users with the option of installing the update and an alert that stated that the update included “bug fixes and improved the security of [the user's] iPhone or iPad.” In February 2017, Apple added more information in a “Read Me” section which stated that the iOS 10.2.1 “improves power management during peak workloads to avoid unexpected shutdowns on iPhone.”215 In a statement released in December 2017, Apple explained that the power management features is intended to “smooth out the instantaneous peaks” by looking at a combination of the device temperature, battery state of charge and battery impedance to avoid unexpected shutdowns, which may have some effects such as longer launch 216 times, lower frame rates, backlight dimming and lower speaker volume. They 2020 CanLIIDocs 3123 argued that the software update was necessary to manage performance and avoid unexpected shutdowns as batteries age. They insist that, in fact, the update was a technique to counter the effects of obsolescence.217
The Apple Batterygate is perhaps not as clear of a case of programmed obsolescence as it was portrayed in the media. It is rather a case of a poorly designed product and lack of transparency about the power management feature included in the update and how it would potentially affect the performance of the device as the batteries age.218 As it was the case in the light bulb affair, consumers did not have the information needed to understand the compromise that the update was intended to reach between durability and efficiency. This lack of transparency is reflected in the decision of the French DGCCRF and the Italian ICA decisions. In the United States,
214 Ibid. 215 “About iOS 10 Updates” (last updated 2 March 2020), online: Apple
Apple agreed to pay up to $500 million to settle the matter and avoid the burdens and costs of litigation but still denied wrongdoing.219 Several lawsuits were filed by consumers against Apple across the United States in response to the throttling controversy. In 2018, the cases pending in different districts were coordinated into pretrial proceedings.220 While the case was settled, the Court ruled that the plaintiffs did not sufficiently plead that Apple fraudulently omitted information about the plaintiffs’ devices under California consumer protection laws. The Court held that they failed to explain precisely how those statements are misleading when consumers knew about “the degradation of batteries and the increasing capability of software.”221 The Court concluded that
[i]n reality, Plaintiffs apparently seek to hold Apple liable for failing to provide a battery that lasted as long as Plaintiffs preferred. Plaintiffs’ own allegations are to that effect. Plaintiffs readily concede ‘‘that iPhones worked as expected when new.’’ In their view, the problem is that ‘‘[a] battery and processor must be designed such that even as the battery ages and loses performance, it will still be capable of meeting the processor’s peak power demands for years to come.’’ … Nevertheless, California courts have been careful to ‘‘cabin the 2020 CanLIIDocs 3123 scope of the duty to disclose to avoid the unsavory result that manufacturers are on the hook for every product defect that occurs at any time, regardless of any time limits contained in their warranties.’’222.
As in the Y2K case, assumptions about knowledge regarding the obsolescence of software-enabled products may tip the balance. However, the Court found that consumers sufficiently stated a claim for trespass to chattels, noting that the allegation is not that the defendant’s actions “merely … shortened the battery life of the [devices],” but “instead, the alleged changes wrought by Apple’s iOS updates in the instant case ‘‘establish a significant reduction in service constituting an interference with the intended functioning of the system.”223 As it was the case with the Y2K episode, the throttling controversy also reflects a fight for the control of intangible embedded in the tangible product. While consumers were the owners of the iPhones, Apple continued to exert some control over the use, performance and durability of the products through software updates. The case forces us to see programmed obsolescence under a new light in a context where consumers use the label of “planned
219 “No. 02”, Westlaw Journal Class Action 07 (17 March 2020); “Apple to pay users $25 an iPhone to settle claims it slowed old handsets”, The Guardian (2 March 2020), online:
IV—Conclusions
The stories of obsolescence, planned or programmed, are inhabited by beliefs and collective narratives about technology. The dominant theme running through the chapters is that technology and its owners confiscate control over the use of a product, expropriating or reducing the capacity of the user to exercise command over it. Obsolescence becomes the diabolus ex machina. Technological objects make it difficult, if not impossible, for users to understand and repair their goods. Manufacturers of these products are in a position to program the use, the performance and the obsolescence of the goods, a power that they can use to trigger further consumption. In Veblen’s account of the emergence of the industrial civilization, technology changes the nature of socio-economic relationships between actors of the industry. Owners of the means of production become the de facto usufructuary and 2020 CanLIIDocs 3123 first beneficiary of the technical knowledge. The knowledge once held by the community of craftsmen is concealed in the machine made by engineers who are hired by investors. In Veblen’s words, “[t]he possession of the material equipment, therefore, placed in the discretion of its owners the utilisation of such technological knowledge and skill as the members of the given crafts might possess. The usufruct of the handicraft community’s technological proficiency in this way came to vest in the owners of the plant”.224 Innovation processes are thereby internalized and the automated functions of machines make the worker’s involvement and skills dispensable. Technical knowledge is bought and reserved for competitive advantages. It is used to design products in accordance with the specifications that the manufacturer sees fit with an eye for maximizing profits. This is perhaps the lesson of our investigation: the narratives of obsolescence omit to take into account how law, and in particular intellectual property law, grants power to owners of production means over the functionality of a product. After all, patents can be filed and obtained on improvements without being used. The GE case is indeed as much a case of unexploited innovation as it is one of monopolistic power. The technology for better and more durable lamps was available and patented, but it was more expensive to manufacture. More than an obsolescence story, the light bulb affair is a patent story. Through this prism, patent law is a perfectly legitimate retention mechanism. Indeed, some inventions will never be commercialized, however useful, socially or ecologically beneficial they may be. In the same way, copyright and trade secrets become strategies to prevent reverse engineering and are key instruments to control access to codes and algorithmic transparency. Settlements were an opportunity for Apple to escape intrusive evidentiary investigations into its internal operations. It is significant that Veblen himself, not a lawyer, became aware of the danger posed by intellectual property for the development of industrial society. On rare occasions, he shared his concerns on the nature of these new forms of property. In 1914, he wrote:
224 Veblen, The Instinct of the Workmanship, supra note 44 at 279. 2020] PROGRAMMED OBSOLESCENCE 111
“[i]n more advanced state of industrial arts, where ownership and the specialization of industry have had their effect, trade secrets, patent and copyrights are often of substantial value, and these are held in segregation from the common stock of technology.”225 In 1919, he explicitly compared intellectual property to classical forms of monopolies such as conspiracy or the combination of ownership. Intangible property, he wrote, “represents a “conscientious withdrawal of efficiency,” an effectual control of the rate or volume of output.”226
At the end of our journey, we are left with the impression that obsolescence sits uneasily with the current prescriptions of the law. It has not been fully characterized as a tort or as a crime. Its current regulation, often short of evidence, is rather limited and indirectly addresses the issue by upholding the obligations incumbent on manufacturers to inform consumers. There are reasons for these hesitations. After all, programmed obsolescence puts capitalism on trial. Modern law, being a creature of the capitalist system and author of the long-standing institutions of contract and property, cannot easily, without great risk of internal conflict, endorse liberal and transformative values while, at the same time, vehemently condemn its natural offspring. 2020 CanLIIDocs 3123
225 Ibid at 105. 226 Thorstein Veblen, The Vested Interests and the State of the Industrial Arts (New York: BW Huebsch, 1919) at 74.
ALGORITHMS AS LEGAL DECISIONS: GENDER GAPS AND CANADIAN EMPLOYMENT LAW IN THE 21ST CENTURY
Anthony Niblett*
Introduction
Should judges and arbitrators in Canada use algorithms to assist with their decision making? Could we ever replace the decisions of judges with the assessments of an algorithm? Some legal scholars and futurists have posited the idea that artificially intelligent algorithms could form the basis of legal decisions.1 This is not merely an issue for the future. Predictive algorithms are already used by lawyers to assist with dispute resolution.2 And artificially intelligent tools are being used as a basis for legal 2020 CanLIIDocs 3123
* Associate Professor at the University of Toronto Faculty of Law. In the interests of full disclosure, I am also a co-founder of a company called Blue J Legal, a start-up that uses machine learning technology to predict legal outcomes (including reasonable notice awards). The dataset used as the basis of the scholarly analysis here was created, in part, by legal researchers and analysts from Blue J Legal. 1 In popular media, see e.g. Christopher Markou, “Are We Ready for Robot Judges?”, Discover Magazine (15 May 2017) online:
decisions in China,3 Estonia,4 and other jurisdictions.5 The question of how much authority the Canadian legal system decides to delegate to algorithms is, therefore, one of paramount importance in the 21st century.
The use of such algorithmic tools to help make decisions raises a number of potential concerns. Algorithmic bias is frequently high on the list of concerns and examples of such bias are plentiful. In the legal context, studies have shown that racial bias infects algorithms used by judges to assess flight risk in bail decisions or risk of recidivism in sentencing hearings.6 These algorithms have been found to assess black defendants more harshly than white defendants, even when race is not one of the variables explicitly considered by the algorithm.7 In other contexts, Amazon recently built an artificially intelligent tool to help with hiring employees, but shut it down after it was discovered to be discriminating against women.8 The prevalence of such biases should caution us against using algorithms in decision making. This would appear to
3 Monisha Pillai, “China Now AI-Powered Judges”, RADII (19 August 2019), online:
114 UNBLJ RD UN-B [VOL/TOME 71 be especially true in the legal context, where decisions can come at the expense of life and liberty.
In this paper, I explore the potential to use one particular type of predictive algorithm in legal decision making. The type of algorithm examined here predicts the “most likely” outcome if a case were to go to court. An algorithm in this mould seeks to predict what would happen if a judge were to decide on the case. The algorithm relies on data generated from previous judicial decisions. It presupposes that the prior decisions of judges provide a good basis for making a prediction in future cases. If the algorithm were to be used by a judge, then the judge is essentially saying that she agrees with the law as decided in previous cases.
Here, I investigate potential bias in these types of algorithms. I focus on one specific legal issue where such algorithms could conceivably be used by judges to assist with decision making in the near future. The legal issue I explore is: what is a reasonable notice period to be awarded to an employee who has been dismissed. Under Canadian employment law, there is an implied term in an employment contract that upon dismissal without cause, an employee is entitled to a reasonable notice 2020 CanLIIDocs 3123 period, or a payment in lieu of such a notice period. What is reasonable will depend on the circumstances. It will depend on the age of the employee, how long they have worked with the employer, the type of job the employee had, what opportunities for similar employment exist, amongst other factors. This can be a difficult exercise for judges and arbitrators. How are they to weigh up all these different factors to arrive at what is reasonable?
Judges and arbitrators frequently look to past cases for assistance in determining the length of a reasonable notice period. They look at prior decisions to see how previous judges and arbitrators have weighed the factors, and—for the most part—try to come up with a reasonable notice period that is in line with the prior law. This, too, is what the predictive algorithm does. The predictive algorithm uses the data describing relevant precedents and provides a best guess about how a new case would fit into the existing body of case law.
But what if the body of case law is riddled with bias? If judges were to use predictive algorithms that replicate the existing law, then any biases currently found in the case law will merely be reinforced. Thus, it is imperative to ensure that the data upon which the predictive algorithm is based—i.e., existing case law—not only reflects the objectives of the law, but also is free from harmful bias that may entrench discriminatory outcomes.
It is thus necessary to ask whether the existing case law does indeed contain biases. Here, I focus on gender bias. I focus on gender bias for two reasons. First, it is relatively easy to determine the gender of plaintiffs in past cases (at least, it is much easier than the determining other characteristics, such as race). Second, there are studies that show there is evidence that legal decisions in reasonable notice cases contain elements of gender bias. Professor Kenneth Thornicroft, for example, has
2020] ALGORITHMS AS LEGAL DECISIONS 115 shown that female plaintiffs are awarded lower reasonable notice awards than male plaintiffs.9 If women receive lower damages then men for wrongful dismissal, holding all other variables constant, then one questions whether women and men are treated equally under the law. It follows that any algorithm using past decisions as the foundation for future decisions will merely perpetuate the gender bias. This, clearly, would be concerning and completely at odds with the objectives of the law.
Here, I re-examine the statistical evidence that reasonable notice awards in Canada reflect gender bias. I take advantage of the fact that there have been many more judicial decisions since Professor Thornicroft’s studies. I also use data from all published decisions, whether they be from arbitrators or judges, over the period 1997 to 2019. My data describe 1,728 legal decisions, over ten times the size of Professor Thornicroft’s dataset.10 Further, my data are more refined, with more variables of interest that enable more detailed description and explanation of the content of the existing law. I perform simple statistical tests to determine whether or not the existing case law reflect a gender gap in reasonable notice awards. If there is correlation between gender and the outcome, holding all other variables of relevance constant, then it would suggest that there may be differences in the way that female and male 2020 CanLIIDocs 3123 employees are treated by the legal system.
In short, I find no direct evidence of a gender gap in the awards of reasonable notice. In these 1,728 cases, there is no statistically significant correlation between the notice periods awarded to female plaintiffs and to male plaintiffs once all other relevant factors are held constant. This is not to dispute the findings of Professor Thornicroft’s study. On the contrary, I am for the most part able to replicate Thornicroft’s results in the subset of cases he examines and using his methodology. The broader point though is that when all available data are used over a longer period with more refined analysis, direct evidence of differential treatment vanishes.
While these results appear promising, they cannot be the end of the story. Gender differences in the law of reasonable notice can—and do—emerge through other channels, such as job type or compensation.11 Both the type of job and the level of compensation are correlated with judicial outcomes in my dataset; they are also correlated with gender. For example, the data show that clerical workers receive shorter reasonable notice awards than other workers, such as those in management. In the dataset, clerical workers are disproportionately female and management disproportionately male. Thus, gender biases may be baked into the legal test.12
9 Kenneth W Thornicroft, “Gender Bias in the Judicial Assessment of ‘Reasonable Notice’ Under Canadian Common Law” (2013) 64:1 Labor LJ 43 at 43–51 [Thornicroft, “Gender Bias”]. 10 The dataset used here is part of a larger dataset that was created for a separate project (and a different purpose) by numerous legal researchers and analysts from Blue J Legal. I am extremely grateful for their hard work and endeavour in putting together these highly detailed data. 11 Sandra Rollings-Magnusson, “Gender Implications of Wrongful Dismissal Judgments in Canada, 1994– 2002” (2009) 41:1 Can Rev Sociology 27. 12 On this point, see Judith MacFarlane, “Acknowledging the Relationship between Job Status and Gender: A Feminist Critique of Cronk v. Canadian General Insurance Company and the Managerial Distinction” 116 UNBLJ RD UN-B [VOL/TOME 71
Further, compensation is positively correlated with the outcome. In the dataset of cases, female plaintiffs receive lower compensation than their male counterparts. To the extent that compensation is correlated with both gender and the legal outcome, then it might be that female plaintiffs receive shorter reasonable notice period awards because they earn less. This has the effect of compounding any gender wage gap, since the final damages awarded to plaintiffs is the multiple of the reasonable notice award and the plaintiff’s compensation.13
This paper has two parts. The paper explores one broad issue (should we use algorithms as legal decisions?) by focusing on a narrower one (is there statistical evidence of gender bias in Canadian employment law?). The answer to the narrower question helps inform our views on the broader question. I answer the narrower question first.
In Part 1, I re-examine the statistical evidence on bias against female plaintiffs in reasonable notice awards. I show that the data do not bear out any direct evidence of gender bias in the case law. I show that there are, however, other channels through which gender bias has subsisted. In Part 2, I explore the potential for judges to use 2020 CanLIIDocs 3123 algorithmic predictions in the decision-making process, arguing that there may be still be concerns about bias, depending on how the algorithm is implemented. In short, great care needs to be taken to ensure that the algorithm does not reinforce hidden biases in the law. I further explore the possibility of alternative types of algorithms that do not rely on judicial decisions as data. A final part concludes.
1. Gender bias and reasonable notice awards
In this Part of the paper, I examine the evidence as to whether the existing case law on reasonable notice awards reflects gender bias against female plaintiffs. First, I will provide a short background of the legal issue. Next, I discuss the prior literature on gender differences in this area of law. I will describe the dataset of 1,728 cases from 1997 to 2019 and discuss my results. In simple linear regression tests, there is no direct evidence of gender bias. But I also explore other channels through which female plaintiffs may have been treated unfavourably.
1.1 The legal background
Under Canadian employment law, workers dismissed without cause are entitled to a reasonable notice period, or payment in lieu of such a notice period. In the seminal case on this legal issue, Bardal v The Globe & Mail, Ltd, McRuer CJHC held that this
(1997) 9:2 CJWL 418 (“[the majority position of the court in Cronk] reaffirmed aspects of the managerial/clerical distinction have an adverse effect of working women” at 420). 13 This point is also noted by MacFarlane. See ibid.
2020] ALGORITHMS AS LEGAL DECISIONS 117 reasonableness standard does not lend itself easily to bright line rules. In an oft-cited passage he noted
[t]here can be no catalogue laid down as to what is reasonable notice in particular classes of cases. The reasonableness of the notice must be decided with reference to each particular case, having regard to the character of the employment, the length of service of the servant, the age of the servant and the availability of similar employment, having regard to the experience, training and qualifications of the servant.14
The four factors mentioned in this passage—character of employment, length of service, age of employee, and availability of similar employment—are typically referred to as the Bardal factors.
The fact that the leading case provides only vague guidance to what is reasonable has generated an enormous wealth of litigation on this legal issue.15 Since the Bardal decision was handed down, there have been literally thousands of cases that have required judges or arbitrators to determine what is reasonable in the circumstances. This can be distinguished with other jurisdictions, where the 2020 CanLIIDocs 3123 determination of notice periods is more rule-like, leading to less litigation on this specific issue.
1.2 Prior literature
Are reasonable notice awards in Canada biased against female plaintiffs? A robust literature in law, business, and economics has sought to answer this question empirically. The evidence is mixed, but scholars have identified some areas where the decisions of judges suggest bias against female employees who bring suit against their former or current employer. In a relatively recent study, Professor Kenneth Thornicroft examined 132 appeal court decisions of reasonable notice cases from 2000 to 2011 from across Canada.16 Professor Thornicroft found that the 27 female plaintiffs in his sample received lower damages awards than the 105 male plaintiffs, holding constant all other variables of relevance.17 The magnitude of this difference was in the range of 1.43 months to 1.55 months.18
14 Bardal v Globe & Mail Ltd (1960), 24 DLR (2d) 140, [1960] OWN 253 (Ont HC). 15 Chenyang Li, “You Can’t Fire Me: The Problems with Wrongful Dismissal Damages in Canada” (2017) 1:1 Western J Leg Studies 1. 16 Thornicroft, “Gender Bias”, supra note 10. 17 Ibid at 47. See also Kenneth W Thornicroft, “The Assessment of Reasonable Notice by Canadian Appellate Courts from 2000 to 2011” (2013) 17:1 CLELJ 29. Thornicroft found a negative correlation between female gender and size of the award. In this particular study, his results suggested that women received 1.5 to 1.7 months’ less notice then their male counterparts. 18 See Thornicroft, “Gender Bias”, supra note 10. The coefficient on female in Table 3, page 48 in specifications (2) and (4). Depending on the empirical specification, the difference was statistically significant at either the 5% or 10% level. The difference is significant at the 10% level in specification (2) and at the 5% level in specification (4). 118 UNBLJ RD UN-B [VOL/TOME 71
This question had been explored before by other scholars in Canada. One other paper—by Professor Sandra Rollings-Magnussen—found evidence of gender bias.19 The majority of the papers, however, uncover little evidence of a gender gap once other factors are controlled for. Some of these papers investigate cases from earlier time periods, or from a particular jurisdiction, or they focus on one level of court or board.20 But a common factor in each of these studies is that the number of observations are relatively small. These studies typically have fewer than 200 observations, and almost always have fewer than 300. This is somewhat of a limitation in the prior literature. It is this gap I seek to fill.
There have been other studies of gender bias and reasonable notice that have not used statistics to uncover bias. Judith Macfarlane, for example, notes that the legal test for reasonable notice is biased against women given that judges make distinctions based on job type. In particular, clerical workers are treated differently to managerial workers.21 More generally, there is, of course, a large and important literature in feminist legal theory exploring and explaining how the effect of law may be to 22 entrench male supremacy. There is a rich literature exploring gender bias in labour 2020 CanLIIDocs 3123 and employment law that goes far beyond mere statistical analysis of reasonable notice awards.23
1.3 The dataset
I use a novel dataset of 1,728 cases where judges or arbitrators decide the issue of an employees’ reasonable notice period over the timeframe 1997 to 2019.24 These cases
19 Rollings-Magnusson, supra note 12. 20 See e.g. Steven L McShane, “Reasonable Notice Criteria in Common Law Wrongful Dismissal Cases” (1983) 38:3 Relations Industrielles 618; Steven L McShane & David C McPhillips, “Predicting Reasonable Notice in Canadian Wrongful Dismissal Cases” (1987) 41:1 Indus & Lab Rel Rev 108; Tim Liznick, “Wrongful Dismissal: Determining Reasonable Notice” (1987) 5:4 Worklife Report 1; Terry H Wagar & Kathy A Jourdain, “The Determination of Reasonable Notice in Canadian Wrongful Dismissal Cases” (1992) 43:1 Labor LJ 58; Terry H Wagar, “Wrongful Dismissal in Small and Medium-Sized Firms: Some Empirical Evidence” (1995) 12:2 J Small Bus Entrepreneurship 94; Terry H Wagar, “Determinants of Just Cause and Reasonable Notice in the Dismissal of Nonunion Employees” (1996) 4:3 Can Bus Economics 36; Terry H Wager & James D Grant, “The Relationship between Plaintiff Gender and Just Cause Determination in Canadian Dismissal Cases” (1996) 34 Sex Roles 534. 21 MacFarlane, supra note 13. 22 See e.g. Judith A Baer, Our Lives before the Law: Constructing a Feminist Jurisprudence, (Princeton: Princeton University Press, 1999); Susan B Boyd & Elizabeth A Sheehy, “Canadian Feminist Perspectives on Law” (1986) 13:3 JL & Soc 283; Ann C Scales, “Towards a feminist jurisprudence” (1980) 56 Indiana LJ 375; Ann C Scales, “The Emergence of Feminist Jurisprudence: An Essay” (1986) 95:7 Yale LJ 1373. 23 See e.g. Monica Boyd, “Feminizing Paid Work” (1997) 45:2 Current Sociology 49; Joanne Conaghan & Kerry Rittich, Labour Law, Work, and Family: Critical and Comparative Perspectives (Oxford: Oxford University Press, 2005). 24 These 1,728 cases include 157 employment cases from Québec. Readers may argue that these 157 cases should not be included in the analysis, given that they do not follow the common law of reasonable notice.
2020] ALGORITHMS AS LEGAL DECISIONS 119 represent all legal decisions found on this issue on CanLII or other databases, such as WestLaw, in this time period. Of these 1,728 cases, 562 of the plaintiffs (32.52%) are female. While cases with male plaintiffs make up the majority of the dataset, this imbalance is smaller than in previous studies of reasonable notice award that contain fewer observations. In those studies, the proportion of female plaintiffs ranged from 10% to 24%.
For each decision, coders track the outcome of the case: how long was the awarded reasonable notice period? This variable is normalized such that it is measured in terms of the number of months’ notice given.25 Data are also collected on many other independent variables of interest that may be explaining variation in these outcomes. Data are carefully extracted from each case. These data are rich and refined.
First, data are collected on the Bardal factors. I have information on the character of employment, the length of employment (measured in years), the age of the employee, and the availability of similar employment. For character of employment, I can determine the type of job (level of management, professional, sales, etc.), whether the employee was a supervisor or not, and whether the position was 2020 CanLIIDocs 3123 deemed to be unique or specialized. Information about whether the decision maker thought that employment opportunities were limited due to travel or economic reasons are included. Data on the plaintiff includes the level of experience in the industry, education levels, whether they were actively seeking other employment before they were dismissed, and whether they suffer from illness or disability that will limit employment opportunities.
Second, factors other than the Bardal factors are included. These factors include compensation, performance on the job, whether the worker was induced away from another job to work for the employer, and whether the employer was in financial difficulty. Further information of legal relevance, such as which province the case was heard and whether the Canada Labour Code is referenced, is also included. Cases from Ontario and British Columbia make up more than half the dataset. 617 cases (35.71%) are from Ontario, while 405 (23.44%) are from British Columbia.
None of my results turn on the inclusion or exclusion of Québec from the data. If I run the specifications on the 1,571 cases from outside of Québec, the results do not change. 25 One important point to note here is that courts after 1997 but before 2008 awarded additional damages called Wallace damages, following the Supreme Court of Canada’s decision in Wallace v Union Grain Growers, Ltd, [1997] 3 SCR 701, [1999] 4 WWR 86. In my dataset, I explicitly do not include Wallace damages. In each case, I merely note the number of months that the court or tribunal held was a reasonable notice period. This practice of awarding Wallace damages largely stopped after the Supreme Court’s decision in Honda Canada, Inc v Keays, 2008 SCC 39. 120 UNBLJ RD UN-B [VOL/TOME 71
1.4 Tests and results
1.4.1 Differences between female plaintiffs and male plaintiffs
Table 1 shows the mean averages for some of the more important variables in the dataset. On average, across the 1,728 cases, the average reasonable notice award given is 10.15 months. I break down the summary statistics by gender. There are strong statistical differences between male plaintiffs and female plaintiffs. Female plaintiffs receive shorter notice periods than male plaintiffs on average. This difference, simply taking the average of the male plaintiffs’ notice periods and the average of female plaintiffs’ notice periods, is a little over one-and-a-half months. This difference is statistically significant.26 This means that there is little chance that the difference between female and male plaintiffs is attributable to chance alone.
Obviously, this observation by itself does not provide evidence that female plaintiffs are treated differently to male plaintiffs. Other variables that affect the legal outcome need to be considered. Importantly, Table 1 also shows that variables such as 2020 CanLIIDocs 3123 length of service, age, and compensation are all different in the two subsets of the data. Female plaintiffs in the sample have shorter tenures with the employers than male plaintiffs (10.16 years, compared to 11.59 years), are typically younger than their male counterparts (44.33 years old, compared to 47.56 years), and earn less significantly less in terms of annual compensation ($62,521, compared to $102,292 in 2010 dollars). These differences could help explain the differences in the outcomes of the cases involving female and male plaintiffs.
Table 1: Mean average of key variables in the dataset
1.4.2 Results
To test whether or not there is a statistically significant difference between the awards given to female plaintiffs and male plaintiffs, I run a simple linear regression model that looks at the correlation between reasonable notice awards and all relevant explanatory variables. Table 2 reports the results. This table may appear to be difficult to read for those without training in economics or statistics, but the overall conclusions are easily interpretable. Put simply, once I take other factors into account, I find no
26 A statistical test known as a t-test reveals that the difference is significant at the 1% level (t = -3.05).
2020] ALGORITHMS AS LEGAL DECISIONS 121 direct evidence of gender differences in the dataset. That is, if you take two plaintiffs— one male, one female—with the same employment profile and characteristics, there would be no statistical difference between the reasonable notice awards given.
Specification (1) simply tests the correlation between the outcome of each case—the notice period awarded by the judge or arbitrator—and the gender of the plaintiff. Here, the coefficient on the variable female is -1.456. The coefficient can be interpreted as meaning that female plaintiffs receive, on average, notice periods that are about one-and-a-half months shorter than male plaintiffs in our dataset. This reflects difference between female and male plaintiffs in Table 1 above. But, as noted above, these differences between the notice awards of female and male plaintiffs are correlated with other variables that do a better job of explaining the variation in notice periods. I include these variables in specifications (2) through (5).
(1) (2) (3) (4) (5) Notice period Notice period Notice period Notice period Notice period (All data) (All data) (All data) (Omit Quebec) (All data)
Female -1.456*** 0.165 0.144 0.104 -0.027 (0.319) (0.178) (0.161) (0.159) (0.143) 2020 CanLIIDocs 3123 Age of employee (years) 0.033*** 0.031*** 0.033*** 0.031*** (0.010) (0.010) (0.010) (0.009) Tenure of employee (years) 0.510*** 0.505*** 0.523*** 0.860*** (0.014) (0.022) (0.014) (0.034) Tenure squared -0.012*** (0.001) Occupation -- sales -0.021 -0.114 0.131 -0.015 (0.315) (0.366) (0.338) (0.258) Occupation -- clerical work -0.531 -0.697* -0.442 -0.761** (0.366) (0.367) (0.353) (0.336) Occupation -- labourer -0.868*** -0.944** -0.725* -1.065*** (0.277) (0.363) (0.338) (0.302) Occupation -- professional 0.287 0.463 0.719*** 0.629* (0.335) (0.307) (0.163) (0.313) Occupation -- lower management 0.547** 0.547 0.776* 0.430 (0.276) (0.445) (0.414) (0.306) Occupation -- middle management 0.651** 0.752* 1.094*** 0.891** (0.290) (0.373) (0.141) (0.310) Occupation -- upper management 0.941** 1.231*** 1.183*** 1.505*** (0.378) (0.293) (0.335) (0.272) ln(compensation) (2010 dollars) 1.316*** 1.280*** 1.194*** 1.115*** (0.159) (0.263) (0.250) (0.228) Controls for education and experience No Yes Yes Yes Yes Control for illness or disability No Yes Yes Yes Yes Controls for economic conditions No Yes Yes Yes Yes Control for inducement of worker No Yes Yes Yes Yes Control for uniqueness of position No Yes Yes Yes Yes Controls for worker performance No Yes Yes Yes Yes Conttrol for province No Yes Yes Yes (no QC) Yes
Observations 1,728 1,728 1,728 1,571 1,728 R-squared 0.011 0.747 0.764 0.788 0.794 Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 Table 2: Factors that are correlated with reasonable notice awards in Canada (1997-2019)
Specification (2) includes a number of explanatory variables that are likely to correlate with the length of the notice period awarded. Importantly, I include variables that measure the Bardal factors: the age of the employee, the length of the employee’s tenure with the employer, the education and experience of the employee, as well as a number of factors that describe both the character of the employment and the 122 UNBLJ RD UN-B [VOL/TOME 71 availability of other work. I also include other (non-Bardal) factors that judges have mentioned in their written opinions, and are hypothesized to correlate with the outcome.
The results are highly intuitive. The variables, for the most part, correlate with notice period as one might expect. For example, the employee’s length of tenure with the employer is strongly correlated with the outcome. Holding all other factors fixed, an increase of one year in tenure suggests an increase in the notice period awarded of just over half a month (the coefficient on tenure is 0.51). The length of tenure is, by far, the most important factor in terms of explanatory power. Age of employee, experience in the industry, and education are all positively correlated with the outcome, as are factors that suggest it will make it difficult for the employee to find new work. If the employee was actively seeking work at the time of dismissal, this is correlated with a reduction in the notice period. If the employee was induced to join the employer, the notice period is longer by over 1.14 months. The more compensation the worker receives, the longer the notice period. All these variables are statistically significant, meaning that there is low likelihood that these differences are attributable to chance alone. 2020 CanLIIDocs 3123
The type of job is also correlated with the outcome. Upper and middle management jobs are strongly correlated with the length of a notice period. Being in upper management, for example, is associated with notice periods that are 0.94 months longer than workers in jobs not covered by the categories mentioned. In this specification, clerical workers are not statistically different from other jobs not specifically included in the data. But in specifications (3) through (5), clerical workers do receive significantly shorter notice awards. These data confirm that a managerial- clerical distinction persists in reasonable notice awards.27
Most importantly for the analysis presented here, notice period awards are not correlated with gender. In fact, the coefficient on female is no longer negative in this specification (= +0.165). The positive coefficient does not mean much, however, as it is not statistically different to zero. I am unable to reject the hypothesis that any putative difference between female plaintiffs and male plaintiffs is not explainable by chance alone.
Specification (3) includes the province in which the case was heard. Provinces vary in their willingness to award reasonable notice.28 New Brunswick, Ontario, and British Columbia are among the more generous to employees; while Québec is less generous. In specification (4), I remove all Québec cases. This does not affect the result regarding gender. There is still no gender gap.
27 See e.g. Cronk v Canadian General Insurance Co (1995), 128 DLR (4th) 147, 25 OR (3d) 505, (CA) where the Ontario Court of Appeal retained the “managerial/clerical” distinction in accessing reasonable notice damages for wrongful dismissal cases. 28 Specifications (3), (4), and (5) also cluster standard errors at the provincial level. This does not significantly impact upon the results, particularly the results on gender.
2020] ALGORITHMS AS LEGAL DECISIONS 123
In specification (5), I include a variable called tenure squared, that is the number of years the employee worked for the employer multiplied by itself. Based on visual inspection of the data, the relationship between tenure and the outcome appears to have a concave shape. This means that an extra year when tenure is short is associated with a greater marginal increase in the notice period than an extra year when the tenure is long. The significant negative coefficient on tenure squared confirms this concavity. This suggests that this may be a better fit than specification (3). While the coefficient on female in this specification is now negative (-0.28 months), it is still not statistically significant at any reasonable level of significance.
The main takeaway from Table 2 is that there does not appear to be a statistically significant relationship between the gender of the plaintiff and the length of the notice period awarded. Controlling for other variables of interest, female and male plaintiffs are not treated differently.
1.4.3 Relationship to prior literature 2020 CanLIIDocs 3123
My empirical findings on gender are broadly consistent with studies of reasonable notice awards from the last century. The findings do, however, differ from those of Professor Thornicroft, who finds a statistical difference between female and male plaintiffs. What can account for the difference between the two findings. Why does Thornicroft find evidence of gender bias, but I do not? The answer lies in the amount of data available in my study and the refinement of independent variables this volume of data allows me to include in my analysis. Thornicroft’s study is limited to 132 appeals cases, whereas I have access to data from 1,728 court and tribunal decisions. This not only gives me a more complete picture of the legal landscape, it also provides a more recent investigation, and the number of observations gives greater degrees of freedom to include a greater number of relevant independent variables. None of this is to say Professor Thornicroft’s study is not correct. Indeed, if I restrict my analysis to appeals courts and the time frame used in Professor Thornicroft’s study, I am able to largely replicate his important findings, using his empirical methodology.29
Three observations are important here:
First, as noted above, the independent variable that explains variation in reasonable notice awards better than any other is the length of the employee’s service. The correlation between tenure and the outcome is very strong. In my dataset, tenure is represented as a continuous variable. This builds on the assumption that employees who have worked at a firm for 2 years will be treated differently to those who have worked 3 years (or even 2.5 years), and workers with 11 years of service are treated differently those with 12 years of service. For each additional year of service, the reasonable notice award is correlated with longer reasonable notice awards. Professor
29 Restricting my dataset to appeals in courts that were heard between 2000 and 2011, I find 133 cases. Professor Thornicroft’s study has 132 observations. This difference is not important to the findings. 124 UNBLJ RD UN-B [VOL/TOME 71
Thornicroft, on the other hand, uses categorical variables for tenure, putting workers into groups of less than 5 years of service, 5–15 years of service, and more than 15 years of service. Given the size of the dataset, this is completely understandable. My data, however, allow for greater precision; it allows for greater explanatory power.
Second, there are a number of variables that are not statistically significant in Professor Thornicroft’s study that are correlated with the outcome in my dataset. For example, there are provincial differences that come alive when seen in the broader and longer dataset.
Third, even when I use Professor Thornicroft’s empirical specification— using a reduced number of independent variables and converting continuous variables into categorical variables30—across all courts and tribunals and across the broader dataset, I find no evidence of gender differences.31 This can be viewed as a positive story. This finding suggests that the spectre of gender bias found in Thornicroft’s study is not present in the broader adjudication system32 and has dissipated since 2011.33
2020 CanLIIDocs 3123 1.4.4 Other evidence of gender bias
My data suggest that there is no direct evidence that courts and tribunals treat female plaintiffs systematically different to male plaintiffs in awarding reasonable notice. But this, by itself, is not evidence that there may be bias or discrimination. It may be, for example, that there are other explanatory variables that are correlated with reasonable notice awards that mask bias. I explore three such channels here. The first is the type of job; the second is the level of compensation received by the employee; and the third is the willingness to litigate.
The type of job is correlated with reasonable notice awards (see Table 2). Upper level management, for example, consistently receive higher awards, holding fixed all other variables. The more senior management position held by the worker, the longer the reasonable notice award. On the other hand, clerical workers receive, on average, less than other workers. Clerical workers are disproportionately female in the dataset (110 of the 127 clerical workers in my dataset are female (86.61%)).
30 The key specification in Thornicroft’s study that I recreate is specification (2) in the table on page 48. This empirical specification includes a constant, age > 50 years, tenure < 5 years, tenure > 15 years, compensation, and female. 31 If I recreate specification (2) of Thornicroft’s study for all 1,728 cases, the coefficient on female is 0.10. It is not statistically significant. 32 Looking at all 871 cases over the period 2000 to 2011 in courts and tribunals, using Professor Thornicroft’s methodology, I find no statistical relationship between gender and reasonable notice outcomes. 33 If I use data to appeals courts over the entire period 1997 to 2019 (n = 248) and use Professor Thornicroft’s methodology, there is no significant relationship between gender and outcomes. Looking at the 78 appeals cases since 2011 reveals no significant relationship.
2020] ALGORITHMS AS LEGAL DECISIONS 125
So, while there may not be any direct findings that female plaintiffs are treated differently to male plaintiffs holding fixed the type of job, there is evidence to suggest that the types of job that female plaintiffs have disproportionately held are awarded shorter reasonable notice periods. All this is to suggest that the legal test itself contains gender bias. The case law reflects deeper problems in gender benefit disparity in the broader Canadian labour market that gets reproduced in the case law. As Judith Macfarlane states: “work that is viewed as ‘women’s work’ is undervalued and underpaid compared to that commonly done by men, and there are distinct barriers for women who attempt to leave the more traditional female occupations” and “the managerial/clerical distinction is based on society’s biased viewed about the value of different jobs—bias that harms women.”34
The level of compensation is positively correlated with reasonable notice awards (see Table 2). The higher the salary of the worker, the longer the reasonable notice awards. This correlation may be capturing unobservable relevant factors about the nature of the work that increase the reasonable notice award that I cannot capture in the data. Further, the difference in compensation between female and male plaintiffs, in terms of compensation, is stark (see Table 1). Here, female plaintiffs earn 2020 CanLIIDocs 3123 less than male plaintiffs, a pattern that is still observed in the broader society.35 To the extent that reasonable notice awards mirror compensation, female plaintiffs will receive shorter reasonable notice periods if they earn less than their male counterparts.
A third channel is the willingness to litigate and insist upon legal rights. All the empirical studies exploring gender bias and reasonable notice awards have noted that male plaintiffs make up the majority of the cases. My dataset is no exception. If male plaintiffs are more willing to take legal action in wrongful dismissal cases, this too may influence ultimate outcomes.36
1.5 Conclusions on the empirical evidence
The data tell an important story. I find no direct evidence that gender and outcome are correlated once other factors are taken into account. Of course, those other factors could indeed—and historically have been—correlated with gender. The findings here cannot be taken to suggest there is no gender bias in the law of reasonable notice.
34 See MacFarlane, supra note 13 at 420. MacFarlane also notes that women “make up a disproportionate percentage of workers in lower status jobs compared to the male workforce” at 420. See also: Rollings- Magnusson, supra note 12; Terry H Wager & James D Grant, “The Relationship between Plaintiff Gender and Just Cause Determination in Canadian Dismissal Cases” (1996) 34 Sex Roles 534. 35 See e.g. Statistics Canada, The gender wage gap in Canada: 1998 to 2018 by Rachelle Pelletier, Martha Patterson & Melissa Moyser, Catalogue No 75-004-M (Ottawa: Statistics Canada, 11 October 2019); Nicole M Fortin, “Increasing Earnings Inequality and the Gender Pay Gap in Canada: Prospects for Convergence” (2019) 52:2 Can J Econ 407. 36 See also James D Grant & Terry H Wagar, “Willingness to Take Legal Action in Wrongful Dismissal Cases: Perceptual Differences between Men and Women” (1992) 74:3 Perceptual & Motor Skills 1073 at 1073–74 (146 business students with full-time work experience participated in a study of dismissal from employment. Men were more likely than women to favour court action in the event of dismissal). 126 UNBLJ RD UN-B [VOL/TOME 71
Indeed, the empirical evidence suggests gender differences have likely manifested themselves through other channels such as job type and compensation.
2. Use of algorithms by judges and arbitrators in reasonable notice cases
2.1 Algorithms based on prior decisions of judges and arbitrators
Can algorithms be used to assist judges and arbitrators in making determinations of reasonable notice awards in the near future?37 Here, I focus on one type of predictive algorithm: one that seeks to replicate what the existing law is. In this way, the algorithm takes all existing case data and make a best prediction about how a court would decide this case.38
An example may prove helpful. Sam has been recently dismissed from a job as a middle manager in Ontario. Sam is 49 years old and has been working at the same firm for 12 years. Sam earned $100,000 last year and received no reports of poor 2020 CanLIIDocs 3123 performance. There are few indications that Sam will find it more difficult to find work than somebody else in the same situation. What is a reasonable notice period for Sam? The firm that has recently dismissed Sam says that 4 months is reasonable. Sam seeks 18 months. In order to determine what is reasonable, a court may look to prior decisions. Human judges and arbitrators are limited in their capacity to read thousands of cases and absorb all the relevant information in a short period of time. An algorithm that is based on all prior decisions, however, could essentially provide a distillation of how much weight to put on each relevant factor.39 The algorithm essentially provides a prediction of how previous courts would have decided. For example, in Sam’s case it might provide a predicted reasonable notice period of 12 months. In some sense, it is as if the prediction is the result of a survey, asking every judge who has decided a previous case what the law should be in Sam’s case.
The potential benefits of such an algorithm are clear. Suppose, hypothetically, that in the future, decision makers delegate decision making authority to the algorithm. That is, the algorithm provides the content of the legal decision. In the example above, the prediction of 12 months’ notice becomes the legal decision.
37 See generally Sourdin, supra note 6 on the recent developments in AI and its profound impact on judges and judging the future. 38 See Andrew Stranieri & John Zeleznikow, Knowledge Discovery from Legal Databases (Dordrecht, NL: Springer, 2005) at 211–26 discussing the prediction of outcomes in ordinary cases that depend on judicial discretion—especially discovering patterns of judicial reasoning—provided that the data collected reflects the reasoning processes. See also Johnathan Jenkins, “What Can Information Technology Do for Law” (2008) 21:2 Harv JL & Tech 589; Naomi Burstyner et al, “Using Technology to Discover More about the Justice System” (2018) 44:1 Rutgers Computer & Tech LJ 1. 39 Stranieri & John Zeleznikow, supra note 39 at 224.
2020] ALGORITHMS AS LEGAL DECISIONS 127
This decision is one that best accords with the decisions in previous cases. This type of algorithm takes the doctrine of precedent very seriously.
The idea of such an algorithm would run contrary to the assertion of McRuer CJHC in Bardal who held that there “can be no catalogue laid down as to what is reasonable.” Indeed, using the experience of hundreds of judges over thousands of cases, the law can better provide a clearer answer as to what is reasonable. This is in line with Justice Oliver Wendell Holmes’s proclamations about how legal cases provide more information and the law evolves from standards (such as “reasonable” notice) towards clearer rules.40 The reduction in uncertainty is potentially enormously beneficial to society. There are also important benefits from the viewpoint of the rule of law. The algorithm provides consistent answers.41 The algorithm’s results do not change depending on the identity of a human judge. The algorithm can wash out biases that currently prevail across different judges and arbitrators, by making the law more consistent. No longer will reasonable notice awards be a function of which particular judge you are randomly assigned to or the whims of what the judge had for breakfast.42
Like cases will be treated alike. Moreover, if all parties know the content of the 2020 CanLIIDocs 3123 algorithm there will be no dispute on substance.43 The litigants will realign their expectations. Sam will not seek 18 months. The firm will not provide a mere 4 months. The litigants will not be litigants.
But this certainty and consistency will, obviously, come at a cost. First, there is a concern that the law is something uniquely human.44 These concerns contain less force with regards to the types of algorithms discussed here, since these algorithms are based on the previous decisions of human judges. But perhaps there is something uniquely human about deciding a new case, that factors in special factors that an
40 Strict application of the doctrine of precedent also introduces economies of scale for standards, but it does so in a way that turns the standard into a rule: see Oliver Wendell Holmes Jr, “The Path of the Law”, (1897) 10 Harv L Rev 457; Anthony Niblett, “Case-by-Case Adjudication and the Path of the Law” (2013) 42:2 J Leg Studies 303. 41 See John Zeleznikow, “Building Decision Support Systems in Discretionary Legal Domains” (2000) 14:3 Intl Rev L, Computers & Technology 341 (“the construction of intelligent legal decision support systems in discretionary domains will enhance consistent decision-making leading to increased confidence in the justice system and provide support for alternative dispute resolution” at 341). 42 Okzan Eren & Naci Mocan “Emotional Judges and Unlucky Juveniles” (2018) 10:3 American Economics J: Applied Economics 171 (finding that unexpected losses in football games increase sentence lengths by judges during the week following the loss); Fabian, supra note 7. 43 Stranieri & Zeleznikow, supra note 39 at 222 (claiming that the development of legal decision support systems would lead to users being aware of the outcome of litigation and therefore, encouraged to avoid the costs and emotional stress of the legal proceedings). 44 W Bradley Wendel, “The Promise and Limitations of Artificial Intelligence in the Practice of Law” (2019) 72:1 Okla L Rev 21 (“[t]he capacity of legal rules and principles to furnish reasons, create obligations, and possess authority all depends on the shared standpoint of mutual respect adopted by free and equal persons” at 48). 128 UNBLJ RD UN-B [VOL/TOME 71 algorithm cannot.45 Second, the law will be less dynamic. The decisions of past decisions are concretized. There is little room for evolution. If what we, as a society, view as “reasonable” changes, the law will not change. Third, and relatedly, the law will continue to replicate any biases that are already in the content of the law. To the extent that individual judges’ have biases, these can be largely washed away by the algorithm, but systemic biases in the way judges have decided cases will continue. All this is to say, if we are uncomfortable mechanically applying the past law to future cases, then one should put less weight on the use of algorithm.
The question of bias is obviously of central importance. For the purposes of the narrow focus of this paper, will the recommendations or predictions be biased against female plaintiffs? In the context of the above example: Will the reasonable notice award be different if “Sam” is a female plaintiff or a male plaintiff? The empirical analysis in Part 1 suggests that there is no evidence of direct gender bias in the reasonable notice cases in Canada.
But, to be very clear, this does not mean that a predictive algorithm will not systematically entrench gender differences. Indeed, the use of such algorithms to 2020 CanLIIDocs 3123 provide legal decisions needs to be done with care, to ensure that these biases are not baked in. Consider three reasons why we cannot simply take the evidence in Part 1 to suggest that algorithmic decision making will not be gender biased.
First, an algorithm that expressly includes gender as one of the variables may adversely affect female plaintiffs. Let’s suppose that the predictive algorithm is based on a simple linear prediction model in specification (5). Here, the coefficient on female is -0.28. This means that the model will provide females with a notice period that is shorter than male plaintiffs by about one week. Even though the coefficient is not statistically significant, a predictive model that uses specification (5) would still factor in the negative coefficient and provide for a shorter reasonable notice award. Note, however, if any of the specifications (2), (3), or (4) are used, female plaintiffs would receive slightly longer notice periods. One takeaway here, to potentially reduce the differential treatment, is to not include gender as a variable in the model.
Second, even if gender were simply removed from the algorithm, in practice the outcomes may still reflect gender differences. “Gender blind” algorithms will not necessarily solve the problem because the law may factor in variables that are correlated with gender. As discussed above, the law awards shorter reasonable notice periods to employees of certain jobs disproportionately held by females (e.g., clerical positions) and awards longer notice periods to employees of job disproportionately held by males (e.g., upper management). The law reflects the gender gap in society. This categorical problem of the law treating jobs differently will continue if we use the predictive algorithm. That is, the algorithm will only reinforce the bias by
45 Frank Pasquale, “A Rule of Persons, Not Machines: The Limits of Legal Automation” (2019) 87:1 Geo Wash L Rev 1.
2020] ALGORITHMS AS LEGAL DECISIONS 129 replicating the biases of, say, the managerial/clerical distinction.46 But the potential concern would be less corrosive if the male-female proportion in management positions, for example, equalizes across job types over time. The same concern is true of differences in compensation. If the algorithm awards longer reasonable notice periods to workers who earn more, then any differences in compensation across gender will be reflected in reasonable notice awards. Again, this will be less of a concern if gender differences in compensation wash away over time.
Third, the linear regression model presented here is overly simple. The model in Part 1 treats each variable as linear relationship to reasonable notice awards.47 But law is not linear. More complex predictive models perform much better at predicting out-of-sample reasonable notice awards (i.e., predicting reasonable notice awards that the algorithm has not seen before). Supervised machine learning models are more flexible than the linear prediction models. They provide additional predictive power. An algorithm that use boosted decision trees, for example, provides predictions that more accurately reflect how judges have decided previous cases.48 But, in order to generate this additional predictive power, the algorithm may find patterns in the data that cannot be captured by a simple linear model. This may uncover additional 2020 CanLIIDocs 3123 relationships that are further correlated with gender.
The takeaway, here, is that even though the data do not necessarily indicate direct evidence of gender bias in reasonable notice awards, one would need to be very careful in tailoring an algorithm to ensure that the law does not treat female employees differently to male plaintiffs. But this Part, so far, has only explored algorithms that describe what the law has been. Are there alternative types of algorithms that can reduce bias of human decision makers? Are there algorithms that can improve the content of the law? I seek to address these questions in the next section.
2.2 Algorithms that correct biases and improve the content of the law
Recent research suggests that algorithms can help reduce bias in the law.49 In bail decisions, for example, research from the United States shows that machine learning algorithms that seek to predict flight risk not only do much better than human judges, but they also mitigate racial bias inherent in the decisions in human judges.50 The
46 See e.g. MacFarlane, supra note 13. 47 One exception to this rule is tenure in specification (5) of Table 2, where a quadratic relationship between tenure and reasonable notice is specified. 48 J Ross Quinlan, “Simplifying Decision Trees” (1987) 27:1 Intl J of Man-Machine Studies 221. 49 See Cass R Sunstein, “Algorithms, Correcting Biases” (2019) 86:2 Soc Research: An Int Q 499. See also Skeem & Lowenkamp, supra note 8; Robert P Bartlett et al, “Algorithmic Accountability: A Legal and Economic Framework” (2020) online, (pdf): Berkeley School of Law
The key to these findings is that the algorithms are not replicating the decisions of human judges. Instead, the algorithms are seeking to achieve a different objective. In bail, the algorithms are seeking to predict flight risk. Human judges make two types of errors: they grant bail to high-risk defendants on bail and they deny bail to low-risk defendants.52 An algorithm that uses hundreds of thousands of real-life observations about the flight risk of defendant is able to reduce both types of error. That is, better information will lead to better predictions, which—in theory—can lead to an improvement in the law.
Could we imagine using a similar type of algorithm in reasonable notice cases? Suppose, instead of seeking to replicate how previous judges have decided what is reasonable, the algorithm seeks to achieve some other objective. In this way, the algorithm can be better tailored to achieve particular goals in the law. For example, suppose that we, as a society, decide that notice periods should accurately reflect the 2020 CanLIIDocs 3123 length of time that it takes a dismissed employee to find a similar position. Here, one might imagine an algorithm that uses data on how long it takes a given worker to find a new position. Algorithms of this type could use far more data than an algorithm based on case law. The algorithm could draw from the experiences of literally millions of Canadian workers who have been dismissed. How long did it take each worker to find a new position that was similar to the previous employment?
Returning to our example of Sam above, this type of algorithm would predict how long it usually takes 49-year-old managers to find a similar job, given the state of the economy and other relevant factors. The answer would better reflect the realities of the employment market, rather than relying on the subjective judgments of human judges to determine what is “reasonable” in a given context.
While these algorithms may offer great promise, they are, of course, not perfect. They are not a panacea. There are, for example, still problems with the “objectivity” of the data. Most problematically for this example, the length of time that it takes workers to find a similar job is a function of our law on reasonable notice. That is, workers who receive a long reasonable notice period are more likely to take longer to find a similar new job. Workers who receive shorter notice periods are more incentivized to take new jobs after a shorter period of time. That is, there is great deal of endogeneity in the algorithm. The data will reflect the current state of the law.
Himabindu Lakkaraju, Jure Leskovec, Jens Ludwig & Sendhil Mullainathan, “Human Decisions and Machine Predictions” (2018) 133:1 Q J Economics 241 [Kleinberg, “Human Decisions”]. 51 David Arnold, Will Dobbie & Crystal S Yang, “Racial Bias in Bail Decision” (2018) 133:4 Quarterly J Economics 1885; Meghan Sacks, Vincenzo A Sainato, & Alissa R Ackerman, “Sentenced to Pre-trial Detention: A Study of Bail Decision and Outcomes” (2014) 40 American J Crim Justice 661. 52 Kleinberg, “Human Decisions”, supra note 51 at 240–66.
2020] ALGORITHMS AS LEGAL DECISIONS 131
Further, there may be great difficulty in determining what a “similar” job is. This may require subjective assessments of what counts as a similar job.
Finally, the discussion here presumes that the objective of the law on reasonable notice is to provide notice periods that accurately reflect how long it takes for a worker to find a similar job after dismissal. But what if decision makers have different objectives? What if a judge believes that the law should be more employer friendly, perhaps allowing employers greater incentive to dismiss workers? An algorithm seeking to meet this objective would provide much shorter notice periods. The objective of the algorithm needs to be agreed upon in advance. To the extent that different decision makers cannot agree on objective, this may counsel in favour of greater flexibility in the law, rather than concretizing one specific goal.
Conclusion
If legal decisions in the 21st century are ever to be based on algorithms that rely on past legal decisions, we, as a society, must be overwhelmingly satisfied with the 2020 CanLIIDocs 3123 content of the existing case law. If judicial decisions reflect biases—such as gender biases—then using an algorithm based on past decisions to decide future cases would run counter to the rule of law. It goes without saying that we would not want to use an algorithm that reinforces and entrenches such bias in future decisions.
In this paper, I showed that there is no direct evidence of a gender gap in reasonable notice awards in Canada. That is, if you took two employees with the same job type, same characteristics, and same employment opportunities, there appears to be no statistical difference between a male plaintiff and a female plaintiff. Unearthing statistical evidence of gender gaps in other areas of law should be a fertile source of academic research in the near future, particularly if Canada follows other jurisdictions and starts using such algorithms to assist with legal decision making.
But just because gender gaps are not directly observable in the area of law under observation does not mean that we can simply turn to an algorithm to help decide all reasonable notice cases in Canada. Gender bias can manifest itself through other channels. Indeed, the legal test used by courts treats different job types differently. And these distinctions have, historically, reflected gender bias, Further, gender bias is only one type of bias with which we may be concerned. Important questions will remain about the degree to which the law reflects the societal values we wish employment law to reflect. And broader questions about how “human” the law should be and how much authority we, as a society, wish to delegate to an algorithm will need to be answered. These are questions that we can no longer leave as hypotheticals or to the realm of science fiction.
A STUDY ON HOW STOCK INDEX FUTURES CONTRACTS CAN BE LEGITIMIZED IN IRAN’S CAPITAL MARKET: A COMPARISON WITH SINGLE STOCK FUTURES CONTRACTS
Mohammad Soltani & Nafise Shooshinasab*
Abstract
Following the 1979 Islamic Revolution in Iran, many Islamic experts believed that capital market transactions should be forbidden because they were similar to gambling practices prohibited by Islam. They argued that gharar (uncertainty) and gaming based on chance, as the constituent elements of gambling, are an integral part of capital market transactions, and such transactions are not allowed in Islam. However, a study 2020 CanLIIDocs 3123 of the history of the Sharia board members’ views in Iran’s capital market shows that, contrary to the dominant Islamic belief, a variety of transactions—including some types of futures contracts—became prevalent in Iran’s capital market. Futures contracts are commitments of parties to buy or sell an underlying asset at a predetermined price on a specific future date. The article reviews the evolution of Islamic jurisprudence towards the legitimacy of “single stock futures contracts” (SSFs) as an example of transactions that were once considered gambling, and compare SSFs with “stock index futures contracts” (SIFs)—a transaction with a similar definition and function that is still prohibited. We argue that the differences between these contracts, including the differences in their underlying assets and settlement procedure, are not fundamental enough to justify Sharia board members’ distinctive approach to gambling and legitimization issues. We propose that SIFs should be legitimized in Iran’s capital market, as the same justifications that were used to legitimize SSFs can partly be applied to SIFs as well. Finally, the paper proposes some additional justifications for SIFs in an effort to cover the distinctive challenges that they create.
Keywords
Futures contracts, gambling, Iran’s capital market, Islamic jurisprudence, Sharia, SIFs, SSFs, transaction legitimacy
* Mohammad Soltani is an Assistant Professor at the Department of Public Law and Economic Law, Faculty of Law, Shahid Beheshti University, Tehran, Iran. Nafise Shooshinasab is a PhD student at the Common Law Section of the University of Ottawa. She did her first PhD in Private & Islamic Law at the Faculty of Law of the University of Tehran, Tehran, Iran.
2020] FUTURES CONTRACTS IN IRAN’S CAPITAL MARKET 133
1. Introduction
After the 1979 Iranian Islamic Revolution, many Islamic experts believed that capital market transactions were similar to gambling practices prohibited by Islam and should thus be banned.1 Contrary to dominant Islamic beliefs, however, a variety of capital market transactions have since become prevalent in Iran. In this paper, we analyze the evolution of Islamic scholars’ attitudes toward a particular kind of futures contract2 called “single stock futures contracts” (SSFs)3—a transaction that was once considered gambling in Islamic jurisprudence. The paper argues that SSFs have a similar structure as “stock index futures contracts” (SIFs), particularly from the perspective of Islamic rules on the elements of gambling.4 We examine the reasons why the former has been legitimized by the Sharia board members of Iran’s Securities Exchange Organization (SEO), while the latter is still considered to be gambling. Considering the major similarities, we ultimately argue that SIFs should a pari be legitimized in Iran’s capital market.
Maysir (gambling) has been banned in Islamic jurisprudence for centuries.5
Gharar (selling something uncertain that cannot be described in accurate detail) and 2020 CanLIIDocs 3123 “gaming” can lead to contracts with gambling features. To eliminate these features from Islamic capital market transactions, we propose that Islamic capital markets should take three main steps for the creation and development of new products: first, establish a Sharia board; second, recognize the necessity and function of new products; and third, justify the new products according to Islamic rules.
Islamic finance, like conventional finance, has a need for risk management against unexpected changes in prices.6 New products and service innovations in Iran’s capital market require approval from a Sharia board of the SEO (established in 2007) in order to access the market. Board members base their decisions on the legal and financial analyses provided by respective experts. There is often an absence of a uniform interpretation of Islamic law, however—especially when there is no clear
1 Shahla Keshavarznejad & Ali Saeedi, Tarikh Bazar Sarmaye Iran, (Tehran: Bourse Press, 2016) at 436 [translated by author]. 2 A futures contract is a contract between two parties to exchange assets at a determined time in the future at a price on which parties have agreed at the time of the contract. Futures contracts are categorized as derivative instruments. See Dennis W Carlton, “Futures Markets: Their Purpose, Their History, Their Growth, Their Success and Failures,” (1984) 4:3 J Futures Markets 237 at 238. 3 See ICE Futures Europe, “Single Stock Futures Contract” (2017) at 1, online (pdf): International Exchange
134 UNBLJ RD UN-B [VOL/TOME 71 ruling in the primary sources of Sharia, such as the Quran. Fortunately, although many product developments in Islamic capital markets gives rise to jurisprudential issues under Sharia, the history of capital market developments indicates that there is always the hope that new products can be legitimized. In our view, legal scholarship can contribute to legitimization efforts by introducing new ways, based on Islamic rules, to facilitate the acceptance of new instruments, such as SIFs, in Islamic capital markets.
In this paper, we employ a comparative methodology to examine the evolution of the SEO’s Sharia board members’ opinions toward the legitimacy of SSFs and SIFs in Iran.7 We supplement the comparative approach with a historical analysis of the legitimization of SSFs. In doing so, we attempt to contextualize why, in our view, SIFs in Iran’s capital markets should be similarly legitimized. It is our hope that this mixed methodological approach might serve to persuade both governing bodies and Islamic scholars alike that transactions with similar structures ought to be treated similarly.
There are a range of articles and books in several different languages, 2020 CanLIIDocs 3123 including Persian, Arabic, and English, on the Islamic challenges of futures contracts. Works by Zainordin et al,8 Khan,9 and Azmat et al10, for example, provide a fairly comprehensive account of the Islamic futures contracts. These works, however, tend to not engage in jurisprudential debates over Iran’s current capital markets, and are largely uncritical of the SEO’s Sharia board members’ perspectives. While there are also some works11 that highlight the similarities between SSFs and SIFs, there have been no efforts, as of yet, to draw upon these similarities in an attempt to legitimize SIFs. Indeed, using inductive reasoning to legitimize SIFs in Iran’s capital market has yet to be considered in legal scholarship.
Our paper proceeds as follows: first, we explain why SSFs and SIFs are structurally similar. We then examine their similarities (and differences) in detail, with particular attention to the gambling elements common in these two types of contracts.
7 Over the last decade in Iran, Islamic scholars have made every effort to use the traditional Islamic institutions to legitimize capital market transactions, but the more they tried to justify these transactions, the less successful they were in introducing new instruments with a different structure from those popular in non-Islamic countries. However, this needs to be scrutinized in a separate essay. 8 Nur Aqilah Zainordin et al, “Challenges and Opportunities of Islamic Banking and Financial Institutions in Malaysia” (2016) 10:1 South East Asia J Contemporary Bus Economics L 1. 9 M Fahim Khan, Islamic Futures and Their Markets: With Special Reference to their Role in Developing Rural Financial Market, 2nd ed (Jeddah: Islamic Research and Training Institute & Islamic Development Bank, 2000). 10 Saad Azmat et al, “The Shariah Compliance Challenge in Islamic Bond Markets” (2014) 28 Pacific-Basin Finance J 47. 11 “Understanding Stock Index Futures” (2013) online (pdf): CME Group
2020] FUTURES CONTRACTS IN IRAN’S CAPITAL MARKET 135
At the following stage and with a legal history method, we demonstrate how the criticisms regarding the gambling elements were justified by the SEO’s Sharia board members when legitimizing SSFs. We then conclude in arguing that analogous criticisms regarding the gambling elements of SIFs can be similarly justified in order to facilitate their legitimization in Iran’s capital market.
2. SSFs and SIFs as Futures Contracts
Futures contracts are a type of derivative securities that contain the following five conditions: an underlying asset, the contract size, due date, price, and the delivery or contract offset. They are called “derivatives” because the value of the contract is derived from the value of an underlying asset. Thus, there exists a mutual relationship between the underlying asset and its derivative regarding the fact that the price of the futures contracts is determined by its underlying asset.12 For example, if we selected gold as the underlying asset, the value of a futures contract would be derived from the market price of gold. However, because a futures contract is a zero-sum transaction13 14 between the parties, the net value of a futures contract is always zero. This means 2020 CanLIIDocs 3123 that whatever gain the buyer achieves is exactly equal to the loss borne by the seller and vice versa.
Historically, in the US, futures contracts were transacted on commodities. However, in the 1970s, wide fluctuation in the prices of financial assets paved the way for futures exchanges on such assets—a reason that once justified the emergence of the commodity futures markets.15 Whereas commodity futures require the delivery of a physical commodity, financial futures require the delivery of a financial instrument. The first financial futures contracts were foreign currency contracts introduced in 1972 at the Chicago Mercantile Exchange (CME). Then, interest rate futures were introduced at the Chicago Board of Trade in 1975. Finally, SIFs were introduced in
12 Johan De Beer, “Changes in the Volatility Level and Structure of Shares Post Single Stock Futures Trading” (2009) 7:2 J Corporate Ownership Control 296 at 296. 13 See Jason Alan Jankovsky, “Basics of Zero-Sum Markets” in Time Compression Trading: Exploiting Multiple Time Frames in Zero-Sum Markets (New Jersey: John Wiley & Sons Press, 2010) at 4–5:
A zero-sum market is one where no transaction results in an exchange of money between the buyer and seller. Transactions are done by either buying or selling the current price in the market. Both the buyer and the seller must be present at a traded price, and both assume the risk of price action for or against the current price in the market. Money actually changes hands when the buyer/seller liquidates the open long/short position by doing an opposite transaction at some later time. No money is ever made or lost in the market; it changes hands based on the difference between the traded prices and the individual’s personal account. In other words, say 100 accounts have a total of $100,000 at the start of trading. After trading is over and the market has changed in price, the total amount of money those 100 accounts have is still $100,000. A certain number of the accounts will have more than they started with; that amount will be exactly the amount that the losing accounts will show as a loss. 14 Jan Röman, “Futures Contracts” (2016) at 4, online (pdf): Analytical Finance
136 UNBLJ RD UN-B [VOL/TOME 71
1982 at the CME and the New York Futures Exchange (NYFE).16 SSFs were not launched in the US until 2002.17 Currently, futures contracts in US capital markets include shares, bonds, currency, commodities, and index, and the contracts follow daily settlement procedures of the clearinghouse. Conversely, the trend was rather different in Iran’s capital market: following the legitimization of the commodities’ futures contracts in 2008, Sharia board members legitimized SSFs in 2010 while SIFs have yet to be legitimized.
2-1 Similarities between SSFs and SIFs
SSFs and SIFs share many common features.18 Here, it is worth noting that SSFs in Iran’s capital market have the same definition and function as in western capital markets.19
2-1-1 Definition
2020 CanLIIDocs 3123
2-1-1-1 SSFs
SSFs or single stock futures contracts, also known as individual equity futures, are commitments to buy or sell the shares of a particular company at a specific price at a future date.20 Put slightly differently, SSFs are “futures contracts on individual stocks”.21 They are contracts for the delivery of a particular company’s stocks with a certain contract size on a specific future date.22 These contracts are transacted based on a single type of share, which is why they are called single stock futures contracts.
16 Röman, supra note 14 at 3–4. 17 Joseph KW Fung & Yiuman Tse, “Efficiency of Single Stock Futures: An Intraday Analysis” (2008) 28:6 J Futures Markets 518 at 519. 18 Similarities in the definitions and functions of SSFs and SIFs does not mean that they are completely similar in these regards. SSFs and SIFs are designed to serve different purposes, which are indicated in the Differences section below. However, since the basics of these contracts are the same, both have the potential to be legitimized from an Islamic perspective. 19 Nevertheless, the justification for the acceptance of SSFs in Iran’s Islamic capital market is different from the western ones. This means that western capital markets have simply accepted that SSFs may not end up delivering the asset to one party in “most” cases—although they have considered the possibility of the delivery—and what mostly happens in these markets is simply the settlement of the prices by the due date. Meanwhile, in Islamic SSFs, parties should “at least” have the intention to deliver the underlying asset when they enter the contract. 20 Iran, Art 1 The Directive of SSFs (2018) [translated by author]. See generally the SEO website, online:
2020] FUTURES CONTRACTS IN IRAN’S CAPITAL MARKET 137
Similar to commodity futures, SSFs entail both the rights and obligations of the parties to take delivery or to deliver the underlying assets on their expiration dates.23
2-1-1-2 SIFs
SIFs or stock index futures contracts are agreements to buy or sell a standardized value of a stock index at a specified price at a future date. These contracts contain an overall reflection of market movements, which is used to predict future price fluctuations of the index.24 Therefore, SIFs are transacted on a particular stock market index as the underlying asset. Investor profit or loss depends on the price of the index on the expiration date of the agreement, or at the time when they close out25 their position.26
In the US, several stock market indices, such as the Standard & Poor’s 500 Index (S&P 500), were developed for investors to trace movements in capital market prices.27 Therefore, through SIFs, traders transfer the expected risk of the price movements of a portfolio of shares to the futures market based on their analysis of the index’s movements. 2020 CanLIIDocs 3123
2-1-1-3 Common Features
SSFs and SIFs have some basic features in common. Both contracts are based on an underlying asset, can be transferred to a third party if one of the parties prefers to close out their position, and are settled by a due date. The contracts are transacted at a specific price and are based on the specific size of the underlying asset.
Also, in both SSF and SIF agreements, brokers require the traders to deposit as collateral a sum of money called a “margin”.28 The function of the margin is similar in both transactions. A margin is settled daily, and the increase or decrease of the market value is applied to the investor’s margin account. The buyer and seller must deposit enough money in the margin account at a brokerage firm to cover the loss at the end of the trading day on a “margin call” from a central clearinghouse, which clears
23 See Art 50 The Directive of SSFs (Iran), supra note 20. 24 Charles MS Sutcliffe, Stock Index Futures, 3rd ed (London, UK: Routledge, 2006) at 3–4. 25 In the US, for example, SIFs are not supposed to be delivered, and parties merely offset their commitments by the due date. These contracts are traded exclusively on electronic trading platforms, such as the CME Globex system in the US. See CME Group, supra note 11 at 1. 26 “Introduction to Index Futures and Options” (2008) at 4, online (pdf): Australian Securities Exchange
138 UNBLJ RD UN-B [VOL/TOME 71 and settles both types of transactions.29 Clearinghouses can guarantee all transactions by requiring the traders to deposit an amount that is large enough to cover their customers’ losses.30
2-1-2 Function
SSFs and SIFs have similar functions including risk management, high liquidity, and low transaction costs.31 Using these instruments, investors can create a wide range of potentially profitable scenarios, regardless of the market direction. SSFs and SIFs provide traders with the flexibility and opportunity to exploit the trends and variations in the market, features absent from traditionally traded contracts.32
2-1-2-1 Risk Management
Capital market investors’ inability to enjoy short positions at low cost is a major challenge leading to market inefficiency. This can be overcome by SSFs and SIFs, 2020 CanLIIDocs 3123 which both enable investors to short more easily and achieve a more efficient outcome in the stock market.33
SSFs and SIFs are derivative instruments that allow investors to cover their risk against a change in the value of their underlying assets. For example, if someone (i) owns a large number of shares, (ii) follows the price movements of the shares, and (iii) believes that the market is going to fall but they do not want to sell for some reason—including high costs or taxation—they can safeguard the value of their shares or portfolio by entering into an SSF or SIF. Doing so protects their portfolio and helps to manage risk.34
29 Chen et al, supra note 28 at 221. 30 Peter Ritchken, “Forward and Futures Markets” (1999) at 16, online (pdf): Weatherhead School of Management at Case Western Reserve University
2020] FUTURES CONTRACTS IN IRAN’S CAPITAL MARKET 139
2-1-2-2 High Liquidity
Of all the different types of futures contracts, SSFs and SIFs are the most liquid and convenient transaction types for risk management.35
Historically, for example, speculators used to purchase and then sell grain in advance, in an attempt to profit from the changes in prices. Most speculators never owned or intended to own the commodity. Their purpose was simply to make a profit by following trading strategies reflecting their predictions of the movement of commodity prices. The high liquidity of futures contracts attracted speculators to these contracts. As a group, they were prepared to buy these contracts from farmers, and to sell them to manufacturers at any time.36
The popularity of SSFs and SIFs in many capital markets is due to their high liquidity, derived from their ability to execute short positions. The liquid market in SSFs and SIFs, however, prevents certain financial market anomalies caused, at least in part, by restrictions on shorting.37 Furthermore, SSFs and SIFs are standardized contracts transacted in organized exchanges. The features of these contracts which 2020 CanLIIDocs 3123 contribute to a liquidity increase include daily settlements, margin requirements, the role of the brokerage firms and clearinghouse, and price limiting moves.
2-1-2-3 Low Transaction Costs
Transaction costs are reduced in SSFs and SIFs for three major reasons. First, SSFs and SIFs are both categorized and organized as exchange-traded contracts. Second, SIFs specifically allow for more efficient transactions because the underlying asset is a diverse basket of securities,38 eliminating the need for transacting on each separate stock. Third, the physical transfer of commodities involves at each stage increased human activities, such as packaging, storage, and transport, which naturally increases transaction costs.39 SSFs and SIFs, however, do not incur such costs as they are contracts in which the investors have the freedom to offset their commitments by the due date without any delivery.
35 Misbahul Islam & Jayanta Chakraborti, “Futures and Forward Contract as a Route of Hedging the Risk” (2015) 5:4 Risk Governance & Control: Financial Markets & Institutions 68 at 69. 36 Ibrahim Uzaimah, Commodity Futures Contract: An Analysis in Islamic Commercial Law (PhD Dissertation, University of Wales, 2000) at 1 [unpublished]. 37 See Frank Partnoy, “Some Policy Implications of Single Stock Futures” (2001) U San Diego L & Econ Research Paper No 10 (“[t]he trading of [SSFs] might ameliorate this anomaly, both by enabling investors to avoid regulations applicable to short sales and by creating a virtually unlimited supply of short futures positions for investors” at 5). 38 Cooper & Mello, supra note 4 at 3. 39 Muhammad Asif Ehsan, “Futures Contracts in Islamic Finance: A Jurisprudential Analysis” (2014) 10:1 J Islamic Economics Banking Finance 156 at 167.
140 UNBLJ RD UN-B [VOL/TOME 71
2-2 Differences between SSFs and SIFs
The differences between SSFs and SIFs lie in the nature of the underlying assets, the settlement processes, and their necessity to capital markets.
2-2-1 Underlying Assets
Both SSFs and SIFs derive their value from that of their underlying assets. However, the assets on which they are transacted are different.
SSFs are issued based on the stocks of a particular company. Therefore, the assets on which SSFs are transacted are shares, which are deliverable if one of the parties intends to deliver or the other party requests this. Thus, the parties to SSF transactions normally deliver the consideration of the agreement, unless they agree to offset by the due date.40
In contrast, SIFs are transacted on an index as the underlying asset. An index, 2020 CanLIIDocs 3123 by its very nature, is comprised of a portfolio of stocks within the stock market index. In other words, it derives its value from the prices of the securities that constitute the index, and is created to represent the predictions and beliefs of the market as a whole, or of a particular sector of the economy.
By comparison, when a buyer invests in a particular company’s stock in SSFs, he or she owns (or can own) the company’s stock, forming the basis of the SSF transactions. In other words, the seller of an SSF is committed to delivering the underlying stock on the expiration date, even if the parties decide not to deliver the underlying asset at the due date, and merely clear their commitments. However, in the case of a SIF, the buyer intends to trace the stock price movements of a large number of companies. For example, one of the most popular SIFs, the S&P 500 Index, includes the movements of the shares of 500 companies. This means that the party who made a wrong prediction of the index movements would pay the price difference to the party who made the right guesses. Because of this, the clearinghouse interferes only to offset the commitments of the parties by the due date, and delivery never happens.41
Therefore, the major difference between SSFs and SIFs is in terms of the underlying assets. This difference is one of the major reasons why Iran’s capital market began to expand its futures market with SSFs rather than SIFs: the subject matter of the contract in SIFs in its western context (index) is fully abstract, while that of SSFs
40 See Sharia Board of SEO, “Detailed Discussions of Sharia Board Regarding SSFs” (5 September 2007) at 1, 9, online: Research, Development and Islamic Studies
2020] FUTURES CONTRACTS IN IRAN’S CAPITAL MARKET 141
(shares) can potentially be delivered electronically.42
2-2-2 Settlement Process
The settlement process is another key feature that distinguishes SSFs from SIFs. SSFs can be settled either by delivery of the underlying commodities or by offsetting transactions at the expiration date.43 SSFs, therefore, have two types of settlements: physical and cash settlement. The physical settlement of SSFs means that upon expiration of the contract, the buyer will receive the underlying shares, and the seller will be required to deliver.44 However, the parties can also agree to a cash settlement. In this model, upon the expiration of an SSF, profits and losses are credited or debited to the account of the relevant party in an amount equal to the difference between the contract price and the final settlement price.45 On the other hand, the western conception of SIFs involves only one type of settlement, called a “cash settlement”. In the cash settlement process, after establishing a long or short position, market participants are subject to a normal “market-to-market” procedure. Since SIFs are not traded on actual stocks but on price movements, the parties need to follow the value 2020 CanLIIDocs 3123 of the index and pay or obtain cash for the price difference in order to close the deal by the due date.46 It is essentially impossible for individual investors in SIFs to deliver the underlying asset because doing so would require handing over a portfolio of shares, such as the shares of 500 companies. Because the delivery cost of such a portfolio is prohibitively high, the subject matter of SIF contracts is never delivered in practice, and the contracts are instead settled in cash.
In summary, the difference in the settlement process between SSFs and SIFs is affected by the issue of “delivery” contained in these contracts, meaning that while SSFs have both types of physical and cash settlements based on what the contract
42 Even some members of the Sharia Board of SEO in Iran once believed that electronic assets could not be considered underlying assets because they are not tangible so they cannot be delivered physically. However, the majority of members decided that tangibility differs from physical delivery of assets, and electronic assets also have the potential to be delivered. See Sharia Board of SEO, Detailed Discussions of Sharia Board Regarding SSFs, supra note 40 at 12. 43 Travis Jones & Robert Brooks “An Analysis of Single-Stock Futures Trading in the U.S.” (2005) 14 Financial Services Rev 85 at 89. 44 David G Downey, “Single Stock Futures: An Alternative to Securities Lending” (2018) at 2–3, online (pdf): US Securities And Exchange Commission
Curiously, trading on organized exchanges in the equivalent of a futures contract for the narrowest of stock indexes—namely, for a single stock—already exists. Futures positions now can be taken indirectly in many stocks if one takes a position in both a put option and a call option. Therefore, why not allow delivery of certificates on narrowly based stock index futures contracts? The shorts would not elect to do so if they deemed it too costly.”
142 UNBLJ RD UN-B [VOL/TOME 71 parties choose, SIFs have cash settlements only. The reason for such an exclusive method of settlement of SIFs is that in western capital markets, which pioneered SIFs, the physical delivery of the underlying asset is not expected, because it is not considered a deliverable asset.
2-2-3 Necessities
2-2-3-1 SSFs
SSFs are designed mostly for individual investors, who tend to own a limited type of shares. In early 1995, SSFs were introduced in Hong Kong to respond to investors’ needs to manage the risk of price fluctuations of their financial instruments.47 As a result, SSFs help traders keep ownership of the underlying asset, while being able to prevent a fall in the value of their financial assets.
There are other reasons for the introduction of SSFs as well. For example, an 2020 CanLIIDocs 3123 investment bank may acquire shares as an underwriter, which it is banned from selling because of its commitment to the issuer not to sell the shares in a certain time. Additionally, the holder of a sell position in a stock option contract who does not have the right to sell may use SSFs due to the stocks having increased in value. In these cases, the investor may wish to hedge, rather than immediately sell their shares, as protection against price falls.48 Short sellers can also benefit from cutting costs associated with the stock loan process by investing in SSFs, since the margin requirement is as low as 20%.49 Finally, in addition to SSFs being an efficient mechanism for opening short positions, they also provide investors with the opportunity to postpone the sale of a particular company’s shares, securing their rights to receive a dividend, as well as vote.
2-2-3-2 SIFs
SIFs are designed mainly for mutual funds that tend to own a wide variety of shares, such as a portfolio of 500 shares, or for investors whose main activity is to follow the index movements in order to speculate.
Many investors prefer SIFs compared to other contracts such as SSFs for a range of reasons. They offer a historical comparison of the returns on money invested in a wide range of stocks, as against investment in some other types of assets such as
47 Fung & Tse, supra note 17 at 519. 48 Chris Brooks, Ryan J Davies & Sang Soo Kim, “Cross Hedging with Single Stock Futures” (2007) 74:4 J Insurance Risk Management 473 at 474. 49 Fung & Tse, supra note 17 at 520.
2020] FUTURES CONTRACTS IN IRAN’S CAPITAL MARKET 143 government debt or gold. They can also be used as a simple standard by which to monitor the performance of investment fund managers. Since the share price is often treated as a measure of the market’s expectations of the cash flows from the company concerned, investors can benefit from the market index, as an indicator of national economic performance, to simply track changes in financial markets.50 All these needs can be responded to through SIFs. Finally, as Stephen Figlewski notes, “[b]oth market timing (predicting the movement of the market as a whole and adjusting the portfolio’s beta accordingly) and stock selection (trying to increase portfolio returns by buying stocks that are undervalued relative to the market) are made easier by the use of index futures.”51
3. Gambling Elements vs. Justifications: Islamic Perspective
Gambling and other terms such as “gaming” and “games of chance” are derived from the Anglo-Saxon “gamen” and “gamon,” which mean sport or play.52.The general use of the word “gamble” regards any activity that involves risk and uncertainty:
“uncertainty is an imaginary situation where nothing is known about anything and 2020 CanLIIDocs 3123 where there are no measurable variables, nor even an idea as to what factors are relevant to a problem.”53 Hence, gambling means to play a game of chance for money, or to take an extremely uncertain position in order to gain some advantage. Gambling has nothing to do with the exchange or production of real goods or services. The gambler merely makes a payment in the hope of gaining a larger sum of money.54 Thus, “uncertainty” and “gaming” are the core elements of gambling.
Although the prohibition against gambling or maysir is very clear in Islam, it is not defined in any of its original sources.55 However, it is very clear that there is an essential element in gambling, namely uncertainty,56 or gharar as it is termed in
50 Sutcliffe, supra note 24 at 3. 51 Stephen Figlewski, “Hedging with Stock Index Futures: Theory and Application in a New Market” (1982) 5:2 J Futures Markets 183 at 183. 52 Hussin Bin Salamon, Mansoureh Ebrahimi & Kamaruzaman Yusoff, “Speculation, The Islamic Perspective: A Study on Al-Maisir (Gambling)” (2015) 6:1 J Leg Soc Sciences 371 at 372. 53 Henry B Arthur, “Economic Risk, Uncertainty and the Futures Market” (1966) at 214, online (pdf): Mimir Publishers
144 UNBLJ RD UN-B [VOL/TOME 71
Islamic law.57 Gharar is not defined precisely by Sharia either. However, it is agreed that whenever there is the risk a contract will have unknown consequences, it is likely that such a transaction could lead to an illegal unpredictable loss of individuals’ earnings,58 and thus, it is gharary (uncertain). Therefore, any contract that includes uncertainty in either the possibility of delivery of the subject matter, the quality or quantity of the subject matter, price, payment time, or existence of the subject matter, would be considered gharary.59
To sum up, from an Islamic point of view, “uncertainty or gharar” as well as “gaming based on chance” in an activity, constitute the core elements of gambling or maysir. It should be noted that these two elements together60 can constitute “gambling,” however one of them separately does not denote “gambling”; they must be found together.
3-1 Comparison between SSFs and SIFs in terms of the Gambling Elements
The challenges from an Islamic perspective with respect to SSFs and SIFs are “gharar 2020 CanLIIDocs 3123 or uncertainty in delivery” and “game of chance”, which have been raised by Sharia board members when these contracts have been proposed for introduction into Iran’s capital market.
3-1-1 Gharar (Uncertainty in Delivery)
Delivery is an important factor in sales contracts from an Islamic perspective. It is the possibility of eventual delivery that determines the prices of financial derivatives.61 However, gharary sales are linked to the ability to deliver the goods subject to sale.62
Historically, the key reason for including any asset in the futures market was the possibility of fluctuations in its prices—in other words, uncertainty.63 Therefore, the reasoning underlying the creation of these markets was to promote the exchange of goods by the redistribution of the risks created by the uncertainty of prices. When futures trading was popular only for commodity markets, and the delivery of goods was binding, this trade served the purpose of distribution and management of risk for
57 Abdullah, supra note 55 at 741. 58 Mohamed Ali El-Gari, “Towards an Islamic Stock Market” (1990) 1:1 Islamic Economic Studies 1 at 6. 59 Mohammad Hassan Najafi, Javahir-al-Kalaam (Qom: al-Milani Foundation Press, 2011) at 387. 60 Contracts that have only one of these elements are void from an Islamic perspective. However, such contracts are not considered “gambling” if one of the elements does not exist. 61 Ehsan, supra note 39 at 166. 62 El-Gari, supra note 58 at 6. 63 Carlton, supra note 2 at 242.
2020] FUTURES CONTRACTS IN IRAN’S CAPITAL MARKET 145 the sake of enhancing production and, thus, exchange.64 However, when delivery was no longer a binding obligation, the primary concern became to manage and manipulate the risk. Over time, risk management became the objective rather than the means.
Gharar is a standard that helps clarify whether the parties’ transaction is based on a certain subject matter, and whether they can reasonably anticipate what they give and what they receive in a certain contract. In this regard, Sharia scholars discuss the legitimacy of each western contract before it is introduced to the market, to make sure that they prevent almost every possible dispute about the subject matter between parties.65 In other words, they try to make a distinction between reasonable and unreasonable gharar, where the former involves “economic benefit” while the latter does not.
Therefore, the concept of gharar focuses on the fact that both parties to the agreement should clearly know what they will give and receive in a contract. The issues surrounding gharar can be studied in the context of the high-risk activities that caused the 2008 financial crisis in the US—an environment in which risk and 66 uncertainty are the norm. As Covitz et al note, “[t]he US asset-backed commercial 2020 CanLIIDocs 3123 paper (ABCP) market erupted in late summer of 2007 and played a pivotal role in the global financial crisis.”67 In such a market, the issuance of papers without any corresponding underlying assets led to the paper crisis.68 In other words, the proliferation of derivative transactions, such as credit default swaps (CDSs), without any underlying assets of the same value, led to gharary transactions. Moreover, there were third party purchasers who gambled on CDSs because they believed that the borrower would not default on the loan payment. These gambling transactions formed a new market for CDS papers without any underlying assets. In these gharary transactions, it was not clear what the sellers paid and what the purchasers received in consideration.
When futures contracts were introduced to Iran’s capital market for the first time in 2007, Sharia board members studied the issue of uncertainty and believed that futures contracts should be forbidden due to uncertainty in delivery, using an Islamic
64 Khan, supra note 9 at 26-27. 65 Ansari, supra note 54 at 186. 66 Stephen C Nelson and Peter J Katzenstein, “Uncertainty, Risk, and the Financial Crisis of 2008” (2014) 68:2 Intl Organization J 361 at 362. 67 Daniel M Covitz, Nellie Liang & Gustavo A Suarez, “The Evolution of a Financial Crisis: Panic in the Asset-Backed Commercial Paper Market” (2009) at 2, online (pdf): Divisions of Research & Statistics and Monetary Affairs Federal Reserve Board
146 UNBLJ RD UN-B [VOL/TOME 71 rule that bans bay-al-kali-bi-kali69 (sale of debt) to justify the exclusion.70 Under this rule, both final payment and item delivery cannot be deferred in a sale contract. Otherwise, it leads to gharar because there is a high risk that one of the parties will fail to fulfill their commitment in the future, or simply avoid it.
In 2007, Sharia board members argued that SSFs were just a tool of speculation and represented a formalistic agreement in which parties do not actually intend to deliver the subject matter of the contract,71 using statistics to bolster their claim. They argued that it had been estimated that approximately 1% of all transactions are actually settled by a customer making or taking delivery of physical commodities,72 strengthening the view that SSFs are merely a means of speculation. Sharia board members held that speculation is unacceptable because it involves high uncertainty, which is one of the main elements of gambling that is strictly forbidden by the Quran.73 They considered SSFs as a high-risk activity based on chance rather than calculation. Therefore, Sharia board members argued the problem of “uncertainty” or gharar in SSFs from two perspectives: first, the final payment and item delivery in a sales contract cannot be deferred until the due date (bay-al-kali-bi-kali), as this increases the risk of parties’ failure to meet their commitments; and second, parties may not 2020 CanLIIDocs 3123 have any real intention to deliver the subject matter at all—rather, their real intention is to arbitrate.
In SIFs, the index, as the underlying asset, is not considered a property and has no value per se from an Islamic law perspective.74 In 2016, Sharia board members argued that the subject matter in SIFs is the movements of the index, and investors try to speculate and benefit from the price differences without performing any real transactions on any particular asset. In other words, Sharia board members believe that SIFs incorporated absolute speculation on prices, and parties merely benefited from the differences in prices, without having any serious intention of exchanging any goods or products. They argued that investors in SIFs intended to make an income simply by making good guesses with no intention to receive, deliver, or produce goods.75
69 Bay-al-kali-bi-kali is the deferment of both final payment and item delivery. See generally Mohammad Hashim Kamali, “Commodity Futures: An Islamic Legal Analysis” (2007) 49:3 Thunderbird Intl Bus Rev 309 at 313, 317. 70 Sharia Board of SEO, “Detailed Discussions of Sharia Board Regarding Futures Contracts” (28 November 2007) at 1, 12. 71 See Ibid. 72 Sharia Board of SEO, “Detailed Discussions of Sharia Board Regarding SSFs” (23 July 2008) at 26. 73 Abdullah M Al-Awadhi & Ahmad Bash, “Islamic Asset Pricing and Speculation Restrictions” (2018) 6:1 J Islamic Banking Finance 45 at 45. 74 Sharia Board of SEO, Detailed Discussions of Sharia Board Regarding SSFs, supra note 40 at 12. 75 Khan, supra note 9 at 46.
2020] FUTURES CONTRACTS IN IRAN’S CAPITAL MARKET 147
Ultimately, Sharia board members concluded that gharar clearly exists in SIFs as sales contracts, because the underlying asset of the contract is undeliverable. They argued that even if the parties intended to deliver the subject matter, it would not be possible to do so, because the movement of the share price or index is not deliverable.76 This means that the parties to a SIF are unable to fulfill their obligation to deliver the underlying asset on the expiration date, simply because the subject matter cannot be possessed.77
In sum, the Sharia board members considered the problem of “uncertainty in delivery” in SIFs from two perspectives: SIFs are a means of speculation, and the subject matter is undeliverable.78
3-1-2 Game of Chance
Another element of “gambling” is its characterization as a “game of chance”. The term “game of chance” refers to an activity where the participant pursues monetary gain without using any skills and information. It is, therefore, appropriate to use the term 2020 CanLIIDocs 3123 “gambling” for games of chance. Thus, if one uses the term “game of chance” to describe transactions on the stock market, it implies that the trader in the transaction did not and could not have any specific skills or use any special information.79
Another feature of the “game of chance” is the lack of mutual benefit, meaning that in such activities, one of the parties is always the loser and the other is always the winner.80 Mutual benefit is any direct or indirect economic benefits, directly or indirectly, that both parties of the contract may enjoy when entering a contract, ranging from monetary profits and loss avoidance, to the mental comfort derived from the risk management that a contract offers. On the other hand, in a “game of chance” both parties are aware of the fact that one of them will ultimately be the loser. In legitimate beneficial contracts, there is the possibility that both parties will benefit, and both parties expect some gain when they enter the contract.81
76 Sharia Board of SEO, Detailed Discussions of Sharia Board Regarding SSFs, supra note 40 at 13. 77 Ehab MM Injadat, “Futures and Forwards Contracts from Perspective of Islamic Law” (2014) 1:2 J Economics Political Economy 241 at 243. 78 The third one is also similar to those two other reasons at heart, which will be explained below in the next section. 79 Reuven Brenner, Gambling and Speculation: A Theory, A History and A Future of Some Human Decisions (New York: Cambridge University Press, 1990) at 90. 80 Abu-al-Qasem Mousavi Khoei, Mesbah-al-Feqahi, vol 1 (Qom: Khoei’s Works Publication, 1996) at 377. 81 T Henry Dewey, Legislation Against Speculation and Gambling in the Forms of Trade: Including Futures, Options, and Short Sales (New York: Baker, Voorhis & Co Press, 1905) at 5.
148 UNBLJ RD UN-B [VOL/TOME 71
However, many Islamic scholars argue that SSFs and SIFs do not serve a mutually useful economic purpose, but rather are a cover for gambling activities.82 The purpose of Islamic financial rules is to ensure that all parties involved in a transaction are protected against losing money in exchange contracts that are risky and uncertain. Therefore, Sharia board members have argued that, in SSFs and SIFs, the contracting parties intend to benefit from the price difference.83 Such contracts are based on chance, contain no intention to deliver, and have one party who loses their money in favour of the winner. Thus, in Sharia board members’ view, they should both be considered gambling.
3-2 Why Allow SSFs While Prohibiting SIFs?
In 2007, out of economic necessity, Iran’s SEO established a Sharia board to decide the legitimacy of capital market transactions.84 Sharia board members employed various Islamic institutions to justify the legitimacy of different transactions. They did this in order to help Iran’s capital market keep pace with modern capital market developments. Sharia board members argued that the diversity of transactions had 2020 CanLIIDocs 3123 become a necessity to attract more investors.85 Thus, they made every effort to expand Iran’s capital market and increase the diversity of contracts.
In essence, the futures market is an institution to help promote the exchange of goods. Since this institution can help improve efficiency in production and trade by redistributing risks, the nature of this market cannot be declared as intrinsically against Sharia. Thus, in 2008, Sharia board members made every effort to legitimize futures transactions as a whole, and SSFs in particular. After many discussions regarding the potential illegitimacy of SSFs, The Directive of SSFs in Iran was enacted in 2010.86 Today, SSFs are recognized—and are quite popular—in Iran’s capital market. A considerable number of SSFs transactions in Iran’s capital market demonstrate the radical shift in beliefs in this regard.
82 El-Gari, supra note 58 at 5. 83 Sharia Board of SEO, Detailed Discussions of Sharia Board Regarding SSFs, supra note 40 at 1. 84 Iran, Art 1-10, The Directive of Establishment and Duties of Sharia Board (2007). See generally, online:
2020] FUTURES CONTRACTS IN IRAN’S CAPITAL MARKET 149
In the following section, we address the two gambling elements mentioned in the previous section in order to explain how Sharia board members tried to justify elements of SSFs to legitimize these contracts. Also, using the comparisons made between SSFs and SIFs in the previous sections, we discuss how these justifications can likewise be applied to SIFs, in part. We discuss the response to “gharar” or “uncertainty in delivery” in the following sub-section, “Interpretation,” and the response to “game of chance” under the subsequent sub-section, “Hedging and Speculation vs. Gaming”.
3-2-1 Interpretation
When SSFs became a necessity in Iran’s capital market, Sharia board members employed “Interpretation,”87 as an Islamic rule, to view the definition of these Islamic institutions from a different perspective.
In 2007, when Sharia board members decided to discuss the legitimacy of futures contracts as a whole and SSFs in particular, they started with the problem of 2020 CanLIIDocs 3123 bay-al-kali-bi-kali (sale of debt), which leads to gharar. They argued that the underlying concept of futures trading is simply to enable commodity exchanges. From an Islamic perspective, commodity exchange always entails some risk elements, and the risk element is particularly significant if the commodity to be exchanged requires time for its production or supply. Their interpretation was that SSFs are promises to sell, rather than sales contracts. In contrast to a sales contract, in which either the payment or the delivery should occur on the contracted date, in a promise to sell, both the payment and the delivery can be postponed until the due date. They concluded that if SSFs are promises to sell and not sales contracts, deferment of both considerations do not expose the contract to gharar, as it is only the parties’ promises to buy and sell and not a concluded sale (bay-al-kali-bi-kali).88
Regarding the second element of gharar, where SSFs are a means of speculation rather than a contract in which the parties have a real intention to deliver the subject matter, Sharia board members resorted to “Interpretation” again. They followed the opinions of some Islamic scholars who offered a narrower definition of gambling.89 The latter group argued that speculation has wrongly been considered
87 But see Usman Hayat & Adeel Malik, Literature Review: Islamic Finance: Ethics, Concepts, Practice (Charlottesville: CFA Institute Research Foundation, 2014) (“[a] frequent criticism is that Islamic finance, at least the manner in which it is currently practiced, is a legalistic phenomenon. The “Islamic” in Islamic finance is “Islamic” in legal form only and “conventional” in economic substance” at 68). 88 Sharia Board of SEO, “Detailed Discussions of Sharia Board Regarding SSFs” (10 October 2007) at 10. 89 Ahmad Abdel Fattah El-Ashkar, “Towards an Islamic Stock Exchange in a Transitional Stage” (1995) 3 Islamic Economic Studies 79 at 82.
150 UNBLJ RD UN-B [VOL/TOME 71 gambling, and is different90 because speculation may involve a great deal of computation, and chance can hardly influence it.91 From some Islamic scholars’ opinions, an SSF agreement is a legitimate type of speculation because it is based on information, analyses, and computations, rather than mere chance.92 They argue that one should make a distinction between the two types of speculations: one is a speculation activity that is not related to any real activity and is meant to be monetary, non-productive, based on chance, and beneficial to only one of the parties (the winner); and the other involves some real activities, which contribute to partly shifting the risks from the traders who believe that they cannot afford to bear all the risk to those who believe that they can afford to bear it.93 Sharia board members concluded that SSFs are contracts of the second type of speculation and are therefore not gharary.94
The same reasoning can be applied in response to the alleged illegitimacy of SIFs. It can be argued that although speculation always involves an attempt to predict the future outcome of an event, the process may or may not be backed by the analysis and interpretation of relevant information. Sharia board members decided that the former case is consistent with Islamic rationality,95 as almost all legitimate business decisions involve some degree of speculation. Therefore, they believed that when 2020 CanLIIDocs 3123 speculation is based on information, it is not only permissible, but desirable. It is only the absence of relevant information, or the conditions of excessive uncertainty, that make speculation resemble gambling. As a result, SIFs, as transactions in which information and precise analyses play an important role,96 can be approved from this perspective.
Another gharary feature of SIFs is the impossibility of delivery. However, SSFs and SIFs are not comparable in this regard. While in SSFs there is the possibility of delivery of the subject matter, this is not possible in SIFs. Accordingly, this paper offers three solutions to this problem—two of which are different from the justifications presented for SSFs97—so that SIFs can be allowed in Iran’s Islamic
90 But see Rafic Yunus Al-Masri, “Speculation between Proponents and Opponents” (2007) 20:1 King Abdulaziz U Islamic Economics Institute 43 (it seems that “speculation is a form of gambling, whereby a minority wins what the majority of investors lose” at 43). 91 Ibid at 83. 92 Al-sheikh Javad Tabrizi, Ershad-al-Taleb, vol 1 (Qom: Mehr Publication, 1979) at 218. 93 Khan, supra note 9 at 46. 94 See Sharia Board of SEO, Detailed Discussions of Sharia Board Regarding Futures Contracts, supra note 70 at 14. 95 Sharia Board of SEO, Detailed Discussions of Sharia Board Regarding SSFs, supra note 40 at 12. 96 See generally Hans R Stoll & Robert E Whaley, “The Dynamics of Stock Index and Stock Index Futures Returns” (2009) 25:4 J Financial Qualitative Analysis 441. 97 The reason why the first two solutions are studied in this part is that these solutions need a “customization” of the SIFs so that they can work according to the rules of an Islamic capital market. We believe that “customization” requires some “Interpretations” of the subject matter, and that is why these two solutions are discussed under the title of “Interpretation”.
2020] FUTURES CONTRACTS IN IRAN’S CAPITAL MARKET 151 capital market. The third solution is one which was previously discussed with respect to the legitimacy of SSFs.
The first way to address the Sharia board members’ concern on the issue of the “impossibility of the delivery of the subject matter” in SIFs is to consider the nature of SIFs more precisely. For example, the subject matter of a SIF on the S&P 500 is not actually the index itself, but the shares of 500 companies. Therefore, although the number is large, it is still limited. As a result, it is not impossible to deliver the subject matter in SIFs if the buyer asks for delivery. Such a delivery will be more feasible, with lower transaction costs, if this role is played by a mutual fund that owns the portfolio of shares and represents the investors in SIFs.
A second approach would be a smaller contract size, such as a portfolio made of 50 stocks, which could reasonably be delivered.98 This could create popularity for SIFs among individual investors in conjunction with mutual funds and, at the same time, make delivery possible. Although SIFs, in their original form in western countries, contain no delivery, changing this contract in Iran’s Islamic capital market to one in which delivery is expected would contribute to the legitimacy of SIFs under 2020 CanLIIDocs 3123 Islamic law. This change would also allow investors with less capital to access SIFs.99 Such investors could benefit from the advantages of SIFs in order to follow price movements, and at the same time, deliver the subject matter in cases where a buyer wishes to do so.
Unlike the above approaches, the third solution is restricted to situations where the parties do not own the portfolio but intend to manage their future risk and profit from price differences. In other words, this solution could help introduce SIFs in their western context to an Islamic capital market. This third solution offers a different interpretation of the nature of SIFs compared to the two previous solutions.
With respect to the problem of gharar in SIFs, due to the “impossibility of delivery” or “lack of parties’ intention to deliver”, this solution suggests that SIFs should not be analyzed based on the regulations governing sales contracts, where delivery plays an important role. Rather, Sharia board members should consider categorizing them as a type of contract under Article 10 of Iranian Civil Law,100 in which parties agree to cover the other party’s risk in cases of fluctuations in prices; a contract in which delivery is not a condition at all. In other words, SIFs are not a
98 See Art 1-38 The Directive of SSFs, supra note 20. Iran’s capital market approved such transactions in 2017. However, the directive does not mention that the transaction on a basket of shares has something to do with SIFs and considered it as regular stock transactions that are based on a portfolio of shares. 99 But see Jędrzej Białkowski & Jacek Jakubowski, “Determinants of Trading Activity on Single Stock Futures Market” (2012) 19:3 J Derivatives 29 at 33: the application of large-size contracts to hedging or speculation can result in compromising a degree of precision in matching positions. On the other hand, reducing the contract size increases a trading cost, as both brokerage commissions and exchange fees are mostly quoted per contract independently of a size. 100 It provides that: “[p]rivate contracts are valid regarding the parties of the contract, unless these contracts are explicitly against the law” [translated by author].
152 UNBLJ RD UN-B [VOL/TOME 71 promise to sell, and are not supposed to be a sales contract either. Rather, they are a contract in which one party pays a sum of money to the other party by the due date in exchange for achieving security for probable future purchases of stocks.
Such contracts are not unprecedented in Iranian Islamic law. For example, insurance agreements have a rather similar role in Iran, where one party pays an amount of money to the other party so that the latter covers the probable risks of the former and provides the former with economic security. In insurance agreements, the indemnity is the effect of the insurance contract and not its consideration.101 Islamic scholars have held that an insurance agreement is not a sales contract, but rather an agreement in which the insured party achieves security in consideration for the money paid to the insurer. They have argued that it is reasonable—and not gharary—if someone, based on some calculations and information, undergoes a smaller loss to cover a probable bigger loss. They also explained the issue of gharar by holding that parties of the insurance agreement are aware of the type of damages that should be compensated by the insurer, the maximum amount to be paid by the insurer, and the amount that the insured party should pay to the insurer. Therefore, parties can anticipate the consequences of the insurance agreement to a certain degree. Finally, 2020 CanLIIDocs 3123 Islamic scholars have found that a reasonable degree of certainty, based on the common practices of the general public, is enough for a contract not to be gharary, and such a degree exists in insurance contracts.102
3-2-2 Hedging and Speculation vs. Game of Chance
Sharia board members responded to another gambling element in SSFs, the “game of chance”, by making a distinction between “hedging and speculation” and “game of chance”. This justification can be similarly applied to SIFs: if Sharia board members justified SSFs because they are not a “game of chance”, but rather a tool for hedging or a legitimate speculative activity, they should accept the same reasoning for SIFs.
Stock prices are highly volatile, with the owners of stocks bearing a substantial risk. Thus, hedging as a function of SSFs and SIFs helps investors reduce their risk.103 As Ghosh notes, “[u]ntil 1982, market participants could not control the market risk of their portfolios. The introduction of [SIFs] offer[ed] them an opportunity to manage the market risk of their portfolios without changing the portfolio composition. The objective of hedging is to minimize the risk of the portfolio for a given level of return.”104 Therefore, hedging uses the potential of futures markets
101 See generally Samir Mankabady, “Insurance and Islamic Law: The Islamic Insurance Company” (1989) 4:3 Arab Law Quarterly 199 at 200. 102 Asadollah Lotfi, “Gharar in Insurance Agreements” (2015) 15:2 Islamic L J 101 at 118 [translated by author]. 103 Figlewski, supra note 51 at 183. 104 Asim Ghosh, “Hedging with Stock Index Futures: Estimation and Forecasting with Error Correction Model” (1993) 13:7 J Futures Markets 743 at 743.
2020] FUTURES CONTRACTS IN IRAN’S CAPITAL MARKET 153 to avoid risk. Figlewski gives as an example that “…by selling futures on the [S&P 500 Index], an investor can hedge against systematic risk by locking in a known return on the market component of his portfolio without selling any of his stocks.”105 Therefore, “[p]ositions are hedged by taking an equal but opposite position in the futures market.”106
With respect to SSFs, Sharia board members have argued that the owner of the shares could reduce their risk through hedging, as hedging makes sense when the seller owns the shares or where there is the possibility they own them.107 Therefore, in this case, there is no “game of chance” because the seller’s real intention is to manage the risk of a fall in their real property prices. In other words, concerns regarding the illegitimacy of speculation do not make sense here. The same reasoning can be applied to SIFs, where mutual funds, which are the most common traders of SIFs in western capital markets, or a retail investor owns the portfolio of shares. Consider a situation where a mutual fund owns a portfolio of 500 shares, the value of which is $100. Subsequently, the seller sells their portfolio for a price of $110. If, by the due date, the value of the portfolio is $120, both parties will benefit. The seller would sell the portfolio for $120, gaining $10 as their own profit, and the buyer would also gain $10 2020 CanLIIDocs 3123 in profit. This can be considered a legitimate beneficial activity, because of the possibility that both parties will benefit at the end of the contract date. In this situation, SIFs act as a hedging tool and not as a “game of chance”, because “mutual benefit” exists.
In cases where the seller does not own the property, or the underlying asset is not regarded as a property at all,108 investors may choose to “speculate”. As Sharia board members determined regarding SSFs where the seller does not own the subject matter, to control or reduce the risk of an asset or a portfolio of shares, the speculator must analyze the information and determine the optimal speculation ratio.109 This ratio can differ significantly depending on the estimation technique used.110 These techniques are so sophisticated that some argue that with the emergence of each new product in the capital market “questions such as whether investors understand the risks of the novel products they are investing in, whether the products are too complicated for retail investors, or whether those selling them or the regulators themselves
105 Figlewski, supra note 51 at 183. 106 Joost ME Pennings & Matthew TG Meulenberg, “Hedging Efficiency: A Futures Exchange Management Approach” (1997) 17:5 J Futures Markets 599 at 600. 107 See Sharia Board of SEO, Detailed Discussions of Sharia Board Regarding SSFs, supra note 40 at 15. However, Sharia board members used “hedging” and “speculation” terms interchangeably and did not pay attention to the subtle difference between these two terms. 108 This is regardless of the fact that it can be argued that an Index can be considered property according to Islamic rules—but this issue needs a separate essay. 109 Sharia Board of SEO, Detailed Discussions of Sharia Board Regarding SSFs, supra note 40 at 15. 110 Ghosh, supra note 104 at 743.
154 UNBLJ RD UN-B [VOL/TOME 71 understand them” are hard to answer.111 Clearly, speculation, under this analysis, is different from a “game of chance”, where information and skills play no role. SIFs not only lack a “gaming” element, but also involve the possibility of mutual benefit. Consider a situation where the index price is $100. Subsequently, the seller sells their position for a price of $110. If by the due date, the value of the index is $120, both parties will benefit. The seller has reduced his risk for probable future purchases, meaning that the seller’s loss would be $10 instead of $20, which can be considered as a gain of $10 for them, and the buyer makes a profit of $10.
4. Conclusion
Using a comparative methodology, this paper discussed the similarities and differences between SSFs and SIFs: they are similar in terms of their definition and function, while different in their subject matter of the contracts, settlement procedures, and their necessities to capital markets. While the subject matter of SSFs are the shares of one particular company, the subject matter of SIFs (in their Western conception) is the price movements of the index. Further, while SSF contracts can be settled by 2020 CanLIIDocs 3123 delivery of the subject matter, delivery of (Western) SIFs never happens. Finally, while SSFs are, among other things, a tool for selling short with less cost, SIFs act largely as an indicator of national economic performance. Therefore, SSFs and SIFs are fundamentally similar except for the fact that while the subject matter in SSFs (the stocks of a particular company) is deliverable, the subject matter in SIFs is the index, which cannot be owned or delivered—even in the future.
Using a legal history method, this paper also analyzed the legitimacy of SSFs and SIFs under Islamic law through a study of the views of Sharia board members. Until 2010, the board believed that SSFs and SIFs were illegal because these types of contracts contained gambling elements, including “gharar” and “gaming” based on chance. Accordingly, board members believed that these contracts were “gharary” (uncertain) because, even if delivery of the subject matter is considered possible, both payment and delivery of the subject matter are deferred in these contracts. This constitutes bay-al-kali-bi-kali—a type of gharary contract in Islam. Indeed, the parties do not really intend to deliver the subject matter in SSFs, as their real intention is to speculate. Thus, the board considered these contracts to be a “game of chance”, in which skill and information play no role, and in which one party of the contract is always the loser and the other the winner.
However, changing needs of capital market investors made Sharia board members change their views on SSFs. Board members’ views evolved to interpret SSFs as a promise to sell rather than a sales contract, and thus, responded to the problems of bay-al-kali-bi-kali. Further, board members used the “Interpretation” rule as an Islamic institution to expand their understanding of the concepts of gharar and
111 Mary Condon, “A Discipline in Search of Itself? Contemporary Challenges for Securities Law in Canada” (2014) 65 UNBLJ 341 at 346–47.
2020] FUTURES CONTRACTS IN IRAN’S CAPITAL MARKET 155 speculation. In their current view, speculation was wrongly considered gambling. It is different from gambling and game of chance because speculation may involve a great deal of computation, and chance can barely influence it. As such, in 2010, Sharia board members approved “The Directive of SSFs in Iran,” which formally provided that SSFs are not a mere tool of speculation.
We have argued in this article that the problem of “gharar” in SIFs can similarly be addressed by “Interpretation” of speculation as a legitimate activity based on skill and information. The only remaining issue is the impossibility of delivery of the subject matter in SIFs. We presented three solutions to remedy this problem: first, mutual funds can act as the parties’ representatives—owning the shares of 500 companies and delivering the shares in case the buyer of the SIF wishes so; second, we proposed customizing these contracts based on Iran’s Islamic capital market needs to launch these agreements based on 50 shares, the ownership and delivery of which, by the seller, is more feasible; and third, with respect to limited cases in which the delivery of the subject matter is impossible, Sharia board members should not consider a SIF agreement as a sales contract or a promise to sell, but rather a contract based on
Article 10 of Iranian Civil Law, the subject of which is to cover the risk of the other 2020 CanLIIDocs 3123 party for future purchases in consideration for a sum of money.
Finally, we addressed the concern of the “game of chance” element in SSFs by proposing they be considered a “hedging tool”, which means that parties cannot make profits unless they own the subject matter, intend to cover the risk of their property, and have certain skills and information so that they can analyze the future price of the share. We argued further that SSFs involve the possibility of mutual benefit and, thus, cannot be gambling. The same argument can be applied to SIFs, as such contracts are a means of hedging, by use of which parties try to manage their risk on a portfolio of shares when they own the portfolio. In cases where the seller does not own the portfolio of shares, it is considered a speculative activity which may entail mutual benefit for both parties, and thus, is not a “game of chance”.
THE EQUITY INCENTIVE CANADIAN STARTUPS NEED (HINT: IT IS NOT STOCK OPTIONS)
Bryce C. Tingle, QC*
Growth companies contribute disproportionately to Canada’s job creation, economic development and innovation. Most growth companies can match neither the salaries nor the security of more established competitors for executive talent. This makes their only advantage—the growth prospects of their equity—particularly important part of their compensation arrangements. Canadian growth companies (and Canadian businesses generally) make less use of equity incentives than their American peers and the kind of incentive they use almost exclusively, stock options, are strongly criticized by politicians, academics, institutional shareholders, and corporate governance experts. 2020 CanLIIDocs 3123 Stock options are accused of contributing to income inequality and creating incentives for value-destroying behaviour in large established corporations, but it is not clear these critiques have much to do with their use by growth companies. As well, it is not clear why these companies should be restricted to the use of stock options as the only equity incentive scheme available to them without adverse tax effects. For example, there are good reasons American growth companies make extensive use of share grants. As Canada enters its fourth round of amendments in this century to the tax rules relating to equity incentives, it is time to consider a tax regime that begins to differentiate between growth companies and their larger, more established counterparts, and that ceases to differentiate between issuing an option to acquire shares and simply issuing shares.
Table of Contents INTRODUCTION ...... 157 I. THE DEBATES AROUND STOCK OPTIONS ...... 158 II. THE USE OF STOCK OPTIONS BY CANADIAN STARTUPS ...... 162 III. THE PROBLEMS WITH STOCK OPTIONS ...... 166 IV. ARE CORPORATE GOVERNANCE FAILURES THE EXPLANATION FOR CANADIAN EQUITY COMPENSATION PRACTICES? ...... 173 V. THE IMPACT OF CANADIAN TAX POLICY ON THE USE OF STOCK OPTIONS ...... 176 VI. THE EQUITY INCENTIVE CANADIAN STARTUPS NEED ...... 181 VII. CONCLUSION ...... 184
* N. Murray Edwards Chair of Business Law, Faculty of Law, University of Calgary.
2020] THE EQUITY INCENTIVE CANADIAN STARTUPS NEED 157
INTRODUCTION
The fastest growing companies in Canada contribute disproportionately to the country’s economic and employment growth.1 They labour under many disadvantages, such as the much-discussed scarcity of institutional capital focused on the earliest stages of their development.2 Some of the disadvantages faced by Canadian startups seem almost intractable, such as the small size of the Canadian market and the paucity of experienced executive talent outside the natural resource sector. This paper, however, is about a problem that is comparatively easy to fix: Canadian growth companies (and Canadian businesses generally) make less use of equity incentives than their American peers and the kind of incentive they use almost exclusively, stock options, are almost universally regarded as inferior. Stock options set up perverse managerial incentives, fail to operate at all in certain conditions such as the current down-market, and make it very difficult to attract badly-needed managerial talent, particularly if already employed in other countries.3
The explanation why Canadian startups continue to use an equity incentive
that many of their executives, directors and investors regard as less than ideal, is 2020 CanLIIDocs 3123 connected to Canadian tax rules. The federal government is in the midst of its fourth attempt to reform the tax regime around stock options in less than twenty years.4 It is difficult to think of any set of technical tax rules that has received such constant attention. Since the tax treatment of options was extensively revised in 2000, the importance of this form of equity to startups has occupied the center of the debate. 5
1 See discussion at text accompanying notes 27-32 below. 2 Industry Canada, Growing the Businesses of Tomorrow: Challenges and Prospects of Early Stage Venture Capital Investment in Canada by Macdonald & Associates Limited (Ottawa: Industry Canada, 2005) (showing that at the height of venture capital flows in Canada these amounted to less than half the amount provided to American peer companies at 28); Reuven Brenner & Gabrielle Brenner, “Venture Capital in Canada: Lessons for Building (or Restoring) National Wealth” (2010) 22:1 J Applied Corporate Finance 86 at 91 (suggesting venture capital investment in Canada is about one-quarter of what one would expect from looking at international comparators).
3 See discussion at text accompanying notes 38-45 below. 4 Income Tax Act, RSC 1985, c 1 (5th Supp) [ITA] as amended by Income Tax Amendments Act, 2000, SC 2001, c 17; Jobs and Economic Growth Act, SC 2010, c 12; An Act to amend the Income Tax Act, SC 2016, c 11; Canada Revenue Agency, News Release, “Minister of National Revenue releases fifth report in the tax gap series” (18 June 2019), online: Government of Canada
158 UNBLJ RD UN-B [VOL/TOME 71
But no one is asking: why stock options? Why should the government decide which compensation structures are used by startups?
I. THE DEBATES AROUND STOCK OPTIONS
Stock options provide their holder with the right to buy a certain number of shares at a price that is set at the time the option is granted. The number of shares that may be purchased vests over time and the option, itself, expires five to ten years after it is issued. Once vested and prior to its expiry, the holder of an option can pay the exercise price to acquire the shares. This is only done when the share price of the issuer has increased above the exercise price, and nearly always only when there is a liquid market for the underlying shares, because the option holder must sell some of the shares to cover the price of exercising the option and, as we will see, to pay the tax owing from exercising the option. The value of the option at the time of exercise is the difference between the exercise price and the then current market price.
There are two different debates about stock options in Canada: one involving 2020 CanLIIDocs 3123 the corporate governance of large, mature companies; the other around income inequality. Neither has much to do with the use of stock options by startups, but startups are adversely impacted by the outcome of these debates all the same.
The corporate governance debate is the oldest. Corporate managers are in the decision-making business: committing investors’ resources to corporate projects and then effecting a multitude of adjustments, large and small, over the course of many years to ensure the organization succeeds in these projects. These decisions can be skewed by managers’ prejudices, inertia, failures of character, indifference, foolish disregard for risk, too extreme fear of risk, self-interest, and inattention. If a corporation can improve the decision-making of its managers in a broad and systematic way, it will achieve materially better returns from its opportunities. Properly calibrating executive pay structures has generally been seen as the best tool available to do this.
Economics is predicated on the idea that people respond to incentives and so a great deal of the debate about corporate governance has revolved around the proper structure of executive compensation.6 This has not been considered a simple task and
6 Michael C Jensen & William H Meckling, “Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure” (1976) 3:4 J Financial Economics 305 at 309–10 (usually credited with beginning the modern discussion of how manipulating the agent’s compensation incentives can maximize the principal’s welfare); Michael C Jensen & Kevin J Murphy, “Performance Pay and Top-Management Incentives” (1990) 98:2 J Political Economy 225 at 227–28; Bengt Holmstrom, “Moral Hazard in Teams” (1982) 13:2 Bell J Economics 324 at 325 (noting that the principal in an agency relationship must administer incentive schemes to control free riding); Martin J Conyon, “Executive Compensation and Incentives” (2006) 20:1 Academy Management Perspectives 25 (“[p]rincipal-agent theory predicts that a firm designs contracts in order to yield optimal incentives, therefore motivating the CEO to maximize shareholder value. In designing the contract, the firm recognizes the CEO is risk averse. Thus, imposing greater incentives requires more pay to compensate the agent for increased risk” at 28–29); Sok-Hyon Kang, Praveen Kumar & Hyunkoo Lee, “Agency and Corporate Investment: The Role of Executive Compensation and Corporate Governance”
2020] THE EQUITY INCENTIVE CANADIAN STARTUPS NEED 159
executive compensation packages have grown increasingly complex over the past several decades.7 An executive compensation advisory industry has grown alongside the changing compensation norms and very few boards now set executive compensation without input from specialized consultants.8 For their part, the advice provided to institutional shareholders by proxy advisory firms such as Institutional Shareholder Services and Glass-Lewis on the subject of executive pay is so complex that it won’t fit in their proxy voting guidelines, requiring separate detailed treatment.9 Even these stand-alone executive pay guidelines don’t include many of the models required to actually evaluate pay practices and generate voting outcomes.10
For reasons we will discuss later, since the dot-com collapse and the Enron era frauds, stock options have been a major issue in arguments amongst managers, boards, proxy advisors, compensation consultants, and the various third parties that concern themselves with corporate governance.11 Well-coordinated shareholder proposal campaigns have been launched against Canada’s largest companies demanding the curtailment of their use of stock options.12 An article in the (usually 2020 CanLIIDocs 3123
(2006) 79:3 J Bus 1127 at 1128; Jared Harris & Philip Bromiley, “Incentives to Cheat: The Influence of Executive Compensation and Firm Performance on Financial Misrepresentation” (2007) 18:3 Organization Science 350 at 350. 7 One survey of executives “paint[s] a grim picture for long-term incentives. Many of their key characteristics—high risk, complex and ambiguous performance conditions, arbitrary and unfair outcomes, multi-year deferral—suggest that individuals will discount them to a fraction of their economic value.” PwC, “Making Executive Pay Work: The Psychology of Incentives” (2012) at 24, online (pdf): PwC
11 See discussion at text accompanying notes 52-100. 12 See e.g. the shareholder proposal of United Brotherhood of Carpenters and Joiners, Local 27 Pension Trust to EnCana Corporation for their April 28, 2004 Annual General Meeting in Shareholder Association for Research Education (SHARE), “2004 Shareholder Proposals Submitted to Canadian Corporations” (4 August 2004) at 3, online (pdf): Shareholder Association for Research & Education
160 UNBLJ RD UN-B [VOL/TOME 71 staid) Canadian Chartered Accountant’s professional publication, can, without much further elaboration, refer to “Shareholder and political controversy over lavish salaries, perks and dubious compensation tools, such as stock options…”, suggesting that their continuing use is a function of managerial intransigence.13 Similar sentiments can be found in Canada’s national newspapers,14 the reports issued by think-tanks,15 and statements by the tribunes of good governance in Canada such as the Shareholder Association for Research and Education (SHARE) and Canadian Coalition for Good Governance.16 In every case the arguments against stock options are couched in terms that suggest the continuing use of options represents a failure of governance.17
The second argument about stock options has revolved around the growth of income inequality in the West over the past forty years or so. The most prominent advocate of the view that stock options are central to this story is Thomas Piketty, who pointed out in Capital in the Twenty-First Century, that stock options are “a form of
Stock Options as a Form of Executive Pay” (2014), online (pdf): Shareholder Association for Research & Education
The executive compensation must be based on solid performance criteria that executives can control and promote the creation of long-term added value for the organization. However, the main objective of stock options as a form of compensation is to reward and motivate executives by linking the executives’ performance to the organization’s share performance. However, such a linkage is far from being established.
2020] THE EQUITY INCENTIVE CANADIAN STARTUPS NEED 161
remuneration that has played an important role in the increase of wage inequality…”.18 Other observers have noted that “The staggering increases in the compensation of corporate executives of US Stock-exchange listed companies stems particularly from stock options and other forms of compensation that are linked to the value of the corporate stock.”19
The politicians have noticed. The 2015 national election included pledges by the Liberals and NDP to increase the taxation of stock options.20 There was an aborted attempt by the Liberals to do so in 2016, that was withdrawn because of protests from Canada’s startup community.21 However, in June 2019, the government introduced a Notice of Ways and Means Motion to effect changes to the taxation of stock options with the express goal of reducing inequality.22 The Department of Finance’s press release noted the benefits of the tax treatment of options, “are disproportionately going to a very small number of high-income individuals.”23 The release goes on to note that six per cent of stock option deduction claimants accounted for almost two-thirds of the cost of the deduction to taxpayers.24
The most recent federal government proposal is well defined, except around 2020 CanLIIDocs 3123 the treatment of startups. This is surprising, because, as the 2019 federal budget notes, “The public policy rationale for preferential tax treatment of employee stock options is to support younger and growing Canadian businesses.”25Canadian-Controlled Private Corporations (CCPCs), the conceptual workhorse of tax policy when it comes
18 Thomas Piketty, Capital in the Twenty-First Century, translated by Arthur Goldhammer (Cambridge, MA: Harvard University Press, 2014) at 381. 19 Yvan Allaire & Mihaela Firsirotu with the collaboration of François Dauphin, “Inequality and Executive Compensation: Why Thomas Piketty is Wrong?” (2014) at 20, online (pdf): Institute for Governance of Private and Public Organizations
162 UNBLJ RD UN-B [VOL/TOME 71 to entrepreneurs, would be excluded from the changes, but the government is not sure what to do about “non-CCPCs [that] could be startups, emerging or scale-up companies.”26 The government is looking for input about how to distinguish this type of company.
II. THE USE OF STOCK OPTIONS BY CANADIAN STARTUPS
Startups or “growth companies” (the terms will be used interchangeably) are relatively young companies with the intention—often not realized—of growing rapidly in terms of employees and revenue. The idealized startup is one that is developing and commercializing a new technology, but in fact Canadian growth companies arise in any industry. However, they all tend to pursue a potentially valuable innovation in the way firms in that industry have previously operated. This strategic dependence on new and untested products, services, or methods, along with their small size, need for outside capital, and often unexperienced management teams, makes them unusually risky.27
2020 CanLIIDocs 3123 Importantly, though, these smaller, high-risk businesses create nearly half of Canada’s net new jobs and a significant percentage of its economic growth.28 For example, high growth firms in Canada accounted for only 4 per cent of Canadian businesses but 45 per cent of the net jobs created during a typical decade.29 Using venture capital-backed companies as a proxy for growth companies generally, research has found they achieve sales growth five times greater than the Canadian private sector as a whole, nine times faster employment growth, and export 70 per cent of their sales—four times the Canadian average.30 Canadian growth companies are also disproportionately innovative, conducting more R&D and filing more patents per dollar invested than more established firms.31
26 Department of Finance Canada, supra note 22. 27 Only about eight percent of startups in Canada actually achieve a high rate of growth: The State of Entrepreneurship in Canada, by Eileen Fisher & Rebecca Reuber (Ottawa: Government of Canada Publications, February 2010) at 9–10, online (pdf): Industry Canada
2020] THE EQUITY INCENTIVE CANADIAN STARTUPS NEED 163
It is worth noting as well, that Canadian growth companies compare favourably with their peers in the United States. Again, taking venture-backed companies as an example, (they are a small subset of growth companies, but one for which statistics are tracked) Canadian growth companies have higher exit values per dollar invested and per dollar of R&D spent, than comparable American companies.32
Fostering growth companies is generally regarded by policymakers as essential to preserve an economy’s competitiveness and productivity.33 This is why, for example, the Canadian government has created various tax programmes, including some relating to the taxation of stock options, designed to foster the flourishing of startups in this country.34
that faster growing early-stage companies innovate at almost twice the rate of slower growing peers); Josh Lerner, “Boom and Bust in the Venture Capital Industry and the Impact on Innovation” (2002) 87:4 Federal Reserve Bank Atlanta Economic Rev Fourth Quarter 25 (Finding that “on average a dollar of venture capital [given exclusively to growth companies] appears to be three to four times more potent in stimulating patenting than a dollar of traditional corporate R&D” at 35); Gilles Duruflé, supra note 30 at 25 (Canadian venture backed companies devote 54% of their employees to R&D and hold an average of 10 patents); 2020 CanLIIDocs 3123 Industry Canada, SME Financial Data Initiative: High Growth SMEs, by Patrick Huot & Christine Carrington (Ottawa: Industry Canada, May 2006) at 6, online (pdf): Industry Canada
164 UNBLJ RD UN-B [VOL/TOME 71
Startups have a very different relationship to equity incentives than larger, more mature companies. Indeed, income inequality and the proper calibration of incentives (the preoccupations of corporate governance elites and the Federal government) are issues that are largely irrelevant to growth companies.
The key concern for these riskier companies with less free cash flow to use for compensation, is the impact of various executive compensation structures in recruiting and retaining managers. This is particularly important for two reasons. First, a well-known feature of rapid corporate growth is that it is accompanied by nearly continual change in senior leadership as founders reach the limits of their competency or burn out, and as new stages of development require new levels and types of managerial experience.35 Second, most growth companies can match neither the salaries nor the security of more established competitors for executive talent. This makes their only advantage—the growth prospects of their equity—a particularly important part of its compensation arrangements.
Equity is regarded as cheap to the company and attractive to high quality employees. Saying that equity is “cheap” does not mean that equity incentives don’t 2020 CanLIIDocs 3123 have a “cost”. They obviously result in shareholder dilution and thus the reduction of corporate cash flows to the pre-existing shareholders. Since 2001, Canadian companies have been required to disclose the cost of outstanding conditional equity grants, such as stock options, in their financial statements.36 But equity awards like stock options don’t take up any of the company’s internal cash and for certain kinds of companies, those with the biggest opportunities, internal cash is extremely valuable. Indeed, studies in the United States find that employment and sales growth are negatively correlated with profitability, underlying the importance of equity being available to recruit talent.37
For highly talented executives of the sort startups desperately need to attract, the chance to own equity is almost the only available inducement to accept an employment offer. On virtually every other measure, employment in a startup is worse than a job in a more established company. A growth company’s business model is untried and changes rapidly, salaries are always lower than they would be elsewhere,
35 Michael Ewens & Matt Marx, “Founder Replacement and Startup Performance” (2018) 31:4 Rev Financial Studies 1532 at 1541 (twenty percent of VC-backed firms experienced a founder replacement event); Warren Boeker & Rushi Karichalil, “Entrepreneurial Transitions: Factors Influencing Founder Departure” (2002) 45:4 Academy Management J 818; Michael Gorman & William A Sahlman, “What Do Venture Capitalists Do?” (1989) 4 :4 J Bus Venturing 231 at 241 (finding that the statistical mean is one termination of a CEO/President for every 2.4 years of a venture capitalist’s investing experience). 36 See Ed Abahoonie & Koen De Grave, “Stock Option Awards Under IFRS: An Analysis of the Potential Impact” (2008), online (pdf): PricewaterhouseCoopers
2020] THE EQUITY INCENTIVE CANADIAN STARTUPS NEED 165 the firm could run out of money at any point, there are frequent management changes, power in the firm is volatile and prone to sudden change, competitors are larger with greater resources, and the startup has a very good chance of failing, which brings reputational risks to its senior management.38 A big reason most senior employees join a growth company is because the equity incentives on offer will make them very wealthy if the company is a success.39
Research in Canada on the differences between faster and slower growing new firms has established that the human resource strategy with the greatest impact on growth is the provision of incentive compensation plans.40 Equity incentives are better predictors of fast growth than training programs or even the recruitment of skilled employees.41 Post-IPO, the more exposure a company’s executives have to various equity incentives, the faster sales, earnings, return on assets and share price grow.42
Start-ups and fast growing companies thus make heavy use of equity in their compensation arrangements.43 In 73% of American venture capital-backed companies every single employee holds stock options.44 At the time successful American companies conduct their Initial Public Offering, employees own approximately 20% 2020 CanLIIDocs 3123 of the shares and the founders (who are often still executives of the company) own a further 10-20%.45 Thus, in the most successful growth companies (the ones that make it to IPO) insiders hold an extremely large percentage of the company, the majority of which is held by individuals joining the business after its initial formation. Equity incentives, whether in the form of stock options or outright grants of shares, are absolutely necessary to startups. Without them, firms cannot attract the
38 Tingle, Start-up and Growth Companies in Canada, supra note 28 at 8–17. 39 John RM Hand, “Give Everyone a Prize? Employee Stock Options in Venture-backed Firms” (2008) 23:4 J Bus Venturing 385.
40 John Baldwin, “Innovation and Training in New Firms” (Paper presented to a conference devoted to “Assessing the Impact of Training on the Permanence of Small and Medium Sized Enterprises” held jointly by the Centre for Business Research, University of Cambridge and the Centre for Small and Medium Sized Enterprises, University of Warwick, March 1998) at 25–26, online (pdf): Statistics Canada
166 UNBLJ RD UN-B [VOL/TOME 71 talent they require, particularly experienced managerial, financial, sales, and marketing executives. For this task, Canadian growth companies use stock options almost exclusively. 46 As an example, the TSX Venture 50 is a ranking of the strongest performers across five industry sectors (ten chosen from each).47 On average the TSX Venture 50 companies delivered an annualized return of 173% over a representative year. Some of them grew five-fold in that time. They serve as good proxy, then, for the kind of company we are most concerned to foster in this country and the kind of company that most depends on equity incentives to attract and retain executives.48 Hand-collected data from the TSX Venture 50 companies’ 2013 proxy season information circulars shows that in the previous three years (for an aggregate of 150 years of executive compensation decisions) every one of the companies had issued stock options, whilst only four had ever used a share grant.49 For only two companies did share grants exceed in value the options granted during those three years. Stock options accounted for approximately 88% of the value of equity issued by these companies over the three-year time period.50
While only anecdotal evidence is available about the practices of private
Canadian growth companies, it seems to suggest they also use stock options almost 2020 CanLIIDocs 3123 exclusively.51
III. THE PROBLEMS WITH STOCK OPTIONS
The use of stock options in Canada and elsewhere exploded in the 1990s, but the seeds had been planted in the previous decade by academics that promoted the value of
46 Antonio Spizzirri, “Canadian Director Compensation Analysis 2009-2012” (March 2014), online (pdf): Rotman School of Management
2020] THE EQUITY INCENTIVE CANADIAN STARTUPS NEED 167
equity components in executive pay as a method of ameliorating agency costs.52 In this early work, employee stock options were explicitly identified as leading to stronger alignment of management and shareholder goals.53 By the 1990s, large institutional shareholders such as CalPERS (the largest public pension fund in the US) and governance bodies like the National Association of Corporate Directors were calling for “a significant amount of director compensation to be paid in the form of stock grants and options”54 By the end of the decade, a Harvard Business Review article could opine: “Options are the best compensation mechanism we have for getting managers to act in ways that ensure the long-term success of their companies and the well-being of their workers and stockholders.”55
Boards responded to these changes to governance norms and began issuing options in large numbers as the decade progressed. In the United States, option compensation comprised 20% of CEO pay in 1992, but rose to a “staggering” 49% in 2000.56 By the beginning of this century, S&P 500 index companies were granting six times more options than they had a decade previously.57 The unprecedented growth in CEO pay over the decade was almost entirely a reflection of this increase in the use 58 of stock options. 2020 CanLIIDocs 3123
Canadian companies followed suit. In 1991, only 33% of the largest 100 Canadian public companies granted stock options to their executives; by 2000, 100% of them did.59 Smaller Canadian companies replicated this increasing use of options.60 While it never reached the size it did in the United States, by 2002, 36% of the
52 See e.g. Holmstrom, supra note 6; Eugene F Fama, “Agency Problems and the Theory of the Firm” (1980) 88:2 J Political Economy 288. See also Jensen & Murphy, supra note 6. 53 Jensen & Murphy, supra note 6 at 226; Kevin J Murphy, “Executive Compensation: Where We are, and How We Got There” in George Constantinides, Milton Harris & Rene Stulz, eds, Handbook of the Economics of Finance, vol 2A (Amsterdam: North Holland, 2013) 211; Rachel Merhebi et al, “Australian Chief Executive Officer Remuneration: Pay and Performance” (2006) 46:3 Accounting & Finance 481 at 482; Brian J Hall & Jeffrey B Liebman, “Are CEOs Really Paid Like Bureaucrats?” (1998) 113:3 QJ Economics 653 at 655. 54 Shamsud D Chowdhury, “Director Compensation: The Growing Popularity of Deferred Stock Units” (2009) 73:1 Ivey Bus J 11 at 11. 55 Brian J Hall, “What You Need to Know about Stock Options” (2000) 78:2 Harvard Bus Rev 121 at 122. 56 Carola Frydman & Dirk Jenter, “CEO Compensation” (2010) National Bureau of Economic Research Working Paper No 16585 at 6. 57 Brian J Hall & Kevin J Murphy, “The Trouble With Stock Options” (2003) 17:3 J Economic Perspectives 49 at 49 [Hall & Murphy, “Trouble With Stock Options”]. 58 Kevin J Murphy, “Explaining Executive Compensation: Managerial Power Versus the Perceived Cost of Stock Options” (2002) 69:3 U Chicago L Rev 847 at 847–48. 59 Ryan Compton, Daniel Sandler & Lindsay Tedds, “Options Backdating: A Canadian Perspective” (2009) 47:3 Can Bus LJ 363 at 363. 60 For example, Zhou found that “Among the TSE300 firms, 90 per cent had an annual bonus plan and used stock options in compensating their executives.” Xianming Zhou, “CEO Pay, Firm Size, and Corporate Performance: Evidence From Canada” (2000) 33:1 Can J Economics 213 at 217.
168 UNBLJ RD UN-B [VOL/TOME 71 compensation received by TSX 60 CEOs was received from stock options.61 This was the largest single component—by far—of Canadian executive pay packages.62
It was about this time that options began to fall out of favour with academics and governance activists. The decline is sometimes linked to accounting changes early in the last decade that required companies to begin expensing options.63 A study in Canada, however, found that while option use decreased around the time of the accounting changes, it did not decrease in the ways we would expect if prior option use had been primarily a function of its invisibility in corporate cash flow statements.64 Instead, the decline is likely due to the growing appreciation that stock options are a sub-optimal way to motivate executives.
First, in the early years of this century, finance academics began pointing out that the structure of options encourages excessive risk-taking.65 Options pay out if the stock price exceeds the exercise price and, obviously, the greater the increase in the share price, the larger the payout. But there is no penalty to the option holder if the share price declines below the exercise price. The options become worthless, but the executive doesn’t lose any money. Indeed, it makes no difference to the executive 2020 CanLIIDocs 3123 option-holder whether the share price declines slightly below the exercise price or dives well below the exercise price. The options are equally worthless in either scenario. This means that the rational thing for an executive holding options to do is swing for the fences—take the big risks. If this strategy pays off, the executive makes a lot on the options; if it fails, she isn’t harmed. From her perspective there is no
61 Allaire, supra note 15 at 22. 62 Ibid. In contrast, salary amounted to 27% and bonuses to 25%. 63 See e.g. Frydman & Jenter, supra note 56 (“[m]oreover, it is likely that both the prior decline in the stock market and the advent of option expensing in 2004 have contributed to the declining popularity of stock options in recent years” at 6). See also Iman Anabtawi, “Explaining Pay Without Performance: The Tournament Alternative” (2005) 54:4 Emory LJ 1557 at 1576. 64 See Karrie Geremia, Carla Carnaghan & Toni Nelson, “Exploring Changes in Management Compensation Structure in Canada: Evidence on the Consequences of Requiring Options Expensing” (Canadian Academic Accounting Association Annual Conference 2010 delivered at the Hyatt Regency Vancouver, 29 May 2010), online: SSRN
2020] THE EQUITY INCENTIVE CANADIAN STARTUPS NEED 169
reason to minimize losses.66 Both the 2001 and 2008 financial crises were blamed on this aspect of stock options.67
In the febrile atmosphere following the discovery of the Enron-era accounting frauds,68 options were also accused of promoting unethical behaviour.69 All equity incentives arguably encourage manipulation of share prices, but because options typically have short vesting periods (three or four years), must be exercised before they expire (usually five to ten years after issue), and can be exercised opportunistically at the discretion of the executive, options will inevitably occasionally function as a short-term incentive. 70 And in the short-term, share prices can be manipulated, through accounting fraud,71 earnings management,72 misleading public
66 Harley E Ryan Jr & Roy A Wiggins III, “The Interactions Between R&D Investment Decisions and 2020 CanLIIDocs 3123 Investment Policy” (2002) 31:1 Financial Management 5 at 27 (noting that option awards to managers encourages them to invest with more risk). 67 Sanjai Bhagat & Brian Bolton, “Financial Crisis and Bank Executive Incentive Compensation” (2014) 25 J Corporate Finance 313 at 313; Hamid Mehran, Alan Morrison & Joel Shapiro, “Corporate Governance and Banks: What Have We Learned from the Financial Crisis?” (2011) Federal Reserve Bank of New York Staff Report No 502 at 7–8, online (pdf): New York Federal Reserve
170 UNBLJ RD UN-B [VOL/TOME 71 business disclosure,73 or the pursuit of appealing short-term business strategies that harm the long-term interests of the company.74
Other problems with options were identified over time.75 Once the bull market of the 1990s was over, it became noticeable that options lose their power to either motivate or retain managers when share prices decline below option exercise prices.76 It also became clear that managers holding stock options tend to reduce dividend payments (in which option holders do not participate) in favour of repurchasing stock (which benefits option holders by increasing share value).77 As a method of distributing cash not immediately needed by the business, dividends are superior to stock repurchases because the latter encourages a variety of practices that treat shareholders unequally78 and can run contrary to the long-term interests of the company.79 These problematic practices arise from the fact that unlike dividends, share repurchases do not treat shareholders equally (the issuer transacts with some shareholders and not with others) which gives rise to the possibility issuers will take
2020 CanLIIDocs 3123 73 David Aboody & Ron Kasznik, “CEO Stock Option Awards and the Timing of Corporate Voluntary Disclosures” (2000) 29:1 J Accounting & Economics 73 at 75. 74 See Bryce Tingle, “Bad Company! The Assumptions Behind Proxy Advisors’ Voting Recommendations” (2014) 37:2 Dal LJ 709 at 736–38 [Tingle, “Proxy Advisors’ Voting Recommendations”]; See also John R Graham, Campbell R Harvey & Shiva Rajgopal, “The Economic Implications of Corporate Financial Reporting” (2005) 40:1/3 J Accounting & Economics 3 at 4 (finding that managers would rather risk negative long-term consequences through economic actions than make a GAAP accounting decision to manage earnings and that the majority of managers have admitted to forfeiting long-term value in order to smooth earnings). 75 For a summary of all the concerns raised over the past decade or so about options, see Richard T Holden, “The Original Management Incentive Schemes” (2005) 19:4 J Economic Perspectives 135 at 142–43. 76 Susan J Stabile, “Motivating Executives: Does Performance-Based Compensation Positively Affect Managerial Performance?” (1999) 2:2 U Pa J Lab & Employment L 227 at 267–68; Viral V Archarya, Kose John & Rangarajan K Sundaram, “On the Optimality of Resetting Executive Stock Options” (2000) 57:1 J Financial Economics 65 at 66–67; Lisa K Meulbroek & Li Jin, “Do Underwater Executive Stock Options Still Align Incentives?: The Effect of Stock Price Movements on Managerial-Incentive Alignment” (2001) Harvard Business School Finance School Working Paper No 02-002 at 9–10, online: SSRN
2020] THE EQUITY INCENTIVE CANADIAN STARTUPS NEED 171 advantage of unsophisticated shareholders by buying their temporarily undervalued shares and engage in price-manipulation around repurchases.80
The obvious alternative to stock options is to issue shares directly to executives in the form of restricted share grants. (These are shares granted to an executive under an escrow agreement that contains certain conditions for their release. The conditions may be as simple as the passing of one or more vesting periods.) Issuing actual shares is considered to be the gold standard of equity incentives since it places managers in exactly the same position as other shareholders.81 Unreasonably high-risk business decisions are discouraged since the executive can lose money through a decline in the value of their shares, as well as make money if the bet pays off. Unlike options, shares do not need to be sold by a certain deadline and thus can function as a long-term incentive in a way options are incapable. Because they don’t “expire”, share grants thus tend to ameliorate the pressures towards unethical or short- term actions intended to manipulate the share price. Shares also remove the bias against dividends and don’t lose their power as an incentive if the share price declines below the exercise price. In the words of professors Hall and Murphy, writing at the beginning of the period share grants began to grow in prominence, “when existing 2020 CanLIIDocs 3123 compensation is adjusted, incentives are maximized through restricted stock grants rather than options.”82
There are other advantages of share grants. Studies show that executives value restricted share grants more highly than options.83 This means less dilution of the shareholders is required when shares are used as equity incentives rather than options. One set of researchers found that it would take a dollar worth of stock options to equal 45 cents in restricted stock.84 In fact, moving from stock options to share grants may be one of the few free lunches in this world. The cost of shares to the corporation and its shareholders (relative to options) may be less than the value of the shares to executives. Thus, a principal of consulting firm Towers Watson, which specializes in executive pay, writes the following:
80 Tingle, “Proxy Advisors’ Voting Recommendations”, supra note 74 at 736. 81 Mary Ellen Carter, Luann J Lynch & Irem Tuna, “The Role of Accounting in the Design of CEO Equity Compensation” (2007) 82:2 Accounting Rev 327 at 355 (finding that firms are using more restricted stock in order to provide longer-term incentives). 82 Hall & Murphy, “Stock Options for Undiversified Executives”, supra note 70 at 26–27. See also Ingolf Dittmann & Ernst Maug, “Lower Salaries and No Options? On the Optimal Structure of Executive Pay” (2007) 62:1 J Finance 303 (demonstrating compensation costs could be reduced by 20% as part of a compensation scheme that involved issuing shares rather than options). 83 Hall & Murphy, “Stock Options for Undiversified Executives”, supra note 70 (“options are, in fact, an unusually expensive and therefore inefficient way to convey compensation to executives” at 16). 84 Lawrence D Brown & Yen-Jung Lee, “Changes in Option-Based Compensation Around the Issuance of SFAS 123R” (2011) 38:9/10 J Bus Finance & Accounting 1053 at 1086–87. See also Dittmann & Maug, supra note 82.
172 UNBLJ RD UN-B [VOL/TOME 71
The economic cost to shareholders of a restricted share often is about equal to that of two options. Participants’ risk tolerances, on the other hand, cause them to place a higher relative value on restricted stock. They may be happy to give up three options they would otherwise receive for one share of restricted stock. A restructured grant policy can create a win-win situation, retaining or improving participants’ satisfaction while lessening economic [and accounting] cost.85
The academic concerns about the over-use of options in the 1990s found voice in the calls of institutional investors and governance activists to eschew them in favour of stock grants. In Canada, the Canada Pension Plan Investment Board indicated its support for stock grants rather than stock options.86 The Canadian Coalition for Good Governance similarly encouraged share grants while criticizing stock options.87 Its founder, Stephen Jarislowsky, a leading institutional investor in Canada, was more blunt: “Stock options are very dangerous for shareholders—they should be banned.”88 A report from the Institute for Governance of Private and Public Organizations urged directors to, “reduce or, preferably, eliminate stock options from the compensation program.”89 2020 CanLIIDocs 3123 In the wake of this opprobrium, the use of options predictably declined sharply even as the percentage of executive pay attributable to equity incentives remained constant, or even increased.90 In the UK stock options were “largely replaced” by performance-contingent stock grants.91 The largest stock option issuer in history, Microsoft, made a stir in 2004 when it announced it would discontinue the use of options.92 As a percentage of total executive compensation in the United States, options fell by 67% between 2001 and 2012.93 By 2006, share grants made up the majority of CEO equity incentives used by the S&P 500.94
85 Ericson, supra note 64 at 28. 86 Chowdhury, supra note 54 at 12. 87 “Executive Compensation Principles” (January 2013), online (pdf): Canadian Coalition for Good Governance
2020] THE EQUITY INCENTIVE CANADIAN STARTUPS NEED 173
In Canada option use declined among the 60 largest public companies, although the decline was shallower and slower than the United States. As a percentage of total compensation, options declined by only 25% between 2000 and 2008—less than half the rate of the American decline.95 For these large companies, options predominated among equity incentives until comparatively recently.96 As a percentage of total CEO compensation, Canadian S&P 60 companies now issue options at more or less the same rate as American companies. The lower over-all use of equity incentives in Canada arises from lower use of share grants.97
The criticisms of stock options, made in the context of large public companies, was always going to have an effect on their use by even private growth companies. The investors, directors and managers of startups also tend to be involved in one capacity or another with larger firms and quickly became socialized in the failings of stock options. In reviewing the critiques of stock options found in the writings of academics, institutional investors, newspapers, and governance experts, I was unable to find a single instance in which private startups or growth companies (much less firms listed on Canada’s public venture markets) were excepted from the criticisms. This has had the effect of putting pressure on these companies to reduce 2020 CanLIIDocs 3123 their use of options as well. Whereas twenty years ago it was common to see growth companies with option pools equal to 20 percent of their outstanding shares, they tend to be much smaller now.98 In this way the turn away from stock options has reduced the currency available to startups to attract employees.
The situation of growth companies can be stated simply: as we have seen they make almost exclusive use of stock options, even though share grants are superior in many circumstances. Because of their dependence on stock options and the growing unpopularity of options, growth companies have come to face constraints on the amount of equity they can offer prospective employees.
IV. ARE CORPORATE GOVERNANCE FAILURES THE EXPLANATION FOR CANADIAN EQUITY COMPENSATION PRACTICES?
Because stock options tend to pay out more generously than other types of equity incentives, it is possible that their continuing use is strongly desired by managers and permitted by captured boards.99 This assumed managerial appetite for risk contradicts, however, the other alleged governance failure surrounding equity incentives in
95 Allaire, supra note 15 at 24. 96 Ibid. 97 Ibid. 98 Tingle, Start-up and Growth Companies in Canada, supra note 28 at 174. 99 Bryce C Tingle, “Framed! The Failure of Traditional Agency Cost Explanations for Executive Pay Practices” (2017) 54:4 Alta L Rev 899 at 901–03 [Tingle, “Framed!”].
174 UNBLJ RD UN-B [VOL/TOME 71
Canada: the under-use of equity incentives generally.100 According to proponents of this view, our executives have persuaded their boards to rely too much on relatively “safe” forms of cash compensation rather than riskier equity compensation that mirrors the long-term outcomes experienced by the shareholders. The remedy is for Canada to “embrace an incentive culture that includes stock options.”101 Surely, however, Canada can’t be suffering from both governance failures—simultaneously too much, and too little, appetite for risk?
In fact, there are reasons to think the patterns of use of equity incentives in Canada is not due to failures of corporate governance. One of the most obvious is that, as the history of the use of equity incentives reveals, boards have generally proven to be extremely responsive to changing academic and governance fads.102 Corporate boards adopted at-risk equity incentives when they were encouraged in the early 1990s, began abandoning stock options when encouraged a decade later, and no doubt will follow the next prescription for one-size fits all compensation practices promoted by the governance community.103 In the UK, as we have seen, stock options have effectively been eliminated from CEO pay packages. In the United States, share grants are almost double the size of option grants (and when cash-settled equity-like 2020 CanLIIDocs 3123 incentives such phantom stock are added, they dwarf the once mighty stock option)104 It would be odd if Canada, with its more conservative boards,105 lower over-all executive pay,106 and a greater proportion of companies with a controlling shareholder,107 turned out to have materially weaker boards than the US or UK.
100 See discussion at notes 87–90. 101 Pav Jordan, “Canada’s Venture Sector Urged to Build on Success”, Financial Post (27 May 2011), online:
2020] THE EQUITY INCENTIVE CANADIAN STARTUPS NEED 175
There are reasons, in any event, to question the near-automatic assumption in the governance community that board capture and rent extraction are the best explanations for corporate pay.108 No doubt this is true in some cases,109 but it is not at all clear that it has general explanatory (and, therefore, predictive) power. Recent research has shown, for example, that private equity-backed companies with shareholder-dominated boards, pay their executives in a way that is “statistically indistinguishable” from the level of pay in widely-held public companies.110 “[O]verall the evidence from private equity offers little direct support for the view that managerial influence leads to excessive CEO pay in public companies.”111 Both the quantum and the type of equity incentives provided to executives in the two sorts of firms are identical.112
Other research finds little evidence that increasing the proportion of independent directors on either the board or the compensation committee has any impact on the proportion of pay that is contingent on managerial performance.113 Surely, if managerial power is a significant determinant of pay structure, increasingly independent boards and compensation committees should have some, even limited,
impact on pay. But there appears to be none. 2020 CanLIIDocs 3123
at 176 (noting that externally-controlled firms with at least one major shareholder are presumed to be diligently monitored). g108 See Tingle, “Framed!”, supra note 99; Mel Perel, “An Ethical Perspective on CEO Compensation” (2003) 48:4 J Bus Ethics 381 at 382 (suggesting that the source of excessive CEO compensation is more complicated than the relationship between CEOs and boards of directors). 109 See e.g. Richard Siklos, “Conrad Black Found Guilty in Fraud Trial”, The New York Times (14 July 2007), online:
176 UNBLJ RD UN-B [VOL/TOME 71
V. THE IMPACT OF CANADIAN TAX POLICY ON THE USE OF STOCK OPTIONS
There is another explanation that accounts for Canada’s under-use of equity incentives and its over-reliance on stock options: Canada’s government has, for some reason, made the decision to privilege stock options and penalize stock grants. This has led to considerable difficulty in adopting share grants, particularly amongst Canadian growth businesses and, when combined with the governance industry’s hostility against stock options, it has produced lower levels of equity incentives in this country generally.
Throughout most of the Twenty-First Century, Canada had one of the most generous regimes in the world in relation to stock options.114 (Canada provides capital gains treatment for options so long as their exercise price is set at the fair market price of the underlying stock at the time of grant.115 In contrast, it is much harder to achieve capital gains treatment under the American regime.)116 The Canadian rules were never straightforward. Intended to assist small, cash-constrained entrepreneurial companies,117 the regime was naturally so recondite that it required the assistance of expensive outside professionals.118 (This self-defeating inefficiency is shared by the 119 similarly targeted SR&ED and IRAP regimes.) 2020 CanLIIDocs 3123
From the standpoint of fostering the use of equity incentives for the startup companies that need it the most, the tax treatment of options for non-CCPC firms got a lot more counter-productive with extensive amendments in the 2010 Federal budget. Previously, the exercise of stock options resulted in a capital gain crystallizing, but this gain could be deferred until the underlying shares were sold.120 This matched the moment tax was payable with the moment the employee sold her shares and received the money needed to pay the tax. The financial crisis of 2008, however, left large numbers of executives holding shares that were worth considerably less than when they had been acquired through the exercise of options. In many cases, the shares were
114 Sandler, “Tax Treatment of Employee Stock Options”, supra note 5 (“most employee stock options are taxed more favourably in Canada than in the United States... In particular, most employee stock option benefits in Canada are taxed at capital gains tax rates, whereas in the United States most are taxed at full rates” at 261). 115 ITA, supra note 4, s 110(1)(d). Note that after January 1, 2020, options issued by all firms that are not CCPCs or “startups, emerging or scale-up companies” do not receive capital gains-like treatment if they exceed at $200,000 cap. 116 Sandler, “Tax Treatment of Employee Stock Options”, supra note 5 at 292–95. 117 Ibid at 296–301. 118 Tingle, Start-up and Growth Companies in Canada, supra note 28 at 166–69. 119 See John Lester, “Benefit-Cost Analysis of R & D Support Programs” (2012) 60:4 Can Tax J 793 at 794 (describing how SR&ED and IRAP tax credit programs fail a cost-benefit analysis due to the high cost of administering and complying with the programs). 120ITA, supra note 4, s 180.01 as amended by Sustaining Canada’s Economic Recovery Act, SC 2010, c 25, s 46.
2020] THE EQUITY INCENTIVE CANADIAN STARTUPS NEED 177
worth less than the deferred tax bill due on their sale.121 The government’s response was not to change the rule that deems a capital gain to arise when an option is exercised (as opposed to having it arise when the underlying shares are sold122), but to remove the deferral.123
For all non-CCPC companies,124 including startups listed on the TSX Venture Exchange and the Canadian Stock Exchange, tax is now calculated and due at the time the option is exercised.125 The only rational choice for executives of these companies is to sell their shares immediately following the exercise of an option. Many of the shares will simply need to be sold to pay the tax. As well, the employee has an incentive to realize the gains on which they have just been taxed. An employee will be conscious of the risk of unfairness if she holds onto the shares for the long-term. If the price of the shares decline before she disposes of the shares (and the prices of growth companies are volatile), her tax bill on exercising the option may exceed the final value of the shares. Thus, a tax regime clearly designed to respond to the need of Canadian startups to have access to equity incentives has been revised to discourage precisely what everyone wants: the long-term holding of equity by managers.126
2020 CanLIIDocs 3123
121 Alan Mcnaughton & Amin Mawani, “Contributions of Employee Stock Options to RRSPs and TFSAs: Valuation Issues and Policy Anomalies” (2008) 56:4 Can Tax J 893 (discussing the problem of “underwater” options at 907); Wayne Tunney, “Underwater Stock Options Sink” Canadian Tax Highlights 11:6 (June 2003) 1, online (pdf):
178 UNBLJ RD UN-B [VOL/TOME 71
It might be objected that the Federal Government has attempted to address this problem by providing a tax regime for CCPC option grants that matches the taxable event to the date of disposition.127 The problem is that CCPCs have always been a poor proxy for growth companies. This is implicitly acknowledged in the Federal Government’s June 2019 Notice of Ways and Means on reforming the treatment of stock options as it expressly asks for public comment on how to identify non-CCPCs that are nevertheless growth companies.
The biggest problem with the CCPC regime is that it does not include growth companies listed on either of Canada’s public venture markets. These markets typically deliver at least as much capital to early stage Canadian companies as institutional venture capitalists, and in some years more than six times as much.128 Canada’s public venture markets also have success rates (measured by both overall returns and new listings on the main market) that exceed those generated by domestic institutional venture capital by a significant margin.129 They cannot be excluded by any tax policy seriously intended to foster startups.
Another problem with the CCPC regime is that many Canadian startups have 2020 CanLIIDocs 3123 a strong international flavour. They have founders resident in more than one country, or they have issued shares to acquire assets or technology developed in another country, or they have raised capital from angels, strategic investors, or venture capitalists in other countries. In aggregate, it is very easy for these sorts of companies, even if based in Canada and led by Canadians, to lose CCPC status because voting control no longer resides in the hands of Canadians. As they currently stand, the CCPC stock option rules exclude many of the growth companies it is intended to help.
However ill-fitting Canada’s current stock option tax regime is for startup companies, at least it doesn’t present an insuperable barrier to their use. In contrast, the tax regime’s treatment of share grants simply makes them impossible. If shares are issued to an employee, the Income Tax Act requires the fair market value of those shares be immediately included in the employee’s taxable income.130 The consequence is that it is virtually impossible for a company to issue shares to executives unless the grant is very small or the executive is very rich (possibly as a result of simultaneous large cash bonuses). This may sound like a blow struck against
127 ITA, supra note 4, s 7(1.1). 128 Cecile Carpentier & Jean-Marc Suret, “The Canadian Public Venture Capital Market” (2010) 19:7/8 Strategic Change 303 at 304. 129 Ibid (noting that the TSX Venture Exchange has a success rate that is “approximately four times the corresponding rate for private VC” and that it provides seven times the number of TSX listings as Canada’s VC firms. It also has a failure rate lower than that found in Canadian private VC portfolios). See also Michele Meoli et al, “Can Spending Time in the Minors Pay Off? An Examination of the Canadian Junior Public Equity Markets” (2018) 56:S1 J Small Bus Management 88 (showing that graduates from the TSX- V outperform VC-backed TSX companies by over 28 per cent in the first three years following their listing on the TSX). 130 ITA, supra note 4, s 7(1).
2020] THE EQUITY INCENTIVE CANADIAN STARTUPS NEED 179 corporate fat cats, but it actually means that growth companies seeking to recruit senior executives have no way of giving them a meaningful stake in the business, except through stock options. It also means that bonuses and long-term incentives come out of retained earnings, the most precious thing in the world to a fast-growing company. Indeed, as we have seen, only the largest Canadian companies can afford significant share grants.
Contrasting the situation in Canada with the tax regime in the United States is instructive. While the Americans have similar rules governing the tax treatment of share grants, the Internal Revenue Service has permitted the evolution of several strategies to deal with the problem. The first of these strategies is used primarily in private companies. Preferred shares are sold to third parties to finance the company while the common shares are deemed to have prices as low as one-tenth (or even as little as one-twentieth) of the value of the preferred shares.131 This permits generous common share stock grants to new executives, with relatively small immediate tax consequences.
The relative valuation of the classes of shares (often there are more than one 2020 CanLIIDocs 3123 series of preferred share issued) is almost certainly a fiction. Generally, the preferred shares used in US financings convert into common shares on a one-to-one basis. As the company approaches a successful exit, such as an IPO, when the conversion of the preferred shares seems likely, the differences in value between the two classes of shares declines and eventually disappears.132 Before this point the vast gulf in value between preferred and common shares is ostensibly due to the enhanced rights attached to those shares, in particular the preference in the event of liquidation.133 The value of this right is dubious: “the priority rights attaching to preference shares are relevant primarily in the event of bankruptcy, at which time neither the preference shares nor the common shares would likely realize anything.”134 While it might be argued that the additional rights often embedded in the terms of the preferred shares, such as anti-dilution protection or redemption rights, contribute to the disparity in value, they are used too infrequently to be sufficient justification.135
It is generally agreed by scholars examining American valuation practices in this area that these valuations are primarily sustained by the IRS’ refusal to challenge
131 See generally Ronald J Gilson & David M Schizer, “Understanding Venture Capital Structure: A Tax Explanation for Convertible Preferred Stock” (2003) 116:3 Harv L Rev 874 at 900–01, n 86. 132 Tingle, Start-up and Growth Companies in Canada, supra note 28 at 153. 133 Douglas J Cumming & Sofia A Johan, Venture Capital and Private Equity Contracting: An International Perspective (Burlington, Mass: Academic Press, 2009) at 5. 134 See Sandler, “Tax Treatment of Employee Stock Options”, supra note 5 at 295. 135 See Tingle, Start-up and Growth Companies in Canada, supra note 28 at 349–50 (in practice, anti- dilution rights are rarely employed); Douglas J Cumming & Sofia A Johan, Venture Capital and Private Equity Contracting, 2nd ed (London, UK: Elsevier, 2014) at 425–26 (examining venture capital exits and finding only 17 out of 187 were redemptions/retractions).
180 UNBLJ RD UN-B [VOL/TOME 71 them rather than by the economic fundamentals of the arrangements themselves.136 There is no parallel assumption in this country that Canadian tax authorities will accede to significant valuation differentials between preferred and common shares. One economist contacted three tax lawyers in three different provinces on the question of whether common shares can be priced at a fraction of existing preferred shares and received widely varying answers.137 The only thing they had in common was a distinct lack of confidence in the answers.
Most Canadian lawyers probably don’t believe the CRA would accept US- style valuation differentials between classes of shares. The evidence for this is that unlike the United States, only a minority of growth company financings in Canada use preferred share structures.138 In fact, the same American venture capital funds that use preferred share structures in the vast majority of their domestic investments, make less than one-quarter of their investments in Canada using these structures.139 Given the centrality of the tax valuation benefits afforded common shares in explanations for American preferred share financing structures, we can only conclude that their absence in Canada means the professionals involved in growth company financings in this 140 country don’t believe the tax benefits are available here. 2020 CanLIIDocs 3123
The second device used in the United States to ameliorate the tax consequences of share grants is to issue equity to executives that is, either on its terms or under a separate agreement, subject to certain conditions. These are referred to as “restricted stock grants” and the most common conditions provide that the executive may not sell, transfer or pledge the shares until some milestone has been reached. The shares are often held in escrow by a third party to ensure compliance with the conditions of the grant. The conditions for release can consist of the achievement of individual or corporate performance targets or simply remaining employed by the company until a certain date.141 Under US tax rules, the executive is not subject to tax
136 Victor Fleischer, “Taxing Founders’ Stock” (2011) 59:1 UCLA L Rev 60 at 74–75, n 58. 137 Douglas J Cumming, “United States Venture Capital Financial Contracting: Foreign Securities” in Mark Hirschey, Kose John & Anil K Makhija, eds, Advances in Financial Economics: Issues in Corporate Governance and Finance, vol 12 (Bingley: Emerald Group Publishing, 2007) 405 [Cumming, “United States Venture Capital Financial Contracting”]. See also Scott Ollivierre, “The Influence of Taxation on Capital Structure in Venture Capital Investments in Canada and the United States” (2010) 68:1 UT Fac L Rev 9 (“[t]here is no jurisprudence on point, and the CRA has not published any administrative practice in this regard, leaving it unclear how vigilant it is in enforcing the valuation of common shares at their true economic value” at 26). 138 Douglas J Cumming, “Capital Structure in Venture Finance” (2005) 11:3 J Corporate Finance 550 at 552 (preferred equity made up only 7.27% of investments in their results). 139 Cumming, “United States Venture Capital Financial Contracting”, supra note 137 at 408. 140 This is particularly the case given the many non-tax advantages preferred shares have over common shares in the context of financing entrepreneurial ventures. See Tingle, Start-up and Growth Companies in Canada, supra note 28 at 331–33. 141 Stikeman Elliott LLP, supra note 51 at 22.
2020] THE EQUITY INCENTIVE CANADIAN STARTUPS NEED 181 on the share grant until the restrictions lapse and even then an election can be made that effectively delays the taxable event until the shares are sold.142
The Canadian tax regime is not as patient. Restricted shares are taxed as income in the year they are issued.143 The CRA will, however, permit a reasonable discount from the fair market value of the unrestricted shares. Based on one report of discussions with the Valuation Section of Toronto’s Central Tax Services Office, “it appears that a discount in the range of 10% to 20% would generally be acceptable in the case of a 2 to 3 year holding period restriction.”144 Obviously this is not sufficient to materially reduce the tax penalty resulting from issuing shares to an employee.
It is a testament to the responsiveness of Canadian boards to evolving norms of corporate governance that share grants occur at all among growth companies on, say, the TSX Venture Exchange. It makes no sense to issue long-term compensation that requires its beneficiaries to pay tax today on stock that is illiquid, might be confiscated or cancelled, and could turn out to be valueless tomorrow. The tax regime in this country has thus made it challenging for companies to use stock options as their primary method of providing equity to employees. 2020 CanLIIDocs 3123
Since stock options are held in aversion by most commentators on executive pay and corporate governance, Canadian companies issue less equity than their international peers. For Canada’s largest companies this is an inconvenience. For the growth companies who need to be able to use their equity to attract talent, the situation constitutes one more drag on their growth prospects.
VI. THE EQUITY INCENTIVE CANADIAN STARTUPS NEED
There are a lot of things that could be done to fix Canada’s rules around equity incentives. Now that it appears the favourable treatment of stock options will largely be confined to growth companies, perhaps the tax regime could change so the moment tax is calculated and paid coincides with the moment the cash to pay the tax is received. The current incentives created by Canada’s tax rules around stock options are the opposite of what we would wish.145
It should be noted, incidentally, that the generalized antagonism towards stock options is probably overdone. Options do have the problems discussed earlier, but there is plenty of research suggesting options should never entirely disappear as a
142 IRC § 83(b). 143 ITA, supra note 4, s 7(1). 144 Julie Y Lee, “Stock Option Plans and Other Equity-Based Incentives” (Paper delivered at the 2010 Tax Law for Lawyers conference, 30 May–4 June 2010) [unpublished] at n 33, online (pdf): Canadian Bar Association
182 UNBLJ RD UN-B [VOL/TOME 71 component of executive compensation.146 Managers have a tendency to be risk averse: their income, personal prestige and access to corporate perquisites depend on their continuing employment by the company. Options can be used to make managers more risk neutral (less risk averse), but the proper balance of options with share grants and other components of executive compensation will depend on the personality of the executive, her circumstances, and the company’s business environment.147 The Canadian government should maintain a tax policy that provides a way for growth companies to use stock options effectively.
It is even more important that the tax rules change to facilitate the ability of startups to use share grants as a means to attract the talent they require to compete on the international stage. Often this means trying to hire individuals with the necessary expertise who currently reside in jurisdictions that do make share grants possible. In some sectors, such as tech, nearly all the necessary experts will have spent at least part of their career working outside of the country. There are simply not enough large software companies in Canada, for example, that a startup is likely to find a Canadian resident with decades of experience in enterprise sales. If a Canadian company cannot match, and exceed, the equity incentives a potential employee has in other countries, 2020 CanLIIDocs 3123 they will not be able to build their business. They will, instead, do as well as they can, then sell the company—usually to a foreign buyer.
There is a kind of precedent in our tax regime for deferring the tax paid on a share grant. We grant rollover treatment for a large number of transactions in which shares are received by a resident, deferring the moment tax must be calculated and paid to the moment when the underlying equity interest is sold.148 This analogy is not perfect, since rollover rules merely preserve a pre-existing tax basis, whereas deferring the tax due on the grant of shares involves ignoring the tax basis of the shares. However, the rollover rules do provide a relevant policy precedent as they exist because they are viewed as essential both for reasons of fairness (it is not fair to tax someone for a transaction that does not produce the means to pay that tax) and so as not to interfere with economically important transactions. Deferring the tax due on the grant of shares by a startup in appropriate circumstances, fits comfortably within
146 WM Gerard Sanders & Donald C Hambrick, “Swinging for the Fences: The Effects of CEO Stock Options on Company Risk Taking and Performance” (2007) 50:5 Academy Management J 1055 at 1074 (suggesting that without option incentives CEOs can tend to be too risk-averse); S Trevis Certo et al, “Giving Money to Get Money: How CEO Stock Options and CEO Equity Enhance IPO Valuations” (2003) 46:5 Academy Management J 643 at 650 (in the context of IPOs, CEO equity ownership combined with CEO stock options can lead to a better balance of proper risk taking); Kuntara Pukthuanthong, Richard Roll & Thomas Walker, “How Employee Stock Options & Executive Equity Ownership Affect Long-Term IPO Operating Performance” (2007) 13:5 J Corporate Finance 695 at 717 (finding that operating performance is better for firms with a balanced combination of equity ownership and stock options). 147 David Larcker & Brian Tayan, Corporate Governance Matters: A Closer Look at Organizational Choices and Their Consequences (Upper Saddle River, NJ: Pearson Education, 2011) at 271–78. 148 ITA, supra note 4, ss 85–87.
2020] THE EQUITY INCENTIVE CANADIAN STARTUPS NEED 183 this logic. It is worth keeping in mind that the majority of growth companies fail, and the deferred taxable value of the shares will, in fact, be zero.149
What would a tax regime that allowed startups to grant shares to employees look like? First, it would apply only to Canadian growth companies. This would have to go beyond CCPCs, which are, as we have seen, a pretty poor proxy for the kind of startups we care about. 150
The class of qualified companies would, for example, have to include firms listed on Canada’s public venture exchanges, as well as companies fortunate enough to attract sizeable investments from foreign angels or venture capital firms. The federal government is currently in the process of determining how to define growth companies in relation to their stock option rules; there is no reason not to adopt their eventual definition.
To qualify, a share grant would have to be made to an employee and it would have to be subject to vesting conditions. Few bona fide share grants are made to employees on an unrestricted basis. Part of their purpose is to induce employees to 2020 CanLIIDocs 3123 stick with the business for a certain period of time, so at a minimum they always require some period of continuous satisfactory employment.
If there was still concern for the possibility of abuse, the tax regime could impose a further condition that the share grant be approved by an independent board of directors. Startups of the type we are interested in here tend to acquire independent boards quickly because of their dependence on external, arms-length sources of finance.
Objections can always be raised that any tax deferral scheme creates an expense to the taxpayer, but there are good reasons to believe that in this case, the objection is overblown. Because startups use stock options as equity incentives now, the public fisc should not be affected by permitting a deferral on qualified share grants. Share grants would simply substitute for what would otherwise be a stock option grant. Indeed, given that shares are subjectively seen by their recipients as more valuable than options, the total amount of equity on which the tax is deferred is likely to go down to the extent share grants supplant option awards.
It seems like a false economy to insist on a tax structure that creates significant barriers to growth of the most important class of businesses in our
149 Industry Canada, Fisher & Reuber, supra note 27 at 9 (only about half of all new Canadian firms survive five years). 150 An example of what is to be avoided is the current CCPC and small business tax regime: Michael Wolfson et al, “Piercing the Veil: Private Corporations and the Income of the Affluent” (2016) 64:1 Can Tax J 1 (discussing the role of CCPCs in income inequality and finding that “as many as 80 percent of the taxfilers in the top 0.01 percent (one hundredth of a percent) were CCPC owners during this decade” at 11); Chen & Mintz, supra note 124 (finding “many small businesses are created to enable individuals to reduce personal tax rather than grow companies…” at 4).
184 UNBLJ RD UN-B [VOL/TOME 71 country.151 Facilitating the growth of the fastest growing, most innovative firms in Canada, should lead to greater tax revenues in much the same way that permitting economically efficient corporate transactions through the rollover regime in the long run benefits the public purse.
VII. CONCLUSION
The Federal government should not try to pick winners when it comes to compensation structures. The tax preference for options arose at the height of the corporate governance industry’s enthusiasm for stock options in the 1990s, but remained on the scene long after this enthusiasm waned. There are circumstances where it is in the best interest of a company to use a variety of equity plans to manage employee incentives and recruit talent; government tax policy should be neutral on what sort of incentives are used.
A historical survey of European and American compensation practices found tax policies in various countries had a major impact on the ebb and flow of stock option 2020 CanLIIDocs 3123 grants.152 The scholars found governments’ reactions tended to closely follow isolated events and controversies and characterized policy making in this area as “knee- jerk.”153 Canada’s government should resist the current fashion of denigrating options and move in the opposite direction to liberalize the tax regime—at least for growth companies—to permit executives to hold onto the shares purchased under an option long after exercise.
As well, this paper is not proposing something radical in relation to the taxation of share grants. The changes to the tax regime it suggests would merely put Canadian growth companies in the same position as their peers in the United States. The share grant regime there has existed for decades without creating intractable problems. It is hard to argue that it would be a mistake to emulate a small part of the regulatory regime that makes the US the most successful incubator of fast-growing and innovative businesses.
151 See earlier discussion at text accompanying notes 28–31. 152 Martin J Conyon et al, “The Executive Compensation Controversy: A Transatlantic Analysis” (2011) Cornell University ILR School Institute for Compensation Studies Working Paper No 2011-002 at 111, online: Institute for Compensation Studies
Review of Alan M. Sinclair & Margaret E. McCallum, An Introduction to Real Property Law, 7th ed. (Toronto: LexisNexis Canada, 2017)
Paul T. Babie∗
Alan M. Sinclair published Introduction to Real Property Law in 1969.1 While the magisterial Anger & Honsberger Law of Real Property was available to the practitioner from 1959,2 until the publication of Bruce Ziff’s Principles of Property Law in 1996,3 Sinclair’s book remained the only text available to Canadian law students on the intricacies of real property law. Through the third edition, published in 1987, Sinclair was the sole author of the book.4 In that third edition, Sinclair stated his aim, which was the book’s hallmark through each of the editions he authored:
2020 CanLIIDocs 3123 During several years teaching the elements of the law of real property to first year law students, it has been somewhat of a chore, not always an unpleasant one, to cover the material considered basic to an understanding of this important facet of the law. This factor, combined with the ever-increasing intrusion of new concepts and additional uses of old ones into the field leads today’s student and teacher to the realization that a succinct statement of the basics of the law is essential. Many of us pay less attention than perhaps we should to the fundamentals; but time controls. This small volume is an attempt to alleviate the difficulty and enable the student to assimilate these basics more readily and for the teacher to move more rapidly on a broader base.5
The third edition contained a mere 105 pages, and it lived up to Sinclair’s aim— to provide a succinct statement of the historical background to the doctrines of tenure and estates, and the rules which had been built around them to form the modern common law of real property. Uncluttered by footnotes or long expositions on the complexities of the historical and contemporary law, Sinclair’s book was the perfect resource for the law student.
∗ ALS Professor of Property Law, The University of Adelaide, Australia. 1 Alan M Sinclair, Introduction to Real Property Law (Toronto: Butterworths, 1969). 2 HD Anger & JD Honsberger, Canadian Law of Real Property (Toronto: Canada Law Book, 1959); HD Anger & JD Honsberger, Law of Real Property, 2nd ed by AH Oosterhoff & WB Rayner (Aurora: Canada Law Book, 1985); HD Anger & JD Honsberger, Law of Real Property, 3rd ed by Anne Warner La Forest (Aurora: Canada Law Books, 2006). 3 Bruce H Ziff, Principles of Property Law (Scarborough: Carswell, 1993); for latest edition, see: Bruce H Ziff, Principles of Property Law, 7th ed (Scarborough: Carswell, 2018). 4 The second edition appeared in 1982: Alan M Sinclair, Introduction to Real Property Law, 2nd ed (Toronto: Butterworths, 1982). 5 Alan M Sinclair, Introduction to Real Property Law, 3rd ed (Toronto: Butterworths, 1987) at vii.
186 UNBLJ RD UN-B [VOL/TOME 71
In 1997, Margaret McCallum joined Sinclair as an author for the fourth edition,6 and since then the book continues to attain succinctness—it remains slim, at just under 200 pages—while adding some light footnoting and a slightly fuller treatment of the complexities surrounding real property law.7 While joined by Ziff and an even more substantial Anger & Honsberger, Sinclair and McCallum’s An Introduction to Real Property Law remains an excellent primer for the Canadian law student.
In this review, I focus on one very important contribution contained in the seventh edition. In the Preface, McCallum offers this admonition:
The aphorism that property is theft, attributed to Pierre-Joseph Proudhon, is a rough summary of the following passage from his book, What is Property? Or, an Inquiry into the Principle of Right and of Government (1840):
If I were asked to answer the following question: What is slavery? and I should answer in one word, It is 2020 CanLIIDocs 3123 murder, my meaning would be understood at once. No extended argument would be required to show that the power to take from a man his thought, his will, his personality, is a power of life and death; and that to enslave a man is to kill him. Why, then, to this other question: What is property! may I not likewise answer, It is robbery, without the certainty of being misunderstood; the second proposition being no other than a transformation of the first?
Agreement with Proudhon does not necessarily follow from recognition that rules governing distribution of property rights do not ensure equal access to, or efficient allocation of, the world’s resources. That is not their purpose. As you read this book, think about why we have the property rules that it explains, and how you might organize the rights and obligations of owners of land if you were beginning anew.8
McCallum’s question challenges those who inquire into the nature and operation of real property law to ask about the reason for its rules and how we might want to change them if we seek a more just and equitable distribution of the earth’s goods and resources. Indeed, so important is McCallum’s admonition that one might wish it had been found not in the Preface, where it risks being overlooked, but in the Introduction, as the organizing theme of the entire book.
6 Alan M Sinclair & Margaret E McCallum, An Introduction to Real Property Law, 4th ed (Toronto: Butterworths, 1997). 7 For latest edition, see: Alan M Sinclair & Margaret E McCallum, An Introduction to Real Property Law, 7th ed (Toronto: LexisNexis Canada, 2017). 8 Ibid at ix.
2020] REVIEW: AN INTRODUCTION TO REAL PROPERTY LAW 187
If we used McCallum’s admonition as the lens through which to view real property law, we would see that it requires both description—what the law is, what it does—and normative argument—how the law should be and what it ought to do. Description would reveal not only the rights which comprise it, but also that property carries the potential to do more than distribute resources. It can, as Proudhon reveals, concentrate power in a few, allowing them also to exploit others.9 The potential for such concentrations requires deep thought, then, about the fact that the rules are not entirely neutral. Instead, they might be politico-ideological responses to the world in which we live. If that is so, then rather than a mere collection of rights, property is also deeply concerned with obligations, duties, and responsibilities that go along with the exercise of its rights. Such obligations do exist, but it often takes effort to find them in the midst of the array of rights thrown at us as we consider the law of real property. This difficulty is compounded for the uninitiated who struggle to come to terms with the doctrines of tenure and estates, the rule against perpetuities, and the contingent remainder rules, to name but four parts of the law of real property. But McCallum challenges all of us—the initiated and the uninitiated—as we work our way through the intricacies of what the law is, to turn our minds to what real property could be, indeed, what it ought to be. For it is only when we consider those obligations that do 2020 CanLIIDocs 3123 exist that we can determine where they might also be lacking, where they need alteration, where they need supplementing, and where we might need to add obligations not yet recognized.
The seventh edition provides three helpful guideposts for describing real property law and for making normative claims about its obligations—for thinking not only about what real property law is, but also about how it might be changed to meet new socio-politico-economic conditions. First, the revised Introduction not only contains an excellent overview of tenure and estates (as has every edition), but also a contextualization of Canadian property law. While focused on Canadian law, this proves valuable to anyone interested in the operation of real property systems in post- colonial common law jurisdictions. This contextualization makes it possible to consider the normative question: how can real property adapt to contemporary challenges so as to ensure that in addition to rights, its holders also bear responsibilities to others in exercising the powers conferred by those rights? The Introduction, then, forms an important adjunct to McCallum’s admonition.
A second guidepost explores how Canadian law has responded to the colonial dispossession of First Nations peoples and recognized rights to land and natural resources of those peoples as part of a response to colonial injustice. This is the single most important issue to face post-colonial legal systems.10 Since the Supreme Court
9 See Paul Babie, “A Great Exploitation: The True Legacy of Property”, Review Essay of The Great Demarcation: The French Revolution and the Invention of Modern Property by Rafe Blaufarb (2018) 31 Intl J Sem L 977. 10 In the Canadian context, see John Borrows, Canada’s Indigenous Constitution (Toronto: University of Toronto Press, 2010); Peter H Russell, Canada’s Odyssey: A Country Based on Incomplete Conquests (Toronto: University of Toronto Press, 2017).
188 UNBLJ RD UN-B [VOL/TOME 71
of Canada’s 1973 decision in Calder v British Columbia (AG),11 common law Aboriginal Title in Canada has grown in importance. The emergence of this form of title in Canada mirrors similar developments in other commonwealth jurisdictions, most notably in New Zealand12 and Australia.13 Any book on post-colonial real property therefore requires careful treatment of this significant topic; this is found in an entirely new chapter in the seventh edition.14 There is much to be learned from this chapter about the way in which real property law operated historically so as both to oppress and to work a grave injustice upon First Nations peoples. Equally importantly, the chapter provides guidance as to how the law of real property forms part of our modern response to past injustices. Indeed, First Nations law itself may contribute to an innovative re-purposing of property law to address contemporary challenges such as climate change. In response to the most recent report of the UN Intergovernmental Panel on Climate Change, for example, Indigenous leaders of 42 countries wrote that “[f]ailure to legally recognize our rights leaves our forests vulnerable to environmentally destructive projects that devastate forests and release massive amounts of carbon into the atmosphere.”15 As we consider the ways in which the rules of allocation might be changed so as to result in more equitable distributions of
resources, in ways that minimize harm to others, the interaction of common law and 2020 CanLIIDocs 3123 First Nations perspectives has the potential to bear much fruit.
Finally—and this may be of greatest importance to first-year law students— learning the law of real property is best achieved by doing; the only way fully to internalize the operation of tricky concepts is by continuous practice. For that reason, and while earlier editions of Sinclair always assisted in an understanding of the difficulties of common law real property, the seventh edition includes an entirely new section of review problems as well as a glossary of terms.16 These additions allow students to internalize the rules of real property law for use in preparing for the immediate focus—exams—and the longer-term goal—legal practice.
In short, the seventh edition, albeit in a slightly different way, achieves Sinclair’s original aim, as stated by McCallum: “to create order from the chaos of real property law, and to show the continuing significance of arcane and seemingly archaic common law doctrines.”17 But it does more than that alone, for it demands that we not
11 Calder v British Columbia (AG) [1973] SCR 313, 34 DLR (3rd) 145. See also Kent McNeil, Common Law Aboriginal Title (Oxford: Clarendon Press, 1989). 12 R v Symonds, [1847] NZPCC 387. 13 Mabo v Queensland (No 2), [1992] 175 CLR 1, 107 ALR 1. 14 See chapter 6 in Sinclair & McCallum, supra note 7. 15 “A statement on the Intergovernmental Panel on Climate Change (IPCC) Special Report on Climate Change and Land from Indigenous Peoples and local communities* from 42 countries spanning 76% of the world’s tropical forests” (2019), online: Indigenous + Community Response to IPCC Report
2020] REVIEW: AN INTRODUCTION TO REAL PROPERTY LAW 189 only seek to understand and bring order to what otherwise might be chaos, but that we also seek ways for ensuring that in the disorder, injustice is not worked. It asks us to think about both what exists, and what ought to exist.
2020 CanLIIDocs 3123
THE FEDERAL COURT OF APPEAL OF CANADA: A VIEW FROM THE INSIDE OF A BIJURAL AND BILINGUAL COURT
The Honourable Richard Boivin*
INTRODUCTION
Canada and Mexico share the continent of North America and are important trading partners. Our countries have strong cultural ties that have been forged through immigration and tourism. In addition, our legal and political systems share a degree of commonality. Against this background, I will provide an overview of the Federal Court of Appeal of Canada with a focus on its bijural and bilingual aspects and with a view to providing an understanding of the specificity of the Canadian judicial landscape.
2020 CanLIIDocs 3123
FEDERAL COURTS IN CANADA
From the outset, some context regarding the creation of the Federal Courts of Canada is required. Canada’s original constitution, which created the federal dominion of Canada over 150 years ago, in 1867, provides under section 101 that Parliament may establish Courts for the better administration of the laws of Canada.1 Under this provision, the predecessor of the Federal Courts, the Exchequer Court of Canada, was created in 1875.2 The Supreme Court of Canada was created at the same time via the same legislation.3
Given that the Exchequer Court was created by statute, it did not have inherent jurisdiction, meaning that it did not have the power to decide just any case; it could only decide cases where jurisdiction had been granted to it by Parliament. While the jurisdiction of the Federal Courts has changed over time, the Exchequer Court’s jurisdiction came to include matters such as actions and revenue cases against the Crown, as well as admiralty, intellectual property, and citizenship matters, just like the jurisdiction of the Federal Courts today. In 1971, the Federal Court, which was
* Judge at the Federal Court of Appeal (Canada). This is an updated, English version of a paper initially presented in English, Spanish, and French in México City on 11 October 2019 as part of a seminar organized by the Instituto Tecnológico Autónomo de México (ITAM). I am grateful to Bradley J Condon, WTO Chair Professor and Founding Director of the Centre for International Economic Law at ITAM, for organizing this event. I am also thankful to my law clerks David McLeod, Meg Jones, and Emily Atkinson for their invaluable assistance in preparing this paper. Any errors or omissions are solely my own. 1 Constitution Act, 1867 (UK), 30 & 31 Vict, c 3, s 91, reprinted in RSC 1985, Appendix II, No 5. 2 See The Supreme and Exchequer Court Act, SC 1875, c 11. 3 Ibid.
2020] A VIEW FROM THE INSIDE 191 comprised of a trial division and an appeal division, succeeded the Exchequer Court, principally for the purpose of addressing confusion regarding judicial review.4
What is meant by judicial review in Canada? It consists of reviewing the decisions of government boards, commissions, tribunals, and other governmental decision makers. In the years of the Exchequer Court, jurisdiction was shared with other courts, which created confusion in the jurisprudence and some ambiguity regarding the territorial scope of the application of provincial decisions across the country. Initially, the newly created Federal Court was therefore granted exclusive jurisdiction over federal judicial review. Later, in 2003, the Appeal Division became a separate Court rather than a division of the Federal Court.5 Since then, there have been two courts: the Federal Court, as well as the Federal Court of Appeal, where I now sit as a judge.
How does the Federal Court of Appeal operate? It can likely be compared in Mexico to los Tribunales colegiados de circuito. Like those Courts, the Federal Court of Appeal is a circuit court where judges sit in panels of three across Canada to rule upon matters of federal jurisdiction. All cases that come through the federal system 2020 CanLIIDocs 3123 are appealed to the Federal Court of Appeal. The only further appeal from our decisions is to the Supreme Court of Canada and parties must be granted leave by the Supreme Court in order for their appeal to be heard. The Supreme Court only hears a limited number of cases per year, usually between 60 and 70 since 2015.6 It follows that in practice, approximately 98% of our decisions are final.
There are 17 justices at the Federal Court of Appeal7 and roughly 500 proceedings are commenced before the Court every year. The judges appointed to the Federal Court of Appeal are representative of Canada’s geography and can be composed, from time to time, of judges coming from six different legal backgrounds representing the two legal systems, in French and in English. It follows that a judge may be as follows:
• Anglophone with a common law legal training; • Francophone with a common law legal training; • Anglophone with a civilian legal training; • Francophone with a civilian legal training; • Anglophone with a common law and civilian legal training; and • Francophone with a civilian and common law legal training.
4 See Federal Court Act, SC 1970-72, c 1. 5 See Courts Administration Service Act, SC 2002, c 8. 6 See “2018: Year in Review” (2019) at 1, 9, 13, online (pdf): Supreme Court of Canada
192 UNBLJ RD UN-B [VOL/TOME 71
The Court’s geographic jurisdiction covers all 10 provinces and three territories. As judges of a national court, we are required by law to be based out of Canada’s national capital, Ottawa, but we all travel across the country to hear cases, from Vancouver in the west, to Halifax in the east, and occasionally far up north to Whitehorse or Yellowknife. From their creation, the Federal Courts have therefore been providing access to justice in allowing litigants to appear before a national court in their home province or territory. This is a unique feature of the Federal Courts, compared to any other courts in Canada.
In addition, at the Federal Court of Appeal, as is the case throughout the federal judiciary in Canada, the justices were formerly members of a provincial bar prior to being appointed to the bench, meaning that they used to be lawyers. Indeed, unlike many countries, Canada selects and appoints its judges from the practice of law, as opposed to training jurists to become judges from the outset. Personally, I view this as a particular strength of the Canadian judiciary, which is composed of highly accomplished lawyers with extensive experience prior to joining the bench.
2020 CanLIIDocs 3123 THE JURISDICTION OF THE FEDERAL COURT OF APPEAL
So, what do we do at the Federal Court of Appeal? Our jurisdiction is broad and varied. For instance, we hear cases pertaining to administrative law, prison law, tax law, labour law, constitutional law, intellectual property law, and maritime law. Much of our docket is dedicated to hearing appeals of applications for judicial review of decisions from federal boards, commissions or tribunals. As noted earlier, a judicial review is when an affected person asks the Court to review the decision of an administrative decision maker, i.e. a decision that was made by a bureaucrat or a public servant, or by a federally created board, but was not decided in a court of law. Courts will show deference to administrative decision makers, given the executive’s prerogative to make these administrative decisions, provided such decisions are not unreasonable or arbitrary and remain within the bounds of legality. Otherwise, judicial intervention is required.
One area of the Federal Court of Appeal’s jurisdiction of interest, given that Mexico and Canada are both parties to NAFTA/USMCA, is over the Canadian International Trade Tribunal (the CITT). The CITT is one of the select administrative tribunals from which affected parties have direct access to judicial review at the Federal Court of Appeal and do not first have to apply for judicial review at the Federal Court. While pure NAFTA issues are not adjudicated by the CITT, many types of trade matters are. Many of the cases in front of the CITT are (1) unfair trading practice cases, i.e., dumping cases, where foreign manufacturers sell their products in Canada for less than the price for which they could sell the products in their home country, and (2) subsidy cases, where the CITT determines whether a given subsidy unfairly distorts trade.
2020] A VIEW FROM THE INSIDE 193
The specificity of Canada in this regard is that jurisdiction over such unfair practice cases is split between two bodies, rather than one. These include the Canada Border Services Agency (the CBSA), which is the agency responsible in Canada for facilitating the flow of both persons and goods and enforcing related legislation, i.e., border enforcement. The CBSA decides whether goods have been dumped or subsidized. In parallel, the CITT determines whether the dumping or subsidizing has injured Canadian producers or threatens to harm them in the future. In other words, the CBSA looks at what occurred, while the CITT looks at the effect of what occurred. If either the CBSA or the CITT reaches a negative determination, the inquiries of both bodies end immediately.8
In coming to its decisions, the CITT considers relevant decisions of the World Trade Organization (the WTO). While WTO decisions are not binding on the CITT, they are persuasive authority and are regularly argued by parties and relied upon by the CITT.
The CITT also has jurisdiction to inquire into issues that are referred to it by the government and to make recommendations. These inquiries notably include what 2020 CanLIIDocs 3123 are called “safeguard inquiries”, which allow the CITT to recommend that Canada temporarily restrict imports to allow Canadian producers to adapt. For example, in April 2019, in the context of the changing international trading environment regarding steel, the CITT recommended that a tariff rate quota be placed on two classes of steel goods because they were “being imported in such increased quantities and under such conditions as to be a principal cause of a threat of serious injury to the domestic industry.”9
Another area of interest is the jurisdiction of the Federal Courts over immigration law matters, more specifically the Immigration and Refugee Board. This is a wide area of jurisdiction for the Federal Court, which is the court of first instance. Immigration matters often make up more than half of all matters that the Federal Court considers in a year.10 A large part of these matters consists of applications for judicial review of refugee decisions.
The context for refugee decisions is usually as follows. When someone comes to Canada and files a claim for asylum, their file is handled by the Refugee Protection Division of the Immigration and Refugee Board. They have a hearing in front of one of the members of the Board, during which the Board member asks the claimant
8 See Canadian International Trade Tribunal, Anti-Dumping Injury Inquiries - Guide (26 January 2017), online (pdf): Government of Canada
9 Safeguard Inquiry into the Importation of Certain Steel Goods: Inquiry No GC-2018-001 (Ottawa: Canadian International Trade Tribunal, 3 April 2019), online: Government of Canada
10 See “Statistics” (last modified 31 December 2019), online: Federal Court
194 UNBLJ RD UN-B [VOL/TOME 71 questions about their case in order to determine whether they should be granted refugee status. Once the hearing is concluded, depending on the circumstances, the Board will either issue the decision immediately or issue a written decision at a later date. If the refugee claimant receives a negative decision, in most cases they can appeal to the Refugee Appeal Division of the Board in writing. If the appeal also results in the rejection of their claim, it is at this point that the claimant can apply for judicial review at the Federal Court. If the application for judicial review is accepted by the Federal Court, the usual remedy is to send the case back to the Immigration and Refugee Board to be re-decided in accordance with the Federal Court’s reasons.
In most cases, the Federal Court is the last stop for immigration matters, including refugee cases. Federal Court decisions on immigration matters can be appealed to the Federal Court of Appeal, but only where the Federal Court has certified the case to move forward because it raises a “serious question of general importance”, i.e., the case will have implications that will extend beyond the confines of the particular facts of the case. Given this requirement, the amount of immigration cases heard by the Federal Court of Appeal is lower than at the Federal Court level, but the cases that we do see often raise interesting, new, and important issues. In addition to 2020 CanLIIDocs 3123 refugee issues, these cases concern other immigration matters such as sponsorship applications, inadmissibility to Canada, and deportation.
Another important area of the Federal Court and the Federal Court of Appeal’s jurisdiction is national security. While the Federal Courts have had jurisdiction over this domain for a very long time, national security cases have taken more prominence and have attracted more interest since the events of September 11th, 2001, which had a serious impact on the way that the Canadian government and society view security and approach security issues.
Jurisdiction in this regard largely stems from the Canadian Security Intelligence Service Act,11 which sets out the law regarding Canada’s intelligence agency (that goes by the name “Canadian Security Intelligence Service”, or “CSIS” for short), as well as the Canada Evidence Act12 and the Immigration and Refugee Protection Act.13 National security cases may, for example, concern whether the government has to disclose information in an individual’s file to that individual, usually in criminal or immigration proceedings where the government has chosen to redact the file for security purposes. There is therefore a need to balance fairness concerns and individual rights with the interests of state security.
11 Canadian Security Intelligence Service Act, RSC 1985, c C-23. 12 Canada Evidence Act, RSC 1985, c C-5. 13 Immigration and Refugee Protection Act, SC 2001, c 27.
2020] A VIEW FROM THE INSIDE 195
BIJURALISM AT THE FEDERAL COURT OF APPEAL
A unique feature of Canada’s legal system, dating back to the Québec Act of 1774,14 is bijuralism. Indeed, two systems of law have coexisted and still coexist in Canada: common law and civil law. Most lawyers are only trained in the legal tradition of their home province, but some lawyers are trained in both civil and common law. In Québec, the French majority-speaking province in Canada that was historically colonized by France, the law is based on the civilian tradition. Therefore, Québec, like Mexico, has a civil code and emphasizes the primacy of written laws. The Civil Code of Québec (CCQ), which was adopted in its original form in 1866,15 is inspired by the 1804 French Code Napoléon. The CCQ serves as a cornerstone of the legal system in Québec. It is noteworthy that the Federal Courts Act provides that at least five judges from the Federal Court of Appeal and 10 judges from the Federal Court must be from Québec, thus ensuring that a reasonable proportion of judges within the Federal Courts have been trained in the civil law tradition.16 The Supreme Court Act also provides that at the Supreme Court of Canada, a statutory Court created at the same time as the Exchequer Court (the ancestor of the Federal Courts) under the same constitutional 17 provision, at least three of the nine judges be appointed from Québec. Hence, the 2020 CanLIIDocs 3123 Supreme Court Act ensures, in the same way as the Federal Courts Act, that a third of its judges will be civilians, thus recognizing the importance of bijuralism at the federal level.
The remaining Canadian provinces follow a common law system that is based on the common law tradition of the other historical colonial power in Canada, namely England. This system, like in the United States, emphasizes the importance of judicial precedent. Common law is built through new cases that follow, build, distinguish, and occasionally overturn, previous case law. While the Mexican system incorporates judicial precedent into its legal rule-making through the concept of jurisprudencia,18 in Canada, case law is binding in a more strict and immediate sense, in that a single decision of a higher Court which articulates a legal principle is binding on decisions of lower Courts.
While federal public law matters are handled uniformly, given that Canada’s national constitution allows Parliament to pass legislation that applies across the country in certain subject matters, such as criminal law, private law matters are generally handled differently depending on the province in question. Even amongst the common law provinces, at times the law can develop quite differently from one
14 British North America (Québec) Act, 1774, 14 Geo III, c 83 (UK). 15 CCLC. 16 Federal Courts Act, RSC 1985, c F-7, s 5.4. 17 Supreme Court Act, RSC 1985, c S-26, s 6. 18 Similar to the French notion of jurisprudence constante. Among other requirements, a court must make five consecutive and uninterrupted decisions (i.e. without a decision in between going the other way) on an issue in order for the principle to become binding on lower courts.
196 UNBLJ RD UN-B [VOL/TOME 71 province to another. Indeed, pursuant to section 92(13) of the Constitution Act, 1867,19 provinces have exclusive legislative jurisdiction over property and civil rights.
There will thus be situations where private law, be it common law or civil law, will apply directly to federal statutes to fill the void not addressed in federal law. This is called the principle of complementarity and it was codified in 200120 by section 8.1 of the Interpretation Act:21
8.1 Both the common law and 8.1 Le droit civil et la common the civil law are equally law font pareillement autorité authoritative and recognized et sont tous deux sources de sources of the law of property droit en matière de propriété et and civil rights in Canada and, de droits civils au Canada et, unless otherwise provided by s’il est nécessaire de recourir à law, if in interpreting an des règles, principes ou notions enactment it is necessary to appartenant au domaine de la refer to a province’s rules, propriété et des droits civils en principles or concepts forming vue d’assurer l’application
part of the law of property and d’un texte dans une province, il 2020 CanLIIDocs 3123 civil rights, reference must be faut, sauf règle de droit s’y made to the rules, principles opposant, avoir recours aux and concepts in force in the règles, principes et notions en province at the time the vigueur dans cette province au enactment is being applied. moment de l’application du texte.
Prior to the enactment of section 8.1, the principle of complementarity had been formulated and applied by courts in a number of decisions, one of the seminal decisions being St-Hilaire v Canada (Attorney General).22 This case concerned whether a Québec resident (the wife) was entitled to survivor benefits under a federal government pension plan to which her husband had contributed, in light of the fact she had pleaded guilty to manslaughter in connection with his death. The federal statute governing the pension plan was silent on eligibility requirements of surviving spouses to receive benefits. As such, the Court had to decide whether the common law or the civil law of Québec should apply to determine the wife’s entitlement to benefits. Justice Décary drew upon the principle of complementarity in holding that when a federal statute refers to a private law concept without defining it, the private law of the province where the dispute arose “fills the void”.23 He also noted that Canada’s Constitution enshrines “the federal principle that a federal law that resorts to an external source of private law will not necessarily apply uniformly throughout the
19 Supra note 1. 20 See Federal Law–Civil Law Harmonization Act, No 1, SC 2001, c 4, s 8. 21 Interpretation Act, RSC 1985, c I-21. 22 St-Hilaire v Canada (AG), 2001 FCA 63. 23 Ibid at para 51, Décary JA.
2020] A VIEW FROM THE INSIDE 197 country.”24 In the circumstances, the civil law of Québec was applied in determining the wife’s eligibility for benefits because her civil rights were at issue. The other Justices hearing the appeal endorsed Justice Décary’s reasoning on the applicability of the civil law, although they disagreed with his interpretation of the relevant legal provisions. They found that the wife was not entitled to any benefits in the circumstances.
More recently, in Canada v Raposo,25 our Court reiterated the principle of complementarity involving the definition of “contract of partnership and association” pursuant to article 2186 of the CCQ. In this case, the Court had to determine whether a member of a gang that was involved in drug trafficking in Québec was jointly and severally liable for sales tax owed from drug sales. This determination turned on whether the gang was a partnership and thus how “partnership” should be defined under the Excise Tax Act26 (the ETA). The parties agreed that since the ETA did not define “partnership”, the principle of complementarity, enunciated in St-Hilaire, applied and that “partnership” had to be defined according to the CCQ. The disagreement revolved around which Code provisions applied. The appellant argued that only one provision that defined contracts of partnership applied and that in the 2020 CanLIIDocs 3123 circumstances, a partnership could be found to exist and the respondent gang member could be jointly and severally liable for the debt. The Court of Appeal preferred a broader approach, noting that provisions regarding contracts in general applied to contracts of partnership. It held that, in this case, federal law did not exclude the application of Code provisions that made void any contract whose object is contrary to law or public order. The Court of Appeal therefore upheld the lower court’s finding that the contract of partnership linking the gang members in the case was void and that the respondent was not jointly and severally liable for the tax debt.
There have been attempts to harmonize the law across provinces and across legal traditions where possible and desirable, but there is also recognition that having diversity in rules across the country is one of the realities (and an acceptable reality) of belonging to a federalist state, and that this brings a sort of vibrancy to our legal system and allows provinces to learn from each other. It has been said by a former Judge of the Supreme Court of Canada that “while civil law and common law complement the private law provisions of federal legislation, at the same time, federal legislation should not be applied uniformly throughout the country in every respect”.27 This is an interesting aspect of being a judge at the federal level in Canada, as opposed to the provincial level: we can be called upon to interpret laws and decide cases based
24 Ibid at para 35, Décary JA. 25 Canada v Raposo, 2019 FCA 208. 26 Excise Tax Act, RSC 1985, c E-15. 27 The Honourable Mr. Justice Michel Bastarache, “Bijuralism in Canada” (delivered at the Department of Justice Lunch and Learn Workshop on Bijuralism and the Judicial Function in Ottawa, Ontario, 4 February 2000), in Department of Justice Canada, Bijuralism and Harmonization: Genesis (Ottawa: Department of Justice Canada, 2001) at 19, online (pdf): Government of Canada
198 UNBLJ RD UN-B [VOL/TOME 71 on both common and civil law notions and understandings and try to unify the Canadian legal system as a whole, while also recognizing some necessary differences in application across the country.
In this spirit, a committee composed of members of the Québec Bar in private practice, lawyers from the Department of Justice of Canada, and members of the judiciary, including judges from the Federal Court, the Federal Court of Appeal, and the Québec provincial superior Court, recently developed a pilot project with a view to allowing lawyers from the Québec Bar to use the Code of Civil Procedure (CCP) in certain proceedings before the Federal Court and the Federal Court of Appeal,28 instead of the Federal Courts Rules,29 which have been drafted with common law principles in mind. This initiative is inspired by a desire for a closer connection with Québec jurists and is consistent with bijuralism. It aims to extend the existing substantive bijuralism of the Courts to include procedural bijuralism and allows members of the Québec Bar to conduct proceedings under the familiar CCP before the Federal Court and the Federal Court of Appeal.
2020 CanLIIDocs 3123 BILINGUISM AT THE FEDERAL COURT OF APPEAL
Canada’s legal and judicial system is also bilingual, reflective of the country’s two official languages: English and French. At the provincial level, there are varying bilingualism policies, with most provinces being unilingual. In some provinces, laws are written in both English and French, legal services are offered in both languages, and citizens can use English or French in the Courts; in others, only English may be used. The Canadian Constitution does not explicitly or implicitly provide for the use of Indigenous languages before courts. However, in some of Canada’s territories, which are located in the north of the country, some Indigenous languages are recognized territorially as official languages, such as Inuktitut, Chipewyan, and Cree.30 Certain Indigenous languages have also been recognized in Québec.
At the federal level, the Parliament of Canada is required by section 133 of the Constitution Act, 1867 to enact legislation in English and in French and both versions are of equal authority, meaning that neither is viewed as a translation of the “true” law—both are regarded as “true”. Section 18 of the Canadian Charter of Rights and Freedoms31 further provides that both versions “are equally authoritative”. Hence, lawyers who appear before the Federal Court and the Federal Court of Appeal must
28 Federal Court & Federal Court of Appeal, Notice to the Parties and the Profession re formal implementation of the pilot project on procedural bijuralism (28 November 2019), online (pdf): Federal Court
2020] A VIEW FROM THE INSIDE 199 provide the judges with citations in both official languages in order to allow judges of the Federal Court and the Federal Court of Appeal to take into account both linguistic versions of a statute. Judges take this very seriously. For instance, a panel of the Federal Court of Appeal has in the past disavowed a decision rendered by a different panel for lack of comparing both versions of a statute.
It is also noteworthy that decisions rendered by the Federal Court of Appeal and the Federal Court have to be translated pursuant to subsection 20(1) of the Official Languages Act,32 which has been recognized by our Court as a quasi-constitutional Act.33 In most cases, the Court’s decision will be issued in one language and translated afterwards. The delay for translation is variable. Exceptionally, translations of decisions of national importance will be prioritized in order to allow the release of the decision simultaneously in both official languages. The Federal Courts Act also provides that “each decision reported in the official reports shall be published therein in both official languages”.34
At the Federal Court of Appeal, we regularly hear cases in both English and
French and sometimes we even hear bilingual proceedings, with translators for the 2020 CanLIIDocs 3123 parties if necessary, during which one party pleads in English and one party pleads in French. Given the protection that both French and English are offered in Canadian institutions, the Court also often hears cases about language rights, bilingualism, and official languages.
Moreover, the right to use either English or French in proceedings at the Federal Courts is a constitutionally protected language right. The Federal Court of Appeal reiterated this principle in Industrielle Alliance, Assurance et services financiers inc v Mazraani.35
In this case, language issues arose at first instance in several ways. Witnesses and counsel for the intervener expressed their preference for speaking in French. The trial judge encouraged them to speak in English because of their English language skills and the fact that one of the parties was unilingual in English. The unilingual party also said he would need an interpreter if the trial were to proceed in French, but no interpreter was provided. Parts of the testimony and counsel’s argument for the intervener were ultimately delivered in French, as the witnesses and counsel struggled to speak in English at times.
32 Official Languages Act, RSC 1985, c 31 (4th Supp). 33 See Canada (AG) v Viola (1990), [1991] 1 FC 373 at 386–87, 123 NR 83 (CA). See also Mazraani v Industrial Alliance Insurance and Financial Services Inc, 2018 SCC 50 at para 25 [Mazraani SCC]. 34 Supra note 16, s 58(4). 35 Industrielle Alliance, Assurance et services financiers inc v Mazraani, 2017 FCA 80, aff’d Mazraani SCC, supra note 33.
200 UNBLJ RD UN-B [VOL/TOME 71
The Federal Court of Appeal found that the trial judge had violated the language rights of the unilingual party and the intervener’s witnesses and counsel. The Court emphasized that any person who appears before or submits written pleadings to a Federal Court has the constitutional right to use the official language of his or her choice and that a person’s ability to use either language does not impact this right. The Court also drew upon federal legislation that requires Federal Courts to provide simultaneous interpretation to any party who requests it and to ensure that any person giving evidence before the court can be heard in the official language they choose to use. The Court found that the trial judge should have adjourned the hearing to arrange for interpretation services, instead of impinging on participants’ language rights in the name of expediency. The Court ordered a new trial before a different judge.
Mr. Mazraani appealed the Federal Court of Appeal decision to the Supreme Court of Canada and leave was granted. The Supreme Court of Canada unanimously upheld the Federal Court of Appeal’s decision. It emphasized that a person's choice of language before Federal Courts must be free and informed and that judges in Federal Courts are required to protect language rights. It concluded that the language rights violations in issue had clearly impacted the hearing, witnesses, and parties and had 2020 CanLIIDocs 3123 also “brought the administration of justice into disrepute.”36
Furthermore, bilingualism amongst judges of the Federal Court of Appeal has been in constant progression in recent years. Although not all judges in our Court qualify as fluently bilingual, there has been much improvement, at least with respect to the judges’ understanding of both official languages, which are both used in Court meetings.
THE FUTURE FOR INDIGENOUS LANGUAGES AND LEGAL TRADITIONS?
Bilingualism and bijuralism are well established in Canada’s legal framework and are emblematic of the Federal Court and the Federal Court of Appeal. It is also fair to say that a lot has been achieved in the past decades in order to enhance these aspects of Canada’s judicial system, for example, by way of the appointment of more bilingual judges to the bench.
The next challenge ahead is to find ways to include Indigenous languages and legal traditions in the judicial system and/or to assist in supporting the use of Indigenous legal systems alongside the system embodied by federal and provincial courts. The reflection has begun, as the following examples from the Federal Courts illustrate:
36 Mazraani SCC, supra note 33 at para 78.
2020] A VIEW FROM THE INSIDE 201
• The Federal Court has an Aboriginal Law Bar Liaison Committee,37 as well as Aboriginal Practice Guidelines38 that aim to provide more flexibility in tailoring proceedings to meet particular circumstances by addressing the use of Indigenous languages, oral history, and Elder evidence.
• A witness testified in Cree in Benoit v Canada.39 A translation of his testimony was introduced into evidence. Testimony in Cree was also received in Montana Band v Canada.40
• The Federal Court has issued written and oral summaries of four decisions since May 2019 in Indigenous languages,41 including Whalen v Fort McMurray No 468 First Nation (in Cree and Dené),42 Jim Shot Both Sides v Canada (in Blackfoot),43 Thomas v One Arrow First Nation (in Plains Cree),44 and Watson v Canada (in Cree).45
• The Federal Court has recognized the validity of Indigenous laws 2020 CanLIIDocs 3123 upon considering custom elections referenced under the Indian Act.46 The Federal Court of Appeal also recently observed that a custom election code formed part of a First Nation’s law.47
37 See “Indigenous Bar Association - Aboriginal Law Bar Liaison Committee”, online: Federal Court
202 UNBLJ RD UN-B [VOL/TOME 71
In addition, there are federal statutes that recognize the power of Indigenous Peoples to develop and implement law, including the following:
• the First Nations Fiscal Management Act;48
• the First Nations Goods and Services Tax Act;49
• the First Nations Jurisdiction over Education in British Columbia Act;50
• the First Nations Land Management Act;51
• the First Nations Oil and Gas and Moneys Management Act;52
• the Indian Act;53 and
• An Act respecting First Nations, Inuit and Métis children, youth, and families.54 2020 CanLIIDocs 3123
Of particular note is the Indigenous Languages Act.55 It aims to support and promote the use of Indigenous languages, support efforts of Indigenous Peoples to reclaim, revitalize, maintain, and strengthen Indigenous languages, and facilitate funding for those efforts, and establish a framework to facilitate the effective exercise of Indigenous Peoples’ rights related to languages, among other purposes.56 It provides for access to federal government services in Indigenous languages,57 as well as translation and interpretation by federal institutions,58 and establishes an Office of a Commissioner of Indigenous Languages.59 Some of these sections have already come
48 First Nations Fiscal Management Act, SC 2005, c 9, ss 5, 9. 49 First Nations Goods and Services Tax Act, SC 2003, c 15, ss 4, 12, 67. 50 First Nations Jurisdiction over Education in British Columbia Act, SC 2006, c 10, s 9. 51 First Nations Land Management Act, SC 1999, c 24, ss 6, 15–16, 20. 52 First Nations Oil and Gas and Moneys Management Act, SC 2005, c 48, s 35. 53 Indian Act, supra note 46, ss 81, 83. 54 An Act respecting First Nations, Inuit and Métis children, youth, and families, SC 2019, c 24, ss 18, 20– 24. 55 Indigenous Languages Act, SC 2019, c 23. 56 Ibid, note 55, s 5. 57 Ibid, s 10.1. 58 Ibid, s 11. 59 Ibid, ss 12–42.
2020] A VIEW FROM THE INSIDE 203 into force, while others are set to follow by October 1, 2020.60
CONCLUSION
Many of the Federal Courts’ practices that uphold bijuralism and bilingualism in relation to the Constitution and Canada’s official languages are well established and robust at an institutional level. However, practices aimed at supporting the use of Indigenous languages and legal traditions tend to be implemented on an ad hoc basis. Academics in the field of Aboriginal law suggest that it is important both to strengthen Indigenous methods of dispute resolution and to facilitate incorporation of Indigenous languages and legal traditions into Canadian courts.61 Hence, although progress has been accomplished to include Indigenous languages and their legal traditions in the judicial system, it remains to be seen at this juncture how Indigenous languages and legal traditions will continue to operate alongside and within Canada’s Federal Courts system.
2020 CanLIIDocs 3123
60 Indigenous Languages Act: Order Fixing the Days on which Certain Sections of that Act Come into Force, SI/2019-93, (18 September 2019) C Gaz II, vol 153, 6336. 61 John Borrows, “The Role of Governments and Courts in Entrenching Indigenous Legal Traditions” in John Borrows, Canada’s Indigenous Constitution (Toronto: University of Toronto Press, 2010) 177 at 206– 18. See also Alan Hanna, “Spaces for Sharing: Searching for Indigenous Law on the Canadian Legal Landscape” (2018) 51:1 UBC L Rev 105 at 108–09, 155.
LOCAL REPRESENTATION AND LITIGANT VOICE: THE STORY OF CLASS ACTIONS IN NEWFOUNDLAND AND LABRADOR
Suzanne Chiodo*
Introduction
Three provinces have dominated the development of class actions in Canada. Québec was the first province to adopt the procedure in 1978, followed more than a decade later by Ontario. In 1996, British Columbia became the third province to enact class actions legislation. The influence of these three provinces on the genesis of class action case law and policy in Canada is widely accepted both in this country and in other jurisdictions.1 All the Canadian provinces, with the exception of Prince Edward Island, 2020 CanLIIDocs 3123 have now enacted generic opt-out class actions legislation.2 Because Ontario, British Columbia, and Québec were the first to enact class actions legislation and have dominated the field since, the presumption is that the other provinces simply followed their lead. The historical record shows this presumption to be false.
Around the turn of the millennium, between the introduction of class actions in British Columbia in 1996 and the Supreme Court of Canada’s decision in Western Canadian Shopping Centres Inc v Dutton3 in 2001, several provinces engaged in lively debates as to whether they should introduce class actions. These debates were influenced by developments elsewhere in Canada, but also arose from developments within and unique to the provinces themselves.4 This phenomenon is clearly seen in
* Assistant Professor, Western Law, London, Ontario. I thank Rande Kostal for his comments on drafts of this paper. All errors are the author’s own. 1 For example, in her book The Class Action in Common Law Legal Systems: A Comparative Perspective (Oxford: Hart Publishing, 2004), Rachael Mulheron looks only at Ontario and British Columbia, even though three other common-law provinces had enacted class actions legislation at that time: Manitoba, Saskatchewan, and Newfoundland and Labrador. 2 I define a class action or class proceeding as a procedural mechanism that is opt-out (i.e. people meeting the class definition are included in the class, and are bound by any judgment affecting the class, unless they take steps to exclude themselves) and generic (i.e. applicable across areas of the substantive law). See also Rachael Mulheron, Reform of Collective Redress in England and Wales: A Perspective of Need (London, UK: Civil Justice Council, 2007). 3 Western Canadian Shopping Centres Inc v Dutton, 2001 SCC 46 [Dutton]. 4 See e.g. the lengthy and considered treatment of class action procedures and their alternatives in Alberta Law Reform Institute, Report on Class Actions (2000), online: (pdf) University of Alberta
2020] LOCAL REPRESENTATION AND LITIGANT VOICE 205 the fourth Canadian province to enact class actions legislation: Newfoundland and Labrador.5
This article demonstrates that, while the debate on class actions in Newfoundland and Labrador was partly the manifestation of a national phenomenon, the province did not enact legislation simply because Ontario, British Columbia, and Québec had. The historical record shows that class actions in the province arose from grass roots activism that reflected the needs of its people. To a large extent, that remains true in the practice of class actions there today. This unique aspect of the province’s procedural story has never been told.
The movement for class actions in Newfoundland was led by two key figures in the province’s recent history: Bill Kelly and Chesley Crosbie. The impetus for change arose from the Hiland Insurance bankruptcy, an insurance scandal that affected tens of thousands of people in the province and led to calls for a public inquiry into the government’s failure to protect policyholders. In November 2001, Bill 34, “An Act To Permit An Action By One Person On Behalf Of A Class Of Persons” was introduced and, after less than a month of debate in the legislature, the Newfoundland 2020 CanLIIDocs 3123 Class Actions Act6 received Royal Assent on December 13, 2001 and came into force on April 1, 2002.
In July 2001, with the Supreme Court of Canada’s decision in Dutton,7 class actions became available at common-law and, within a few years, most provinces had enacted legislation. Since then, class action activity in Newfoundland has been dwarfed by that of Ontario, British Columbia, and Québec. This is not simply because of the population difference, but also because national opt-out classes (commenced largely in Ontario) have proven far more lucrative than local actions. Nevertheless, despite the prevalence of national opt-out class actions, local class actions have continued and are, in many ways, more representative of provincial concerns. An investigation of the history of class actions in these provinces illustrates the importance of local representation and litigant voice, two crucial considerations that need to be borne in mind as class actions in Canada become increasingly national in scope.
The history of class actions in Canada has, until recently, only been investigated at a fairly superficial level. My book on class actions in Ontario was the
5 Class Actions Act, SNL 2001, c C-18.1. Throughout this article, references to Newfoundland will include Newfoundland and Labrador. Since its passage, the only amendment to the Class Actions Act has been on 16 December 2004, when the wording of section 2(ii) was changed from ‘common but not necessarily identical issues of law that arise’ to ‘common but not necessarily identical issues of law that arise from common but not necessarily identical facts’. This appears to be more of a clarification than a substantial amendment. 6 Ibid. 7 Dutton, supra note 3.
206 UNBLJ RD UN-B [VOL/TOME 71 first serious historical treatment of this area of the law,8 and my research has now turned towards the history of all the Canadian provinces and territories. This includes Newfoundland, where the history of class actions has not been investigated at all.9
This article will proceed in three parts. Part I looks at the representative action, which forms the roots of class actions in Canada. The limitations of the representative action gave rise to concerns in Canada that significant claims, especially consumer claims, were going uncompensated. Part II investigates this phenomenon in Newfoundland, where the Hiland Insurance bankruptcy and the inability of consumers to obtain redress in the courts led to calls for class action reform. Part III looks at the evolution of class actions in Newfoundland since the passage of the legislation, and the nature and effect of the class actions that have been litigated in the province in the past two decades.
Part I—The Representative Action
The representative action is a creature of equity with its roots deep in English history. 2020 CanLIIDocs 3123 Following the fusion of the courts of law and equity, representative actions could be brought at common law in England and in the Canadian jurisdictions that had inherited the English system: that is, all the provinces except Québec.10 However, the restrictive interpretation of the “same interest” test that the English courts applied after Markt11 was also applied by the courts of the Canadian provinces. Few representative actions could satisfy the three-part test articulated in Ellis:12 common interest; common grievance; and relief beneficial to all. In particular, actions that were based on separate contracts, or where damages were individual to each class member, were generally not allowed to proceed as representative actions.13
8 Suzanne Chiodo, The Class Actions Controversy: The Origins and Development of the Ontario Class Proceedings Act (Toronto: Irwin Law, 2018). 9 There is a rich vein of legal historical writing about the province, although the historiography of Newfoundland’s legal history has, until recently, been non-existent: Christopher English, “The Legal Historiography of Newfoundland” in Christopher English, ed, Essays in the History of Canadian Law: Two Islands, Newfoundland and Prince Edward Island (Toronto: Osgoode Society for Canadian Legal History, 2005). 10 When the courts of common law and equity were fused, the representative action rule was enacted in the Rules of Procedure Schedule to the Supreme Court of Judicature Acts of 1873 and 1875: The Supreme Court of Judicature Act, 1873 (UK), 36 & 37 Vict 8, c 66, Sch Rule of Procedure, s 10 and The Supreme Court of Judicature Act, 1875 (UK), 38 & 39 Vict 10, c 77, Order XVI, First Sch, s 9. 11 Markt & Co Ltd v Knight Steamship Co Ltd, [1910] 2 KB 1021, 7 WLUK 25 (CA). 12 Duke of Bedford v Ellis (1900), [1901] AC 1 (HL), [1900] 12 WLUK 42 (HL). 13 See e.g. Walker v Billingsley, [1952] 4 DLR 490 at 493–500, 5 WWR (ns) 363 (BCSC) where the BC Supreme Court held that a representative action by members of a trade union for damages could not be brought because each group member had individual and differing damages claims. The claim for an accounting, however, was allowed to proceed.
2020] LOCAL REPRESENTATION AND LITIGANT VOICE 207
This remained the same throughout Canada until the 1970s, when the courts in British Columbia14 and Ontario began to be more receptive to group litigation following the introduction of class actions south of the border and in Québec, and as the need for a mechanism to provide legal redress for groups became more apparent. The more expansive interpretation of the representative action rule took two forms. First, the bifurcated approach whereby a declaration of liability was obtained that could then form the basis for subsequent actions for monetary damages; and second, the liberal interpretation of the “same interest” test.15
In Chastain v British Columbia Hydro and Power Authority,16 the plaintiff sought a declaration that the defendant had no authority to request a security deposit from new customers with low incomes, as well as an injunction forbidding the defendant from keeping deposits already collected. The Court found that the group members all had the “same interest” in the success of the action even though different amounts of money would be returned to each customer.17 In Shaw v Real Estate Board of Greater Vancouver,18 the group members’ common interest in the success of the action was found to meet the “same interest” requirement (even though calculating their individual entitlements by way of an accounting would be “long, detailed and 2020 CanLIIDocs 3123 difficult”),19 and the Court of Appeal held that the money owed to class members could be calculated following a finding of liability.20
This approach influenced the limited case law in Newfoundland, where Rule 7.11, the Newfoundland rule that permitted representative actions, was worded in a similar way to its Ontario and BC equivalents, with the same requirement that “numerous persons have the same interest” in a proceeding.21 One of the few
14 Representative actions in BC were governed by Ordinance 16, Rule 9 of the Rules of the Supreme Court (1961), which was very similarly worded to its counterparts in other provinces: “[w]here there are numerous persons having the same interest in one cause or matter, one or more of such persons may sue or be sued, or may be authorized by the Court or a Judge to defend in such cause or matter, on behalf or for the benefit of all persons so interested.” In 1976, this rule became Rule 5(11) of the Supreme Court Rules, but the substance of the rule remained the same. 15 Chiodo, supra note 8 at 27–28. 16 Chastain v British Columbia Hydro and Power Authority (1973), 32 DLR (3d) 443, [1973] 2 WWR 481 (BCSC) [Chastain cited to DLR]. 17 Chastain, supra note 16 at 443. 18 Shaw v Real Estate Board of Greater Vancouver (1973), 36 DLR (3d) 250, [1973] 4 WWR 391 (BCCA) [Shaw cited to DLR]. 19 Ibid at 254–55. 20 Ibid at 255. 21 Newfoundland, Rules of the Supreme Court 1986, Rule 7.11(1). Until 1986, the Newfoundland rule was identical to the English rule that required “numerous persons having the same interest in one cause or matter”: see Hillyard v St John’s (City) (1982), 30 CPC 176, 1982 CarswellNfld 31 (TD) [Hillyard cited to CPC]. Rule 7.11(1) was slightly different in that it provided for the discontinuation of a representative action: “[w]here numerous persons have the same interest in a proceeding … the proceeding may be begun,
208 UNBLJ RD UN-B [VOL/TOME 71 representative actions to be brought around this time was allowed to proceed, based on the test of the British Columbia Court of Appeal in Shaw. In Hillyard v St John’s (City),22 the plaintiff purported to represent 775 taxpayers in an area of St. John’s, who had formed themselves into a group called “The Anspach Neighbourhood Citizens’ Committee”. He sought an injunction against the Municipal Council of St. John’s to restrain it from issuing a permit for the construction of town houses and other buildings in the area.23 The Newfoundland Supreme Court (Trial Division) applied the reasoning in Shaw and found that, “the intended plaintiffs have a common interest and a common grievance, [so that] a representative suit is in order because the relief being sought is in its nature beneficial to all whom the plaintiff proposes to represent.”24 Because the group did not claim monetary damages, and success for the plaintiff meant success for the rest of the group, a representative action was found to be appropriate in this case.25
This reasoning also influenced the evolution of the case law in Ontario,26 where representative actions were available under Rule 75. GM (Canada) v Naken27 similarly attempted to circumvent the barriers of individual contracts and individual damages. The representative action was brought on behalf of nearly 5,000 purchasers of the 1971 and 1972 models of the General Motors Firenza, which were notoriously 2020 CanLIIDocs 3123 unreliable. The plaintiffs sought damages for the reduced resale value of each vehicle on a pro-rata basis in the amount of $1,000 per class member. The defendants moved to strike the action on the basis that reliance on the manufacturer’s representations was an individual issue, and that damages would have to be individually assessed. The Court of Appeal allowed the case to proceed as a representative action, holding that aggregate damages could be awarded because each class member’s monetary claim was the same, and that individual issues regarding reliance could be determined in subsequent individual trials.28 and, unless the Court otherwise orders, continued, by or against any one or more of them as representing all or as representing all except one or more of them.” 22 Hillyard, supra note 21. 23 The plaintiff’s claims for a declaration against the Council that its actions in issuing the permit were null and void, as well as for damages, had previously been struck out by the Court: ibid at 179–80. 24 Ibid at 182–83. 25 Ibid at 180–83. 26 Chiodo, supra note 8. See also Cobbold v Time Canada Ltd, (1976), 71 DLR (3d) 629, 13 OR (2d) 567 (HCJ). 27 Naken et al v General Motors of Canada Ltd et al (1975), 66 DLR (3d) 205, 11 OR (2d) 389 (HCJ), rev’d (1977) 79 DLR (3d) 718, 17 OR (2d) 193 (Div Ct), rev’d (1979) 92 DLR (3d) 100, 21 OR (2d) 780 (CA) [Naken OCA], rev’d [1983] 1 SCR 72, 144 DLR (3d) 385 [Naken cited to SCR]. See also Chiodo, supra note 8 at 29–31. 28 Naken OCA, supra note 27. However, two subsequent decisions from the Court of Appeal called into question Arnup JA’s analysis of the aggregate damages issue: Seafarers International Union of Canada v Lawrence (1979), 97 DLR (3d) 324, 24 OR (2d) 257 (CA); and Stephenson v Air Canada (1979), 103 DLR (3d) 148, 26 OR (2d) 369 (HCJ). The Court of Appeal for Ontario gave its decision in Stephenson on 10 February 1981 without reasons. Its decision is cited in Ontario Law Reform Commission, Report on Class Actions (Toronto: Ministry of the Attorney General, 1982) at 24 [OLRC Report].
2020] LOCAL REPRESENTATION AND LITIGANT VOICE 209
Naken proceeded to the Supreme Court of Canada,29 where those expecting the Court to reform the law on class actions were sorely disappointed. Justice Estey held, as prior case law had, that representative actions under Rule 75 could not proceed where, as here, they involved separate contracts and individualized damages30 because they did not meet the “same interest” test.31 He concluded that Rule 75, “consisting as it does of one sentence of some 30 words, is totally inadequate for employment as the base from which to launch an action of the complexity and uncertainty of this one.”32 The creation of this “new and distinct method of proceeding” was a job not for the Court, but for the legislature.33
The Naken decision effectively stymied class actions not only under Rule 75, but also under the representative action rules of other Canadian jurisdictions. Just as in Ontario and BC, pressure for class action reform in Newfoundland arose from the perception that the representative action rule was being interpreted far too restrictively, so that claims were going uncompensated and defendants were not being held accountable for their actions. There was a general perception in Canada, both among law reformers34 and the general public,35 that the Supreme Court of Canada had let an opportunity pass by and that the onus was now on the provincial legislatures to enact 2020 CanLIIDocs 3123 class actions legislation.36 While the Ontario and BC legislatures answered this call in 1993 and 1996 respectively, Newfoundland and Labrador did not pass class proceedings legislation until December 13, 2001—just six months after the Supreme
29 Naken, supra note 27. 30 Ibid at 78. 31 Ibid at 99, 103–04. 32 Ibid at 105. 33 Ibid at 102. 34 WA Bogart, “Naken, the Supreme Court and What Are Our Courts For?” (1984) 9:3 Can Bus LJ 280; Jennifer K Bankier, “Class Actions for Monetary Relief in Canada: Formalism or Function?” (1984) 4 Windsor Yearbook of Access to Justice 229; J Robert S Prichard, “Class Action Reform: Some General Comments” (1984) 9:3 Can Bus LJ 309. Bogart, Bankier, and Prichard were all involved in the OLRC Report, supra note 28. 35 Steve Thorne, “Court decision undermines use of class action suits”, Halifax Mail Star (25 February 1983) at 14; Marina Strauss, “Supreme Court bars Firenza buyers from launching a class-action suit”, Globe and Mail (9 February 1983) at 2; Jacob Ziegel, “Class action dealt a legal blow”, Globe and Mail (29 March 1983) at 7; Editorial, “We need class-action suits”, Toronto Star (10 February 1983); all these articles can be found in Ontario Law Reform Commission files: Class Actions (November 1976–December 1982), Toronto, Archives of Ontario (RG 4-66, BA77, Box No B380537). See also Editorial, “Make class actions easier”, Toronto Star (12 July 1984) in Ontario Law Reform Commission files, Class Actions (November 1976–December 1982), Toronto, Archives of Ontario (RG 4-66, BA77, Box No B380543). 36 Ontario, Legislative Assembly, Standing Committee on Administration of Justice, Estimates, Ministry of the Attorney General, Official Report of Debates (Hansard), 32-1 (17 June 1983), J-173 in Ontario Law Reform Commission files, Class Actions (November 1976–December 1982), Toronto, Archives of Ontario (RG 4-66, BA77, Box No B380537).
210 UNBLJ RD UN-B [VOL/TOME 71
Court of Canada overturned Naken and allowed class actions at common law to proceed.37
Following Naken, there were very few attempts to use Rule 7.11 in Newfoundland, and most of the representative actions that were brought in the province or on behalf of Newfoundlanders were dismissed on the basis of the Supreme Court of Canada’s decision.38 In West End Electronics Ltd v St John’s (City), the Newfoundland Supreme Court (Trial Division) refused to permit an action by the plaintiff and 22 of its customers for flood damage allegedly caused by the defendant.39 The Court held the following:
The practical problems which can spring up on all sides of a proceeding like this were considered in G.M. (Canada) v Nakin [sic] and found to be insurmountable. While a class action sounding in damages is not, for that reason alone, inappropriate, nevertheless, in many cases it is. In this proceeding, apart from the inconvenience of trying twenty–two claims for distinct damages by twenty–two claimants, other problems exist. These include liability, if any, in respect of mitigation of damages, discovery and costs.40 2020 CanLIIDocs 3123
Courts in Atlantic Canada continued to interpret the representative action rule very narrowly. In 1988, the New Brunswick Court of Queen’s Bench interpreted Naken and held that an action would have to meet the following requirements in order to proceed:41
1) the class must be properly defined; 2) all members must have a common interest; 3) there must be a wrong common to all; 4) damages suffered must be the same to all except in amount; 5) the relief sought must be beneficial to all; 6) none of the members of the class may have an interest antagonistic to the other members; and
37 Dutton, supra note 3. However, as noted below, the government of Newfoundland and Labrador announced its intention to pursue class action reform one week before the Supreme Court of Canada released its reasons in Dutton. 38 West End Electronics Ltd v St John’s (City) (1992), 100 Nfld & PEIR 76, 9 CPC (3d) 393 (NL SC) [West End Electronics cited to Nfld & PEIR]; Inshore Fishermen’s Bonafide Defense Fund v Canada (AG) (1994), 117 DLR (4th) 56, 132 NSR (2d) 370 (CA). 39 The Court instead allowed the 22 customers to be joined as plaintiffs: West End Electronics, supra note 38 at para 18. 40 Ibid at para 14. 41 Guarantee Co v Caisse Populaire de Shippagan et al (1988), 86 NBR (2d) 342 at 349, [1988] NBJ No 42.
2020] LOCAL REPRESENTATION AND LITIGANT VOICE 211
7) there must be created in the course of the action or as a result thereof a fund or a pool of assets which is isolated and subject to pro rata distribution should the need arise.
This test is much more stringent than the Ellis test, but the additional requirements— proper class definition, same damages, no conflicts of interest among the class, and a common fund—all relate to the modern practice of bringing class actions for damages and in the absence of a pre-existing group. As I have argued elsewhere, these factors make the modern opt-out class action an entirely different creature from its predecessor in equity, which involved groups that pre-existed the litigation (such as trade union members, residents of a parish, or tenants of the same landlord) that sued for declarations of rights and other equitable remedies.42
In addition to the narrow interpretation of the representative action rule, plaintiffs in several high-profile actions were ordered to pay significant costs when their actions failed, giving rise to arguments that the system was unfairly stacked against them.43 In Newfoundland, this phenomenon manifested itself in the form of the Hiland Insurance bankruptcy and the litigation that followed. 2020 CanLIIDocs 3123
Part II—Bill Kelly and the Kelly Bill
Just as in Ontario, Alberta, and other provinces, the catalyst for class action reform in Newfoundland took the form of a high-profile event that affected a large number of people, and for which they were denied legal recourse. In Ontario, it was Naken;44 in Alberta, it was the pine-shakes litigation;45 and in Newfoundland, it was an insurance scandal that affected tens of thousands of people and led to calls for a public inquiry into the government’s failure to protect policyholders.46 It was the grass-roots campaign for class action reform that arose from this scandal, however, that marked out Newfoundland’s story from that of the other provinces.
42 Chiodo, supra note 8, at 17–19. 43 Palmer v Nova Scotia Forest Industries (1983), 2 DLR (4th) 397, 60 NSR (2d) 271 (SCTD); in that case, approximately $250,000 in costs was awarded against the 15 plaintiffs: “Ruling Shakes Spray Plaintiffs”, Halifax Mail-Star (16 September 1983) at 1. 44 Naken, supra note 27. 45 Holtslag v Alberta, 2000 ABQB 351. The case involved homeowners whose homes were installed with pine shakes, a roofing material that grew fungus and deteriorated quickly. The Court struck out the representative action because it would require individual damages assessments; there were also issues regarding the class definition and the existence of common issues. In the month following this decision, approximately 2,000 individual actions were commenced, to be case managed by the court and litigated by way of test cases. In December 2000, the Alberta Report noted the “costly, complex and cumbersome” nature of the pine shakes litigation: see Alberta Report, supra note 4 at xx. 46 Newfoundland and Labrador, House of Assembly, 42-2, vol XLII, No 61 (15 November 1994) (Scott Simms). Mr. Simms referred to hundreds of policyholders who were demonstrating outside the provincial legislature and raised a Parliamentary petition asking for a public enquiry.
212 UNBLJ RD UN-B [VOL/TOME 71
In October 1994, Hiland Insurance Limited, a provider of automobile and fire insurance, had its licence cancelled and went into provisional liquidation. The following month, Newfoundland’s Superintendent of Insurance informed Hiland’s approximately 22,000 policyholders in the province that their policies were cancelled effective December 6, 1994. Their unrefunded premiums totalled approximately $20 million. JJ Lacey, the agent and broker that had sold the Hiland policies, subsequently went into bankruptcy.47
One of the policyholders was Bill Kelly, a recently-retired journalist who had until the previous year hosted the Newfoundland-based television show Land & Sea, a staple of popular culture in the province. Kelly was no stranger to campaigning for change; when Land & Sea was taken off the air in 1990 due to funding cuts in regional broadcasting, Kelly organized rallies and petitions to get the show back on the air, even though he was in hospital recovering from quadruple-bypass surgery.48
In the wake of the Hiland bankruptcy, Kelly did not remain retired for long. He formed the Hiland Consumers’ Association in an attempt to obtain redress on 49 behalf of the policyholders, and sought out another high-profile Newfoundlander, 2020 CanLIIDocs 3123 lawyer Chesley Crosbie.50
The Crosbie family was prominent in provincial politics.51 Chesley’s great- grandfather, Sir John Chalker Crosbie, was the third Prime Minister of the Dominion of Newfoundland.52 His grandfather, Chesley A Crosbie, was a delegate to the Newfoundland National Convention and was opposed to confederation with Canada.53 Chesley’s father, John Carnell Crosbie, was a provincial and federal cabinet minister who also served as Lieutenant-Governor.54 Chesley Crosbie himself was a Rhodes Scholar who, upon his return from Oxford University, was admitted to the bar and in
47 Kelly et al v Lacey (JJ) Insurance Ltd (Trustee of) (1997), 154 Nfld & PEIR 1 at paras 4–11, 15 CPC (4th) 114 (NL SC) [Lacey]. 48 CBC NL - Newfoundland and Labrador, “Remembering Bill Kelly: Land & Sea” (2012), online (video): YouTube
2020] LOCAL REPRESENTATION AND LITIGANT VOICE 213
1991 established his own law firm.55 He was later to become leader of the Progressive Conservative Party in Newfoundland.56
As a personal injury lawyer, Chesley Crosbie was a member of the American Trial Lawyers Association,57 and became aware of class actions through reading ATLA material and attending conventions in the US.58 In November 1995, he represented Bill Kelly in the latter’s efforts to obtain leave from the Bankruptcy Court to commence a representative action against JJ Lacey’s insurer. That action was commenced in March 1996, for damages equivalent to the amount of unrefunded premiums for each policyholder.
The representative action was dismissed in July 1997.59 The Newfoundland Supreme Court (Trial Division) held that the claims did not involve the “same interest” because they would each necessitate individual enquiries into when Lacey and the policyholders knew or ought to have known of Hiland’s precarious financial situation, as well as the manner in which Lacey handled each class member’s premium payments.60 In addition, the Court held that many class members might have damages exceeding the amount of unrefunded premiums, and that the res judicata effect of the 2020 CanLIIDocs 3123 representative action would effectively bar those claims, leading to injustice.61 As a result, the representative aspects of the plaintiff’s claim were struck out.62
Costs were ordered against Bill Kelly in the amount of $15,000.63 The costs were taxed on the higher scale because the action was taken in a representative
55 Aaron Wherry, “Why the Tories said no to a star candidate in Newfoundland”, Maclean’s (9 July 2015), online:
214 UNBLJ RD UN-B [VOL/TOME 71 capacity, and the costs therefore reflected the total amount in issue for all the class members—not just Kelly—as well as the total exposure to the defendants.64
The defendants pursued Kelly for the recovery of those costs, to the point where he undertook a campaign as determined as his efforts to save Land & Sea. Kelly took to the streets to ask people to contribute to his costs and sign a petition calling for class actions legislation in the province.65 As acknowledged by the politicians who later introduced the Class Actions Act, he almost single-handedly “embarked on a crusade … he took to the streets and the malls, and the villages and the towns all over Newfoundland and Labrador to persuade the government to bring in appropriate legislation so that this would never happen again.”66
Kelly’s public notoriety and determination earned him a meeting with Liberal Premier Roger Grimes, where he made the case for class action reform.67 In December 2000, Kelly petitioned the provincial Parliament for the enactment of class actions legislation.68 The petition was signed by “thousands and thousands of people”69 and referred to the class actions legislation that was available in Québec, Ontario, and
British Columbia. Bill Kelly’s personal exposure to costs was also a motivating factor, 2020 CanLIIDocs 3123 as the preamble to the petition demonstrates:
Whereas class action proceedings contain a “no cost” or minimum cost clause which enables ordinary citizens to initiate court action without facing the prospect of tens of thousands of dollars in legal costs, thereby putting them on a “more or less” equal footing with large corporations, government agencies
64 Kelly costs judgment, supra note 63 at paras 11, 15, 20. 65 Newfoundland and Labrador, House of Assembly, 44-3, vol XLIV, No 34 (20 November 2001) (second reading: Jack Harris); Interview with C Crosbie, supra note 50. See also “Supreme Court says no to class action”, CBC News (7 June 1999), online:
2020] LOCAL REPRESENTATION AND LITIGANT VOICE 215
and other organizations with virtually unlimited funding.70
On July 6, 2001, Grimes announced his government’s plans to draft a class actions statute, acknowledging that, “it is Mr. Kelly’s efforts in this area that prompted government to consider this legislation.”71 The Hiland Insurance case was cited as evidence of the need for such legislation,72 with supporters stating that, “may it never happen again what had happened to approximately 20,000 of our residents in the ordeal that they had to go through.”73
Importantly, the announcement pre-empted the Supreme Court of Canada’s reasons in Dutton,74 which were released one week later. The government of Newfoundland and Labrador therefore pursued class action reform on its own initiative, and before the Supreme Court gave the green light to class actions at common law.
The government rushed to draft the statute, basing it on the Uniform Law Conference of Canada’s Uniform Class Proceedings Act.75 The statute therefore
included a provision that Newfoundland residents would be included in class actions 2020 CanLIIDocs 3123 unless they opted out, whereas non-residents would have to opt-in to a Newfoundland class action.76 As detailed further below, this would put Newfoundland at a
70 See Class Proceedings Act, RSBC 1996, c 50 [BC CPA], which contains what is effectively a “no-costs” provision (s 37). The Québec provisions (CCP, CQLR c C-25.01, Book VI, Title III) contain a provision that dictates the losing party pay costs, but its tariff of costs payable in a class action is vastly reduced—this may be the “minimum cost” clause to which the petition refers. However, the Ontario Class Proceedings Act, 1992, SO 1992, c 6, applies the normal “loser-pays” costs rules, with some exceptions (s 31) [Ontario CPA]. The Uniform Law Conference of Canada’s model Class Proceedings Act (1995) [ULCC Act] contained two alternative clauses at s 37(1), one stating that the normal “loser-pays” costs rules apply with some exceptions, and the other stating that costs may not be awarded to any party to a class action (again, with some exceptions). The petition’s statement that class actions proceedings are “no-cost” as a matter of course is therefore incorrect. 71 Newfoundland and Labrador Executive Council (Justice), News Release, “Government to draft new class action legislation” (6 July 2001), online:
216 UNBLJ RD UN-B [VOL/TOME 71 disadvantage compared to provinces in which national opt-out class actions would be permitted. Importantly, the statute also stated that costs could not be awarded in class actions, barring exceptional circumstances.77 The adverse costs award against Bill Kelly, and the desire not to deter representative plaintiffs in future, had a significant influence upon this provision.78 On November 19, 2001, Bill 34, “An Act To Permit An Action By One Person On Behalf Of A Class Of Persons” was introduced.79 It was nicknamed the “Kelly Bill”.80
The focus was primarily on providing access to justice for the people of Newfoundland and Labrador,81 and it was recognized that class actions would make justice more affordable.82 Unlike class actions reform in other jurisdictions, the Newfoundland bill appeared to have almost unanimous support,83 including that of the Conservative opposition party (although Conservative politicians warned of the increased litigation against public authorities that had resulted from the Ontario legislation).84 The consultation process included favourable submissions from the Canadian Bar Association, the Atlantic Trial Lawyers’ Association, the Law Society, and the Federation of Municipalities.85
2020 CanLIIDocs 3123
actions: Uniform Law Conference of Canada, Class Proceedings Act (Consolidated 2006), s 16; Uniform Class Proceedings Act (Amendment) 2006. 77 Class Actions Act, supra note 5, s 37; Newfoundland and Labrador, House of Assembly, 44-3, vol XLIV, No 44 (6 December 2001) (third reading: Jack Harris). 78 Newfoundland and Labrador, House of Assembly, 44-3, vol XLIV, No 34 (20 November 2001) (second reading: Yvonne Jones, Jack Harris). 79 Newfoundland and Labrador, House of Assembly, 44-3, vol XLIV, No 32 (19 November 2001) (introduction and first reading: Kelvin Parsons). See also Newfoundland and Labrador Executive Council (Justice), News Release, “Class Actions Act” (21 November 2001), online:
2020] LOCAL REPRESENTATION AND LITIGANT VOICE 217
Supporters distanced the class actions statute from its American counterpart, stating clearly that, “We are not adopting the American model … we are adopting a well tried and tested Canadian model which will ensure the access to justice for thousands while protecting potential defendants from frivolous and vexatious lawsuits.”86 Class actions were seen as a much-needed replacement for an outdated, inflexible, and nebulous representative action rule.87 The class action in Ontario against the Inco nickel refinery and various provincial and municipal government bodies was held up as an example of when it would be appropriate to bring a class action.88
The fact that the three largest provinces already had class actions was also mentioned, with the implication that Newfoundland was missing out on the benefits of such legislation.89 The experience of other jurisdictions was also relied upon to show that class actions were not a risky proposition, but had been effective elsewhere, and that Newfoundland could learn from their experiences in tailoring its own statute.90 The debate on class actions reform in Alberta, Manitoba, and the Federal Court was referred to, as well as the fact that Saskatchewan was on the verge of introducing its 91 own legislation. 2020 CanLIIDocs 3123
However, the debate was not always as well-informed as it could have been. Supporters of reform referred to the experience in England92 and Lord Woolf’s wide- ranging review of the civil litigation system there. As part of this review, Lord Woolf recommended that opt-out (and opt-in) class actions legislation be enacted.93 However, the Newfoundland debaters were incorrect in stating that those recommendations led to the “adopt[ion of] a class action regime along the lines of those of British Columbia and Ontario.”94 At the time of Lord Woolf’s Report, there were in fact very few voices calling for generic opt-out class actions in England.95 Lord Woolf focused on the case
86 Newfoundland and Labrador, House of Assembly, 44-3, vol XLIV, No 34 (20 November 2001) (second reading: Yvonne Jones). 87 Ibid. 88 Newfoundland and Labrador, House of Assembly, 44-3, vol XLIV, No 36 (22 November 2001) (second reading: John Ottenheimer) [Ottenheimer]. Ironically, this particular class action failed on a trial of the common issues because the court held that the plaintiffs suffered no actionable damage: Smith v Inco Limited, 2011 ONCA 628. 89 Ottenheimer, supra note 88. 90 Newfoundland and Labrador, House of Assembly, 44-3, vol XLIV, No 35 (22 November 2001) (second reading: Roland Butler) [Butler]. 91 Ibid. 92 By “England”, I mean the jurisdiction of England and Wales. 93 Lord Woolf, Access to Justice: Final Report to the Lord Chancellor on the Civil Justice System in England and Wales (London: Her Majesty's Stationery Office, 1996), Chapter 17, paragraph 87, recommendation 9. 94 Butler, supra note 90. 95 Either opt-out class actions were recommended for certain sectors (as in the National Consumer Council’s 1988 Group Actions: Learning from Opren report) or generic reform was recommended on an opt-in basis
218 UNBLJ RD UN-B [VOL/TOME 71 management of multiple individual claims (the opt-in approach), and this was the approach that England adopted, in the form of the Group Litigation Order (GLO) framework.96 To this day, England has not adopted an opt-out class action regime, except in the area of competition law.97 The incorrect perception that England had enacted class actions legislation was not limited to Newfoundland, but was also reflected in the Supreme Court of Canada’s Dutton decision that made class actions available at common-law across Canada.98 As far as reform in Canada relied on the English experience, then, it relied on a misperception.
The Kelly Bill received first reading on November 20, 2001, passed second reading two days later and then went to a Committee of the Whole House on December 6, 2001. The Bill passed a week later without amendment.99
The whole process took place just a few months after the Supreme Court of Canada’s decision in Dutton,100 although, oddly, the decision does not seem to be mentioned in the debates or elsewhere in the historical record (neither is the Newfoundland debate mentioned in the Dutton decision). Dutton involved the mismanagement of investment funds belonging to participants in the federal 2020 CanLIIDocs 3123 government’s Business Immigration Program. The Court of Appeal for Alberta had allowed the representative action to proceed; the Supreme Court of Canada, in a unanimous decision written by Chief Justice Beverley McLachlin, not only upheld the Court of Appeal decision101 but also overturned Naken. The Court found that, when interpreting the representative action rules of the various provinces, attention should
(as in the Law Society Civil Litigation Committee’s 1995 Group Actions Made Easier report). Lord Woolf’s interim report, released in June 1995, did not even mention class actions. It was the publication of Group Actions Made Easier in September 1995 that influenced his thinking in this area, although this focused not on opt-out class actions but on the case management of multiple individually viable claims arising from similar circumstances. Lord Woolf published an issues paper in January 1996, and this led to his recommendations on opt-out and opt-in class actions in Chapter 17 of his final Access to Justice report (supra note 93). Nevertheless, those recommendations attracted barely any attention compared to Lord Woolf’s wider proposals; his recommendation on opt-out class actions appeared almost as an afterthought, and as a corollary to the opt-in approach. 96 Civil Procedure Rules (UK), Part 19.III: Group Litigation. 97 See the Consumer Rights Act 2015 (UK), 2015 c 15, Schedule 8, amending s 47 of the Competition Act 1998 (UK). 98 Dutton, supra note 3. Part of the Supreme Court’s reasoning for doing so was the fact that “many jurisdictions have enacted comprehensive class action legislation” (at para 30). England’s GLO framework was included in this list, even though the GLO framework does not, as explained above, constitute comprehensive class action legislation. 99 Newfoundland and Labrador, House of Assembly, 44-3, vol XLIV, No 44 (6 December 2001) (third reading). 100 Dutton, supra note 3. 101 See ibid at paras 59–60. The Court allowed the plaintiffs’ appeal regarding the discovery of individual class members, finding that such discovery at this stage of the proceedings would be premature.
2020] LOCAL REPRESENTATION AND LITIGANT VOICE 219 be paid to the criteria and procedures set out in class actions legislation in Ontario, British Columbia, Québec and elsewhere.102
It further held that representative actions should not be refused based on the existence of traditional bars to their use, such as separate contracts or individual damages assessments.103 In doing so, the Court modified the requirements for bringing a representative action:104
1. the class is capable of clear definition; 2. there are issues of law or fact common to all class members; 3. success for one class member means success for all; and 4. the proposed representative adequately represents the interests of the class.
The Court held that countervailing considerations that may outweigh the benefits of a class action also had to be considered, with a balance to be struck between efficiency and fairness.105 While these requirements overlap somewhat with the certification requirements in the Newfoundland and Labrador Class Actions Act,106 the latter was 107 108 influenced much more obviously by the ULCC Act and the BC CPA. 2020 CanLIIDocs 3123
Three further points of interest can be drawn from the decision. First, the court justified its departure from Naken in two ways. It distinguished the previous decision on its facts.109 More importantly, it found that the modern class action was no longer, as it had been in 1983, an untested procedure, and that “[i]n the intervening years, the importance of the class action as a procedural tool in modern litigation has
102 Ibid, at paras 46–47, 51. See also Alberta Law Reform Institute, Joining Claims and Parties, Including Third Party Claims, Counterclaims and Representative Actions (2004) at 77. 103 Dutton, supra note 3 at para 43. 104 Ibid at paras 38–41, 43, 48. 105 Ibid at paras 42, 44. This was the “cost-benefit” test that had already been rejected by the Ontario and British Columbia legislatures. It is not to be confused with the guidance, contained in the BC CPA (supra note 70) and the Newfoundland Class Actions Act (supra note 5), on whether a class action is the “preferable procedure”. 106 Class Actions Act, supra note 5, s 5. 107 BC CPA, supra note 70, s 4. 108 Ibid, s 4, which provides further guidance on when a class proceeding might be considered the “preferable procedure” for the fair and efficient resolution of the common issues. This guidance is not provided in the Ontario or Québec class proceedings provisions. 109 The Court found that Naken would have required particularised fact-finding at both the liability and damages stages, which “would have unnecessarily complicated the resolution of what amounted to 4,600 individual claims” (Dutton, supra note 3 at para 47). Given the fiduciary duty claims in Dutton, it is not entirely clear that the resolution of that case would be any less complicated: see Picard JA’s dissent in the Dutton case at the Court of Appeal of Alberta: Western Canadian Shopping Centres Inc v Bennett Jones Verchere, 1998 ABCA 392 at paras 22–33.
220 UNBLJ RD UN-B [VOL/TOME 71 become manifest.”110 Ontario and BC had passed class actions legislation since Naken, but those were not the only changes that had taken place.
In the three decades since, numerous legal, political, and social changes had occurred in Canada that had changed the workings of the civil justice system as well as the perception of the courts’ role in society. I have detailed these changes elsewhere.111 They include the growing movement for consumer protection and environmental rights, and legal procedures to enforce those rights; the acceptance of contingency fees, which would help to promote access to justice in general and class actions in particular; and the advent of the Canadian Charter of Rights and Freedoms. This latter development entailed a more activist judiciary, as well as a challenge to the classic two-party model of adversarial litigation, with the rise of public interest standing, intervention, and the court’s ability to strike down legislation. All of these developments accelerated the acceptance of class actions.
In Newfoundland, the debate on class actions coincided with a growing feeling that Ottawa was leaving the province behind. The cod fisheries were closed in July
1992, ending one of the main sources of economic activity in the province; a decade 2020 CanLIIDocs 3123 later, unemployment remained twice the Canadian rate, and the province’s population had fallen by 10 per cent.112 In 2003, the findings of the Royal Commission on Renewing and Strengthening Our Place in Canada were released, which revealed the depressed state of the economy, as well as the fact that hydroelectricity resources in Labrador had primarily benefited Québec.113 While class action legislation was a fairly minor issue compared with these fundamental problems, it was one way in which Newfoundland could catch up with the rest of Canada. As one MPP said in the debates on the legislation:
It is a bill that we felt ought to be consistent with at least a number of other Canadian jurisdictions. As it turns out, they are in fact the largest three Canadian jurisdictions, namely the Provinces of Québec, Ontario and British Columbia. I think it is only fitting that we, in this Province, have similar legislation.114
Bill Kelly, too, declared that such legislation was necessary so that, “[n]ever again will
110 Dutton, supra note 3 at para 46. 111 Chiodo, supra note 8 at 47–69. 112 William E Schrank, “The Newfoundland fishery: ten years after the moratorium” (2005) 29:5 Marine Policy 407. 113 Victor Young et al, Report of the Royal Commission on Renewing and Strengthening Our Place in Canada (St. John’s: Office of the Queen’s Printer, 2003). 114 Newfoundland and Labrador, House of Assembly, 44-3, vol XLIV, No 36 (22 November 2001) (John Ottenheimer).
2020] LOCAL REPRESENTATION AND LITIGANT VOICE 221 ordinary Newfoundlanders and Labradorians be left on the sidelines”.115 His grass- roots campaign made his province’s experience of class action reform unique. While reform efforts in other provinces were led by politicians with legal backgrounds (such as Attorney General Ian Scott in Ontario, and Minister for Social Development Pierre Marois in Québec), Bill Kelly was an ordinary person. He was a local celebrity who had help from Chesley Crosbie, but essentially he orchestrated his own campaign, mobilised his own army of volunteers, and got his petition signed by thousands of people.116 Reform would not have taken place when it did without him. Because of his activism, class actions came about in Newfoundland years ahead of the other Maritime provinces.117
Since then, Newfoundland’s Act has also proved influential on the common- law development of class actions in Prince Edward Island. In King & Dawson v Government of PEI,118 the province’s Supreme Court allowed a representative action to proceed under Rule 12.01 of the PEI Rules of Civil Procedure (the wording of which is substantially similar to Newfoundland’s representative action rule). In an effort to provide a procedural framework for the prosecution of the action, the Court appended the wording of Newfoundland’s Class Actions Act to its decision, while at the same 2020 CanLIIDocs 3123 time calling for the PEI legislature to enact its own statute.119
In comparison to the larger provinces of Ontario, British Columbia, and Québec, however, Newfoundland has found itself overshadowed in the two decades since it enacted class proceedings legislation.
Part III—Class Actions in Newfoundland and Labrador
Since the passage of the Class Actions Act, class actions in Newfoundland and Labrador have been much fewer in number than those commenced in Ontario and Québec.120 This is not simply due to the novelty of the statute. It is also because the
115 Newfoundland and Labrador Executive Council (Justice), News Release, “Justice Minister announces Class Actions Legislation” (21 November 2001), online:
222 UNBLJ RD UN-B [VOL/TOME 71
Class Actions Act is provincial by nature; for class actions commenced under that statute, people from outside the province are required to opt in. Because of inertia, or lack of information, or other barriers to proceeding, opt-in classes tend to be much smaller than opt-out classes (where people meeting the class definition are included in the class until they opt out). Class actions in Newfoundland have therefore been largely provincial, both in class member composition and in subject matter.
This has had its advantages and disadvantages. The disadvantages are that Newfoundland has not been able to compete with provinces such as Ontario when it comes to class actions that are national in scope. In the first class action to be commenced under the new legislation, concerning the cholesterol-lowering drug Baycol,121 the Newfoundland action was certified at the same time that a settlement agreement was reached in Ontario, on behalf of a national class.122 The Newfoundland court stayed the certification application until the settlement agreement in Ontario was approved;123 thereafter, the Newfoundland case was wrapped into the national settlement out of Ontario. However, the Ontario settlement only included class members who suffered from rhabdomyolysis, a muscle condition that can cause kidney 124 failure, and only one Newfoundland class member qualified. In cases such as this, 2020 CanLIIDocs 3123 Newfoundland residents can find their rights decided in another province or another part of the country entirely, which greatly restricts the ability of residents to object to a settlement or otherwise take part in proceedings.125 This was the reasoning of the Newfoundland Trial Division when it took jurisdiction over Newfoundland residents in a later class action that arose out of events in New Brunswick.126
On the other hand, the advantages of the Class Actions Act are demonstrated when the class actions are truly provincial. Provincial class actions are, in many ways, more representative of local interests and tied more closely to local events that figure more largely in the public consciousness; they therefore tend to be “class actions that the population supports”.127 The most prominent example is the class action that arose from the breast cancer testing scandal at Eastern Health in St. John’s.128 In that case, more than 400 patients from 1997 to 2005 received inaccurate results from hormone
been reported on CanLII—it is likely that more have been commenced, although the number would still be far less than the number of class actions in Ontario). 121 Pardy et al v Bayer Inc-Class Actions Act, 2003 NLSCTD 109. 122 Pardy et al v Bayer Inc, 2004 NLSCTD 72. 123 Ibid at para 180. 124 Daryl-Lynn Carlson, “Clash Actions” The National (April/May 2005) at 31–33. 125 Ibid at 31. 126 Ring v Canada (AG), 2007 NLTD 213. 127 Interview with C Crosbie, supra note 50. 128 Doucette v Eastern Regional Integrated Health Authority, 2007 NLTD 138 [Doucette 1].
2020] LOCAL REPRESENTATION AND LITIGANT VOICE 223
receptor testing, which meant they were not given the correct treatment.129 While class actions arising from corporate wrongdoing or negligence are generally commenced after a public enquiry or regulatory finding, in the case of the breast cancer testing scandal it was the other way around.130
In that case, Chesley Crosbie was once more to play a key role in the development of class actions in his province. Women who had been affected by the faulty testing initially went to him to commence individual medical malpractice lawsuits. As more and more women came forward, Crosbie realized there was an extensive problem and began investigations to start a class action lawsuit.131 In response to these investigations, Eastern Health produced documents that revealed the extent of the crisis. This provoked intense criticism in the provincial legislature, which eventually led the Health Minister to announce a judicial inquiry.132 The Cameron Inquiry began in 2008, which found numerous mistakes in the diagnosis and treatment system, leading to the publication of a condemnatory report in 2009.133 The class action eventually settled;134 together with the Cameron Report, it mandated numerous systemic changes at Eastern Health,135 thereby fulfilling the behaviour modification 136 goal of class proceedings. 2020 CanLIIDocs 3123
While the number of actions commenced since the enactment of the Class Actions Act has not been high, these actions have been reflective of local concerns. In fact, of the 22 reported actions commenced under the Act to date, 19 have involved facts arising out of provincial occurrences or issues, including the following:
129 Barb Sweet, “Cameron Report: Breast cancer testing scandal propelled passionate activism”, The Telegram (9 April 2019) online:
224 UNBLJ RD UN-B [VOL/TOME 71
• Indigenous groups seeking status and recognition;137 • medical negligence at local clinics and hospitals;138 • impugned decisions of provincial administrators;139 • claims against educational institutions in the province;140 • sexual assault claims against provincial institutions;141 • privacy breaches against locally-based corporations;142 • claims against a payday loan company in St. John’s;143 • residents suing corporations for flooding and property damage;144 • high rates of vehicle collisions caused by moose on the roads;145 and • allegedly addictive video lottery gaming terminals.146
The following table also illustrates the numbers of class actions by category compared to those commenced in Ontario.147 As can be seen, Crown liability, privacy, and environmental class actions are much more common in Newfoundland than they are in Ontario. Of the Crown liability actions, one-third involved local Indigenous rights; two-thirds of the privacy actions involved localized breaches; and both of the environmental class actions involved localized flooding. Conversely, class actions 2020 CanLIIDocs 3123 involving national or cross-provincial issues such as competition law, securities, and franchise law were non-existent. Despite these differences, the rate of certification of
137 Davis v Canada (AG), 2008 NLCA 49; Best v Nunatsiavut Government, 2015 NLTD(G) 83. 138 Doucette 1, supra note 128; Doucette 2, supra note 134; Rideout v Health Labrador Corp, 2007 NLTD 150. 139 HP Management Inc v R, 2007 NLCA 65; Sundance Saloon Limited v Newfoundland and Labrador (Finance), 2014 NLCA 15; Brown v Newfoundland and Labrador (Workplace Health, Safety and Compensation Commission), 2009 NLTD 106. 140 Memorial University of Newfoundland v Lee and Acreman, 2011 NLCA 55; Thorne v The College of the North Atlantic, 2017 NLCA 30; Chen v Memorial University of Newfoundland and Labrador, 2019 NLSC 193. 141 Jane Doe (#7) v Her Majesty in Right of Newfoundland and Labrador, 2019 NLSC 170; Anderson v Canada (AG), 2015 NLTD(G) 181. 142 Hynes v Western Regional Integrated Health Authority, 2014 NLTD(G) 137; Butt v Kiewit Energy Corporation, 2019 NLSC 119. 143 Bellows v Quik Cash Ltd, 2004 NLSCTD 191. 144 Dewey v Corner Brook Pulp and Paper Limited, 2019 NLCA 14; Chiasson v Nalcor Energy, 2019 NLSC 133. 145 George v Newfoundland and Labrador, 2016 NLCA 24. 146 Piercey Estate v Atlantic Lottery Corporation Inc, 2008 NLTD 202; Rice v Atlantic Lottery Corporation Inc, 2012 NLTD(G) 58; Atlantic Lottery Corporation Inc-Société des loteries de l’Atlantique v Babstock, 2018 NLCA 71. 147 The categories and numbers for Ontario are taken from the LCO Report, supra note 120 at 15. The Ontario numbers represent class actions from 1993–2018. The numbers in this table do not include the ‘other’ category in Ontario, which represents 3% of class actions filed in that province. The numbers for Newfoundland are those reported on the CanLII website, as indicated above.
2020] LOCAL REPRESENTATION AND LITIGANT VOICE 225 class actions is similar in Newfoundland as compared to Ontario and Québec,148 indicating that the provincial nature of the Class Actions Act is not an impediment to the working of the statute.
Category Number of Percentage of Percentage of NFL class NFL class ON class actions actions actions Securities 0 0 16 Product Liability 1 5 15 Privacy 2 10 3 Negligence 2 10 3 Mass Torts 0 0 6.5 Insurance 0 0 6.5 Franchise 0 0 2 Environmental 2 10 2 Employment & 3 14 12 Pensions 2020 CanLIIDocs 3123 Crown Liability 9 37 7 Consumer Protection 3 14 12 Competition Act 0 0 15 Total 22 100 100
Actions arising out of local circumstances, and spearheaded by local counsel,149 are much more likely to be representative of class members’ interests, and thereby provide them with a voice and greater opportunities for participation. An empirical study of this phenomenon—for example by comparing the opt-out rates in class actions arising out of purely provincial events with those in class actions arising from national events—is beyond the scope of this paper. However, it would provide an interesting avenue for future research.
Conclusion
Since the passage of class actions legislation in Ontario and elsewhere, class action activity has migrated towards the more financially fruitful national opt-out class action. This is not the case in Newfoundland and Labrador, however, which has been
148 Of the 18 actions in Newfoundland in which the result of certification is reported on CanLII, 13 (72%) were certified, while 5 (28%) were not. The percentage of cases that have been certified vs not certified are: for Ontario, 71% vs 29%; for Québec, 71% vs 29% (see LCO Report, supra note 120 at 39). I note that my numbers for Newfoundland are significantly higher than those reported by The Honourable Ward Branch in Class Actions in Canada, 2nd ed (Toronto: Carswell, 2019), which reports a 58% certification rate (at para 22.80). Many of the Newfoundland certification decisions reported on CanLII are relatively recent, which may explain the discrepancy. 149 Almost half of the actions listed involved Chesley Crosbie.
226 UNBLJ RD UN-B [VOL/TOME 71 forced by the terms of its statute to adopt a more local approach. A review of the case law indicates that this local representation has given rise to more opportunities for litigant participation. This is a hypothesis that would benefit from further empirical testing. The historical record is clear, however: the story of class actions in Newfoundland and Labrador is one of grass roots activism that arose from and reflected the needs of the people of the province. To a large extent, that remains true in the practice of class actions there today.
2020 CanLIIDocs 3123
THE ABOLITION OF ADVERSE POSSESSION OF CROWN LANDS IN NEWFOUNDLAND AND LABRADOR
By Gregory French*
Introduction
This paper will analyse the impact of the Act to Further Amend the Crown Lands Act, SN 1975–76, No 20, on modern land title in Newfoundland and Labrador. The legislation marked a dramatic change in real property law in Newfoundland and Labrador, where title had historically been grounded on adverse possession of Crown lands. The author argues that the legislative intent behind the 1976 legislation, as evidenced by House of Assembly records, was misconstrued by the Courts, resulting 2020 CanLIIDocs 3123 in a subversion of the initial intention of the law, which has persisted to this day. The author examines the issues that have arisen since the passage of the 1976 statute, and suggests that the Province must address the issue of land, in order to fulfill the original intention of the legislature, and why changes are necessary.
It is important to understand the background of land settlement in Newfoundland and Labrador to appreciate the significance of the legislative intervention and the impact that the subsequent evolution of the law has had on Newfoundlanders and Labradoreans. A significant amount of title in this province is grounded in “adverse possession” against the Crown: possession that is not grounded on a conveyance or grant of title from the Crown, but instead based on open, notorious, continuous and exclusive occupation of the land. Such land has never been sold or demised by the Crown, but was instead occupied by settlers without any documentation granting such a right. “Adverse possession” permits such title to vest in the occupier of the land after a defined period.
Part I of this paper canvasses the history of settlement and treatment of real property in Newfoundland and Labrador from the 16th century to the 1970s, which gave rise to the legislative intervention in 1976. Part II discusses the land reform debates of the 1970s and uncovers the original intention of the legislature in implementing changes. Part III critically examines the first case addressing the new amendments, which the author argues undermined the legislative intent of the 1976 amendments and gave rise to the problems encountered today. Part IV examines the evolution of adverse possession against the Crown subsequent to that case. Part V critiques the legislative and judicial approach since 1982, and identifies the practical problems that have arisen for Newfoundlanders and Labradoreans since that time.
* Associate lawyer at Mills, Pittman & Twyne, Clarenville, NL.
228 UNBLJ RD UN-B [VOL/TOME 71
Finally, Part VI looks to the future to determine possible solutions.
I. An Introduction to the history of Newfoundland and Labrador land possession
One must first appreciate the significance of adverse possession to the history and society of Newfoundland and Labrador to truly understand the significance of its abolition.1
From its initial discovery, Newfoundland’s value was tied to the Grand Banks and its rich fishing grounds.2 Its value to its European colonizers was for the rewards it could produce for the mother country, and not for the value it could produce for itself. As is well known to Newfoundlanders, its climate is often harsh and its weather unrelenting, and its rocky landscape is poor for most agricultural uses. The earliest of settlers would no doubt have found it a most unwelcoming environment. Because its value lay in the fishing grounds off its shores, its only practical value for many years following its initial discovery was as a port for the British fishing fleets. Its best ports were prized by the captains of these British ships, and as the annual fishery would 2020 CanLIIDocs 3123 begin underway from the coasts of England, it became a race to secure favourable grounds. Through the 16th century and in earnest by the 17th century, settlement of the island of Newfoundland began as English crewmen remained in ports at the end of the fishing season in order to preserve access for the following season. As the settler population began to grow in Newfoundland through the 17th century, concern developed in Britain that a permanent population on the island of Newfoundland would result in competition for access to the Grand Banks and its fishing grounds as against the British fleet. A permanent population in closer proximity to those grounds would have a greater advantage against British interests. As a result of these concerns, the Imperial Parliament in London passed laws in the 17th century restricting access to the Island of Newfoundland and ordering occupied lands to be relinquished to British fishing fleets.3
These restrictive laws prevented settlers from obtaining title to their lands, and such titles were expressly not recognized by the British government. These laws
1 For a most thorough and well-researched history of Newfoundland land ownership, see Alexander Campbell McEwen, Newfoundland Law of Real Property: The Origin and Development of Land Ownership (PhD Dissertation, University of London, 1978) [unpublished]. This paper will not delve as deeply into the origin and development of land tenures, except as background. 2 The following paragraph is a condensed summary of the settlement of Newfoundland from the author’s earlier work on the topic: see Gregory French, “Property Interests in Resettled Communities” (2015) 66 UNBLJ 210 at 211–14. More on the history of settlement of Newfoundland and Labrador can be found in the seminal historical text by DW Prowse, A History of Newfoundland from the English, Colonial, and Foreign Records (London, UK: Eyre and Spottiswoode, 1896). 3 French, supra note 3 at 211, nn 2–3. The particular legislation at issue was the Western Charter of 1634, amended further in 1670, and ultimately An Act to Encourage Trade to Newfoundland, 1698 (UK), Imp Act 10 & 11 Will III, c 25, which expressly prohibited settlement along the shoreline and imposed penalties for so doing.
2020] ADVERSE POSSESSION OF CROWN LANDS 229 would persist until the 19th century, by which point the settler population of Newfoundland had grown to such a level that its presence could no longer be ignored.4
The root cause of modern title issues can be traced back to the Imperial Parliament’s refusal to acknowledge the reality of the settler situation for centuries. Settlers would have understood themselves to have a proprietary interest in the land which they occupied, even if that interest was not formally recognised by the Crown. One must bear in mind as well the unfortunate state of functioning civil society in Newfoundland until the 19th century—laws were propagated from England, with an explicit effort to disavow the interests of Newfoundland’s settlers. It was not until 1791 that Newfoundland was first granted a formal civil court system5, and not until 1832 that a legislature was established.6 For the average fisherman in a rural outport community in the 18th or 19th century, obtaining a grant to land would have been impossible, if any need existed for obtaining one. Undisturbed possessory title was sufficient for the daily lives of most Newfoundlanders, whose forefathers had obtained their title by the same degree of possession and which was otherwise unfettered by an absentee colonial government.7 To the extent that the settlers faced interference from English interests, those interferences were transitory at best, given the lack of 2020 CanLIIDocs 3123 institutional support for enforcement of the English enactments and the mere seasonal attendance by English authorities.8
Once a functioning judiciary was established in the Colony of Newfoundland, the settler population had some recourse and defence to the otherwise arbitrary decrees from a distant monarch. The history of Newfoundland through the 16th, 17th and 18th centuries is replete with examples of the thorough disregard for the interests of the settler population of Newfoundland and particularly its claims to ownership of land.9
It is noteworthy that the Supreme Court of Newfoundland predates the establishment of the Newfoundland legislature by some 40 years. During this period, the laws of England were to be enforced by the newly established Court “as far as the
4 French, supra note 3 at 212. Crown grants became available in the settlement at St. John’s by the Saint John’s, Newfoundland Act, 1811 (UK), 51 Geo III, c 45, and became available throughout Newfoundland by the Newfoundland Fisheries Act, 1824 (UK), 5 Geo IV, c 51, which abolished the 1698 restrictions. 5 Legislative History of the Judicature Act, 1791–1988 (St. John’s: Newfoundland Law Reform Commission, 1989) at 1–10. 6 EM Archibald, Digest of the Laws of Newfoundland (St. John’s: Henry Winton, 1847; republished by the Law Society of Newfoundland and Labrador, 2018) at 40; Buyer’s Furniture Ltd v Barney’s Sales & Transport Ltd (1983), 43 Nfld & PEIR 158 at para 8, 2 DLR (4th) 704 (CA); Roy v Newfoundland (Legal Aid Commission) (1994), 116 Nfld & PEIR 232 at 10, 363 APR 232 (Nfld SC (TD)). 7 See French, supra note 3 at 213–14. 8 See generally Prowse, supra note 3. 9 Ibid covers this period of history in great detail from the colonial and imperial records.
230 UNBLJ RD UN-B [VOL/TOME 71 same can be applied” to the circumstances of Newfoundland.10 This anomalous circumstance vested an almost legislative power in the judiciary to establish how far English law would be enforced in Newfoundland, as a matter of policy and practice.11 Early caselaw of the Supreme Court of Newfoundland shows a few examples of the Crown’s efforts to control settlement and use of land into the 19th century that were brought before the Court for enforcement.12 In the earliest of those cases, Rex v Kough in 1819, the Court established that possession of land for a period of sixty years would act as a bar to recovery actions by the Crown, based on the limitation laws in force in England at the time.13 While the standard of general occupancy was the best and only interest available for centuries, the Kough decision finally allowed the settlers a defence to the Crown’s efforts to maintain ownership over the lands of the Colony.
The situation of possessory land titles in rural Newfoundland was concisely summarized by Justice Kent in Murphy v Moores and Government of Newfoundland in 1938:
The history of this country shows that a large portion of the public or Crown
lands was originally occupied by settlers, who took possession of it without 2020 CanLIIDocs 3123 any documentary title and having thus taken possession of it they cleared it, cultivated and lived upon and “worked” it in undisturbed possession for generations and still continue to do so. Much of the land around the coast is at present held by no title other than possession for the statutory period of limitation. Such of them as are fortunate enough to be in a position to prove that exclusive and continuous possession by themselves and their predecessors in title for the statutory period of sixty years may feel secure so long as that proof remains in existence and is available to them. But the proof of such possession depends upon the survival of residents in the vicinity whose memory enables them to describe the facts relating to the land and the possessors and the use of it as they were for sixty years or more before they are called upon to testify; but those who are not so fortunate have no security of title other than that of bare possession which did not avail against the Crown; it being well settled law that nothing short of proof of a grant or of exclusive uninterrupted possession for the statutory period of sixty years raises any presumption against the Crown.14
This is the situation as it stood until 1976, and it is against this backdrop that the government decided to take action to reform and modernize the law of adverse possession against Crown lands.
10 An Act for Establishing Courts of Judicature in the Island of Newfoundland and the Islands Adjacent, (1792) 32 Geo III, c 46, s 1. 11 See discussion per curiam in Chancey v Brooking (1823), 1 Nfld LR 314 at 316–17 (SC). 12 See e.g. Rex v Patrick Kough et al (1819), 1 Nfld LR 172 (SC) [Kough]; The King v Cuddihy (1831), 2 Nfld LR 8 (SC); The King v Luke Ryan (1831), 2 Nfld LR 47 (SC). 13 See Kough, supra note 13 at 176–78. 14 Murphy v Moores and the Government of Newfoundland (1938), 14 Nfld LR 161 at 163–64 (SC), Kent J [Murphy].
2020] ADVERSE POSSESSION OF CROWN LANDS 231
II. The Beginning of Land Reform
By the 1970s, Newfoundland had developed considerably from its colonial roots, and was a modern society in line with its sister provinces in the Canadian Confederation. However, the legacy of centuries of government inaction and laissez-faire settlement patterns had created significant uncertainty in land titles. For generations, settlement and development of land required little formality, particularly in rural “outport” communities. Such communities were small, scattered coastal settlements of no more than a couple of hundred residents, with limited public services and no central government—a legacy of a pattern of settlement tied to securing unoccupied ships’ harbours year after year.15
Modernization of Newfoundland and Labrador, particularly its rural areas, began in earnest under Premier Joseph R. Smallwood, after Newfoundland joined Confederation in 1949. Smallwood’s government was pressed to act to develop rural Newfoundland and to deliver on the promises made to sell Confederation to the people of Newfoundland.16 Because of a widespread population, separated by islands, bodies of water and rough terrain, it was impossible to modernize rural Newfoundland in its 2020 CanLIIDocs 3123 existing form of thousands of tiny settlements without incurring unfathomable cost. So began the era of government-sponsored “Resettlement”, beginning in 1953, when Smallwood’s government provided financial incentives to encourage the relocation of the populations of isolated communities into nearby larger centres, to which government could focus its provision of services.17 The “Fisheries Household Resettlement Programme” officially ended in 1975.18
It may be that the Resettlement programme is what caused the first major issues with regard to rural land title. Those whose communities were resettled were provided government money to relocate elsewhere, but no other support in moving or in particular in obtaining land elsewhere.19 This led to an influx of people into communities where land title was often no better than the community they had just left—the same principles of longstanding adverse possession governed the acquisition of new residential property by the resettled. Indeed, it is common for the rural real estate practitioner to encounter title which is grounded on nothing more than a homemade Bill of Sale or form deed from a Justice of the Peace which dates from the
15 French, supra note 3 at 214. 16 One can note on a district map of Newfoundland that it was rural Newfoundland that voted overwhelmingly for Confederation, and those areas closest to the City of St. John’s that voted for continued independence: see “The 1948 Referendums”, online: Heritage Newfoundland and Labrador
232 UNBLJ RD UN-B [VOL/TOME 71 era of resettlement of the surrounding area.20 As had been common practice throughout history, one acquired a vacant parcel of land from its putative owner, and would place a house there, and would remain in undisturbed possession thereafter. The difference under Resettlement was the volume of such activity—instead of an isolated house being constructed or moved, communities received an influx of dozens at once. The number of new people moving into these communities undoubtedly gave rise to conflict at some level, be it in the form of unaddressed property disputes coming to light or land speculation by community residents.
Continuing into modern development required overhauling a system of land title that had long rested on adverse possession. By the 1970s, Newfoundland’s informal system of settlement was at odds with modern practice. This led to inhibitions in the ability to develop land, because of uncertainty regarding title—land claims often did not rest on formality, and may or may not be readily ascertainable, creating conflicts between those seeking title in the proper route via Crown lands applications, and those who asserted private property rights to ungranted land.
The debate on the matter came to the floor of the House of Assembly on May 2020 CanLIIDocs 3123 31st, 1976, under the Progressive Conservative government of Premier Frank Moores. At issue was a proposed amendment to the Crown Lands Act21, which proposed a radical change to the law by abolishing adverse possession of Crown land. The changes were brought forth by then Minister of Forestry and Agriculture Joseph Rousseau, and, according to Minister Rousseau’s statements in the House of Assembly, the goal of the legislative changes was to facilitate obtaining land title:
What we would like to do now, Mr. Speaker, is to reduce the amount of time [for obtaining title by adverse possession] from sixty years to twenty years and give those people who have had what we call squatters rights, as defined, the right to that land, get this backlog of people off the list of people looking for the land and not accept any adverse possession after January 1, 1977….22
The shortening of this period to twenty years was intended “to provide a method of confirming possessory titles in appropriate instances.”23
Edward Roberts, then Liberal Party leader and Leader of the Opposition, endorsed the changes. Part of the issue from Mr. Roberts’ perspective was the problem with Newfoundlanders obtaining funding from the Canada Mortgage and Housing
20 The author speaks from personal experience in practice on this point, as many communities in Trinity Bay, Bonavista Bay and Placentia Bay were resettled in the 1960s. 21 Crown Lands Act, RSN 1970, c 71. 22 “Bill No 21, An Act Further To Amend The Crown Lands Act”, 2nd reading, Newfoundland and Labrador, House of Assembly, Verbatim Report (Hansard) 37-1, No 74 (31 May 1976) at 9197 (Hon Joseph Rousseau). 23 Ibid at 9198.
2020] ADVERSE POSSESSION OF CROWN LANDS 233
Corporation (CMHC), which required residents to have title to land as a precondition to obtaining funding.24 Possessory titles were an antiquated relic that could not meet the demands of modern society. And more practically, they posed a challenge to a government that was trying to straighten out matters of land title and take stock of its land holdings. In many communities, the Crown Lands Administration could not say with any certainty who owned what land, or whether or not it had been alienated from the Crown. Figuring out whether any given parcel of land on a grant application was Crown land or not was time consuming and costly for government. Hundreds of millions of dollars had been spent on surveying and aerial photography and other control and recording mechanisms required to keep track of over 400,000 square kilometers of land.25
Two problems with government’s absolutist approach to terminating adverse possession were noted in debate by the Liberal Reform Party, then the third party in the House of Assembly.
Firstly, this change was proposed for the first time in 1976, proposing to end adverse possession of Crown Lands within less than a year. To those who had settled 2020 CanLIIDocs 3123 on the land for a period of less than twenty years, but whose occupation had started before the proposed legislation, what would become of their interests?26 Without the legislative changes, those individuals would one day expect their interests to vest, presuming that they were aware of any such challenge to their claims to begin with. To Minister Rousseau’s view, the government of the day would give such claimants consideration morally, but not legally, should those individuals apply for Crown grants.27 Liberal member Tom Rideout raised particular concern in that context that the number of people potentially impacted by this change would number well into the thousands, and who would now be at risk of prosecution for unlawful possession of Crown land.28 Minister Rousseau responded to that concern:
[I]t would not be our intention, for example, if a man was continuously, whatever you call that term, Mr. Speaker, in open, notorious, continuous and exclusive possession of Crown lands for a reasonable period. I do not think that would apply. Something would be worked out.29
A second, related, problem was foreseen by Liberal Reform Party leader (and former Premier) Joseph R. Smallwood. Recalling the particular informal methods by
24 Ibid at 9201. One should note that “the familiar loan and mortgage” was a foreign concept to rural Newfoundland, whose impoverished population had neither the means nor opportunity to obtain one (see French, supra note 3 at 213–14 and sources cited therein). 25 House of Assembly, 31 May 1976, supra note 23 at 9229–30 (Hon Joseph Rousseau). 26 Ibid at 9225 (Roderick Moores: Liberal Reform-Carbonear). 27 Ibid. 28 Ibid at 9225–26. 29 Ibid at 9227.
234 UNBLJ RD UN-B [VOL/TOME 71 which land was transferred in centuries past, Mr. Smallwood had concern that “there must be many, many thousands in Newfoundland who acquired the land, you know, twenty, thirty, fifty, a hundred years ago, and yet no form of written title. Is that not so? It is not just those who have squatters land on January 1 next?”30 Smallwood’s concern appears to regard past transactions of land that are not necessarily “squatters rights” claims, but which would otherwise have been recognized title historically, and whether or not these changes would impact those not in possession. Minister Rousseau responded as follows:
The most important principle of this bill is that nobody is trying to do anybody out of their land. What we are saying is, “look if you have had squatters rights for twenty years, why do it for another forty years? Let us clear it up so we know that that land belongs to John Jones or John Q. citizen. We are going to give it to you after twenty years. If you do not have it up to 20 years by January 1, 1977 you go through the normal path, the normal procedure that everybody else goes through for a crown land application.” That plus the fact that we hope to have our surveys in so we know where the markers are, where we can say who owns what land across
the Province, we hope to greatly expedite the question and the problems that 2020 CanLIIDocs 3123 arise in respect to crown lands.31
With that, the bill passed without amendment, and entered the law of Newfoundland and Labrador on June 11th, 1976, as the Act to Further Amend the Crown Lands Act.32 For the purposes of this paper, the operative part of this statute would become section 134B.(2) of the Crown Lands Act of 1970, which read as follows at the time of its enactment in 1976:
The period of possession of Crown lands prior to the 1st day of January, 1977, which would, by the application of the law pertaining to the acquisition of an interest in land based upon open, notorious and exclusive possession existing prior to the enactment of this section, have been necessary to confer upon any person an interest in such land is deemed to be, and always to have been, twenty years.33
Subsection 134B.(3) of the 1976 Act provided that the Minister may issue a Crown Grant “upon being satisfied that a person has acquired an interest in Crown lands pursuant to the provisions of subsection (2)”.34
Taking the statute at face value, it is an apparent retroactive change to the law of adverse possession of Crown lands. The period of possession, which had previously
30 Ibid at 9230. 31 Ibid at 9231. 32 Act to Further Amend the Crown Lands Act, SN 1975–76, No 20, s 3. 33 Ibid. 34 Ibid.
2020] ADVERSE POSSESSION OF CROWN LANDS 235 been held to be 60 years, “is deemed to be, and always to have been” 20 years. On the plain reading of the statute, any period of twenty years prior to the enactment of that section would seem to amount to a statutory dispossession of the Crown. What changed was not the nature of adverse possession, but the length of same. The period required is deemed to always have been twenty years, much as any period of sixty years’ prior occupation had previously been exercised. Shortening the time period to twenty years would also align with the general limitation period for recovery of land in effect at the time.35 This interpretation can be confirmed by reviewing Minister Rousseau’s statements on the debate of the bill—the goal of the legislation was to make it easier for those in possession of their land to obtain title, as a compromise of sorts to the termination of adverse possession of Crown lands going forward from 1977.
As can be seen from the review of the transcripts from the House of Assembly, the legislative intent behind the 1976 amendments was not just to put an end to the centuries-old “free for all” system of simply taking Crown land. Instead, it was to usher in an era of certainty not just for government, but also for landowners, who could have some certainty in their title, and could have that certainty more 2020 CanLIIDocs 3123 quickly. The legislature’s intention was ameliorative—to ensure that those in possession of land can have title vested and to implement some certainty on land tenure, and to ensure that the Crown could better control its own land holdings for future use and development. Together these intended outcomes would lead to the better administration of land title in Newfoundland and Labrador.
III. Early Judicial Consideration of the 1976 Amendments—Ball v Day
Knowing what the legislature intended by review of the statements of the Members of the House of Assembly in debate, we must now look at how that legislation took effect in practice. Dealing only with the language of the statute and not with the express statements made in the House, how would the Courts interpret the legislative change? Regrettably, rather than furthering the ameliorative purposes of the 1976 amendments, the Court of Appeal took a much narrower view of the law, which set in motion a restrictive interpretation of the statute which continues today.
The matter came before the Newfoundland and Labrador Court of Appeal in 1982 in the case of Ball v Day.36 This is the first appellate case dealing with the 1976 legislative amendments. That case was an appeal from the District Court of Newfoundland on an action for trespass and recovery of land between two competing private claimants. The Appellant had constructed a house in 1973–74 and had applied for a Crown grant to the land on which he had built his house. As noted in the 1976 debates, Crown lands faced significant backlogs in making title determinations and
35 Limitation of Actions (Realty) Act, RSN 1970, c 207. 36 Ball v Day (1982), 38 Nfld & PEIR 365, [1982] NJ No 50 (CA) [Ball cited to Nfld & PEIR].
236 UNBLJ RD UN-B [VOL/TOME 71 issues of ascertaining potential private claims during this era. When the Appellant renewed his application for a Crown grant, the Respondent entered an adverse claim as successor in title to a previous occupant of the same land, who had been there for a period from 1917 to 1947. The Respondent’s predecessor in title (one Heber Porter) had had a dwelling house on the parcel for a period of 27 years, from 1917 to 1944, and had kept vegetable gardens on the same land for three more years, for a total of 30 years’ physical possession. The land remained vacant thereafter, but Heber Porter had sold the land at issue to the Respondent in 1963, who staked off the boundaries but did nothing else with the land. At the District Court level, the Respondent was successful. Barry DCJ held that the operation of the 1976 amendments operated retrospectively, shortening the period of adverse possession against the Crown from 60 years to 20 years. As the land had been possessed by Heber Porter for at least 20 years, the land had been adversely possessed and the Crown could not grant an interest to the Appellant.37
A unanimous Court of Appeal overturned Judge Barry’s decision, finding that “the learned trial judge ... misconstrued the amendment to the Crown Lands Act
... and he erred in his application of that Act to the factual situation before him.”38 The 2020 CanLIIDocs 3123 reason for the disagreement with Judge Barry’s decision lay in the Court’s interpretation of the doctrines of adverse possession generally. Citing to Halsbury’s Laws of England, the Court endorses the principles that a person in adverse possession has only a transmissible interest which is good against all but the rightful owner until “the statutory period has elapsed”; and if the property is abandoned before the statutory period has elapsed, then a subsequent occupier cannot rely on his predecessor’s adverse possession—the clock effectively runs de novo.39 On the facts of Ball v Day, the Court held that the thirty years of possession by Heber Porter was insufficient to divest the Crown during the era of that possession. While the Crown Lands Act amendments were retrospective, “it does not operate to revive the title of one who held possession for less than 60 years and abandoned possession before January 1st, 1977.”40
This interpretation invites some criticism. Firstly, this interpretation is at odds with the plain language of the 1976 amendments, which expressly deems the period of possession “to be, and always to have been, twenty years”. If the “statutory period” is retroactively deemed to have always been twenty years, would not the thirty year occupation of the parcel in Ball v Day meet that requirement? No issue seems to have been taken with the quality of Porter’s use of the land or the duration of his occupation, just with the timeframe in which it occurred and the legal significance of it. One could argue that the general presumption against retroactivity of legislation was applied in this instance. However, the Court of Appeal agreed with the trial judge that, in fact,
37 Ibid at para 2. 38 Ibid at para 3. 39 Ibid. 40 Ibid.
2020] ADVERSE POSSESSION OF CROWN LANDS 237 the law was intended to apply retrospectively.41 The scope of the retrospective application is where the Court of Appeal and trial level parted ways. While there is a general presumption at law against retroactivity, an express statement that the law is deemed “always to have been” would seem to be a plain statement of retroactive intent. For the law to take effect as of January 1st, 1977, no statement is required that adverse possession is deemed to always have been twenty years. One would fairly interpret the language chosen by the legislature being more expansive in effect, particularly when read in light of the legislature’s concerns and intentions.
Secondly, one notes that the facts of Ball v Day are peculiar. The original occupant of the land, Heber Porter, expressly renounced an animus possidendi in his trial evidence. Upon leaving the land in 1947, Mr. Porter expressly abandoned it. Yet there appears to be little discussion of how it was that Porter came to sell parcels of that same land in the 1960s, if he had disavowed his ownership of same.42 The evidence at trial indicates that the land remained fenced for some time thereafter, although in disrepair, and boundary markers were put in place by the Respondent upon his acquisition of same in 1963. The Appellant’s construction of a dwelling house on the lot would seem, on that evidence, to have occurred notwithstanding a visible 2020 CanLIIDocs 3123 adverse presence on the land at issue. The appeal does not explore the root of the Appellant’s title prior to his Crown Grant application, or why the Appellant felt entitled to construct his home where he did. The Applicant having applied for a Grant only after building his house in 1974, predating the Crown Lands Act amendments of 1976, would seem to invite a question of the possession as it may have existed for the 60 years prior thereto, and not just for the 1956–1977 period.43 Perhaps more importantly, it would invite some inquiry into the Appellant’s root of title to begin with, noting that the Appellant appears to have had none, whereas the Respondent’s claim was grounded in historical occupation. The appellate decision makes no reference to what evidence, if any, was adduced regarding use prior to 1917, which may or may not have been available at the time and may have been relevant to determining whether the Crown had already been dispossessed by 1973, as the law stood at that time. This may be a case of bad timing and bad facts, as it relates to the principles to draw.
Thirdly, this is a case between two private individuals. The Provincial Crown does not appear to have played any role in the proceedings or advanced any interpretation of the Act. One is left to wonder if the outcome of this case truly represented the expectations of the Crown Lands Administration, particularly in light of the Ministerial statements in the House of Assembly and the concerns expressed on all sides in the House at the time. It appears from a review of the Court of Appeal decision that the Appellant’s argument that the Crown had superior title to the Respondent would be an improper jus tertii defence to the Respondent’s assertion of
41 Ibid. 42 Ibid at para 2. 43 See ibid.
238 UNBLJ RD UN-B [VOL/TOME 71 ownership.44 The slightest possession by the Respondent, on the limited facts presented in the appellate case, should have been sufficient to repel the Appellant’s claim, as there is no indication whatsoever of the Appellant having colour of title or any other root of title himself at the time the Appellant entered into possession of the land at issue.45
Fourth, the facts of this case are almost directly analogous to the facts of Murphy v Moores and the Government of Newfoundland.46 In that case, Moores claimed damages and an injunction against Murphy for interfering with Moores’ fence, which was on land which was leased by Moores from the Crown. Murphy countersued seeking to set aside the Crown lease on the basis of Murphy’s prior possession of the land. The land at issue had been cultivated by Murphy’s great-grandfather, but had laid fallow for approximately 45 years prior to the Court action.47 However, Murphy had maintained the boundaries with wooden pegs and made sporadic use of the land for grazing. Of particular importance is the holding of Justice Kent: “mere discontinuance without entry by someone else does not amount to abandonment”.48 This is based on the continuation of the interest in the land by Murphy via inheritance, notwithstanding the limited activity on the land by Murphy himself. The maintenance 2020 CanLIIDocs 3123 of the boundary and occasional grazing was sufficient to maintain possession, and “actual and legal possession are the same thing for the purpose of the Statutes of Limitation”.49 No reference to this case is made in Ball v Day, but instead to the principles of adverse possession as stated in Halsbury’s Laws of England and as applied to the then-recent statutory amendment. Nevertheless, these principles had stood in the jurisprudence of Newfoundland and Labrador for over 40 years prior to the Ball decision, and in all likelihood to the reliance of the practicing bar and the public. It invites some question of how the doctrine of “colour of title” might impact such a claim, if the Respondent had possessed a written instrument from the original occupier of the land.50
Reviewing Ball v Day overall, the correctness of the decision can be debated insofar as the application of the statute is considered. The decision does not follow the clear language of the statute. Perhaps it is a case of bad facts making bad law, as the Court was faced with a review of peculiar circumstances—apparent abandonment of
44 House v Town of Glovertown (1977), 17 Nfld & PEIR 416 at paras 69–70, [1977] NJ No 131 (TD). 45 Ibid at paras 72–73. This principle was more recently discussed and approved by our Court of Appeal in House v Toms, 2017 NLCA 40. 46 Murphy, supra note 15. 47 Ibid at 164–65. 48 Ibid at 167. 49 Ibid at 166 [emphasis added]. However, cf Wickham’s Estate, infra note 50 at paras 184–85. Note, though, that the legislative changes at issue predate the Wickham Estate decision by about five months. 50 Wickham’s Estate v Estates of Wickham and Wickham (1977), 17 Nfld & PEIR 452 at paras 85–91, 99– 101, 129–30, [1977] NJ No 134 (TD).
2020] ADVERSE POSSESSION OF CROWN LANDS 239 the land some thirty years prior, followed by a sale to the Respondent and house construction by the Appellant, all occurring before the law changed. Nevertheless, the Court’s interpretation of the statute law in this case would shape the law of adverse possession in Newfoundland and Labrador going forward.
IV. Evolution of Land Reform—The Legislature Reconsiders
We must now look at the law not just by the actions of the legislature, but by the interpretation of the Courts. It was not just Ball v Day that led to further legislative amendment, but other developments in the jurisprudence after 1976. The dialogue between the Courts and the legislature becomes apparent as the developing caselaw appears to shape the legislature’s response and the law becomes increasingly restrictive.
Within a year of the Ball v Day decision, the issue of adverse possession of Crown land returned to the House of Assembly. While the Progressive Conservative party remained in power, the government was now led by Premier Brian Peckford, and 2020 CanLIIDocs 3123 the new head of Crown Lands was Minister of Forest Resources and Lands Charles Power. On December 21st, 1983, Bill No. 74 came before the House, to further amend the Crown Lands Act. This time, the amendments take on a very different tenor. Gone are the discussions of facilitating title to confirm existing possession, and in its place is a more punitive tone. The amendments propose to tighten up restrictions on adverse possession even further, and to specifically limit the eligibility for application for a grant based on adverse possession to the period “immediately prior to” January 1st, 1977. It is evident from the Hansard transcript that the concerns of the House have been impacted by jurisprudence. Although there is no reference to Ball v Day, there is some discussion about an unnamed Provincial Court case at Goose Bay before Judge Kean, involving a Crown land claim at Tom Luscombe’s Brook in Lake Melville, Labrador.51 It appears from the discussion that the decision was a prosecution for unlawful construction of a cabin on Crown land, and the accused had successfully beaten the charge, much to the consternation of other constituents of the Labrador Members.52 Of note in the decision is that, according to the debate, former Liberal Party leader (and then ordinary opposition Member of the House of Assembly) Edward Roberts had represented the accused in that case. Given the awareness of jurisprudence by the Members of the House, one could surmise that a Court of Appeal decision on Crown Lands would not have escaped notice.
51 The author has attempted to find out more about this case, but the decision is unreported and no details are given about the parties in the Hansard. 52 “Bill No 74, An Act To Amend The Crown Lands Act”, 2nd reading, Newfoundland and Labrador, House of Assembly, Preliminary Unedited Transcript (Hansard) 39-2, No 83 (21 December 1983) at 9479–84. One should take note of the discussion of enforcement in Newfoundland v Collingwood (1996), 138 Nfld & PEIR 1, [1996] NJ No 33 (CA) [Collingwood]. The Court of Appeal made note, in reviewing the trial evidence, that the Crown Lands representative at trial gave evidence of taking action against Mr. Collingwood to “maintain credibility with the residents of Labrador” by pursuing a claim against the “high profile cabin” of “businessmen from the island” (ibid at para 23).
240 UNBLJ RD UN-B [VOL/TOME 71
Minister Power commented on the proposed amendments as follows:
One other major one we are doing, Mr. Speaker, is to make adverse possession, I guess squatters’ rights in many persons’ terminology, make that land—if occupancy has been twenty years immediately prior to January of 1977, then certainly that land will be deemed to be, I guess, removed from the Crown lands through adverse possession. We have also put in a small amendment to make sure that persons who have occupied Crown land illegally between 1957 and 1977 are punishable by law because they have committed an offence. There is now a sort of loophole there, and unless we can prove that occupancy took place before 1957, or after 1977, it is somewhat difficult to prove our case, even though people have adversely occupied Crown land.53
With that, the Bill passed, and the operative provisions for obtaining a grant by adverse possession under section 134B were amended as follows:
(3) The Lieutenant-Governor in Council may, upon being satisfied that
(a) a person has acquired an interest in Crown lands 2020 CanLIIDocs 3123 pursuant to subsection (2); and (b) the land has been in continuous use for agricultural, business or residential purposes or for any of the purposes set out in section 14 for a twenty year period immediately prior to the first day of January, 1977, cause the Minister to issue a Crown grant to that person in respect of such land, and such Crown grant may be issued subject to those charges, exceptions or qualifications as the Lieutenant-Governor in Council may direct.
(4) Where the Crown lands affected by this section contain not more than twenty hectares, the Minister may issue the Crown grant, upon being satisfied that
(a) a person has acquired an interest in Crown lands pursuant to subsection (2); and (b) the land has been in continuous use for agricultural, business or residential purposes or for any of the purposes set out in section 14 for a twenty year period immediately prior to the first day of January, 1977, and the grant may be issued subject to those charges, exceptions or qualifications as the Minister may decide.54
The tide had turned. No longer could there be any debate about whether or not ancient possession could apply; the legislature expressly stated that it must be a
53 House of Assembly, 21 December 1983, supra note 53 at 9478. 54 Crown Lands (Amendment) Act, SN 1983, c 80, s 6.
2020] ADVERSE POSSESSION OF CROWN LANDS 241 twenty year period “immediately prior to the first day of January 1977”. The legislative change bolstered the Court of Appeal’s decision in Ball v Day by confirming the Court’s interpretation. Indeed, the Courts of Newfoundland would go on to confirm consistently that the twenty year period would have to cover the period immediately prior to January 1st, 1977, in reliance on Ball v Day.55
Of note, though, is that notwithstanding the addition of subsections 134B.(3) and (4) to the Crown Lands Act in 1983 that specify the period must be “immediately prior to” January 1st, 1977, subsection (2) was not so amended until 2016.56 Up until that period, the 1976 language had continued in effect, being that the period of possession prior to January 1st, 1977, was “considered to be, and always have been” twenty years.57 The Court of Appeal in Ring v Newfoundland and Labrador remarked on the distinction between subsections 36(2) and 36(3) of the 1991 Lands Act on this point, but remarked that the decision in Ball v Day must have been understood by the legislature to be a sufficient interpretation and thus did not require legislative intervention.58
With the Ring decision, no longer could there be a debate or 2020 CanLIIDocs 3123 misunderstanding. The twenty year period had solidified as being the twenty year period immediately prior to January 1st, 1977, being December 31st, 1956 to December 31st, 1976.59
The decisions made by both the Courts and the legislature appear to have disconnected from one of the two intended outcomes of the 1976 amendments. An increasing focus on the enforcement of Crown rights and abolition of adverse possession against the Crown overtook the secondary reason for the amendments—to confirm existing land titles up to 1976.
55 Crowley v Crowley (1984), 51 Nfld & PEIR 140, [1984] NJ No 201 (TD); Collingwood, supra note 53; Ring v Newfoundland and Labrador, 2013 NLCA 66 [Ring]. 56 An Act to Amend the Lands Act, SNL 2016, c 53, s 14. 57 Note the minor language shift in subsection (2) from Act to Further Amend the Crown Lands Act, SN 1975–76 No 20: “deemed to be, and always to have been, twenty years”; which became “considered to be, and always to have been, 20 continuous years” in the new replacement Lands Act, SNL 1991, c 36, s 36(2). This change would not appear to affect the substance of adverse possession, which had always required continuity. 58 Ring, supra note 56 at para 15. 59 Note that several cases remark on the period being from 1957–1977. However, possession which began at any date later than January 1st, 1957, would not meet the twenty year requirement. For instance, if one began occupancy in the summer of 1957, the period of possession by January 1st, 1977 would be 19.5 years. Further, it is the period “immediately prior to” January 1st, 1957, which implies an exclusion of the date of January 1st. All possession should therefore be traced back to December 31st 1956 at a minimum.
242 UNBLJ RD UN-B [VOL/TOME 71
V. Criticism of the Modern Interpretation
Looking at the end result of legislative changes and caselaw since 1982, Newfoundland and Labrador finds itself afflicted with a peculiar problem. For a province where title for centuries had rested on adverse possession against the Crown, no person can obtain title based on such adverse possession unless it occurred during a particular defined twenty year span in the mid-20th century, regardless of historical use, and regardless of one’s understanding and intention in the pre-1976 period. For a jurisdiction which traces its occupied history back as far as the 16th century, such a restrictive view of title ignores centuries of use and occupation as had long been established by the time of the legislative changes.60
The Court of Appeal’s decision in Ring cannot be said to be wrong. It reflects the holding in Ball v Day, and it reflects the statutory amendments made in 1983. However, that it is right does not mean that it is immune from criticism. The decision did appear to come as a shock to many practitioners in Newfoundland.61 And indeed, it was the legislature that followed the Court in making alterations to these provisions—Ball v Day predated 134B.(3) and (4) of the Crown Lands Act; Ring 2020 CanLIIDocs 3123 predated the amendments to 36(2) of the Lands Act.
Of note in particular in Ring is the Court of Appeal’s interpretation of the original statutory changes of 1976:
In addition, the legislative objective of section 36 may be gleaned from the two changes made to section 134B in 1976. These were to abolish thenceforth adverse possession as against the Crown and to change the required period of open, notorious, exclusive and continuous possession from sixty to twenty years. The first of the changes, abolishing claims where possession occurred after December 31, 1976, indicates an intention by the legislature to significantly limit future claims by preventing an adverse claim from succeeding based on possession occurring after the critical date. The second change, reduction from sixty to twenty years of possession, must be interpreted in this context. It could not reasonably be inferred that the legislature intended to make establishing adverse possession as against the Crown easier. That is the result that would follow if the claimant had only to demonstrate adverse possession for any
60 Prowse, supra note 3. Prowse noted from historical records that there was evidence of permanent settlement by 1522, when some forty to fifty houses were documented on the island of Newfoundland, but that it would be impossible to determine the exact date when settlement began because of the lack of recordkeeping regarding early settlement (ibid at 59). Even by the first English-sanctioned settlement efforts under the Royal Charter issued to John Guy in 1610, a settler population had established itself in Newfoundland with its own established mores (ibid at 99–100). One notes that in that early time, the interests of the settled population of this island were thoroughly disregarded by the English and treated as a nullity and Newfoundland itself treated as a tabula rasa. 61 The Law Society of Newfoundland and Labrador expressed its concerns about the Ring decision in the Lands Act Review process of 2015, highlighting this case: see Krista Connolly, Tracy Freeman & Paul Pope, “Lands Act Review Final Report” (August 2015), online (pdf): Government of Newfoundland and Labrador
2020] ADVERSE POSSESSION OF CROWN LANDS 243
continuous twenty rather than sixty years prior to January 1, 1977. The only reasonable interpretation consistent with the legislative intention restricting future adverse possession claims is that the twenty-year period means those years immediately preceding January 1, 1977. 62
As is apparent from a review of the 1976 Hansard, the intention of the legislation was to put a stop to the continuation of squatters’ claims on Crown lands going forward, but not to work injustice to those whose claims existed prior to the changes. With respect, the Court’s interpretation that the 1976 amendments were intended to make it more difficult to obtain title would seem to be incorrect. The period was shortened to allow existing title to be confirmed. The holding in Ball v Day is vulnerable to criticism itself, but ultimately that decision rested more on an interpretation of adverse possession generally more than it did on the interpretation of the statute, given that the issue in Ball was the explicit abandonment of the land and whether the Act would operate retrospectively to revive such title.
And indeed, this interpretation could be seen to work injustice in a way that defied the initial intentions. Take, for instance, the facts of the Ring case.63 Ms. Ring 2020 CanLIIDocs 3123 and her ex-husband acquired title to four acres of land from one Joan Dillon in 1975. Ms. Dillon had acquired ten acres of land from the Estate of Patrick Lawlor in 1972.64 Appended to the 1972 deed was an affidavit of possession covering the period from 1913 to 1933 by the late Patrick Lawlor. The land had remained in use for general woodcutting subsequent to Patrick Lawlor’s death in 1933. In 1986, the land adjacent to Ms. Ring’s parcel, which had also been part of the ten acres purchased by Joan Dillon in 1972, was successfully quieted in a Quieting of Titles proceeding.65 Ms. Ring’s quieting failed, and the failure was upheld on appeal, in no small part because of the absence of use from 1957 to 1977.
The decision in R v Ring, while legally correct, is problematic on a practical level. One must consider the parties expectations at the time of acquisition of the land. In 1975, when Ms. Ring acquired her land, the general expectation of sixty years of
62 Ring, supra note 56 at para 16. 63 Facts taken from the trial level decision: Ring v Newfoundland and Labrador (2012), 328 Nfld & PEIR 119, [2012] NJ No 336 (TD) [Ring (trial decision)]. 64 Deed from the Estate of Patrick Lawlor to Joan Dillon, dated 8 June 1972, registered at the Newfoundland Registry of Deeds at Vol 1310, Fol 261. The deed was registered contemporaneously, and prior to the 1976 legislative amendments. 65 Re Tweeddale Quieting of Titles, Court File No 1986 St J No 1054, certificate registered at the Newfoundland and Labrador Registry of Deeds at Roll 300, Frame 1155. A review of the Newfoundland and Labrador Land Use Atlas indicates that the 1986 quieting is immediately adjacent to Ms. Ring’s piece (in Court File No 2010 01G 4699) and of the same size and shape, and the property description in the 1986 quieting as registered indicates a common boundary with the Ring parcel of 978.5 feet. The property description in the 1986 quieting certificate expressly notes that the land is part of the conveyance from the Estate of Patrick Lawlor to Joan Dillon.
244 UNBLJ RD UN-B [VOL/TOME 71 possession would have been sufficient.66 An affidavit of possession was prepared which covered the history back to 1913, being 62 years prior to the date of purchase by Ms. Ring. In line with the ratio of the 1938 Murphy decision of the Supreme Court, the fallow woodland would seem to have continued in the legal possession of the Lawlor family until such time as the Lawlor family deeded it out, given that the Lawlors appear to have made analogous sporadic use of the land as in the Murphy case. At the time of purchase in 1975, Ms. Ring’s title may well have been satisfactory. The legislative enactments in 1976 and their subsequent legal interpretation thus had the effect of defeasing individuals whose title may well have been acceptable before that date. One cannot read Ball v Day as revoking titles which were based on 60 years’ possession against the Crown prior to the legislative changes. Ball only held that the title would not be revived if not possessed for the statutory period. As noted above, there is good reason to critique the Ball decision on its application of this principle in light of the legislative intent.
The retrospective application of the law reducing adverse possession to a period of twenty years makes sense as an ameliorative provision, to give certainty to those seeking title they already had. Looking at the Hansard, concerns were raised 2020 CanLIIDocs 3123 about Newfoundlanders applying for CMHC loans and needing to confirm title in order to obtain financing. A retroactive application of a restrictive provision is significantly more troubling. Land title effectively became a game of “musical chairs” that nobody knew they were playing—the result being that when the proverbial music stopped on January 1st, 1977, only those who had maintained use and occupation of land for the preceding twenty years could rely on their title, and not those who could trace their roots of title back for hundreds of years but had the misfortune of allowing a lapse in use in that period. This is the same concern that Joseph Smallwood raised in 1976, and which Minister Rousseau adamantly denied would be an issue. The government was not “trying to do anybody out of their land”, according to the Minister at that time. However, that is exactly the outcome that occurred. Indeed, Ms. Ring and her predecessor’s absence of use from 1972 to 1977 was noted by the trial judge.67 Many individuals acquire vacant land with an eye to building on it—would the legislature truly intend that the acquisition of vacant land, appearing to accord with the standards of the day, be undone by legislative fiat? Such an outcome does a disservice to landowners in Newfoundland and Labrador. Indeed, it would appear to become that even contemporaneous registration of title documents throughout history would still fail if not accompanied by actual possession for a period which was determined ex post facto. Such would appear to have been the case in Ball v Day, and such was the case in Ring. An individual buying land in the 1950s, 1960s or early 1970s could hardly be faulted for failing to realize that in 1976, their ownership would be judged
66 Prior to amendments to the Quieting of Titles (Amendment) Act, RSN 1976, No 19, the Crown could be defeased in a proceeding upon presentation of satisfactory proof of possession for sixty years (see Quieting of Titles Act, RSN 1970, c 324, s 12(1) and (2)). Note that amendments in SN 1975–76, No 19 introduced a specific requirement to notify the Minister responsible for Crown Lands, separate and apart from the general notice requirements, in order to defease the Crown. 67 Ring (trial decision), supra note 64 at para 58.
2020] ADVERSE POSSESSION OF CROWN LANDS 245 solely on the set period spanning 1956 to 1977, and not on the centuries of occupation and records preceding it. Smallwood’s concerns have metastasized into reality.
These concerns affect everyday Newfoundlanders and Labradoreans, and communities throughout the province. The intended outcome in 1976, which would have provided greater certainty to land title held up to that date, was lost and replaced with a punitive enforcement provision, the effects of which would not have been known to those who would be impacted by it going forward.
VI. The Modern Problem—Where do we go from here?
Considering the matters addressed above, it is apparent that the current provisions of the Lands Act regarding adverse possession of Crown lands have not been satisfactory to the people of this province, especially in areas of longstanding settlement and in small communities. The Crown’s position has become adverse to the interests of many residents, and puts an increasingly onerous burden on the public.
2020 CanLIIDocs 3123 The 1956 to 1977 period is fixed in statute law, and each passing year takes us further and further from the period of possession that must be proven. Leaving aside the question of the morality and effects of the legislative changes as interpreted over the years, a practical problem arises with the passage of time. It is simply becoming harder to prove possession back to 1956 as people with such a memory age and die off. Note as well that such witnesses to the title history would have to have been present in the community since 1956, and of such an age to appreciate the nature and quality of occupation of the land at that time. With the rural exodus from such small communities in the 1990s and continuing today, it is difficult to find individuals who meet sufficiently the requirements to prove dispossession for the required period. Indeed, in communities that may have only been a hundred or so people at their peak, it may be impossible in some instances to satisfactorily prove possession by affidavit—there simply is not anyone left alive and of sound mind who could do so. To real estate practitioners in rural Newfoundland, the problem of resolving title proves more difficult with each passing year.
In 2015, the Government of Newfoundland and Labrador appeared to take action on the problems that had developed. A review committee was established that year to solicit input from the public and from various stakeholders in matters of real estate and real property throughout the province. The goal was to determine what changes, if any, should be made to address the problems that have arisen regarding the administration of Crown lands, including the effects of the defined twenty year adverse possession period.
The final report of the Lands Act Review Committee was released in August of 2015.68 The Committee reviewed the possibility of legislative change, and in
68 Supra note 62.
246 UNBLJ RD UN-B [VOL/TOME 71 particular proposals which were advanced by interested members of the public, which included revising the fixed date to a later period (i.e. from possession ending in 1977 to possession ending at a later year), or reintroducing adverse possession for a set period of years (as had been the case before 1976). Those options were rejected by the Committee, which recommended maintaining the 1977 period.69 Nevertheless, the suggestions made to the Review Committee seem to hearken back to the initial intention of the 1976 amendments:
a) Implanting a registration “recognition” period, whereby documentation registered before a set date or for a set period of years will be deemed to have divested the Crown, as a sort of constructive possession. b) Setting a running period of years for adverse possession (as proposed by the Law Society, 20 years of possession in a municipality and 30 years of possession outside of a municipality). c) Adjust the cutoff date to another, later, period, which would cover possession made prior to the 1976 legislative change but not
possession thereafter. 2020 CanLIIDocs 3123 d) Institute a “grace period” for those in unlawful possession of Crown Lands to obtain a grant without paying market rates for same.
For various reasons these recommendations were rejected by the Provincial Government, and the status quo has remained in effect. Indeed, there may be as many negative effects as positive effects to making such changes now, given that the law and consequently the public has followed a given trajectory for over 40 years. However, with each passing year it becomes increasingly difficult to obtain affidavits of possession from individuals with knowledge of land possession from the 1950s, leading to either a casual disregard of the provisions of the Lands Act or recourse to costly remedies such as Quieting of Title applications. Given that historical title is generally accepted in certain areas (particularly in older areas of St. John’s), or that a chain of deeds may be found which predate the 1976 amendments, there is a temptation to assume that title would be acceptable, but the provisions of the Lands Act are of universal application across the province. If a parcel of land cannot be traced directly to a Crown Grant, then possession from 1956 to 1977 must be proven, regardless of whether the land is in the centre of St. John’s or a rural outport. That Crown Lands may not actively investigate every conveyance and land claim in the province does not mean that the issue does not exist—it is rather like the proverbial tree falling in the forest with nobody around to hear it. When land conveyancing becomes a matter of either ignoring the law for administrative efficacy, or investment of thousands of dollars to confirm a title already long held and recognized as against the Crown, the law and the policy cannot be viewed as successful. Changes are needed today for the good of the public and for the better administration of real property law in our province.
69 Ibid at 78, 81–87.
2020] ADVERSE POSSESSION OF CROWN LANDS 247
The issues raised in this paper are not merely esoteric questions of academic interest. Land tenure is a foundational issue for those looking to invest in our province. Buyers of land need certainty that they have a full and unencumbered title to their land before they are willing to invest in it. Those in current possession of land must have certainty of title to sell or mortgage, and development is hindered by the inability to prove clear title.70 Outstanding questions of title to those in occupancy of land is thus a matter which remains as unsettled and tenuous today as it was some 45 years ago, when the untenable status quo of the day forced the legislature to act. We find ourselves at a similar crossroads today, as rural practitioners of property law would attest.
It would not be the place of this paper or its author to recommend a particular course of action, except to highlight the historical circumstances that must be considered, including in particular the apparent intentions of the legislature in 1976. The current state of the law has failed to realize on the balance that was intended at that time—adverse possession of Crown Lands was abolished going forward, but those with historical possession at that time would have their efforts at obtaining title facilitated. This was an expressly intended goal at the time of the original amendments, 2020 CanLIIDocs 3123 as CMHC funding for Newfoundlanders and Labradoreans turned on the ability of the landowner to prove satisfactory title. Indeed, the same issue exists today for those who sell and mortgage their land, except the legislation has worked to undermine the legitimate past expectations of the citizenry. Any modern amendment to the statute must consider the interests of the people in obtaining title to their own lands and recognize that such title and such claims existed long before the idea of abolishing adverse possession entered the political landscape.
70 See e.g. the works of noted economist Hernando de Soto, in particular his 2000 book The Mystery of Capital: Why Capitalism Triumphs in the West and Fails Everywhere Else (New York: Bantam Press, 2000). De Soto’s treatise highlights the connection between poverty and unsettled property rights, which prevents individuals from leveraging property interests into a source of capital. One should consider the parallels to both the land title situation and economic situation in rural Newfoundland and Labrador.
LEGAL ETHICS AND JUDICIAL LAW CLERKS: A NEW DOCTRINAL ACCOUNT
Andrew Flavelle Martin*
Abstract
Judicial law clerks are largely overlooked in the Canadian legal literature. This article provides a new doctrinal account of the ethical obligations of law clerks that is rooted in the fact that at least some of the major work of law clerks constitutes the practice of law—and thus that law clerks’ ethics are lawyers’ ethics. It argues that the lawyer’s duty to encourage respect for the administration of justice transposes some of the ethical obligations of the judge into professional obligations of the law clerk. The article also argues that the law societies’ regulatory and disciplinary jurisdiction over law clerks is at least largely incompatible with judicial independence. 2020 CanLIIDocs 3123
Contents
Introduction……………………………………………………………………….249
Part 1: The nature of clerks’ duties: Judicial law clerks practice law…………….252
Part 2: The nature of the clerk-judge relationship: A lawyer-client relationship...257
Part 3: The ethical duties of law clerks…………………………………………...259 3A. The ethical duties of law clerks: Impartiality and outside interests……….. 261 3B. The ethical duties of law clerks: Post-service restrictions…………………. 263 3C. The ethical duties of law clerks: The duty to report fellow lawyers to the law society…………………………………………………………………………...266
Part 4: Law Society jurisdiction and judicial independence………………………268
Reflections and conclusion………………………………………………………..272
* Of the Ontario Bar; Assistant Professor, Schulich School of Law, Dalhousie University. Thanks to David Sandomierski, Alexander Corley, Ian Stedman, and Kim Brooks for comments on a draft.
2020] LEGAL ETHICS AND JUDICIAL LAW CLERKS 249
Introduction
Judicial law clerks have become an accepted and apparently integral part of the Canadian judicial apparatus. Nonetheless, there is no compelling doctrinal account of the functions of law clerks. These functions have important implications for how law clerks are regulated and how their ethical obligations compare to those of lawyers, on the one hand, and judges, on the other.
The last decade has seen an explosion of Canadian legal literature on public sector lawyers.1 However, there is little Canadian legal literature on judicial law clerks, and only some of that is recent.2 Most of this literature focuses on articulating the role and duties of law clerks,3 and most specifically deals with law clerks at the Supreme Court of Canada.4 In the only Canadian article that explicitly addresses ethics for law clerks,5 Josh Wilner argues that law clerks’ ethical duties are “derivative” of those of their judges because their functions are derivative of those of their judges.6 However,
1 Jennifer Leitch, “A Less Private Practice: Government Lawyers and Legal Ethics” (2020) 43:1 Dal LJ 315; 2020 CanLIIDocs 3123 Andrew Flavelle Martin, “The Implications of Federalism for the Regulation of Federal Government Lawyers” (2020) 43:1 Dal LJ 363; Andrew Flavelle Martin, “The Government Lawyer as Activist: A Legal Ethics Analysis” (2020) 41 Windsor Rev Leg & Soc Iss 28; John Mark Keyes, “Loyalty, Legality and Public Sector Lawyers” (2019) 97:1 Can Bar Rev 756; Andrew Flavelle Martin & Candice Telfer, “The Impact of the Honour of the Crown on the Ethical Obligations of Government Lawyers: A Duty of Honourable Dealing” (2018) 41:2 Dal LJ 443; Elizabeth Sanderson, Government Lawyering: Duties and Ethical Challenges of Government Lawyers (Toronto: LexisNexis Canada, 2018); Andrew Flavelle Martin, “Legal Ethics and the Political Activity of Government Lawyers” (2018) 49:2 Ottawa L Rev 263; Adam Dodek, “The ‘Unique Role’ of Government Lawyers in Canada” (2016) 49:1 Israel LR 23; Kerry Wilkins, “Reasoning with the Elephant: The Crown, Its Counsel and Aboriginal Law in Canada” (2016) 13 Indigenous LJ 27; Micah B Rankin, “The Trials, Tribulations and Troubling Revelations of Government Lawyers in Canada” (2014) 17:2 Leg Ethics 303; Michael H Morris & Sandra Nishikawa, “The Orphans of Legal Ethics: Why Government Lawyers Are Different—and How We Protect and Promote that Difference in Service of the Rule of Law and the Public Interest” (2013) 26 Can J Admin L & Prac 171; Patrick J Monahan, “‘In the Public Interest’: Understanding the Special Role of the Government Lawyer” (2013) 63 SCLR (2d) 43; Malliha Wilson, Taia Wong & Kevin Hille, “Professionalism and the Public Interest” (2011) 38:1 Adv Q 1; Adam M Dodek, “Lawyering at the Intersection of Public Law and Legal Ethics: Government Lawyers as Custodians of the Rule of Law” (2010) 33:1 Dal LJ 1 [Dodek, “Intersection”]; John Mark Keyes, “The Professional Responsibilities of Legislative Counsel” (2009) 3 JPPL 453; Joshua Wilner, “Service to the Nation: A Living Legal Value for Justice Lawyers in Canada” (2009) 32:1 Dal LJ 177; Allan C Hutchinson, “‘In the Public Interest’: The Responsibilities and Rights of Government Lawyers” (2008) 46:1 Osgoode Hall LJ 105. 2 Joshua Wilner, “To Be or Not to Be?: Some Legal Ethics for Judicial Law Clerks” (2011) 89:3 Can Bar Rev 611; Lorne Sossin, “The Sounds of Silence: Law Clerks, Policy Making and the Supreme Court of Canada” (1996) 30:2 UBC L Rev 279; Mitchell McInnes, Janet Bolton & Natalie Derzko, “Clerking at the Supreme Court of Canada” (1994) 33:1 Alta L Rev 58; Michael John Herman, “Law Clerking at the Supreme Court of Canada” (1975) 13:2 Osgoode Hall LJ 279. 3 McInnes, Bolton & Derzko, supra note 2; Herman, supra note 2. 4 McInnes, Bolton & Derzko, supra note 2; Herman, supra note 2; Sossin, supra note 2. 5 Wilner, supra note 2. Wilner covers bench memos, confidentiality, impartiality, post-clerkship employment, and “assessment of prospective law clerk applications”. I will address the middle three issues. 6 Wilner, supra note 2 (“[b]ecause a law clerk assists with a judge’s performance of her public functions, the clerk’s ethical duties are derivative of the judicial function” at 623). While Wilner, supra note 2
250 UNBLJ RD UN-B [VOL/TOME 71
Wilner is vague and non-committal as to the precise nature of the clerk’s status and role. Moreover, while the Canadian Judicial Council has jurisdiction over all federally appointed judges, it has no derivative jurisdiction over law clerks. Wilner’s account thus implies that law clerks are accountable, if at all, only to their judges as superiors and to the government as the employer. These kinds of accountability are weak in their protection of the public interest, particularly insofar as most law clerks are future lawyers.
In this article, I propose a new doctrinal account or conceptual framework for the ethical duties of law clerks. I argue that law clerks practice law and thus share the professional duties applicable to all lawyers. However, the identity and nature of the court as an institutional client has important implications for those professional duties. Following and elaborating on my work elsewhere, I argue that the duty to encourage respect for the administration of justice transposes some of the ethical obligations of judges into professional obligations of their law clerks as lawyers.7 While the end result is somewhat similar as under Wilner’s approach, insofar as the duties of law clerks are similar to or derived from the duties of judges, my framework has a clearer and sounder doctrinal and legal footing and has important regulatory implications. 2020 CanLIIDocs 3123
This article is organized in four parts. In Part 1, I demonstrate that at least some of the major functions of law clerks constitute the practice of law—in particular, recommendations for dispositions of appeals and applications for leave to appeal and drafting and substantive editing of decisions. Then, in Part 2, I explain that the clerk- judge relationship is best understood as a lawyer-client relationship. While the formal client is the court as an institution, for practical purposes the client is the judge or judges to whom the clerk is assigned. In Part 3, I set out the special duties of law clerks. It is here I make my core argument: that the lawyer’s duty to encourage respect for the administration of justice transposes at least some of the ethical obligations of the judge into professional obligations of the law clerk. I argue in Part 4 that purported law society jurisdiction over law clerks impedes to some extent judicial independence. While the role of law clerks constitutes the practice of law, courts are thus free to employ clerks other than lawyers or students-at-law. Finally, I conclude by reflecting on the implications of my analysis.
I begin with three key points. The first introductory point is a definitional one. I recognize that law clerks are often articled students, and I include them within
sometimes uses the phrase “legal ethics”, he does not appear to consider law clerks as lawyers and thus to consider law clerk’s ethics as squarely legal ethics. 7 See Federation of Law Societies of Canada, Model Code of Professional Conduct, Ottawa: FLSC, last amended 19 October 2019, online (pdf):
2020] LEGAL ETHICS AND JUDICIAL LAW CLERKS 251
my meaning of “lawyer”—as do the rules of professional conduct.8 For the purposes of this article, the professional duties of lawyers and articled students are identical.
Second, if judicial law clerks among other things practice law (as I will argue in Part 1), and the clerk-judge relationship is among other things a lawyer-client one (as I will argue in Part 2), why are a law clerk’s professional duties as lawyers defining and dominant—as opposed, for example, to their duties as employees? The professional duties as lawyers are defining and dominant because they are the most immutable and definitive: lawyers cannot contract out of them and, with few exceptions, clients cannot relieve lawyers from them. They are also stringent. Moreover, the fact that lawyers provide legal advice alongside other services to the same client does not negate, and indeed may complicate, their professional duties as lawyers. My analysis is grounded in the functions performed by law clerks, not their status. I argue that the defining characteristic of law clerks’ ethical duties is that all law clerks practice law, even though not all law clerks are lawyers or articled students. Thus, law clerks’ ethics are not severable from the definition of the practice of law.
The third introductory point is that my focus here is on the ethics of Canadian 2020 CanLIIDocs 3123 law clerks under Canadian law. While there is abundant literature on law clerks in other common law jurisdictions (particularly the US),9 like the Canadian literature, it focuses more on the roles and duties and selection of law clerks and their legitimacy,10
8 See FLSC Model Code, supra note 7 at r 1.1-1 (“‘lawyer’ means a member of the Society and includes a law student registered in the Society’s pre-call training program”). 9 But see e.g. Katharine G Young, “Open Chambers: High Court Associates and Supreme Court Clerks Compared” (2007) 31:2 Melbourne UL Rev 646; Abhinav Chandrachud, “From Hyderabad to Harvard: How US Law Schools Make It Worthwhile to Clerk on India’s Supreme Court” (2014) 21:1 Intl J Leg Profession 73. 10 See e.g. Paul R Baier, “The Law Clerks: Profile of an Institution” (1973) 26:6 Vand L Rev 1125; Eugene A Wright, “Observations of an Appellate Judge: The Use of Law Clerks” (1973) 26:6 Vand L Rev 1179; Lynn K Rhinehart, “Is There Gender Bias in the Judicial Law Clerk Selection Process?” (1995) 83:2 Geo LJ 575; Christopher Avery et al, “The Market for Federal Judicial Law Clerks” (2001) 68:3 U Chicago L Rev 793; David R Stras, “The Supreme Court’s Gatekeepers: The Role of Law Clerks in the Certiorari Process” (2006) 85 Tex L Rev 947, reviewing Todd C Peppers, Courtiers of the Marble Palace: The Rise and Influence of the Supreme Court Law Clerk (Stanford: Stanford University Press, 2006) and Artemus Ward & David L Weiden, Sorcerers’ Apprentices: 100 Years of Law Clerks at the United States Supreme Court (New York: New York University Press, 2006); Mark W Pletcher & Ludovic C Ghesquiere, “In Restraint of Trade: The Judicial Law Clerk Hiring Plan” (2007) 78:1 U Colo L Rev 147; Christopher Avery et al, “The New Market for Federal Judicial Law Clerks” (2007) 74:2 U Chicago L Rev 447; Todd C Peppers & Christopher Zorn, “Law Clerk Influence on Supreme Court Decision Making: An Empirical Assessment” (2008) 58:1 DePaul L Rev 51; Todd C Peppers, Michael W Giles & Bridget Tainer-Parkins, “Inside Judicial Chambers: How Federal District Court Judges Select and Use Their Law Clerks” (2008) 71:2 Alb L Rev 623; Carolyn Shapiro, “The Law Clerk Proxy Wars: Secrecy, Accountability, and Ideology in the Supreme Court” (2009) 37:1 Fla St U L Rev 101; William E Nelson et al, “The Liberal Tradition of the Supreme Court Clerkship: Its Rise, Fall, and Reincarnation?” (2009) 62:6 Vand L Rev 1747; Todd C Peppers, Michael W Giles & Bridget Tainer-Parkins, “Surgeons or Scribes?: The Role of United States Court of Appeals Law Clerks in Appellate Triage” (2014) 98:1 Marq L Rev 313; Timothy R Johnson, David R Stras & Ryan C Black, “Advice from the Bench (Memo): Clerk Influence on Supreme Court Oral Arguments” (2014) 98:1 Marq L Rev 21; Ryan C Black, Christina L Boyd & Amanda C Bryan, “Revisiting the Influence of Law Clerks on the U.S. Supreme Court’s Agenda-Setting Process” (2014) 98:1 Marq L Rev 75; Stephen L Wasby, “The World of Law Clerks: Tasks, Utilization, Reliance, and Influence” (2014) 98:1 Marq L Rev
252 UNBLJ RD UN-B [VOL/TOME 71 especially at apex courts, than on their ethics.11 Without denying the value of comparative work, this article uses a Canadian-centric approach.12
Part 1: The nature of clerks’ duties: Judicial law clerks practice law
The existing Canadian legal literature on law clerks does not explicitly address the question of whether law clerks practice law, or the related question of whether the legal and professional obligations of lawyers apply to law clerks. In order to identify and understand the ethical obligations of law clerks, and indeed whether legal ethics— including the law of lawyering—applies to them, these questions must first be answered. I do so in this part. I start by considering definitions of the practice of law. I then identify how, based on existing accounts in the legal literature, at least some of the major work of law clerks constitutes the practice of law. Indeed, it is this functional characterization that grounds my doctrinal account. That is, I argue that the defining characteristic is that all law clerks practice law, even though not all law clerks are lawyers or articled students.
2020 CanLIIDocs 3123 There is some variation among the provinces and territories around the definition of the practice of law. Most statutes on the legal profession provide a definition, or examples, or both. The most concise definition comes from Nova Scotia:
The practice of law is the application of legal principles and judgement with regard to the circumstances or objectives of a person that requires the knowledge and skill of a person trained in the law, and includes any of the following conduct on behalf of another:
(a) giving advice or counsel to persons about the persons legal rights or responsibilities or to the legal rights or responsibilities of others; (b) selecting, drafting or completing legal documents or agreements that affect the legal rights or responsibilities of a person;
111; Albert Yoon, “Law Clerks and the Institutional Design of the Federal Judiciary” (2014) 98 Marq L Rev 131; John J Szmer, Erin B Kaheny & Robert K Christensen, “Taking a Dip in the Supreme Court Clerk Pool: Gender-Based Discrepancies in Clerk Selection” (2014) 98:1 Marq L Rev 261; Christopher D Kromphardt, “Fielding an Excellent Team: Law Clerk Selection and Chambers Structure at the U.S. Supreme Court” (2014) 98:1 Marq L Rev 289; Lawrence Baum, “Hiring Supreme Court Law Clerks: Probing the Ideological Linkage between Judges and Justices” (2014) 98:1 Marq L Rev 333; Tony Mauro, “Diversity and Supreme Court Law Clerks” (2014) 98:1 Marq L Rev 361; Adam Bonica et al, “The Political Ideologies of Law Clerks” (2017) 19:1 Am L & Econ Rev 96. 11 But see e.g. John Paul Jones, “Some Ethical Considerations for Judicial Clerks” (1991) 4:4 Geo J Leg Ethics 771; David Lane, “Bush v. Gore, Vanity Fair, and a Supreme Court Law Clerk’s Duty of Confidentiality” (2005) 18:3 Geo J Leg Ethics 863; Charles W Sorenson, Jr, “Are Law Clerks Fair Game? Invading Judicial Confidentiality” (2008) 43:1 Val U L Rev 1; Chandrachud, supra note 9 at 80. 12 See e.g. Adam Dodek, “Canadian Legal Ethics: Ready for the Twenty-First Century at Last” (2008) 46 Osgoode Hall LJ 1 (“Canadian legal ethics must also attempt to situate legal ethics within a distinctly Canadian context” at 7).
2020] LEGAL ETHICS AND JUDICIAL LAW CLERKS 253
(c) representing a person before an adjudicative body including, but not limited to, preparing or filing documents or conducting discovery; (d) negotiating legal rights or responsibilities on behalf of a person.13
Legislation in the other provinces and territories is to similar effect but tends to provide more examples.14
From what little has been written in the Canadian legal literature, the role and duties of a judicial law clerk are fairly clear, at least at the Supreme Court of Canada. Michael John Herman lists eight duties of Supreme Court of Canada law clerks: “[p]reparation of ‘bench memoranda’”, i.e. “a memorandum of fact and law… the Justice will carry it with him onto the bench for purposes of reference”;15 “[a]ttendance at the oral hearing”;16 “[p]articipating in the judicial decisions”;17 “[l]egal research”;18 “[p]reparing draft reasons”;19 and “[e]diting the draft judgment”.20 Herman emphasizes that the last function, editing, “is perhaps the most valuable service that the clerk can perform.”21 He also includes a ninth potential duty: oral or written recommendations on applications for leave to appeal.22 2020 CanLIIDocs 3123
Similarly, Mitchell McInnes, Janet Bolton & Natalie Derzko observe that “[t]he primary duties [of law clerks] continue to include the preparation of leave memoranda and bench memoranda, and the provision of assistance in the drafting of reasons for judgment.”23 In the same vein, Sossin writes that “[p]robably the most important role of law clerks at the Supreme Court—and the one least susceptible to
13 Legal Profession Act, SNS 2004, c 28, s 16(1) [emphasis added]. 14 See e.g. Law Society Act, RSO 1990, c L.8 (“[f]or the purposes of this Act, a person provides legal services if the person engages in conduct that involves the application of legal principles and legal judgment with regard to the circumstances or objectives of a person”, s 1(5); s 1(6) gives examples that include “[g]ives a person advice with respect to the legal interests, rights or responsibilities of the person or of another person” and “[s]elects, drafts, completes or revises, on behalf of a person, … a document that affects the legal interests, rights or responsibilities of a person”). 15 Herman, supra note 2 at 280–81. See also Wilner, supra note 2 (who refers to bench memos as “the main work product of law clerks” at 633). 16 Herman, supra note 2 at 281–82. 17 Ibid at 282–83. 18 Ibid at 283–84. 19 Ibid at 284–85. 20 Ibid at 285–86 21 Ibid at 285. 22 Ibid at 286. 23 McInnes, Bolton & Derzko, supra note 2 at 61.
254 UNBLJ RD UN-B [VOL/TOME 71 generalizations—is the drafting of bench memoranda”.24 He also notes that, where the judge writes reasons in a given case, “the law clerk who prepared the bench memorandum will be asked to undertake some supplementary research on particularly complex issues, to edit a draft of an opinion prepared by the Justice or to work on a draft of the reasons in the case.”25
In some respects, the work of law clerks clearly does not constitute the practice of law. To the extent that leave memos or bench memos are merely summaries of the materials filed, this clerical work, while requiring knowledge and skill, is not the practice of law. Likewise, to the extent that editing judgments is a matter of style and proofreading, it too is not the practice of law. Moreover, a bare statement of the law is arguably legal information and not legal advice, and thus pure legal research may not constitute the practice of law.
However, without explicitly stating that law clerks practice law, the Canadian literature on law clerks reveals that they typically exceed this clerical work. For example, in his discussion of leave memos, Sossin generalizes that “[t]he memo 26 concludes by recommending whether or not the case should be granted leave.” As 2020 CanLIIDocs 3123 Sossin also explains, the test for leave is set out in the Supreme Court Act.27 In applying subsection 40(1) to the facts of the leave application, and expressing a conclusion (recommendation or not) as to whether the legal test is met, the law clerk is providing legal advice and thus practicing law. Sossin is correct that this test is “extremely vague and largely subjective” and that “[f]ew clerks have enough experience, or are given enough guidance, to credibly and consistently apply a standard of ‘national importance’.”28 It is nonetheless a legal test and the law clerk is providing a legal conclusion. In the words of the Nova Scotia legislation, the law clerk is applying “legal principles and judgement with regard to the circumstances or objectives of a person that requires the knowledge and skill of a person trained in the law” and more
24 Sossin, supra note 2 at 292. 25 Ibid at 296. 26 Ibid at 289. See also McInnes, Bolton & Derzko, supra note 2 (“the clerks pre-screen the applications and provide recommendations as to dispositions…. The most important, and typically longest, part of the memo examines whether or not the application should be granted [emphasis added]” at 71–72). 27 See Supreme Court Act, RSC 1985, c S-26, s 40(1):
[T]he Supreme Court is of the opinion that any question involved therein is, by reason of its public importance or the importance of any issue of law or any issue of mixed law and fact involved in that question, one that ought to be decided by the Supreme Court or is, for any other reason, of such a nature or significance as to warrant decision by it. See also e.g. for the Court of Appeal for Ontario, Courts of Justice Act, RSO 1990, c 43 (“[a]n appeal lies to the Court of Appeal from … an order of the Divisional Court, on a question that is not a question of fact alone, with leave of the Court of Appeal as provided in the rules of court”, s 6(1)(a)). 28 Sossin, supra note 2 at 289, 291.
2020] LEGAL ETHICS AND JUDICIAL LAW CLERKS 255 specifically “giving advice or counsel to persons about the persons legal rights or responsibilities or to the legal rights or responsibilities of others”.29
Likewise, in his discussion of bench memos, Sossin explains that bench memos not only “summarize the relevant arguments” but also “offer recommendations based on a thorough review of the applicable law.”30 While the former is clerical work, the latter is the development and expression of a legal opinion and thus the practice of law.
There is a similar dichotomy in law clerks’ drafting and editing of judgments. Where editing is simply proofreading, it is clerical work. However, where editing involves confirming that the statements of law are correct, including that the cases relied on are still good law, that editing is arguably the practice of law. Sossin asserts that law clerks often go beyond editing and actually draft reasons.31 He elaborates that clerks’ work goes beyond a statement of the law to provide an “opinion”, which using the criteria above I would characterize as a legal opinion about the legal rights and responsibilities of a person:
2020 CanLIIDocs 3123 First, clerks are expected to give accurate statements of the law in a particular area or with respect to particular circumstances. Second, clerks are expected to give an opinion as to whose view of the proper law—the appellant’s, the respondent's or an intervenor’s—should carry the day. In other words, clerks are asked to both express an opinion of their own and to summarize objectively the opinions of others.32
A court’s decision is a legal document and is, at least in the Canadian common law jurisdictions, law as much as a statute is law. Thus, drafting a decision likewise involves the application of the law to the facts and the giving of a legal conclusion. Even where the conclusion has been specified by the judge, the clerk in providing the draft is implicitly stating that the analysis supports the conclusion.
Based on this application of the definition of the practice of law to the existing descriptions of law clerks’ duties, at least some of the major duties of law clerks constitute the practice of law. For the reasons I discussed in the introduction, legal ethics should thus predominate in law clerks’ ethics.
29 Legal Profession Act, supra note 13, s 16(1). 30 Sossin, supra note 2 at 292 [emphasis added]. See also Wilner, supra note 2 (“[i]t is commonly recognized that when clerks prepare bench memos for their judges they are giving them legal ‘advice’” at 622). See also McInnes, Bolton & Derzko, supra note 2 (a bench memo “assists a judge in determining the more difficult question of how an appeal should be decided” at 74). See also Wilner, supra note 2 (“the purpose of a bench memo is not to convince the judge as how to decide the case but instead to lay a foundation for a decision, to facilitate a decision” at 633). 31 Sossin, supra note 2 at 296. 32 Ibid at 302.
256 UNBLJ RD UN-B [VOL/TOME 71
How, then, can this conclusion be reconciled with the fact that at least some courts do not require judicial law clerks to be articled students or lawyers?33 I certainly do not suggest that these courts are condoning the unlawful unlicensed practice of law. There are two possible answers. The more conservative answer is that such law clerks, i.e. those who are not articled students or lawyers, are implicitly acting under the supervision of a lawyer at the court, for example the executive legal officer or a staff lawyer.34 The FLSC Model Code recognizes that lawyers may delegate functions to non-lawyers other than articled students, though the lawyer maintains “complete professional responsibility” and “must directly supervise” those non-lawyers.35 This answer is problematic insofar as non-lawyers cannot, among other things, “give legal advice”, “act finally without reference to the lawyer in matters involving professional legal judgment”, or “sign correspondence containing a legal opinion”.36 Moreover, as a factual matter, the actual extent of this implicit supervision is unclear.
A bolder and better answer is that legislation on the legal profession, and the codes adopted by law societies under that legislation, do not apply to courts, including their law clerks in the performance of their duties, because such application is at least largely inconsistent with judicial independence. I will return to this argument below 2020 CanLIIDocs 3123 in Part 4.
Admittedly, if the legislatures were convinced that a judicial clerkship is a sufficiently special public role, they would be free to amend legislation on the legal profession to add an exception to the definition of the practice of law.37 (Indeed, for the purpose of law society regulation, the practice of law is whatever the legislature defines it as.) For example, legislation in some provinces explicitly provides that politicians may advocate on behalf of their constituents without practicing law.38 However, given the long history of judicial clerkships in Canada, one can infer that
33 See e.g. the Supreme Court of Canada, the Federal Court, and the Ontario Court of Appeal. 34 While the duties of these lawyers also warrant further study, they are beyond the scope of this article. 35 FLSC Model Code, supra note 7, r 6.1-1. See also ibid (“[in] this rule, a non-lawyer does not include a student-at-law”, r 6.1-2); Black v Law Society (Alberta), [1989] 1 SCR 591, 58 DLR (4th) 317, La Forest J (“[i]t must be remembered that the member lawyer takes full responsibility for the work of a non-member, whether that non-member be a student, a paralegal, or a lawyer from another jurisdiction, and to fulfill his or her obligations to the law society the member lawyer must adequately supervise all work done in his or her name” at 628–29 of SCR). 36 FLSC Model Code, supra note 7, r 6.1-3(b), (d), (k). 37 In some jurisdictions this could be done by regulation. See e.g. Legal Profession Act, supra note 13 (“this Act does not prohibit … (l) any other person or class of persons permitted by the regulations made by the Council and approved by the Governor in Council to carry on one or more of the activities referred to in subsection (1)”, s 16(4)). 38 See e.g. ibid (“[the Legal Profession Act] does not prohibit … (j) a member of (i) the House of Commons of Canada, (ii) the House of Assembly, or (iii) a council of a municipality, from acting as an advocate or representative of a person in the member’s capacity as an elected representative; (k) a member of the Senate of Canada from acting as an advocate or representative of a person in that member's capacity as a Senator”, ss 16(4)(j), (k)).
2020] LEGAL ETHICS AND JUDICIAL LAW CLERKS 257 the legislatures have declined the opportunity to do so. If, however, the implications of my analysis are undesirable, then such an amendment would be appropriate.
Part 2: The nature of the clerk-judge relationship: A lawyer-client relationship
The clerk-judge relationship is amorphous in the existing Canadian literature. However, given my conclusion that law clerks practice law, I argue that it is correctly understood as a lawyer-client relationship.
I start by considering the legal or technical nature of the clerk-judge relationship. Wilner notes that “[t]he [judge-clerk] relationship has several facets, including employer-employee, teacher-student, client-lawyer, and lawyer-lawyer…. The relationship between judges and clerks is an odd, paradoxical mix of partnership and subservience, collaboration and subordination.”39 However, understanding the ethics and professionalism obligations of law clerks requires precision about the dominant characteristic of their relationship with judges. The court may resemble a firm—with the judges as partners and the clerks as associates—or, potentially, a 2020 CanLIIDocs 3123 collection of firms sharing space, with each firm consisting of a judge and her clerks. Judges, however, are not lawyers. As the clerk is providing, among other things, legal advice, there must be a lawyer-client relationship and thus a client. Therefore, the situation is best understood as the law clerks being in-house counsel either to the court as an institutional client or to the individual judge.
While the clerk-judge relationship may be an unusual or even non-traditional lawyer-client relationship, that does not mean it cannot be a lawyer-client relationship. Wilner argues that the relationship is not a typical lawyer-client one because “clients are not normally more experienced in the law than their lawyer is” (which is true, but does not preclude the judge being a client) and that “it is inaccurate to describe a judge as a client, at least a traditional kind of client…. [A] clerk assigned to a particular judge is employed to assist the judge in performing a public function, rather than to represent the judge’s personal interests.”40 Interestingly, Wilner concludes that “the clerk’s professional role seems more akin to that of a government lawyer than to that of a lawyer representing a private client.”41 (He goes on to imagine that the client may be “even perhaps ‘the law’ itself”,42 a characterization that is so abstract as to be unhelpful.) I note that despite the differences between private practice and government practice, a government lawyer is a lawyer, and the client of a government lawyer is a client—i.e., to say that a law clerk is like a government lawyer is to say that she practices law in a lawyer-client relationship. The law clerk has a specific client, and
39 Wilner, supra note 2 at 620. 40 Ibid at 622. 41 Ibid. 42 Ibid.
258 UNBLJ RD UN-B [VOL/TOME 71 that client remains free to provide instructions to the law clerk and accept or reject the law clerk’s advice.
As Wilner notes, however, there is “an ambiguity as to whether a law clerk’s ‘client’ is his particular judge or rather the court as an institutional whole.”43 The nature of, and parties to, the relationship have implications for the ethical duties of law clerks, such as the law clerk’s duty of confidentiality.44 As Wilner notes, confidentiality is essential to the role of the law clerk.45 If each law clerk is counsel to the individual judge, then lawyer-client confidentiality applies as among the law clerks and as among the judges, with the possibility of a joint retainer where a law clerk works on a single matter for more than one judge.46 On the other hand, if each law clerk is counsel to the court as a whole, there is no confidentiality as among the law clerks and as among the judges, with the possibility of different lawyers within the same “firm” representing both sides to a matter.47 (Of course, even if the court is a collective, individual judges can impose additional obligations of confidentiality on their law clerks as a matter of contract or undertakings.48) For example, Herman notes that “[t]he Judges promote the clerks’ practice of discussing cases amongst themselves…. But the employers’ thoughts are disclosed only to aid the fuller 2020 CanLIIDocs 3123 discussion of the case and the better formulation of a legal solution. This limited sharing of the judicial confidence is ordinarily permissible.”49 Similarly, Sossin notes that “[p]erhaps as influential as the discussions between clerks and their Justices are the discussions among the law clerks themselves.”50 Wilner, however, observes that “[s]ome judges encourage this, while others do not permit it.” 51
Here a comparison to government lawyers is helpful. The government lawyer represents an institutional client, the Crown. However, on a day-to-day basis, the government lawyer receives instructions from, and provides legal advice and services to, a particular ministry or department, and confidentiality typically applies as against
43 Ibid. 44 Ibid addresses confidentiality at 637–39. Ironically, the law clerk has an explicit obligation of confidentiality, but the judge does not. See Stephen GA Pitel & Liam Ledgerwood, “Judicial Confidentiality in Canada” (2017) 43:1 Queen’s LJ 123. 45 Wilner, supra note 2 at 637. 46 See FLSC Model Code, supra note 7, r 3.4-5. 47 See ibid (“[w]here there is no dispute among the clients about the matter that is the subject of the proposed representation, two or more lawyers in a law firm may act for current clients with competing interests and may treat information received from each client as confidential and not disclose it to the other clients, provided that…”, r 3.4-4). 48 See also Wilner, supra note 2 (“the law clerk’s duty of confidentiality may be in principle be rooted in a number of sources, including moral obligation, professional responsibility, fiduciary or contractual obligation, and court rule or statute” at 637). 49 Herman, supra note 2 at 279. 50 Sossin, supra note 2 at 294. 51 Wilner, supra note 2 at 638.
2020] LEGAL ETHICS AND JUDICIAL LAW CLERKS 259 other ministries or departments. Likewise, though the client of the law clerk is the court as an institutional whole, he or she is assigned to provide services to a particular judge or judges who are the de facto clients. As the formal client is the court itself, it is ultimately up to the court to determine how confidentiality and other considerations will apply when clerks are assigned to specific judges.
Part 3: The ethical duties of law clerks
Having established that law clerks practice law, and noted that the clerk-judge relationship is a lawyer-client one, I can now explain how and why the ethical duties of law clerks are professional duties of lawyers. At first glance, this may seem like a hybrid set of duties, combining the professional duties of a lawyer and the ethical duties of a judge. However, law clerks are not judges.52 The law clerk practices law in a lawyer-client relationship and thus shares the professional duties applicable to all lawyers. But the identity and nature of the client (whether the court, formally, or the individual judge, in practice) has important implications for those professional duties or imposes additional professional duties: at least some of the duties applicable to 2020 CanLIIDocs 3123 judges are transposed onto their lawyers via the duty to encourage respect for the administration of justice.
This approach has some overlap with Wilner’s approach, under which the positions and functions of the lawyer are “derivative” of those of the judge,53 and so the duties of the law clerk are likewise “derivative” of the duties of the judge:
The nature of their [clerks’] participation is derivative of their professional relationship with their judge and their judge’s duties. Clerks’ functions are completely dependent upon their judges’ functions. That clerks are judicial agents explains the similarities between legal ethics for law clerks and judicial ethics.54
He later notes that “[i]t is this aspect of the relationship—the fact that law clerks are ‘personal extensions’ of the judge—which explains the link between judicial ethics and law clerks’ ethics… Clerks are the trusted agents of ‘their’ judge, and as such clerks may be bound by the judicial standards binding their principal.”55 Wilner argues
52 See e.g. ibid at 612, 621, 622. 53 See e.g. ibid at 612:
That law clerks’ ethics tracks some of the basic concerns in judicial ethics reflects a fundamental tension inherent in the clerkship institution, namely, that clerks are not judges even though their functions overlap with those of judges. That tension arises not because clerks are “parajudges,” but because their functions are derivative of the judicial function, arising out of and dependent upon the judicial office held by the judge they serve [citation omitted]. 54 Ibid at 616. See also ibid (“[t]he derivative character of the law clerk’s responsibilities, being dependant on the judicial function, explains the interrelated and overlapping nature of law clerk ethics and judicial ethics. The agent must adopt and live up to the ethical standards of his principal” at 641). 55 Ibid at 621.
260 UNBLJ RD UN-B [VOL/TOME 71 that “clerks themselves have public duties flowing from the judicial office held by the judge whom they serve.”56 Indeed, he is explicit that “[t]he agent must adopt and live up to the ethical standards of his principal”.57 He argues that “[t]he law clerk’s duty is derivative of the judge’s duty, and the judge is tasked with enforcing it”58 and refers to “ethical obligations” and “public obligations” interchangeably.59
I argue instead that it is the duty to encourage respect for the administration of justice that does the deriving or extending of judges’ duties into lawyer’s duties, within a relationship that is not solely agent-principal but more specifically lawyer- client.60 In general, duties of the client do not always, or even often, automatically become duties of the lawyer.61 The lawyer’s duty to encourage respect for administration of justice is the mechanism by which ethical duties of the judge become professional duties of the law clerk.
The Model Code of the Federation of Law Societies of Canada provides that “[a] lawyer must encourage public respect for and try to improve the administration of justice.”62 While this duty is not often the subject of disciplinary proceedings, they typically involve “unsupported allegations of bias or wrongdoing against judges or 2020 CanLIIDocs 3123 courts.”63 In contrast, law clerks by their actions may discourage public respect for the administration of justice not because they allege wrongdoing but because clerks and their functions are closely connected to their courts and judges and unavoidably reflect on them in the perception of the legal profession and the public. Thus, “[i]n effect, the rules and laws governing … judges and members of courts and tribunals are transposed into professional obligations of their counsel.”64
In order to determine the content of these additional professional duties, the relevant question is whether the absence or breach of such duties would discourage respect for the administration of justice in the mind of a reasonably informed member of the public, given the substantive nature of the work done by law clerks and their close working relationships with their judges. As Wilner puts it, “[e]verything a clerk
56 Ibid at 623. 57 Ibid at 641. 58 Ibid at 621. 59 Ibid at 623. 60 Martin, supra note 1 at 292. 61 See e.g. Deborah MacNair, “In the Service of the Crown: Are Ethical Obligations Different for Government Counsel?” (2005) 84:3 Can Bar Rev 501 (“[i]t is easy to confuse the ethical obligations of government counsel with the legal obligations of clients” at 530), quoted in Andrew Flavelle Martin & Candice Telfer, “The Impact of the Honour of the Crown on the Ethical Obligations of Government Lawyers: A Duty of Honourable Dealing” (2018) 41:2 Dal LJ 443 at 457. 62 FLSC Model Code, supra note 7, r 5.6-1. 63 Martin, supra note 1 at 291. See also the list of cases with descriptions (ibid, n 106). 64 Ibid at 292.
2020] LEGAL ETHICS AND JUDICIAL LAW CLERKS 261 does reflects back on his or her judge.”65 For example, Wilner notes that “[t]he justification for the requirement that law clerks be impartial… would seem to be rooted entirely in the detrimental manner in which a biased clerk would reflect on the perception of the judge’s role in administering justice.”66
The archetype here is political activity. Judges are restricted from engaging in political activity in order to promote their impartiality and public confidence in that impartiality.67 Political activity by law clerks would appear, to a reasonably informed member of the public, to detract from the impartiality or the appearance of impartiality of the judges who those lawyers serve.68 Thus, under the analysis set out above, judges’ ethical duty to refrain from political activity is transposed into a professional duty of the their law clerks to likewise refrain.
Before proceeding further, I note that one rule of professional conduct has particular importance given the nature of the clerk’s function. Because not all of a clerk’s duties constitute the practice of law or giving legal advice, clerks should be particularly careful to distinguish their legal advice from their non-legal advice.69
2020 CanLIIDocs 3123 I acknowledge that many of the professional duties of law clerks are the same as those of all lawyers. However, some of these duties have special implications for law clerks. Some of these special duties apply to the conduct of law clerks after they have left the court. But one of the most important—a bar on political activity and constraints on other civic and charitable activity—applies only during their service to the judiciary.
3A. The ethical duties of law clerks: Impartiality and outside interests
Like political activity, other outside interests raise problems for law clerks in the same way that they raise problems for judges.70 Law clerks share their judges’ obligations
65 Wilner, supra note 2 at 621. 66 Ibid at 623. 67 Canadian Judicial Council, Ethical Principles for Judges, Ottawa: CJC, 2004, at 28, principle d(1), online (pdf):
In addition to opinions on legal questions, a lawyer may be asked for or may be expected to give advice on non-legal matters such as the business, economic, policy or social complications involved in the question or the course the client should choose. In many instances the lawyer’s experience will be such that the lawyer’s views on non-legal matters will be of real benefit to the client. The lawyer who expresses views on such matters should, if necessary and to the extent necessary, point out any lack of experience or other qualification in the particular field and should clearly distinguish legal advice from other advice. 70 Wilner, supra note 2 addresses impartiality at 639–40.
262 UNBLJ RD UN-B [VOL/TOME 71 of impartiality—they “must not only be impartial but also be seen to be.”71 Partiality or partisanship by law clerks calls into question the impartiality of their judges.
An important example of these problems is fundraising. Fundraising by judges raises several concerns, primarily about buying influence with, or preferential access to, the judges involved.72 As Pitel & Malecki put it: “[a] common concern… is that donors (which may include lawyers and potential litigants appearing before a judge) may contribute with the expectation that they will receive future favours from the soliciting judge in return for their generosity.”73 There is a further concern that judges will become “publicly identifie[d] … with the objectives of the organization for whom the solicitations were made…. this heightens the risk that the public could question the judge’s neutrality.”74 Similarly there is concern that solicitation by a judge “may cause people to feel intimidated or coerced into donating.”75 Thus, the Canadian Judicial Council’s Ethical Principles for Judges provides that “[j]udges should not solicit funds (except from judicial colleagues or for appropriate judicial purposes) or lend the prestige of judicial office to such solicitations.”76 Fundraising by their law clerks raises similar concerns. Here law clerks should also be guided by the rules of professional conduct on outside interests: “[a] lawyer must not allow involvement in 2020 CanLIIDocs 3123 an outside interest to impair the exercise of the lawyer’s independent judgment on behalf of a client.”77
Wilner’s conclusion is that law clerks “must be prudent in any civic, charitable, and political activities in which they take part.”78 I have observed elsewhere that in the context of political activity, however,79 mere prudence seems insufficient for law clerks, and they are better advised to refrain completely. It is worth emphasizing here that Ethical Principles for Judges treats political activity, and particularly partisan political activity, differently than other civic and charitable activities. Such civic and charitable activities may be permissible for judges, and so for those activities prudence is an appropriate and sufficient approach for law clerks.
71 Ibid at 639. 72 See e.g. Stephen GA Pitel & Michal Malecki, “Judicial Fundraising in Canada” (2015) 52 Alta L Rev 519. 73 Ibid at 531. See also Ethical Principles, supra note 67 at 36, commentary c(6). 74 Pitel & Malecki, supra note 72 at 531. See also Ethical Principles, supra note 67 at 36, commentary c(6). 75 Pitel & Malecki, supra note 72 at 533. 76 Ethical Principles, supra note 67 at ch 6(c)(1)(b). 77 FLSC Model Code, supra note 7, r 7.3-2. 78 Wilner, supra note 2 at 640. 79 Martin, supra note 1 (“I would argue that no political activity is prudent for a law clerk” at 292, n 109).
2020] LEGAL ETHICS AND JUDICIAL LAW CLERKS 263
3B. The ethical duties of law clerks: Post-service restrictions
Judges face restrictions on their return to the practice of law.80 Although there is variation among the provinces and territories, the FLSC Model Code provides that a former judge cannot, in the three years following her departure from a court, appear before that court or any court or tribunal over which that court has jurisdiction.81 Should their law clerks also face restrictions when they leave a court?
Of course, the general rule on appearing in front of judges with whom a lawyer has a close relationship will apply to former law clerks:
When acting as an advocate, a lawyer must not: … appear before a judicial officer when the lawyer, the lawyer’s associates or the client have business or personal relationships with the officer that give rise to or might reasonably appear to give rise to pressure, influence or inducement affecting the impartiality of the officer, unless all parties consent and it is in the interests of justice[.]82
The application of this rule turns on the nature of the relationship between the judge 2020 CanLIIDocs 3123 and the law clerk. Wilner notes that some judges impose a “cooling-off period” during which they will not allow their former clerks to appear before them.83 But what about other judges of the same court, or judges of inferior courts?
On one level, law clerks are differently situated than judges. Unlike the long- term career that comes from a judicial appointment, service as a law clerk is transient, even fleeting.84 Law clerks are near the beginning of their legal careers.85 In contrast, retired judges receive considerable pensions.86 Restrictions on post-service court appearances would be a considerable deterrent from taking a clerkship, particularly for potential clerks whose career plans contemplate litigation. At the same time, some of the reasons to be wary of judges returning to practice do also apply to their law clerks. Parties and their counsel might be at a perceived or actual disadvantage when facing opposing counsel who recently served the Court in which they are appearing. As Pitel & Bortolin put it, “[o]ne concern
80 FLSC Model Code, supra note 7, r 7.7-1; Stephen GA Pitel & Will Bortolin, “Revising Canada’s Ethical Rules for Judges Returning to Practice” (2011) 34:2 Dal LJ 483. 81 FLSC Mode Code, supra note 7, r 7.7-1. There is an exception where the law society “approves on the basis of exceptional circumstances”. For an example of “exceptional circumstances”, see LaForme v Law Society of Ontario, 2020 ONLSTH 112. 82 Ibid, r 5.1-2(c). Pitel & Bortolin, supra note 80 at 488–90 refer to these kinds of rules as “situation-specific restrictions” and emphasize that they “focus on the relationship between the former judge and the judge before whom he or she is appearing… not… on conflict of interest or confidentiality concerns.” 83 Wilner, supra note 2 at 640. 84 Ibid. 85 Ibid; McInnes, Bolton & Derzko, supra note 2 at 61, 77. 86 Pitel & Bortolin, supra note 80 at 522.
264 UNBLJ RD UN-B [VOL/TOME 71 associated with former judges is that their arguments before a court will be more persuasive than they ought to be.”87 While Pitel & Bortolin argue that such special influence is unlikely in reality, they note that the public might perceive it, and so the restriction is necessary to protect the appearance of impartiality.88 The same concern, although lesser in degree, would seem to apply to law clerks appearing in front of the court they served. Lawyers who have recently acted as the court’s trusted advisors conceivably have more credibility and persuasiveness than other lawyers.
On the other hand, there is little reason to think that this concern would apply to judges of courts inferior to or superior to the court in which a law clerk served. These judges have had no interaction with, and no relationship with, the former law clerk.
Given the short-term nature of law clerks’ service, and the danger of deterring applicants from such service, a cooling-off period would be preferable to a permanent prohibition. Pitel & Bortolin argue that such periods for former judges are an arbitrary and absurd compromise between a permanent prohibition and no prohibition.89
However, for former law clerks, such an arbitrary period, perhaps of one year, has an 2020 CanLIIDocs 3123 important symbolic value and promotes public confidence in the administration of justice. For example, the Supreme Court of Canada’s Confidentiality and Conflict of Interest Declaration provides that “[f]ormer law clerks shall not, within a period of two years after terminating their clerkship, without consent of a judge, appear as counsel in any proceeding before the Supreme Court of Canada.”90 Such a period would be more important for clerks at lower courts, given the rarity of a recently-called lawyer appearing at the Supreme Court of Canada.
In contrast, prohibitions on soliciting post-service practice opportunities are equally warranted—though not equally feasible—for law clerks as they are for judges.91 As Pitel & Bortolin note, there is nothing specific in Ethical Principles for Judges that “would prohibit judges from negotiating employment with parties to a dispute or their lawyers… [a]lthough this is already an obvious violation of judicial ethics principles”.92 As Wilner notes, law clerks doing the same creates “possible conflicts of interests and the appearance of impropriety”.93 While Wilner does not advocate an absolute prohibition, he suggests that “[w]hen in doubt, during their
87 Ibid at 512. 88 Ibid at 513. 89 Ibid. But see also ibid (“[a] time-limited restriction on appearances would make more sense if it was limited to the same court in which the former judge had served” at 514). 90 Email from Samson Rémi, Senior Legal Officer, Supreme Court of Canada (13 March 2019) [on file with author]. 91 Wilner, supra note 2 addresses “post-clerkship employment” at 640–41. 92 Pitel & Bortolin, supra note 80 at 520. 93 Wilner, supra note 2 at 640.
2020] LEGAL ETHICS AND JUDICIAL LAW CLERKS 265 employment it would be prudent for clerks to inform their judges about any potential conflicts of interest raised by their job hunting.”94 Here too a cooling-off period would promote public confidence in the administration of justice. But in the age of the mega- firm, a law clerk serving for a single year would be de facto prevented from seeking employment. While I appreciate the concurrence by Cory J in MacDonald v Martin Estate that “[n]either the merger of law firms nor the mobility of lawyers can be permitted to affect adversely the public's confidence in the judicial system.”95 I acknowledge that such a cooling-off period for law clerks would be impractical in practice and destructive to the institution of clerking.
Law clerks also face restrictions on public comment following their service. During their service, law clerks have a clear duty to avoid public comment because of their duties of loyalty and confidentiality.96 But what about after their departure from the court?
It is generally understood that judgments speak for themselves and judges should not comment on their previous cases: “The rule that judges should speak, or explain themselves, only once, through their judgments, is a wise and salutary one, 2020 CanLIIDocs 3123 based on the long experience of the common law…. They are expected to speak only through their Reasons for Judgment, and thereafter never to explain their judgments.”97 Indeed, this is one of the dangers considered by Pitel & Bortolin for judges returning to court and potentially arguing against one of their previous decisions.98 A similar concern appears to apply to former law clerks who become litigators or academics, which are two very common careers for former law clerks.99 Can a litigator, or an academic, argue or write about decisions that they worked on? What about decisions by the same judge during the law clerk’s service? What about decisions by other judges of the same court? This situation invokes both the duty of confidentiality and the broader duty of loyalty to clients, both present and former.100
94 Ibid. 95 MacDonald Estate v Martin, [1990] 3 SCR 1235 at 1265, 77 DLR (4th) 249. 96 See e.g. Wilner, supra note 2 at 638–39. As I have explained, under my approach these duties of confidentiality and loyalty are a lawyer’s duties. 97 Report to the Canadian Judicial Council of the Inquiry Committee Established Pursuant to Subsection 63(1) of the Judges Act at the Request of the Attorney General of Nova Scotia (Ottawa: CJC, 27 August 1990) (Allan McEachern CJ, Chairman), at 23–25, online (pdf): Canadian Judicial Council
266 UNBLJ RD UN-B [VOL/TOME 71
The identity of the judges who wrote a decision forms part of the reasons and is readily apparent on its face in a way that the identity of the law clerks who worked on a decision is not. However, the identity of a judge’s clerk and their dates of service are readily available. While technically the identity of a client and the nature of the services provided are privileged and confidential,101 many law clerks are open with the names of the judge(s) for whom they worked, both during and after their service.
Here an analogy can be drawn to the duty of a lawyer not to comment on the strength of a client’s case.102 The former law clerk has, or may be perceived as having, inside knowledge about the intended meaning and interpretation of a decision on which she worked as well as its potential weaknesses. Attacking the credibility or applicability of such a decision undoes the work for which the lawyer was responsible.103
On the other hand, a litigator is not free to simply ignore such decisions or pretend that they do not exist; indeed, such an approach would violate the duty of competence to the current client.104
2020 CanLIIDocs 3123 The best course for the former law clerk, in any situation where a decision issued by the court during their service is relevant, is to choose their words carefully, avoid speculating as to the judge’s intended meaning, and avoid insofar as possible commenting on the merits of the decision. This is most important when the decision is one in which the law clerk was involved, but is also advisable for any decision of the judge(s), and even the court, for which the law clerk worked during that service.
3C. The ethical duties of law clerks: The duty to report fellow lawyers to the law society
I have argued above that law clerks have some special professional duties, flowing from their professional duty to encourage respect for the administration of justice.
101 Ibid (“[g]enerally, unless the nature of the matter requires such disclosure, a lawyer should not disclose having been: (a) retained by a person about a particular matter; or (b) consulted by a person about a particular matter, whether or not the lawyer-client relationship has been established between them”, r 3.3-1, commentary 5). 102 Ibid (“[a] lawyer should refrain from expressing the lawyer's personal opinions on the merits of a client's case to a court or tribunal”, r 5.1-1, commentary 5). 103 See e.g. Stewart v Canadian Broadcasting Corp (1997), 150 DLR (4th) 24, 32 OTC 321 (Gen Div) (“Mr. Greenspan not only revisited the future benefits and protections he had worked to provide to Mr. Stewart as his counsel, he undermined them…. He re-visited and undermined the future benefits and protections which he had provided to Mr. Stewart as his counsel” at paras 278, 280 of DLR). 104 See FLSC Model Code, supra note 7, r 3.1-1(a):
‘Competent lawyer’ means a lawyer who has and applies relevant knowledge, skills and attributes in a manner appropriate to each matter undertaken on behalf of a client and the nature and terms of the lawyer’s engagement, including: (a) knowing general legal principles and procedures and the substantive law and procedure for the areas of law in which the lawyer practises.
2020] LEGAL ETHICS AND JUDICIAL LAW CLERKS 267
Before moving on, it is important to emphasize that, and how, another of the professional duties of all lawyers applies to law clerks: the duty to report fellow lawyers to the law society. This is an example of how the duties of law clerks diverge from, rather than converge with, the duties of their judges. Put another way, the duties of judges may supplement, but cannot negate, the professional duties of law clerks. (Another example is that while judges lack an express duty of confidentiality,105 law clerks are doubtlessly bound by the lawyer’s duty of confidentiality.)
Rule 7.1-3 of the FLSC Model Code requires lawyers to report other lawyers to the law society in a range of circumstances:
Unless to do so would be unlawful or would involve a breach of solicitor- client privilege, a lawyer must report to the Society:
(a) the misappropriation or misapplication of trust monies; (b) the abandonment of a law practice; (c) participation in criminal activity related to a lawyer’s practice; 2020 CanLIIDocs 3123 (d) conduct that raises a substantial question as to another lawyer’s honesty, trustworthiness, or competency as a lawyer; (e) conduct that raises a substantial question about the lawyer’s capacity to provide professional services; and (f) any situation in which a lawyer’s clients are likely to be materially prejudiced.106
This duty applies to all lawyers individually. Thus, for example, an articled student or junior lawyer has the duty to report regardless of whether the senior lawyer for whom they work, or the firm’s managing partner, agrees that a report is necessary or appropriate.
Judges do not have the same duty to report lawyers to the law society—if in fact they have such a duty at all. Ethical Principles for Judges provides the following guidance:
It is a delicate question whether and in what circumstances a judge should report, or cause to be reported, a lawyer to the lawyer’s professional governing body…. [G]enerally a judge should take, or cause to be taken, appropriate action where the judge has clear and reliable evidence of serious misconduct or gross incompetence by a lawyer.107
105 See above note 44. 106 FLSC Model Code, supra note 7, r 7.1-3. For a discussion see John Chapman, “Am I My Partner’s Keeper? The Duty to Report a Colleague” (2013) 92:3 Can Bar Rev 611. 107 Ethical Principles, supra note 67 at 22, commentary 14. The commentary also deals with when such a report should be made, i.e. during a proceeding versus after the proceeding has concluded.
268 UNBLJ RD UN-B [VOL/TOME 71
Judges may witness firsthand “conduct that raises a substantial question as to another lawyer’s honesty, trustworthiness, or competency as a lawyer” or “conduct that raises a substantial question about the lawyer’s capacity to provide professional services”. They also may learn of such conduct secondhand from the record before them, be that conduct that is the subject matter of the litigation, for example in a malicious prosecution matter, or conduct by counsel below, for example in an appeal based on inadequate assistance by counsel. But reporting is only triggered when it reaches the threshold specified in Ethical Principles. Moreover, Ethical Principles itself is explicitly not a binding code,108 and thus it is inaccurate to refer to this as a reporting duty.
In contrast, law clerks—who may become aware of the same conduct when their judge does, or before or after—do have a duty to report it to the law society, with a lower threshold than judges’ reporting. Lawyers are required to report not only “serious misconduct” or “gross incompetence”, but also “conduct raising a substantial question as to” integrity or capacity. In the same way that a senior partner cannot decide that his juniors will not make a report, a judge has no ability to absolve a law clerk of this duty. Even if the judge does not want to make a report, indeed even where 2020 CanLIIDocs 3123 the judge determines that the conduct would not merit a report under FLSC Model Code rule 7.1-3, the law clerk has a duty to use her own judgement and make a report if she determines that one is required. The only exception to this duty is where the information for the report is privileged.109
Part 4: Law Society jurisdiction and judicial independence
In this part I argue that, while law clerks practice law and law clerks’ ethics are legal ethics, the law societies’ regulatory and disciplinary jurisdiction over law clerks is at least largely incompatible with judicial independence. More specifically, the requirement that law clerks be articled students or lawyers under the jurisdiction of, and in good standing with, the law societies impedes the constitutionally protected administrative independence of the judiciary. This is because each law society is external to the judiciary and exercises power, albeit independently, delegated through legislation.110
108 Ibid at 3, principles 1–2:
The Statements, Principles and Commentaries describe the very high standards toward which all judges strive…. The Statements, Principles and Commentaries are advisory in nature. Their goals are to assist judges with the difficult ethical and professional issues which confront them and to assist members of the public to better understand the judicial role. They are not and shall not be used as a code or a list of prohibited behaviours. They do not set out standards defining judicial misconduct. 109 FLSC Model Code, supra note 7 (“[u]nless to do so would be unlawful or would involve a breach of solicitor-client privilege…”, r 7.1-3). 110 See e.g. Legal Profession Act, supra note 13.
2020] LEGAL ETHICS AND JUDICIAL LAW CLERKS 269
In the PEI Judges Reference, Lamer CJ identified the three core characteristics of judicial independence as “[s]ecurity of tenure, financial security, and administrative independence”, and the two dimensions of those characteristics as individual and institutional.111 Chief Justice Lamer quoted approvingly from Valente v The Queen on the meaning of administrative independence: “the administrative decisions that bear directly and immediately on the exercise of the judicial function…. assignment of judges, sittings of the court, and court lists—as well as the related matters of allocation of court rooms and direction of the administrative staff engaged in carrying out these functions”.112 He also specified that “administrative independence… only attaches to the court as an institution (although sometimes it may be exercised on behalf of a court by its chief judge or justice.)”, and thus has only an institutional and not individual dimension.113
While the principle of judicial independence does not preclude legislative and executive power over the justice system, it does constrain the exercise of that power. For example, “[l]egislatures have the constitutional power over the creation, transformation and abolition of judicial offices. However, legislatures must exercise this power in a way that complies with the constitutional principle of judicial 2020 CanLIIDocs 3123 independence.”114 This constraint applies specifically to protect administrative independence:
The legislative powers exercised by the province under s. 92 [of the Constitution Act, 1867] are subject to constitutional requirements and, in particular, are limited by the principles of judicial independence. Accordingly, the provincial Legislature's authority over the administration of justice in Ontario, and the Attorney General's statutory duties regarding the administration of justice, must be exercised consistently with the principles of judicial administrative independence.115
Thus, both legislative and executive power are constrained by the administrative independence of the judiciary.
The question thus becomes whether the “direction of the administrative staff engaged in carrying out these functions”, noted by Lamer CJ as a facet of
111 Reference re Remuneration of Judges of the Provincial Court (PEI), [1997] 3 SCR 3 at paras 119–20, 150 DLR (4th) 577 [PEI Judges Reference]. See also Provincial Court Judges’ Assn of New Brunswick v New Brunswick (Minister of Justice); Ontario Judges’ Assn v Ontario (Management Board); Bodner v Alberta; Conférence des juges du Québec v Québec (AG); Minc v Québec (AG), 2005 SCC 44 at paras 5– 7. 112 PEI Judges Reference, supra note 111 at para 117, quoting from Valente v The Queen, [1985] 2 SCR 673 at 709, 712, 24 DLR (4th) 161. 113 Ibid at para 120. 114 Conférence des juges de paix magistrats du Québec v Québec (AG), 2016 SCC 39 at para 1 [Conférence des juges]. 115 Ontario (Ministry of the AG) v Ontario (Information and Privacy Commissioner), 2011 ONSC 172 (Div Ct) at para 33.
270 UNBLJ RD UN-B [VOL/TOME 71 administrative independence, encompasses the power to select law clerks and the exclusive power to discipline them. If it does, the specific interference with judicial independence is the ability of the executive, acting through the self-regulating law society, to restrict the judiciary’s choice of law clerks through the law society’s regulatory and disciplinary powers. That is, legislation on the legal profession may purport to provide that the judiciary may only employ clerks who are lawyers admitted to the bar (or articled students admitted to the licensing process), and the law society can discipline these lawyers and articled students for conduct in the course of their judicial service. In other words, the court can only direct law clerks who are in good standing with the law society and so long as they remain in good standing with the law society.
Despite the extensive and obvious differences between law clerks and prosecutors, at a constitutional level there are similarities between prosecutorial independence, as discussed in Krieger v Law Society of Alberta, and judicial independence as it relates to law clerks.116 The court in Krieger recognized that the constitutional principle of prosecutorial independence restricted, but not eliminated, the ability of both courts and law societies to interfere with the conduct of Crown 2020 CanLIIDocs 3123 prosecutors:
It is a constitutional principle that the Attorneys General of this country must act independently of partisan concerns when exercising their delegated sovereign authority to initiate, continue or terminate prosecutions. So long as they are made honestly and in good faith, prosecutorial decisions related to this authority are protected by the doctrine of prosecutorial discretion.117
It followed from this principle that the law societies could not, and the courts would not, interfere with conduct of Crown prosecutors acting within prosecutorial discretion.118 Likewise, judicial independence is a constitutional principle that, I argue, restricts law society disciplinary and regulatory jurisdiction.119
Krieger, however, was fundamentally about the relationship between the law society and the Attorney General. Unlike prosecutors, the question for law clerks is about the respective roles of the law society and the judiciary. In Krieger, the
116 Krieger v Law Society of Alberta, 2002 SCC 65 [Krieger]. 117 Ibid at para 3. 118 Ibid at paras 3–4. 119 See PEI Judges Reference, supra note 111 at para 83:
Notwithstanding the presence of s 11(d) of the Charter, and ss 96–100 of the Constitution Act, 1867, I am of the view that judicial independence is at root an unwritten constitutional principle, in the sense that it is exterior to the particular sections of the Constitution Acts. The existence of that principle, whose origins can be traced to the Act of Settlement of 1701, is recognized and affirmed by the preamble to the Constitution Act, 1867. See more recently Conférence des juges, supra note 114 at para 31.
2020] LEGAL ETHICS AND JUDICIAL LAW CLERKS 271 unsuccessful argument was that the Attorney General should have sole control over the regulation of Crown prosecutors.120 However, the law society’s regulatory role was narrowed to conduct outside prosecutorial independence, including dishonesty and bad faith of a Crown prosecutor.121
An argument parallel to that in Krieger would be stronger in the context of judicial independence than in the context of prosecutorial independence.122 Law societies’ regulatory and disciplinary jurisdiction over law clerks, in the performance of their professional duties, is to some extent contrary to judicial independence. However, like in Krieger, that jurisdiction is narrowed but not necessarily eliminated by the relevant constitutional principle. In a similar way as law societies maintain jurisdiction over Crown prosecutors outside prosecutorial discretion, law societies may maintain jurisdiction over law clerks outside the course of their duties. For example, a law clerk’s extraprofessional misconduct could arguably be subject to law society regulation, as could bad-faith conduct, if he or she happened to be an articled student or a lawyer.
It would also follow from my analysis that law clerks, during their clerkships, 2020 CanLIIDocs 3123 are exempt from any regulatory or disciplinary consequences for not complying with basic regulatory requirements for good standing that apply to all lawyers, such as mandatory continuing professional development.
I acknowledge here that my analysis may appear to undercut itself. I have argued that law clerks practice law and are subject to the professional obligations of lawyers, but I have also argued that law clerks, in the course of their practice, are beyond law society jurisdiction. I have two responses. The narrower response is that law society jurisdiction over law clerks may merely be restricted, not eliminated. The broader response is that an exclusive focus on enforcement and discipline reflects a thin conception of legal ethics.
I have argued that it is the lawyer role in the lawyer-client relationship, as opposed for example to the employee role in the employer-employee relationship, that is determinative of law clerks’ ethics. Nonetheless, as in Krieger, the relevant constitutional principle may mean that no professional or regulatory consequences, only employment consequences, are available for misconduct within the scope of the constitutional principle. In Krieger, prosecutorial independence in effect meant that the only possible consequences for the good-faith exercise of prosecutorial discretion were internal employment consequences. For law clerks, judicial independence in
120 Krieger, supra note 116 at paras 2, 50. 121 Ibid at para 55. 122 I note that insofar as law clerks are hired as public servants, legislation and policies on hiring also constrain judicial independence. See e.g. Wilner, supra note 2 (“clerks are normally hired as public servants who are paid with public funds [citation omitted]” at 622; “[c]lerks may also be accountable to a public service manager” at 623). If the law society’s jurisdiction over law clerks violates this independence, then presumably this human resources power does as well.
272 UNBLJ RD UN-B [VOL/TOME 71 effect means that the only possible consequences for the good-faith exercise of law clerks’ duties rest with the employer. In this respect, law society requirements may be largely unenforceable by the law society—but that does not mean they are hollow or merely aspirational. Law clerks should strive to meet the spirit and letter of these obligations, and judges and courts should expect them to do so. Indeed, it is public confidence in the judiciary that may suffer consequences, if not the clerk himself or herself, for a breach of these rules. Moreover, for those law clerks who are articled students and those who wish to become lawyers, it would be open to law societies to consider any such breaches as evidence against good character when it comes time for the former law clerk to apply for admission to the bar.
A major exception here is post-service restrictions. Just as post-service restrictions on former judges are a matter for the law society and not the Canadian Judicial Council, and do not impede judicial independence, so too are post-service restrictions for former judicial law clerks a legitimate matter for the law society that does not impede judicial independence.
Of course, it is open to the courts to impose law society membership, as a 2020 CanLIIDocs 3123 lawyer or a student-at-law, and compliance as a condition of employment (although, as mentioned above, many courts do not do so123). By doing so, they would in effect be accepting any implications for judicial independence. Moreover, there may indeed be reasons that law clerks may wish to be articled students or lawyers, despite the fact that judicial independence provides that such status cannot be required of them. The most obvious reason is to complete articling requirements, and thus to become licensed as a lawyer, sooner rather than later—or, for lawyers, to accrue a year of seniority at the bar.
Reflections and conclusion
In this article, I have offered a new doctrinal account of the ethical duties of law clerks. This account is rooted in the fact that law clerks practice law, and thus in the lawyer- client relationship between the law clerk and the judge, the professional duties of lawyers, and specifically the duty to encourage respect for the administration of justice. These ethical duties are professional duties, i.e. duties as lawyers that are enforceable by the law society. However, I have also argued that law society jurisdiction over law clerks is restricted though not necessarily eliminated by the constitutional principle of judicial independence.
Under my account, law clerks are practicing lawyers and so share the professional duties of all practicing lawyers, but because of the identity and the nature of the client, the duty to encourage respect for the administration of justice incorporates at least some duties of the judge. I have focused on impartiality and outside interests. In this respect, the duties of lawyer are the same as the duties of the
123 See supra note 33 and accompanying text.
2020] LEGAL ETHICS AND JUDICIAL LAW CLERKS 273 judge. I have also focused on post-service restrictions. In this respect, the animating principles are the same as those animating restrictions on judges returning to practice, but their application is different. However, the duties of the judge supplement, but do not detract from, the professional duties of the law clerk. I have focused here on the duty to report a fellow lawyer to the law society. Another example is that while judges lack an express duty of confidentiality,124 law clerks are doubtlessly bound by the lawyer’s duty of confidentiality.
My analysis has focused primarily on judicial law clerks at the Supreme Court of Canada, as these are the only group specifically described in the existing literature. My conclusions should be applicable to all judicial law clerks. But further empirical research is needed on the role and duties of law clerks at other courts. Further research is also warranted on other lawyers to the judiciary, such as staff lawyers and executive legal officers.
Given my analysis, how should law clerks be situated among public sector lawyers? Law clerks, if indeed they are properly understood as lawyers, are not properly considered government lawyers under typical definitions. Government 2020 CanLIIDocs 3123 lawyers are generally understood to be lawyers for the executive branch of government.125 Law clerks are unique in that they may be employees of the executive, but their client is the judiciary.126 However, there are some similarities between the two. Sanderson explains government lawyers as having three layers of duties, as lawyers, as delegates of the Attorney General, and as public servants: “The first set of professional duties comes with their membership in provincial and territorial law societies. The second set of public law duties comes with their responsibilities to the Minister of Justice and ex officio, Attorney General. The third set comes with their employment in the public service.”127 Law clerks share two of these layers. They are lawyers (or articled students) practicing law and are thus subject to the professional duties of all lawyers. They are also public servants.128 But they are not delegates of the Attorney General—indeed, as the government is the most common litigant, this would be untenable.
Under Sanderson’s analysis, these layers of duties have conceptual and practical implications because each layer is imposed, and therefore adjudicated, by a
124 See supra note 44 and accompanying text. 125 See Sanderson, supra note 1 at xxiii; Martin, supra note 1 at 271. 126 Martin, supra note 1 at 271. 127 Sanderson, supra note 1 at 48. 128 See e.g. Wilner, supra note 2 (“clerks are normally hired as public servants who are paid with public funds [citation omitted]” at 622). Their status as public servants is in itself problematic: law clerks present a microcosm in the tension between judicial independence and executive control over court resources. Law clerks provide services to the judiciary, with the executive controlling their pay and benefits.
274 UNBLJ RD UN-B [VOL/TOME 71 separate authority. In particular, the law societies only have jurisdiction over the professional layer of duties, not the others.129
Law clerks unquestionably have special duties because of the nature of their work and the nature of their client. These duties might appear to be a third layer— perhaps a “public interest” lawyer or “public duties” layer,130 like that of government lawyers as delegates of the Attorney General. I argue, however, that these special duties are best understood as part of the professional duties layer and are derived from the duty to encourage respect for the administration of justice. Thus, they are enforceable by the law society subject to the considerable limits imposed by judicial independence.
2020 CanLIIDocs 3123
129 Sanderson, supra note 1 at 3. 130 See e.g. Wilner, supra note 2 at 622:
Like that of government lawyers, the law clerk’s role may raise questions about who clerks are meant to be serving, and how the public interest component of their functions can inform their day-to-day duties. To say that clerks serve the ‘public interest’ raises an ambiguity as to whether a law clerk’s ‘client’ is his particular judge or rather the court as an institutional whole, or even perhaps ‘the law’ itself [citation omitted].
TO PARTICIPATE OR NOT TO PARTICIPATE: JUDICIAL INVOLVEMENT IN THE COMMUNITY*
The Honourable Mr. Justice Malcolm Rowe, QC, & Dahlia Shuhaibar+
I. Introduction
Judges have long had to grapple with the limits of their community involvement. Few professionals would ask themselves whether they are to “behave like a monk [or] be a eunuch [or] live in silent solitude [or] embark…to live on a planet other than the earth.”1 Yet, for many years, judges have, in effect, been asking such questions of themselves. Often, judges are chosen in recognition of their significant contributions to their communities. Yet, upon appointment, many have been concerned with whether 2020 CanLIIDocs 3123 continuing to make these same contributions would call into question their impartiality, their independence, and even their integrity.
In considering the proper limits of judges’ involvement in the community, we will not focus on a judge’s public behaviour. That is simple; judges should never undermine the offices they hold by their behaviour in public. Instead, this paper examines the degree to which judges should engage, if at all, in voluntary associations, in political organizations, in matters of public controversy, and in advocating their private interests. Guidance is available, usually in written codes of conduct put in place by bodies overseeing judicial conduct. As well, there is the less formal but quite important guidance provided by one’s chief justice. But, much is left to one’s own good judgment. Judges have much to offer their communities, but a “line”… however difficult to draw…exists between acceptable and unacceptable community involvement. This line is “not capable of mathematical determination”2—yet, each judge must take care not to cross it.
* Adapted from a speech presented by Justice Malcolm Rowe at the Commonwealth Magistrates’ and Judges’ Association Annual Conference in Port Moresby, Papua New Guinea, in September 2019. + Justice Malcolm Rowe is a judge of the Supreme Court of Canada. He previously served as Secretary to the Cabinet and head of the public service in Newfoundland and Labrador. He was also a member of Canada’s diplomatic service and a partner at Gowlings. He was appointed to the Supreme Court of Newfoundland and Labrador in 1999, to the Court of Appeal in 2001, and to the Supreme Court of Canada in 2016. Dahlia Shuhaibar practices public law and civil litigation in Ottawa. She is a former law clerk to Justice Malcolm Rowe at the Supreme Court of Canada (2019–20) and to Justice Patrick Gleeson at the Federal Court (2018–19). 1 Simon H Rifkind, “The Public Concern in a Judge’s Private Life” in George H Williams & Kathleen M Sampson, eds, Handbook for Judges: An Anthology of Inspirational and Educational Readings (Chicago: American Judicature Society, 1984) 59 at 62. 2 Ibid at 66.
276 UNBLJ RD UN-B [VOL/TOME 71
We will proceed in three stages, beginning with a brief overview of judges’ security of tenure in Canada. Next, we will describe key elements of the ethical principles established by the Canadian Judicial Council (CJC). Finally, we will examine various types of community involvement through case studies.
II. Security of Tenure
For the sake of simplicity, we will focus on superior courts. However, the discussion applies with slight modifications to other courts, notably provincial courts.
Canadian judges have security of tenure. This has a long history, dating back to the Act of Settlement of 17013 and the Commissions and Salaries of Judges Act of 1760.4 These statutes established that judges would remain in office “during good behaviour,” rather than at the pleasure of the monarch, and that judges would continue to hold office notwithstanding the death of the monarch.5 Security of tenure allows judges to decide cases independently, notwithstanding government or public 6 disapproval. This independence allows judges “to stand above the political fray, 2020 CanLIIDocs 3123 immune to the pushes, pulls, and swings of popular opinion.”7
Superior court judges can be removed only by a vote of the Senate and the House of Commons.8 No judge since 1867 has been removed by this process, albeit a few have resigned in the face of this eventuality.9
In recent decades, Parliament has made clear that it will act only on the recommendation of the CJC. This body is composed of chief justices from the superior courts in all the provinces and is chaired by the Chief Justice of Canada.10 The CJC is created by the Judges Act and derives its authority to investigate complaints and
3 Act of Settlement (UK), 1700 & 1701, 12 & 13 Will 3, c 2. 4 Commissions and Salaries of Judges Act, 1760, 1 Geo 3, c 23. 5 J van Zyl Smit, The Appointment, Tenure and Removal of Judges under Commonwealth Principles: A Compendium and Analysis of Best Practice (Report of Research Undertaken by Bingham Centre for the Rule of Law) (London, UK: The British Institute of International and Comparative Law, 2015) at 59–60; Shimon Shetreet, Judges on Trial: A Study of the Appointment and Accountability of the English Judiciary (New York: North-Holland Publishing, 1976) at 10–11. 6 Martin L Friedland, A Place Apart: Judicial Independence and Accountability in Canada (Canadian Judicial Council, 1995) at 41. 7 Robert J Sharpe, Good Judgment: Making Judicial Decisions (Toronto: University of Toronto Press, 2018) at 251. 8 Constitution Act, 1867 (UK), 30 & 31 Vict, c 3, s 99(1), reprinted in RSC 1985, Appendix II, No 5. 9 Martin L Friedland, “Appointment, Discipline and Removal of Judges in Canada” in HP Lee, ed, Judiciaries in Comparative Perspective (New York: Cambridge University Press, 2011) 46 at 58–59. 10 Judges Act, RSC 1985, c J-1, s 59(1) [Judges Act].
2020] TO PARTICIPATE OR NOT TO PARTICIPATE 277 recommend removal from that statute.11 It has authority over more than 1100 federally appointed judges.12
Through a committee structure, the CJC investigates complaints made against federally appointed judges; such complaints can be made by a member of the public or by the Attorney General.13 Most complaints are ill-founded and are speedily dismissed. Those that warrant investigation trigger procedures aimed at ensuring fairness to the judge. In serious cases, the committee can recommend to the CJC that a judge be removed for reasons of (a) infirmity; (b) misconduct; (c) failing to undertake his or her duties; or (d) otherwise having been placed in a position incompatible with the due execution of his or her office.14 The CJC has interpreted its task as involving two steps:
1) Determining if the judge has become “incapacitated or disabled from the due execution of their office”; and 2) Determining if public confidence in the judge’s ability to discharge the duties of his or her office has been undermined to such an extent that a recommendation for removal is warranted.15 2020 CanLIIDocs 3123 The test at the second step is the following: “Is the conduct so manifestly and profoundly destructive of the concept of the impartiality, integrity and independence of the judicial role, that public confidence would be sufficiently undermined to render the judge incapable of executing the judicial office?”16 Given the stringency of that test, it is difficult to imagine how a judge’s involvement in a community organization or the like could warrant a recommendation for removal—however, as we will see, reasonable disagreements can easily arise as to the appropriateness of extra-judicial involvement, and some judges have come close to being removed.
While the Judges Act does not contemplate lesser sanctions than recommending a judge’s removal, in practice such sanctions are imposed. These can be quite mild, in effect words of guidance or caution set out in correspondence from the CJC. It may call for counselling or additional training. In more serious circumstances, a reprimand may issue. In almost all instances, the goal is to restore the judge to the proper conduct of his or her role. The approach is strongly oriented to be
11 Ibid, ss 59–70. 12 “Our Mandate”, online: Canadian Judicial Council
278 UNBLJ RD UN-B [VOL/TOME 71 supportive and remedial, while at the same time providing clear guidance as to conduct.
Of course, the vast majority of situations where conduct may be questioned never get to the CJC. They are dealt with informally by the chief justice. But, if more formal measures appear warranted, they are undertaken by the CJC as a group and not by the individual chief justice.
III. Ethical Guidelines and Principles
Guidance for federally appointed judges is provided in the Judges Act and in a set of principles developed by the CJC called Ethical Principles for Judges (“Principles”).17 The Principles are currently under review, and a revised set was initially expected to be released in the spring of 2020.18 A draft set of revised principles was released in November 2019 (“Draft Revised Principles”).19 Provincial judicial councils operate in a parallel way for provincially appointed judges; they have adopted similar principles 20 to guide conduct (see, e.g., Ontario Court of Justice, Principles of Judicial Office; 2020 CanLIIDocs 3123 Provincial Court of British Columbia, Code of Judicial Ethics21).
Section 55 of the Judges Act contains an overarching rule:
No judge shall, either directly or indirectly, for himself or herself or others, engage in any occupation or business other than his or her judicial duties, but every judge shall devote himself or herself exclusively to those judicial duties.22
The CJC’s Principles effectively expand on what this rule means. The Principles are advisory: they are not a list of prohibited behaviours.23 However, they
17 Canadian Judicial Council, Ethical Principles for Judges, Ottawa: CJC, 1998, online (pdf):
The ethical principles are to be applied in light of all the relevant circumstances and consistently with the requirements of judicial independence and the law. Setting out the very best in Ethical Principles does not preclude reasonable disagreements about their application or imply that
2020] TO PARTICIPATE OR NOT TO PARTICIPATE 279 are a “useful touchstone of generally accepted ethical standards in the judicial community guiding judges in how they should act on and off the Bench” and set out a general framework that is relevant to assessing allegations of improper conduct.24
The limitations set out in the Judges Act and the Principles stem from the twin requirements of independence and impartiality. As a former Chief Justice of Canada put it, judicial independence is not a privilege that appertains to the holder of a judicial office; rather, it is a guarantee to citizens that there is an impartial adjudicator to resolve their disputes or hear their challenges to abuse of authority.25 The limitations on judicial conduct call for a higher standard than is expected from other citizens. A judge’s conduct must be “free from impropriety or the suggestion of impropriety; it should be, as far as is humanly possible, beyond reproach.”26 Judges are expected to tolerate restrictions on the rights of the individual that other citizens do not.27 However, they are entitled to expect as few restrictions on their freedoms as is possible and consistent with proper conduct, given their office.28
IV. The Tension Underlying Community Involvement 2020 CanLIIDocs 3123
When it comes to community involvement by judges, two competing considerations are at play. On the one hand, a judge should not be isolated from his or her community. On the other hand, community involvement must be modulated so as to avoid negatively affecting the standing of the judge and the judiciary. Striking the right balance is somewhat contextual.
In 2015, in the context of an allegation of judicial bias, the Supreme Court of Canada wrote that judges can and should participate in their communities: “Membership in an association affiliated with the interests of a particular race,
departures from them necessarily warrant disapproval… [Ethical Principles] is not intended to be a code of conduct that sets out minimum standards of behaviour. 24 In the Matter of Section 65 of the Judges Act, R.S., 1985, c. J-1, and of the Inquiry Committee convened by the Canadian Judicial Council to review the conduct of the Honourable Theodore Matlow of the Ontario Superior Court of Justice: Report of the Canadian Judicial Council to the Minister of Justice (Ottawa: CJC, 3 December 2008) (Catherine A Fraser, Chairperson) at paras 95, 99, majority reasons, online (pdf): Canadian Judicial Council
280 UNBLJ RD UN-B [VOL/TOME 71 nationality, religion, or language is not, without more, a basis for concluding that a perception of bias can reasonably be said to arise.” 29
Indeed, the Supreme Court recognized the value of such experiences:
A judge’s identity and experiences are an important part of who he or she is, and neither neutrality nor impartiality is inherently compromised by them. Justice is the aspirational application of law to life. Judges should be encouraged to experience, learn and understand “life”—their own and those whose lives reflect different realities.30
Withdrawal from the community can actually hinder a judge’s ability to carry out his or her duties. Judging is not “an abstract or mechanical process. … [It] is an intensely human process.”31 Judges are called on to make decisions mindful of the standards of the community; thus, to render decisions regarding fundamental freedoms requires an understanding of societal attitudes and competing interests.32 Assessing damages, sentencing offenders, and giving effect to the “public interest” all require knowledge of the community.33 A judge who is withdrawn from the community would undertake all these tasks with less understanding. Put another way, isolation can result 2020 CanLIIDocs 3123 in “judicial shortsightedness and unresponsiveness to the changing needs of society.”34 The Draft Revised Principles also recognize that judges are often called upon to make decisions in areas in which they have little or no familiarity; consequently, judicial isolation “does not promote wise or just judgments.”35 They therefore encourage judges “to take up opportunities to engage with and learn from the wider public, including communities with which the judge has little or no life experience.”36
Community involvement has other positive effects. First, it fulfils the public’s expectation that professionals should be active members of their communities. Second, it “personalize[s] judges as sincere and caring family members, volunteers, and community leaders.” Finally, it contributes to a judge’s well-being and consequently to improved judicial demeanour and performance.37
29 Yukon Francophone School Board, Education Area #23 v Yukon (AG), 2015 SCC 25 at para 61 [Yukon Francophone School Board]. 30 Ibid at para 34. 31 Sharpe, supra note 7 at 264. 32 CJC, Commentaries, supra note 28 at 8–9. 33 Justice JB Thomas, Judicial Ethics in Australia, 2nd ed (Sydney: LBC Information Services, 1997) at 93. 34 Shetreet, supra note 5 at 324. 35 CJC, Draft Revised Principles, supra note 19 at 37, commentary 5.B.9. 36 Ibid. 37 Raymond J McKoski, Judges in Street Clothes: Acting Ethically Off-the-Bench (Madison: Fairleigh Dickinson University Press, 2017) at 4–6.
2020] TO PARTICIPATE OR NOT TO PARTICIPATE 281
Nonetheless, inappropriate involvement can be detrimental. The overarching concern is to ensure an independent and impartial judiciary. The range of issues that can come before the courts is as diverse as is life; judges must be careful not to indicate predispositions on potential controversies. As a result, involvement in causes or organizations likely to be involved in litigation is to be avoided.38 Similarly, judges should avoid involvement in groups whose purposes include exerting pressure on government or attempting to effect social change.39 Community involvement also runs the risk of causing frequent recusals.40 Additionally, it may lead to improper use of the prestige of the judge’s office.41
In practice, how does one decide which activities are acceptable and which are not? The case studies that follow illustrate inherent difficulties in this demarcation; they demonstrate that new and unforeseen situations continue to arise.
V. The CJC’s Principles and Case Studies
The CJC’s overarching guidance on extra-judicial activity is this: subject to limitations 2020 CanLIIDocs 3123 imposed by the Judges Act and the nature of the office, judges can participate in activities that do not detract from the performance of their duties.42 The Principles then provide advice on various types of community involvement.
A. Political activity
The most definitive of the principles relates to political activity: it is prohibited. Judges must not be members of political parties; engage in political fundraising; attend political events; make contributions to political parties or campaigns; take part publicly in controversial political discussions, except in respect of matters directly affecting the operation of the courts, the independence of the judiciary, or fundamental aspects of the administration of justice; or sign petitions to influence a political decision.43 This is an absolute ban; judges can have no association with any political
38 CJC, Principles, supra note 17 at 28, principle C.1(c); CJC, Draft Revised Principles, supra note 19 at 37, commentary 5.B.11. 39 Wilson, supra note 26 at 8; Thomas, supra note 33 at 97. 40 McKoski, supra note 37 at 46–47; CJC, Principles, supra note 17 at 34, commentary C.3. 41 McKoski, supra note 37 at 48. 42 CJC, Principles, supra note 17 at 18, commentary 2; CJC, Draft Revised Principles, supra note 19 at 21, commentary 3.A.3. 43 CJC, Principles, supra note 17 at 28–29, principle D.3; CJC, Draft Revised Principles, supra note 19 at 34–35, commentary 5.B.2.
282 UNBLJ RD UN-B [VOL/TOME 71 group, nor can they publicly express any political opinions.44
This prohibition stems from concerns relating to the separation of powers between the three branches of the state: the legislature, the executive, and the judiciary. The judiciary “provide[s] an impartial check to other powers at work in the rule of law.”45 The recurring concerns of impartiality and independence are in play. A judge who engages in partisan political activity or makes out-of-court statements on issues of public controversy is “by definition…choosing one side of a debate over another.”46 Moreover, if the judge’s activities attract criticism and/or rebuttal, judicial independence is undermined.47 Political activity of any kind is to be avoided.48
B. Civic and charitable activities
The Principles approve of involvement in civic and charitable activities, with certain limitations:
Judges are free to participate in civic, charitable and religious activities 2020 CanLIIDocs 3123 subject to the following considerations:
a) Judges should avoid any activity or association that could reflect adversely on their impartiality or interfere with the performance of judicial duties. b) Judges should not solicit funds (except from judicial colleagues or for appropriate judicial purposes) or lend the prestige of judicial office to such solicitations. c) Judges should avoid involvement in causes or organizations that are likely to be engaged in litigation. d) Judges should not give legal or investment advice.49
This guidance provides useful parameters within which judges can organize their affairs. Yet, it is not always clear, as the cases of Justices Ted Matlow and Donald McLeod demonstrate.
44 Wilson, supra note 26 at 7. 45 Jonathan Soeharno, The Integrity of the Judge: A Philosophical Inquiry (Surrey: Ashgate Publishing, 2009) at 94. 46 CJC, Principles, supra note 17 at 39, commentary D.2; CJC, Draft Revised Principles, supra note 19 at 35, commentary 5.B.3. 47 CJC, Principles, supra note 17 at 39, commentary D.2; CJC, Draft Revised Principles, supra note 19 at 35, commentary 5.B.3. 48 The Draft Revised Principles also advise judges to recuse themselves when their close family members are politically involved in such a way as to “adversely affect the public perception of [the] judge’s impartiality.” See CJC, Draft New Principles, supra note 19 at 35, commentary 5.B.5. 49 CJC, Principles, supra note 17 at 33, commentary C.1. The Draft Revised Principles expand on these points: see CJC, Draft Revised Principles, supra note 19 at 36–40, commentaries 5.B.8–14.
2020] TO PARTICIPATE OR NOT TO PARTICIPATE 283
Justice Matlow was a judge of the Ontario Superior Court of Justice. In 1999, he became involved in a public controversy relating to a municipal development project (the “Thelma Road Project”) near his home in Toronto. Justice Matlow and his neighbours formed a neighbourhood community group called The Friends of the Village (the “Friends”) whose purpose was to oppose the project. Justice Matlow became one of the group’s leaders. He sought standing in a hearing before the Ontario Municipal Board with respect to the legality of the proposed development. He also engaged in email correspondence and meetings with elected officials and city employees. He made comments to the media setting out his reasons for opposing the project; in these, he expressed his views on legal issues relating to the project. His correspondence indicated he was a judge. His language could be considered inflammatory.50 He continued these activities for several years.
In 2005, a development project was being contemplated elsewhere in Toronto relating to streetcars (the “St. Clair Project”). The project was controversial; a community organization called SOS-Save Our St. Clair Inc. (“SOS”) formed in opposition to it. SOS applied to the Divisional Court for a declaration that the project breached municipal planning laws and failed to accord with environmental 2020 CanLIIDocs 3123 assessments.51
Justice Matlow was assigned to the panel hearing the application. The panel unanimously granted the application, in effect stopping the project. The City of Toronto became aware of Justice Matlow’s involvement in the Thelma Road Project, which was quite similar to the St. Clair Project. The City brought a motion requesting that Justice Matlow recuse himself and that there be a new hearing. The court granted the motion, with Justice Matlow dissenting. The City then made a complaint to the CJC.52
An inquiry committee of the CJC found that Justice Matlow had placed himself in a position incompatible with the office of judge, that he was guilty of misconduct, and that he had failed in the due execution of his office. It found that, taken together, Justice Matlow’s actions had rendered him incapable of executing his judicial office and recommended his removal.53
The report of the inquiry committee was considered by the full CJC. The majority of the CJC found that Justice Matlow’s involvement in the Friends, his meetings with officials on behalf of the Friends, and his interactions with the media
50 CJC, Matlow Report, supra note 24 at paras 4–8, 16–25, majority reasons. 51 Ibid at paras 27–28, majority reasons. 52 Ibid at paras 29–30, 35–37, 39, 41, majority reasons. 53 Ibid at para 45, majority reasons.
284 UNBLJ RD UN-B [VOL/TOME 71 were not problematic on their own.54 However, the manner in which he carried out these activities was inappropriate. The majority of the CJC concluded as follows:
In summary, while judges who have personal interests, such as home ownership, that can be affected by government action have the right, in their private capacity, to contest, as do other Canadians, decisions that affect those interests as do other Canadians, there are limits as to what a judge might do. A judge is not entitled to use the prestige of judicial office to advance his or her private interests. Nor should a judge use intemperate language where others would likely know, or could be expected to know, that he or she was a judge. And under no circumstances is a judge entitled to act as a legal advisor for individuals opposing government action.55
The majority of the CJC found that some of Justice Matlow’s actions constituted judicial misconduct and placed him in a position incompatible with the due execution of his office. However, it found that removal was not warranted given his expressions of regret, his 27-year career on the bench without other incidents, and the fact that the misconduct did not occur during the performance of his judicial duties.56
The majority did, however, strongly disapprove of his actions and directed him to 2020 CanLIIDocs 3123 apologize to specified individuals, attend a judicial ethics seminar, and obtain prior approval should he wish to participate in a public debate in future.57
A minority of the CJC found that the conduct was sufficiently serious to undermine public confidence and, accordingly, would have recommended Justice Matlow’s removal.58
Justice Matlow’s situation is interesting in that he was advocating for his property rights. His involvement in a community organization was not per se to contribute to the community but rather to advocate for his interests and those of his neighbours. The majority of the CJC concluded that judges should be able to advocate for their own interests as private citizens—but they must truly do so as private citizens, without identifying themselves as judges or providing any kind of legal advice. Moreover, as was stated explicitly, a judge should act in line with the dignity of his or her office, avoiding intemperate language.
A recent case before the Ontario Judicial Council (OJC) illustrates the difficult distinction between community involvement that is educational and that which is advocacy. Justice Donald McLeod is a judge of the Ontario Court of Justice. As a Black judge who grew up in subsidized housing and with limited resources, he sought to assist others in overcoming similar barriers; he wished to be a leader for
54 Ibid at paras 107–14, majority reasons. 55 Ibid at para 123, majority reasons. 56 Ibid at paras 176, 179–84, majority reasons. 57 Ibid at para 186, majority reasons. 58 Ibid at para 9, minority reasons.
2020] TO PARTICIPATE OR NOT TO PARTICIPATE 285
Black youth. Before his appointment, he was involved in community initiatives relating to education and mentorship of Black youth.59 One can readily conclude that this commendable commitment was a factor favouring his appointment. After his appointment, he sought to create a national organization called the Federation of Black Canadians (FBC) with a mandate to advance the social, economic, political, and cultural interests of Black Canadians. He became the Chair of the FBC’s Interim Steering Committee.60 Was this compatible with his new role as a judge?
The Associate Chief Justice of the Ontario Court of Justice expressed concerns. She and Justice McLeod agreed to consult the court’s Judicial Ethics Committee. The Committee approved his involvement with limitations: notably, he was not to be involved in fundraising or lobbying. However, after a time, the FBC began to engage in activities in which members, including Justice McLeod, would meet with politicians and government officials. Critiques started to emerge in the media about his participation, and a complaint was made to the OJC. Justice McLeod ultimately stepped down from his position at the FBC.61
The OJC found that his conduct had been incompatible with judicial office 2020 CanLIIDocs 3123 but did not amount to judicial misconduct.62 It recognized that the FBC’s goals and Justice McLeod’s motivations were laudable:
Justice McLeod is rightly seen as a leader in his community. As a racialized judge, he has a moral obligation as a leader and role model in the Black community. As he noted in his response to the complaint, his community involvement was an important factor when he was appointed. There is no reason why it should have entirely ended when he assumed judicial office. He is to be commended for leaving his court room and judicial chambers from time to time in order to present to the public a positive and inspiring vision of what young Black Canadians can aspire to.63
However, his activities had crossed the line into advocacy on public policy. In his meeting with government officials and politicians, he “not only provided information but also advocated specific policy changes and the allocation of government resources to achieve those policy changes.”64 He was also publicly identified as a judge. These activities were not, the OJC found, “merely educative or intended to inform politicians
59 In the Matter of a complaint respecting The Honourable Justice Donald McLeod (20 December 2018) at paras 4–6, online: Ontario Judicial Council
286 UNBLJ RD UN-B [VOL/TOME 71 of the difficulties facing Black Canadians.”65 The OJC emphasized that “[i]t is incompatible with the separation of powers for a judge to enter the fray and ask political actors for policy changes and the allocation of resources, however worthwhile the judge’s motivating cause.”66 His actions were therefore incompatible with judicial office. The OJC did note, however, that Justice McLeod likely would not have crossed the line “had he restricted his efforts to educating members of the public about these issues.”67
The OJC held that Justice McLeod’s actions were incompatible with judicial office; however, they did not amount to misconduct.68 It noted that there was no evidence that Justice McLeod had been involved in partisan political activity or fundraising. He had also contacted his court’s Ethics Committee and spoken with his Associate Chief Justice.69 Overall, he had been motivated to promote public confidence in the judicial system, which was “relevant precisely because the aim of judicial misconduct proceedings is to maintain public confidence in judicial institutions.”70
Justice McLeod’s situation illustrates that judges inevitably sacrifice certain 2020 CanLIIDocs 3123 aspects of full citizenship. However, the OJC made clear that a judge need not be a recluse. Rather, judges can contribute to their communities in many ways, including education as to policy issues. The OJC’s decision is notable in that it explicitly evaluated Justice McLeod’s conduct in light of the racial dynamics in Ontario.71 It noted that many Black people mistrust the criminal justice system, that they are overrepresented in that system, and that they are disproportionately arrested and searched by police. Justice McLeod’s life experiences, the OJC held, made him uniquely aware of these issues, and his presence on the bench promotes the public’s confidence in the administration of justice.72 This explicit acknowledgment by the OJC of the context of the complaint recognizes the legitimate and important role that judges can play in addressing inequities.
Since the drafting of this paper, a further complaint has been made against Justice McLeod relating to the evidence he gave at the first hearing.73 As this matter is still before the OJC, we make no further comment on it.
65 Ibid at para 76. 66 Ibid at para 84. 67 Ibid at para 88. 68 Ibid at para 94. 69 Ibid at paras 95–100. 70 Ibid at para 103. 71 Ibid at para 102. 72 Ibid at paras 102–103. 73 “Notice of Hearing into a Complaint about the Conduct of the Honourable Justice McLeod”, online: Ontario Judicial Council
2020] TO PARTICIPATE OR NOT TO PARTICIPATE 287
C. Fundraising
The Principles are clear that fundraising should be avoided by judges. They prohibit judges from soliciting funds “except from judicial colleagues or for appropriate judicial purposes.”74 Further, judges must not lend the prestige of their office to such solicitations.75
Fundraising is different from other forms of community involvement: “It moves beyond one’s own individual support to the active seeking out of expressions of support from others. It encompasses an advocacy function, championing the cause and seeking to rally others to it.”76 Fundraising also raises a real risk that the public will consider a judge to be affected by contributions made in response to their solicitation.77 The perceived effects on the impartiality and independence of the judge require a complete ban.
D. Involvement in boards, universities, and other institutions
2020 CanLIIDocs 3123 Judges are prohibited from serving on the boards of commercial enterprises.78 They are also discouraged from membership on boards of universities, dioceses, schools, hospitals, charitable foundations, and the like.79 While such positions may seem harmless, it is not uncommon for such bodies to be involved in litigation, as well as in matters of public controversy. Being on such a board could disqualify a judge from sitting; in addition, it could raise concerns about impartiality.
Justice Georgina Jackson of the Court of Appeal for Saskatchewan has provided a helpful set of questions that judges can ask themselves when considering whether to be a member of a board:
1) Would association with this board reflect adversely on the judge’s impartiality? 2) Would this activity interfere with the performance of his or her judicial duties?
74 CJC, Principles, supra note 17 at 28, principle C.1(b); CJC, Draft Revised Principles, supra note 19 at 38, commentary 5.B.14. 75 CJC, Principles, supra note 17 at 28, principle C.1(b); CJC, Draft Revised Principles, supra note 19 at 38, commentary 5.B.14. 76 Stephen GA Pitel & Michael Malecki, “Judicial Fundraising in Canada” (2015) 52:3 Alta L Rev 519 at 530. 77 Ibid at 531. 78 CJC, Principles, supra note 17 at 36, commentary C.7. 79 CJC, Principles, supra note 17 at 37, commentary C.9; CJC, Draft Revised Principles, supra note 19 at 37–38, commentaries 5.B.11–12; Wilson, supra note 26 at 8.
288 UNBLJ RD UN-B [VOL/TOME 71
3) Is the judge being asked to join this board to lend the prestige of the judicial office to fund-raising? 4) Is this board likely to be involved in litigation? 5) Is the judge being asked to play a role in the expectation that he or she will give legal or investment advice?80
These questions provide a useful framework. However, as we will see, this too is an area where judges can find themselves uncertain as to what is appropriate.
In the Yukon Francophone School Board case referred to above, the Supreme Court of Canada addressed a bias claim against a judge. The judge had heard a case in which the school board was suing the territorial government for deficiencies in the provision of minority language education. The judge ruled in the board’s favour on most issues. The Court of Appeal found a reasonable apprehension of bias based on incidents at trial and the judge’s involvement as governor of a philanthropic Francophone community organization.81
The Supreme Court upheld the bias findings based on the incidents at trial, but it set aside the findings relating to the judge’s position in the philanthropic 2020 CanLIIDocs 3123 organization.82 The Court explained:
Judicial impartiality and neutrality do not mean that a judge must have no prior conceptions, opinions or sensibilities. Rather, they require that the judge’s identity and experiences not close his or her mind to the evidence and issues. There is, in other words, a crucial difference between an open mind and empty one.83
As this case did not arise in the context of a complaint to the CJC, it is unclear what the CJC’s views on the judge’s involvement in the organization would be. However, the Principles suggest that the main considerations would be whether the organization was likely to be involved in litigation or in matters of public controversy, both of which arose here. The case presents an interesting perspective on the interaction of the Principles with the law of reasonable apprehension of bias.
This kind of participation is at the heart of a case that was recently before the courts. Justice Patrick Smith of the Ontario Superior Court of Justice was asked to be the interim dean of Bora Laskin Faculty of Law at Lakehead University, following the unexpected resignation of its dean.84 That faculty has, as an important part of its
80 The Honourable Georgina R Jackson, “The Mystery of Judicial Ethics: Deciphering the ‘Code’” (2005) 68:1 Sask L Rev 1 at 9. 81 Yukon Francophone School Board, supra note 29 at paras 3, 5. 82 Ibid at paras 55–56. 83 Ibid at para 33. 84 Report of the Review Panel Constituted by the Canadian Judicial Council Regarding the Honourable Patrick Smith (Ottawa: CJC, 5 November 2018) (RS Veale, Chairperson) at para 1, online (pdf): Canadian
2020] TO PARTICIPATE OR NOT TO PARTICIPATE 289 mandate, the study of Aboriginal and Indigenous law.85 The law school cited Justice Smith’s experience with Indigenous communities and publications on Aboriginal law as its reasons for asking him to accept the position.86
Justice Smith spoke to his Chief Justice about the invitation to serve as interim dean. They sought approval from the federal Minister of Justice for a six- month leave of absence, noting that Justice Smith’s responsibilities would be confined to academic leadership, that he would delegate administrative responsibilities to other personnel, and that he would not be remunerated (beyond his judge’s salary). The Minister granted the leave of absence, noting that she had “no concerns.”87
The Executive Director of the CJC wrote to Justice Smith expressing concerns about the appointment and informing him that the situation may warrant consideration by the CJC. The matter was ultimately referred to a review panel.88
The review panel concluded that s 55 of the Judges Act “requires judge to devote themselves exclusively to their judicial duties and to abstain from businesses and occupations falling outside the judicial sphere.” (As noted, s 55 states that “[n]o 2020 CanLIIDocs 3123 judge shall…engage in any occupation or business other than his or her judicial duties, but every judge shall devote himself or herself exclusively to those judicial duties” [emphasis added].) The meaning of “occupation,” the panel held, was to be “broadly interpreted to capture all non-judicial activities, whether paid or unpaid, that interfere with the judicial role.” Moreover, the leave-of-absence mechanism in s 5489 did not permit a judge to take on a business or occupation outside the judicial sphere, with limited exceptions. Justice Smith had also violated his ethical obligation “to avoid involvement in public debate that may unnecessarily expose him to political attack or be inconsistent with the dignity of judicial office,” and his appointment had raised concerns that the prestige of his office was being misused. The review panel concluded
Judicial Council
54 (1) No judge of a superior court shall be granted leave of absence from his or her judicial duties for a period (a) of six months or less, except with the approval of the chief justice of the superior court; or (b) of more than six months, except with the approval of the Governor in Council.
290 UNBLJ RD UN-B [VOL/TOME 71 that Justice Smith had violated s 55.90 However, it held that the conduct was not serious enough to warrant removal.91
In a recent judicial review, the Federal Court held that the review panel’s interpretation of s 55 was unreasonable. The court held that, properly interpreted, s 55 prohibits engaging in any occupation or business other than judicial duties—not any extra-judicial activities.92 It further concluded that the panel’s interpretation of s 54 was unreasonable: the provision did not prevent a judge from taking up extra-judicial activities that are not inconsistent with his or her judicial duties while on leave.93 In the result, Justice Smith had not violated s 55.94
The court further held that Justice Smith had not violated his ethical obligations. The panel’s references to potential litigation involving the law school were “entirely speculative and unworthy of…consideration,” and in any case, Justice Smith could simply recuse himself if any such case arose. Moreover, the panel had exaggerated the media coverage generated by Justice Smith’s appointment, and its concerns about improper use of his judicial reputation were not supported by the 95 Principles. 2020 CanLIIDocs 3123
Finally, the court held that the proceedings were procedurally unfair, amounting to an abuse of process.96 There had been insufficient grounds to refer the matter to the review panel, materials had not been disclosed to Justice Smith, and the procedures were pursued, at least in part, for an improper purpose.97 The court declared that Justice Smith had not contravened s 55, quashed the decision, and ordered that its judgment be posted on the CJC’s website and referenced in every mention to Justice Smith’s case.98 Following the decision, the CJC has decided that it is not in the public interest to appeal the decision and has stated that “[a]ll aspects of the order will be complied with promptly.”99
90 Ibid at para 76. 91 Ibid at paras 77–78. 92 Smith v Canada (AG), 2020 FC 629 at paras 75–76. 93 Ibid at paras 104, 110–12. 94 Ibid at para 114. 95 Ibid at paras 124–31. 96 Ibid at para 136. 97 Ibid at paras 145–46, 155–57, 160–63, 167–69. 98 Ibid at para 176, Order. 99 “Canadian Judicial Council responds to the Federal Court decision regarding Justice Patrick Smith” (25 May 2020), online: Canadian Judicial Council
2020] TO PARTICIPATE OR NOT TO PARTICIPATE 291
This decision provides a useful analysis of the meaning of ss 54 and 55 of the Judges Act. It confirms that judges can hold other positions while on a leave of absence, as long as they are not incompatible with the judicial role. Notably, while the interpretation of these sections arose in a judicial review, the court held that this was one of the rare situations in which there is only one reasonable interpretation of the provisions.100 The decision also suggests that the mere possibility that a university or other institution may become involved in litigation is insufficient to render a judge’s involvement incompatible with the judicial role.
E. Matters of public controversy
The Principles and Draft Revised Principles advise judges to be cautious as to their involvement in certain groups and organizations. In particular, “consideration should always be given to the possibility or likelihood that the group or organization could become involved in public controversy or litigation.”101 The Draft Revised Principles further counsel that “[t]he more likely the organization is to be embroiled in 102 controversy or litigation, the more cautious the judge should be.” The oft-cited case 2020 CanLIIDocs 3123 of Justice Thomas Berger provides a prime example of what can happen when a judge does become involved in such matters.
Justice Berger was a judge of the Supreme Court of British Columbia. Before his appointment, he was a leading lawyer in Aboriginal law and had represented Indigenous people in landmark cases. In 1981, he made a speech and wrote an article in a newspaper criticizing proposed constitutional changes for their failure to protect Aboriginal and treaty rights. He also criticized the absence of a veto by the province of Québec on future constitutional changes, a matter of intense controversy.103
Prime Minister Pierre Elliott Trudeau stated that it was improper for a sitting judge to make such comments. A judge of the Federal Court made a complaint to the CJC, and an inquiry panel was formed.104 The panel discussed the independence of the judiciary in detail, noting the following:
The history of the long struggle for separation of powers and the independence of the judiciary, not only establishes that the judge must be free from political interference, but that politicians must be free from
100 Ibid at para 181. 101 CJC, Draft Revised Principles, supra note 19 at 38, commentary 5.B.13. See also CJC, Principles, supra note 17 at 36–37, commentary C.8. 102 CJC, Draft Revised Principles, supra note 19 at 38, commentary 5.B.13. 103 Kent Roach, “Judges and Free Speech in Canada” in HP Lee, ed, Judiciaries in Comparative Perspective (New York: Cambridge University Press, 2011) 175 at 178. 104 Ibid.
292 UNBLJ RD UN-B [VOL/TOME 71
judicial intermeddling in political activities. This carries with it the important and necessary concomitant result—public confidence in the impartiality of judges—both in fact and appearance.105
The inquiry panel held that Justice Berger had “intervened in a matter of serious political concern and division when that division or controversy was at its height.”106 It rejected Justice Berger’s argument that his remarks related to matters of conscience rather than politics. The panel suggested that he should resign as a judge if he wanted to comment on such matters.107 It concluded: “Judges, of necessity, must be divorced from all politics. That does not prevent them from holding strong views on matters of great national importance but they are gagged by the very nature of their independent office, difficult as that may seem.”108 However, the inquiry panel refrained from recommending that Justice Berger be removed, as this was the first situation of this kind to come before the CJC.109
A subsequent resolution of the full CJC was less critical than the inquiry panel. It concluded that Justice Berger’s actions were “indiscreet” but did not warrant
removal. The CJC did, however, state that judges should not take part in controversial 2020 CanLIIDocs 3123 political discussions, except as they relate to the operation of the courts.110
The Chief Justice of Canada at the time, Bora Laskin, made the following comments about Justice Berger in a speech: “[U]nbelievably, some members of the press and some in public office in this country, seem to think that freedom of speech for the judges gave them the full scope of participation and comment on current political controversies, on current social and political issues. Was there ever such ignorance of history or principle?”111 Chief Justice Laskin further stated that any judge wanting to comment on political issues should resign from the bench.112 Both Chief Justice Laskin’s comments and the views expressed by the CJC have been commented on unfavourably by academics.113
105 John J Robinette, “Report and Record of the Committee of Investigation into the Conduct of the Hon. Mr. Justice Berger and Resolution of the Canadian Judicial Council” (1983) 28:2 McGill LJ 378 at 389. 106 Ibid at 389. 107 Ibid at 390–92. 108 Ibid at 391. 109 Ibid at 392. 110 Ibid at 379. 111 As cited in Roach, supra note 103 at 179. 112 Ibid. 113 Ibid at 179–80.
2020] TO PARTICIPATE OR NOT TO PARTICIPATE 293
F. Social media
The Draft Revised Principles include guidance on social media usage by judges. Most lawyers have an online presence before appointment to the bench. After appointment, is an online presence comparable to involvement in a community group or organization?
As Sossin and Bacal explain, different considerations may apply when considering a judge’s physical and digital involvement in a given activity:
Unlike social clubs or civic groups that one must join and maintain membership in, being part of a social-media network does not imply any particular commonality or shared values. For example, if a judge joins a civil-liberties association, that may be taken to imply sympathy with the association’s goals and activities. If a judge instead follows that organization on Twitter, or receives a feed from its Facebook page, however, it is unclear whether and in what circumstances this might be taken to imply support or “membership”.114 2020 CanLIIDocs 3123 Drawing from Justice Matlow’s situation, they suggest that “it is clear that judges who use social media to express their engagement in the community and to further their involvement in community affairs are not in violation of any ethical guideline per se”; rather than the medium, the focus will be on the context in which a judge discusses an issue, that is, as a private citizen or as a judicial officer.115
The Draft Revised Principles have provided further guidance. They state that social media activities “are subject to the overarching principles that guide judicial behaviour”; judges must consider how their activities may “reflect on themselves and upon the judiciary and should be attentive to the potential implications for their ability to perform their judicial role.”116 They should also be alert to how social media usage by family members may reflect adversely on them.117 Moreover, judges should be mindful of certain social media realities, including the ease and speed with which communications by social media can be transmitted and the potential for inappropriate communications directed to judges.118 There is also the reality that “[j]udges’ communications and associations with others are a common basis upon which claims of lack of impartiality are based”; as a result, “[j]udges should be vigilant in minimizing reasonable apprehension of bias arising from these communications and
114 Lorne Sossin & Meredith Bacal, “Judicial Ethics in a Digital Age” (2013) 46:3 UBC L Rev 629 at 634. 115 Ibid at 640. 116 CJC, Draft Revised Principles, supra note 19, 5.B.15. 117 Ibid. 118 Ibid, 5.B.16.
294 UNBLJ RD UN-B [VOL/TOME 71 associations,” which is “all the more important, and difficult, in the age of social media.”119
As social media use increases, this will undoubtedly be an area in which judges and judicial councils will face new and difficult issues on the limits of judges’ involvement.
VI. Conclusion
In summary, in Canada the key ethical considerations for judges regarding involvement in the community are the following:
• Judges must weigh the benefits of community involvement against the potential risks to impartiality and independence; • Judges must not engage in political activity of any kind; • Judges may engage in civic and charitable activities, provided they
do not reflect adversely on their impartiality or interfere with the 2020 CanLIIDocs 3123 performance of their judicial duties; • Judges should be wary of involvement in organizations or causes that are likely to be engaged in litigation or matters of public controversy; • Judges may advocate for their own private interests, provided they do not use the prestige of their office inappropriately; • Judges must not engage in fundraising; • Judges should be cautious when becoming involved with boards of schools, charitable foundations, and the like, as they are increasingly involved in litigation and matters of controversy; and • Judges should refrain from commenting on matters of public controversy.
Judges occupy a unique role in society. As a leading Canadian jurist puts it, “[t]o sit in judgment on one’s fellow citizens is a weighty responsibility. To carry it out, the judge must be worthy of public respect and confidence and be trusted by the bar.”120 We expect our judges to act with integrity, impartiality, and independence so as to ensure the fair resolution of disputes. Judges are called upon to show a restraint and propriety beyond that which is expected of their fellow citizens. Determining the appropriate limits of their personal conduct is not an exact science. Fortunately, much guidance exists from judicial councils, courts, and commentary. Ultimately, the genius of the law is experience. As is life, the law is ever changing.
119 Ibid, 5.B.17. 120 Sharpe, supra note 7 at 249.
“RECOVERING” FROM COVID-19: INSURING THE PANDEMIC
Ryan Chute*
Executive Summary
The full scale of COVID-19’s impact is not yet known, but it has already had devastating reverberating effects globally. All commercial enterprises, even those considered essential, are experiencing business interruptions as a result of the global pandemic. How commercial insurance policies respond to COVID-19 may determine whether certain businesses, and economies, weather the catastrophe. This article examines the anticipated commercial insurance response to policy claims stemming from the COVID-19 Pandemic. In particular, it examines four types of commercial insurance policies likely to be significantly impacted by COVID-19: liability, 2020 CanLIIDocs 3123 commercial property, cancellation, and business interruption insurance.
This article explores the ways in which commercial insurers and insurance policies have adapted in response to past pandemics and the impact that that evolution will have in responding to COVID-19. In general, policy amendments have taken the form of exclusionary provisions, stricter definitions, and higher evidentiary thresholds designed to reduce insurer exposure in a pandemic. Consequently, the more wide- spread the impacts of COVID-19, the narrower the circumstances in which commercial coverage will be sufficient. Speciality pandemic policies have emerged to fill the gap in coverage but remain uncommon in the Canadian insurance market.
Commercial liability and property insurance policies are not intended to cover the type of losses expected from an extended pandemic. Both policies include exclusionary provisions and require evidence of either bodily injury or physical property damage caused by an insured peril in order to trigger coverage. As COVID- 19 becomes more pervasive, it will become difficult or impossible to demonstrate causality even if widespread disease is covered under the policy. Cases where both coverage and causality can be demonstrated will be exceedingly rare.
While claims under commercial event cancellation and dedicated business interruption policies are more likely to succeed, they will, at best, provide only partial compensation. Exclusionary provisions and precise evidentiary requirements also expose policy holders to a potential denial of coverage and claims under these policies will likely be decided on a case by case basis.
* Ryan Chute received his JD from the University of New Brunswick in 2020.
296 UNBLJ RD UN-B [VOL/TOME 71
Only dedicated pandemic policies provide full coverage for losses sustained as a result of an extended pandemic. However, insurers warn that these policies are likely to contain fine print limiting or excluding coverage in certain circumstances. As with other forms of commercial insurance, recovery is not certain due to variations in policy language, coverage, and evidentiary requirements.
Recovering from COVID-19
This article will examine the anticipated commercial insurance response to policy claims stemming from the COVID-19 pandemic. While insurers of personal and health-related policies such as life and accident and sickness will be particularly hard hit, widespread illness and quarantine of the scale experienced in response to COVID- 19 will permeate all areas of the industry. From supply chain disruption to mandatory closures, commercial policy holders will face a significant number of ordinarily covered losses complicated by the fact that COVID-19 is an exceptional peril. In the wake of the COVID-19 pandemic, policy holders will want to know whether they are covered; claimants will want to know whether they can sue; and insurers will want to 2020 CanLIIDocs 3123 know how much this is going to cost.
The good news for commercial insurers, and the bad news for the majority of policy holders, is that pandemics like COVID-19 are not unanticipated. The 2002- 2003 severe acute respiratory syndrome (SARS) epidemic gave insurers a window into what could have been.1 In the years that followed, policies of all kinds were drafted with pandemics in mind to include exclusions for known widespread illnesses.2 Other policies, which I will refer to as pandemic policies, anticipate the risk of widespread illness and have adjusted premiums accordingly.3 At the same time, policy holders are not likely to have opted for the remarkably more expensive pandemic policies leaving insurers significantly more prepared for this disaster.4 This article will demonstrate that the more wide-spread the impacts of COVID-19, the narrower the circumstances in which coverage will be sufficient due to express exclusions and an insufficiency of evidence. In particular, this article will examine four types of commercial insurance policies likely to be significantly impacted by the COVID-19 pandemic: liability, commercial property, cancellation, and business interruption insurance.
1 Michael Ha, “Will Insurance Cover SARS Cleanup?”, National Underwriter 107:21 (26 May 2003) 14. 2 See e.g. Greg Meckbach, “Interrupted by Disease”, Canadian Underwriter 81:12 (December 2014) 30 at 32. 3 Mark A Hofmann, “Colleges’ insurance program developed to cover pandemics”, Business Insurance 43:19 (11 May 2009) 21 [Hofmann, “Insurance Program Developed”]. 4 Ibid at 21.
2020] “RECOVERING” FROM COVID-19 297
Background
The World Health Organization (WHO) has defined “pandemic” as the “worldwide spread of a new disease”.5 On March 12th, 2020, WHO declared COVID-19 a pandemic. Historically, flu pandemics of this scale are not uncommon. In 1918, an H1N1 virus referred to as Spanish Flu killed an estimated 50 million people worldwide.6 Forty years later, a mutant strain of H2N2 spread across the globe causing at least 1.1 million deaths.7 Only ten years later, Hong Kong flu (H3N2) killed nearly half a million people.8 In 2009, an H1N1 virus commonly known as Swine Flu infected roughly 700 million people and may have contributed to the deaths of up to 500 thousand. While the human cost of flu pandemics has decreased over the last century, the continued uncontrollable spread of novel diseases demonstrates the vulnerability of our society to pandemic threats. As the world becomes more interconnected through globalization, these pandemics begin to have reverberating economic effects the likes of which have never before been seen. Until COVID-19, Swine Flu was considered the last modern pandemic. However, only 18 thousand deaths were confirmed as attributable to Swine Flu and attempts to contain the spread of the virus were limited in comparison to COVID-19 and SARS. 2020 CanLIIDocs 3123
Categorized as an epidemic, the SARS outbreak of 2003 narrowly avoided classification as a pandemic due to its successful containment. Barrett Hubbard, an insurer with MINT Canadian Specialty has said that, “there is a difference between insuring a pandemic and a contagion outbreak”.9 He defines a contagion outbreak as “a random, localized event for which it is easier to price and write specific coverage”.10 In contrast, Hubbard says that a pandemic is an inevitable occurrence affecting an entire economy.11 SARS, which was successfully contained to the initial communities of infection, is more properly categorized as a contagion outbreak, and as such, did not pose as significant a threat to commercial insurers.
5 “What is a pandemic?” (24 February 2010), online: World Health Organization
298 UNBLJ RD UN-B [VOL/TOME 71
In total, SARS infected approximately eight thousand people worldwide, killing just under 800 of those infected.12 Its impacts, both financial and in terms of lives lost, pale in comparison to modern pandemics. However, the SARS outbreak provided the insurance industry with crucial insight into the potential costs of widespread illness. Although only twenty thousand Canadians were quarantined, Toronto Dominion Bank estimated the net cost of the SARS outbreak to the national economy at between $1.5 billion and $2.1 billion.13 Many of those losses were sustained in the form of ordinarily covered perils such as event cancellations and business interruptions.
As settlement is common in the insurance industry, there are no concrete statistics as to the success of commercial insurance claims stemming from the 2003 SARS outbreak. In fact, no case law exists on the subject in Canada. While there were certainly claims made as a result of SARS, one can only speculate as to the nature of their settlement. However, we do know that insurers responded to the 2003 outbreak by adopting exclusionary policies and holding insureds to strict evidentiary demands. Between 2002 and 2004, it was not uncommon for a wide variety of standard form insurance policies to contain provisions expressly excluding coverage for losses 2020 CanLIIDocs 3123 related to SARS.14 Many policies already contained inclusionary provisions covering only losses stemming from named illnesses. These exclusionary and inclusionary provisions were expanded to address a potential future pandemic. Since 2003, it has become standard for commercial insurance policies to contain some form of provision excluding coverage in the event that losses are the direct result of a pandemic.15
But some policy holders remained concerned in the wake of SARS and some insurers capitalized on the gap in coverage. By the time Swine Flu emerged in 2009, many specialty insurers had developed policies intended to address pandemics.16 Some pandemic policies are general, covering any insured object in the event of an unknown pandemic. Other policies aim to address known outbreaks such as H1N1.17 Despite the availability of these policies, they remain uncommon in commercial settings due to the prohibitive cost and rarity of pandemics.
12 “SARS Basics Fact Sheet” (06 December 2017), online: Centre for Disease Control and Prevention
2020] “RECOVERING” FROM COVID-19 299
Commercial Liability
With forced closures looming, and many already underway, a large number of employers fought to keep their doors open for as long as possible. Others closed initially but have reopened without any marked change in circumstances in an attempt to mitigate losses. However, this borrowed time comes at the cost of potential commercial liability. Employees, customers, venders, and contractors may all have claims against a commercial liability policy holder in the event that they contract COVID-19 as a result of the policy holder’s decision to remain open during the pandemic. Additional liability may arise if the policy holder fails to take necessary precautions to prevent the spread of the virus in their place of business. With thousands of new confirmed cases of COVID-19 being announced each day, it is almost inevitable that businesses remaining open during the pandemic will be exposed to commercial liability. Unfortunately, many of those businesses will discover too late that their commercial liability insurance policies do not cover losses stemming from COVID-19.
While other forms of commercial insurance policies exclude specific viruses 2020 CanLIIDocs 3123 either by name or broadly, general liability policies do not typically include such express exclusions for widespread illness.18 Instead, commercial insurers rely on pollution exclusion clauses excluding contaminants or irritants, arguing that such provisions capture pathogens, viruses and microorganisms.19 This approach was taken in response to both SARS and Swine Flu with mixed success in the United States.20 Again, due to the frequency of settlement, there is no precedent with which to determine the likelihood of pollution exclusions applying in the case of widespread illness in Canada. Nevertheless, whether or not pollution exclusions apply, they are merely the first line of defence against COVID-19 claims for commercial liability insurance providers.
Regardless of the applicability of exclusions, a policy holder must demonstrate causality between the occurrence and the impugned property damage or bodily injury.21 This evidentiary requirement may be the biggest hurdle facing policy holders attempting to recover under a commercial liability policy post-pandemic. Demonstrating property loss as a result of an illness is no easy task and will be discussed in greater detail below. In contrast, losses resulting from a serious and life- threatening physical injury are more easily demonstrated. Where claimants will encounter difficulty is in establishing a causal connection between the actions of the policy holder and the impugned injury.
18 Mark A Hofmann, “Liability policies vary on coverage provisions for claims related to H1N1 virus: Experts”, Business Insurance (Chicago) 43:19 (11 May 2009) 21 [Hofmann, “Liability Policies”]. 19 Ibid. 20 Ibid. 21 Ibid.
300 UNBLJ RD UN-B [VOL/TOME 71
By definition, pandemics involve worldwide spread of a new disease. Where the virus is pervasive, it will be difficult, if not impossible, to demonstrate causality. Mark Noonan, national workers compensation practice leader for Marsh Inc. in Boston said the following:
If you look at swine flu where it’s already indigenous in the United States, in several states it would be difficult to establish a workers comp claim beyond health care or public safety. […] When you go back a few years to avian flu, where it was in other countries but not in the United States, it would have been easier to track it back to a work exposure if a person was traveling.22
Although Noonan is clearly referring to workers compensation, his remarks are equally true for commercial liability insurance or any form of insurance for that matter. With liability insurance, the more opportunities a claimant has to be exposed to the insured peril, the more difficult it will be to provide evidence establishing a causal link between the policy holder’s actions and the loss. In the case of a pandemic, there can be no causal apportionment. Infection has a single cause and the onus is on the claimant to provide evidence that the infection or bodily injury was caused by the 2020 CanLIIDocs 3123 policy holder’s actions and could not have just as likely been caused by any other source.
Claims established earlier in the pandemic are more likely to be successful due to the comparatively low rate of infection. Evidence of causality is more easily collected if the infection can be tracked to a specific individual. However, as infection becomes more pervasive and sources of infection become less certain, circumstances in which causality, and ultimately, coverage can be demonstrated become narrower. Noonan said in 2009 that, “It's going to be a harder proof of claim for HlNl because it's already here in the United States-not impossible, but harder to establish causality”.23 SARS, as a contained contagion outbreak could be more easily tracked to a specific individual who had travelled outside of the country, whereas sources of Swine Flu were too numerous to ascertain.24
At the time of writing, there are confirmed cases of COVID-19 on every continent and in every Canadian province and territory with the exception of Nunavut. A lack of testing kits early on has made tracking community infection with certainty an impossibility. Making matters worse, individuals infected with COVID-19 are commonly asymptomatic, carrying the virus and infecting others without leaving evidence.25 While early liability claims may succeed if they are not caught by pollution
22 Ibid. 23 Ibid. 24 Ibid. 25 “Information on Novel Coronavirus (COVID-19) for Physician and Health Care Professionals” (20 March 2020), online: Ottawa Public Health
2020] “RECOVERING” FROM COVID-19 301 exclusion clauses, except in rare circumstances, the window for meeting the necessary evidentiary burden has long been closed.
Commercial Property
Many commercial property insurance policies cover not only direct losses incurred as a result of damage to an insured property, but those losses which result from the closure of the facility during renovation and repairs. While closed under a mandatory order, policy holders may attempt recovery under a commercial property insurance policy for losses related to the disruption. Viruses are not a covered cause of loss under standard commercial property insurance policies.26 In fact, many insurers have shored up policy definitions in the wake of recent modern pandemics to exclude coverage.27 Typically, property insurance policies trigger only where direct physical damage to or loss of an insured’s property can be demonstrated to be the result of a covered peril.28 Healthcare facilities or specialty businesses may already have policies specifically designed to address communicable disease contamination based on their needs.
However, those policies may require the order of an authorized governmental agency 2020 CanLIIDocs 3123 before the policy is triggered.29 A mandatory closure order such as those resulting from a pandemic would not be sufficient to trigger the policy as it still requires confirmation of property contamination resulting in losses.
Blanket closure orders, such as those ordered in response to the COVID-19 pandemic, are not responding to confirmed contamination, but are preventative measures designed to slow the spread of disease. Consequently, policy holders will not receive confirmation that their commercial properties have experienced contamination, nor will they be able to demonstrate that the mandatory closure was related to a presumed contamination. As with commercial liability insurance, the insured’s evidentiary burden cannot be met regardless of exclusionary provisions.
You will recall that insurer Barrett Hubbard described a pandemic as “an inevitable occurrence affecting an entire economy”.30 You will also recall that the primary difference between a contagion outbreak and a pandemic is containment. Commercial property insurance policies which include coverage for communicable
26 Dominic Anthony, “How to Protect Your Members and Your Fitness Business during the Coronavirus Pandemic”, Club Industry (12 March 2020), online:
302 UNBLJ RD UN-B [VOL/TOME 71 disease contamination often address issues related to containment of a contagion outbreak while leaving the policy holder to shoulder the primary costs of an extended pandemic.31 Where triggered, such policies cover additional expenses such as decontamination and additional staff to ensure a business is capable of reopening after a closure order is lifted.32 These policies are not aimed at addressing widespread illness, but rather, localized contamination. Decontamination, in the event of a pandemic such as COVID-19, would not be sufficient to reopen a non-essential business closed by a governmental authority. That property would remain closed, incurring losses not covered under a standard commercial property insurance policy. Consequently, recovery under a commercial property insurance policy is likely to be extremely limited in the event of losses stemming from COVID-19.
While limited recovery may be possible under specialized policies, these sums will likely be minimal and insufficient to cover extended losses. The traditional property market is not intending to cover such losses.33 Costs of that nature are more appropriately addressed through dedicated business interruption policies which, as we will see, have their own shortcomings in addressing widespread illness.
2020 CanLIIDocs 3123
Cancellation
Another major area of commercial insurance likely to be impacted by the COVID-19 pandemic is cancellation. As stated above, this article will address strictly commercial policies. While personal cancellation policies such as travel cancellation will be greatly impacted by COVID-19, these claims will also be inseparably linked to secondary events such as flight cancellations and travel bans. For that reason, this section will focus primarily on commercial event cancellation.
Cancellation insurance generally covers all unforeseen perils. Consequently, it is one of the few areas of commercial insurance that may expressly include widespread illness.34 During the 2003 SARS outbreak, many cancellation policies did provide coverage for events cancelled due to the illness.35 Although not universally accepted, SARS was typically covered until it became a known, recognized peril and exclusions were introduced.36 Early event cancellations due to COVID-19 may lead to successful claims for this same reason. Standard form policies covering widespread illness will not yet expressly exclude coverage for COVID-19 while policies excluding “mysterious disease outbreaks” will no longer apply following COVID’s recognition
31 Meeson, supra note 16. 32 Ibid. 33 Hofmann & Veysey, supra note 27. 34 Aldred, supra note 14. 35 Ibid. 36 Ibid.
2020] “RECOVERING” FROM COVID-19 303 and designation as a pandemic.37 Unless expressly excluding widespread illness or disease, the policy will likely pay out for early cancellations owing to the COVID-19 pandemic, though later cancellations will not fare as well. This is one of few areas of insurance law where COVID-19’s designation as a pandemic helps, rather than hinders, policy holder claims.
Despite coverage, widespread illness or even the declaration of a pandemic may not be sufficient to ensure recovery in the event that a policy holder chooses to cancel their event. Event organizers must pay close attention to advisories from WHO or the appropriate authorized governmental agency and wait for announcements recommending the cancellation of events. Cancellations prior to such advisories may be considered premature and lead to denial of coverage by insurers.38 Danny Burns, contingency underwriter for Lloyd's of London syndicate 102, explains that, "[f]ear of a peril is not covered”, adding that coverage of losses in anticipation of a pandemic would lead to systemic exposure for insurers.39 Consequently, it is a public policy argument that may prevent an event organizer from recovering losses due to cancellation where cancellation is, itself, in the public interest.
2020 CanLIIDocs 3123 Nevertheless, event cancellation insurance remains the silver lining for commercial policy holders looking to recover losses sustained during the COVID-19 pandemic. Unlike other commercial policies, designation of COVID-19 as a pandemic does not prevent recovery of cancellation insurance and provincial and federal lockdown and social distancing orders meet the policy holder’s evidentiary burden. While these policies will not come close to covering the total commercial losses of policy holders, high costs to insurers stemming from commercial event cancellation policies should be expected.
Business Interruption
With a significant portion of the Canadian population quarantined or self-isolated and many businesses closed or operating remotely in a limited capacity, there is no doubt that insurers will see a remarkable number of business interruption claims resulting from the COVID-19 pandemic. Anticipating future flu pandemics, Nick Beecroft, emerging risks and research manager at Lloyd's stated that “disruption [resulting from a pandemic] will cascade throughout the economy and could generate some very challenging exposures for carriers and the potential for widespread business
37 Margo McCall, “SARS Raises Cancellation Insurance Questions”, Tradeshow Week (Los Angeles) 33:19 (12 May 2003) 8. 38 Aldred, supra note 14. 39 Ibid.
304 UNBLJ RD UN-B [VOL/TOME 71 interruption”.40 He added that “global disease outbreaks-can dampen demand for goods and services, thereby disrupting supply chains”.41 Beecroft’s remarks were made in 2015 but underscore the mindset of insurance underwriters in the years following SARS, Swine Flu, and the escalation of Ebola.42 For the past two decades, underwriters have been anticipating and preparing for the next modern pandemic and the prevalence of dedicated business interruption policies in the Canadian insurance market demonstrates that the same is true for policy holders.
As with other standard commercial policies, infectious diseases are often expressly, impliedly, or conditionally excluded from coverage.43 However, for commercial enterprises seeking business interruption insurance, epidemics like SARS and Swine Flu may be a primary concern. As a result, insurers began to include provisions providing coverage only for “notifiable diseases”.44 The Public Health Agency of Canada defines “nationally notifiable diseases” as “infectious diseases that have been identified by the federal government and provinces and territories as priorities for monitoring and control.45 In the early months of the COVID-19 outbreak, insurers grappled with the issue of whether or not the then unnamed virus constituted 46 a notifiable disease within the intended meaning of the policy. While policies 2020 CanLIIDocs 3123 covering nationally notifiable diseases are not the norm, they are common enough for insurers to be concerned. The 2003 SARS outbreak demonstrated how a relatively contained contagion outbreak could quickly cause substantial financial losses.
Many insurers received their answer on March 11th, 2020 when WHO declared COVID-19 a pandemic. Some countries, including the United Kingdom, declared COVID-19 a notifiable disease prior to that declaration.47 However, at the time of writing, COVID-19 has not yet been added to the register of nationally notifiable diseases in Canada.48 SARS, similarly, was not added to the register of nationally notifiable diseases in Canada until after the outbreak had ended. While it is certain that COVID-19 will be added to the list eventually, a delay in that
40 Lori Chordas, “Fever Rising: Infectious disease outbreaks and epidemics have carriers and reinsurers on high alert.”, Best’s Review (Oldwick) 116:3 (July 2015) 28 at 30. 41 Ibid. 42 See e.g. Douglas McLeod, “Ebola Spread Triggers Insurance Concerns: Ace Ltd issues exclusion for some policies”, Business Insurance (Chicago) 48:22 (27 October 2014) 1. 43 Alex Newman, “How much trouble lies ahead for insurers?”, Investors Chronicle (London) (18 March 2020) 11, online:
2020] “RECOVERING” FROM COVID-19 305 announcement could cause some smaller insured companies to go out of business before business interruption policies are triggered.
While some business interruption policies expressly exclude infectious diseases and others require classification as a notifiable disease in order to trigger coverage, all business interruption policies will require evidence linking the interruption to an insured peril. Unlike other forms of insurance such as commercial property or liability, business interruption policies will not require that the policy holder demonstrate damage to property or bodily injury in order to trigger.49 Since these policies are intended to support the policy holder through disruptions to the ordinary course of business, coverage is triggered once the policy holder is able to demonstrate that a covered peril has caused business interruption. Evidence of financial loss is not required at the time that a claim for business interruption is submitted. Anticipated losses are sufficient provided that the policy holder is able to demonstrate disruption.
The bright side for business interruption policy holders is that the evidentiary threshold for demonstrating a business interruption in the event of a pandemic after 2020 CanLIIDocs 3123 coverage of an insured peril has been shown is quite low. In nearly all cases, an order from an authorized governmental agency in response to the pandemic and impacting the insured business will be sufficient.50 However, as with cancellation insurance, businesses may encounter difficulties recovering for business interruption losses if they close their doors prematurely or choose to remain closed after orders have been lifted. Interruptions caused by other businesses, such as suppliers or independent contractors, may be covered prior to a governmental declaration if evidence can be deduced demonstrating a link to the covered infectious disease or pandemic. In that case, coverage would depend largely on the language used when drafting the provision.
Pandemic Policies
As demonstrated, it can be difficult to recover damages stemming from pandemics under traditional commercial policies. Consequently, some businesses in recent years have begun to seek insurance for these specific circumstances. Opportunistic insurers have rose to the occasion and developed policies designed to cover business interruption and other losses in the event of a pandemic. For the purposes of this article, I will refer to the totality of these policies, regardless of the scope of coverage, as pandemic policies.
49 Hofmann, “Liability Policies”, supra note 18. 50 AHC Media, supra note 28.
306 UNBLJ RD UN-B [VOL/TOME 71
At the time of the Swine Flu pandemic in 2009, many insurers noted the failure of the traditional market to provide coverage for widespread infectious disease.51 Dave Finnis, national property practice leader for Willis HRH in Atlanta noted that the marketplace had not significantly changed in this regard post-SARS.52 This is not strictly true as post-SARS exclusionary provisions have made standard coverage for infectious diseases even less common. However, even in 2009, specialty policies covering damages related to widespread disease were available at a cost.53 Many of these policies were requested by and underwritten for specific employers rather than being a standard offering.54 Despite the availability of custom-tailored policies, post-Swine Flu, the majority of commercial policies continued to exclude coverage for losses stemming from serious infectious disease.
However, some insurers have made pandemic policies a standalone offering.55 These insurers, many of whom offered business interruption and continuity policies prior to SARS, Swine Flu, and COVID-19, have adapted their standard policies to include coverage specifically for pandemics. These insurers have looked at modern pandemics and contagion outbreaks, studying the way other countries responded to widespread illness and the steps taken to contain an outbreak and 2020 CanLIIDocs 3123 establishing an estimated cost of pandemic related disruptions.56 Underwriters then consider their findings in determining the necessary cost of risk transfer. As stated above, pandemic policies are significantly more costly than standard market policies due to the increase in risk being transferred and are not pervasive in Canada.
Although pandemic policies are designed to cover damages stemming from widespread infectious disease, recovery is not guaranteed. As with any insurance policy, coverage is dependent on the language used in the policy and interpretation of provisions will require significant attention to fine print.57 Since, in theory, all of these policies triggered as soon as WHO declared COVID-19 to be a pandemic, insurers offering pandemic policies can expect significant costs in the months ahead. With all of the pandemic policy holders attempting to recover at once, it is likely that insurers will be resistant and turn to definitions, exclusions, and evidentiary thresholds for assistance. Ultimately, coverage for these policies will depend on a case by case analysis, and even where coverage is triggered, it may not be sufficient to compensate policy holders for the totality of their losses.58
51 Hofmann & Veysey, supra note 27. 52 Ibid. 53 Hofmann & Veysey, supra note 27. 54 James Fernyhough, “Coronavirus an Insurance Catastrophe”, The Australian Financial Review (12 March 2020) 23. 55 Meeson, supra note 16. 56 Ibid. 57 Anthony, supra note 26. 58 Hofmann & Veysey, supra note 27.
2020] “RECOVERING” FROM COVID-19 307
Insurance Response
The full scale of COVID-19’s impact is not yet known, but it has already had devastating reverberating effects globally. Nearly all developed nations have declared a state of emergency and most have shuttered non-essential businesses and government services. All commercial enterprises, even those considered essential, are experiencing business interruptions as a result of the global pandemic. There is worldwide concern that markets will collapse, and insurers are not immune. How commercial insurance policies respond to COVID-19 may determine whether certain businesses, and economies, weather this catastrophe.
Commercial liability insurance policies are unlikely to protect policy holders from costs stemming from COVID-19. Where policies do not include an express exclusion for infectious disease, insurers will turn to pollution exclusions to deny coverage. Policy holders will also experience difficulty demonstrating the required physical property damage or bodily injury due to evidentiary shortcomings. The more widespread the virus, the more difficult it will be to show causation, and ultimately, to recover. 2020 CanLIIDocs 3123
Commercial property is not intended to cover losses related to widespread infectious disease. Policy holders will be required to demonstrate both that COVID- 19 is an insured peril and that it caused physical damage to the insured property. Some policies may cover the costs of decontamination but will not cover business interruption costs if the business cannot be immediately reopened. Nevertheless, policy holders are unlikely to meet the evidentiary threshold required to trigger coverage.
Commercial event cancellation policies are more likely to pay out than liability or commercial property insurance. With the exception of pandemic policies, it is the only area of commercial insurance where designation as a pandemic will help policy holders recover on standard market policies. However, event organizers must be careful not to cancel prematurely.
Dedicated business interruption policies could cover losses related to a pandemic. However, standard policies exclude infectious diseases and many modified policies require the virus to be declared a notifiable disease. COVID-19 has not yet been designated a nationally notifiable disease in Canada. If this declaration is made, business interruption policies with this inclusion may provide coverage, but the majority of standard market policies will not.
Companies with the forethought to insure their commercial interests with pandemic policies will be best protected from the impacts of COVID-19. However, insurers offering these policies will be at the highest risk. Pandemic policies contain the lowest evidentiary burden for policy holders, but coverage is still not guaranteed due to variations in policy language and coverage.
308 UNBLJ RD UN-B [VOL/TOME 71
With the majority of pandemic policies insured by a minority of insurers, widespread claims could cause these insurers to go out of business before their financial obligations are met. There are no publicly available statistics indicating the prevalence of these policies or their distribution amongst insurers. Premiums collected, even if collected since the 2009 pandemic, may not be sufficient to cover business continuity costs for larger corporations for the length of the pandemic and recovery period.
However, the outlook for commercial insurers remains positive despite dark clouds on the horizon. Although claims on personal insurance policies, such as life and accident and sickness, will cost insurers a remarkable sum of money, commercial policies, on the whole, are unlikely to provide coverage for the most significant losses as the virus becomes more pervasive. In fact, collected and ongoing premiums on commercial insurance policies may provide larger insurance companies with the capital needed to weather the storm.
2020 CanLIIDocs 3123