The pending NAFTA-based investor-State dispute settlement case Einarsson v. Canada has been noteworthy for matters of both procedure and substance. The claimants are a father and two sons, all holding US citizenship, and their Canadian-incorporated company, Geophysical Services Incorporated (GSI). They claim that by violating copyright and trade secret protections of GSI’s seismic data, which allegedly led to the company’s effective demise, Canada committed several breaches of NAFTA’s Chapter 11. This post engages with the most novel aspect of the substantive matters in Einarsson v. Canada: it is the first investor-State dispute settlement case about data. It analyses this novelty from the perspective of investor-State arbitration by addressing the protection of data as an ‘investment’ in the case. The post concludes by examining the implications of Einarsson v. Canada for tribunals’ future assessment of data, including data crucial to the digital economy, like social media companies’ user data.
Summary of the Dispute
The Claimant company, GSI, emanated from US-based companies that for decades, had collected seismic offshore data in Canada to provide to the oil and gas industry. GSI was incorporated by Theodore David ‘Davey’ Einarsson in Canada in 1993, who subsequently involved his two sons Paul and Russell in leading positions in the company.
GSI was engaged in extensive litigation in Canada from 2007 to 2017 against authorities and third parties, alleging violations of copyright and trade secrets by Canadian authorities’ disclosure of GSI’s seismic data to third parties (for a more extensive account of the facts, see here). Court decisions, upheld on appeal, established on the one hand, GSI’s copyright over the data. On the other hand, they found the Canadian authorities’ disclosure of data to be legal, with Canada’s regulatory regime for the oil and gas industry constituting lex specialis supplanting general copyright protections.
The Einarssons and GSI initiated investor-State arbitration in October 2018. Their key allegations are that Canada, by the court decisions finding authorities’ disclosure of data to be legal, breached NAFTA’s prohibition of performance requirements and indirectly expropriated GSI without compensation. So far, the tribunal has not made decisions on jurisdiction or merits publicly available.
‘Data’ in Einarsson v. Canada as an Asset in the Perspective of Investor-State Arbitration
In Einarsson v. Canada, seismic data is the claimants’ fundamental asset. This section explores the core issue for purposes of the investor-State arbitration: whether such data is protected under an investment treaty as an ‘investment’.
The claimants in Einarsson v. Canada do not explicitly make the case that GSI’s data was protected as an investment. Instead, among the assets listed in NAFTA’s definition of investments in Article 1139, they refer to GSI as an “enterprise”, “equity and security” through the Einarssons owning all of GSI’s shares, the Einarssons’ loans to GSI, and the interests from profit-determined remuneration by GSI (paras. 147-163). Especially their expropriation claim, however, emphasises that GSI’s very existence depended on its seismic data and that data’s protection as intellectual property (section IV.A.).
Thereby, Einarsson v. Canada is one of the few examples so far of investor-State arbitration cases concerning intellectual property rights. Among those rights, data-related copyright is a novel matter for an international investment tribunal to assess. That assessment could rely however on the Canadian court’s finding that GSI’s data was copyright-protected as evidence that even the respondent recognised the data under its domestic law to be the claimants’ property. The coverage of assets by host States’ law on intellectual property rights has been a critical issue in the previous cases of investor-State arbitration tribunals’ addressing the merits of claims related to such rights: patents in Eli Lilly v. Canada and trademarks in Philip Morris v. Uruguay and Bridgestone v. Panama.
In Einarsson v. Canada, the recognition of GSI’s data as property under municipal law supports that the claimants’ data and intellectual property rights would be “intangible property” covered by NAFTA’s definition of assets protected as an investment (see here, section II.1. and here, section 3.3.). As a Canadian-incorporated company’s seismic information from the Canadian offshore submitted to Canadian authorities, the data would presumably also meet the requirement of a territorial link for such protection.
GSI’s data being fundamental for a company that the Einarssons owned all shares of, made loans to, aimed to expand, and incorporated in Canada, it could even conform to each of the requirements of the much-debated Salini ‘test’: “contributions, a certain duration of performance of the contract and a participation in the risks of the transaction” together with the “additional condition” of a “contribution to the economic development of the host State” (para. 52). Including this “additional” and most controversial part of the ‘test’ would be fulfilled by the data’s provision to oil and gas companies operating in Canada.
This analysis leads to two conclusions about Einarsson v. Canada as the first investor-State arbitration case about data. First, with respondent courts’ recognition of claimants’ copyright and the conformity of GSI’s data to requirements for protection as an investment, the tribunal would not venture into unchartered territory when assessing whether those assets are protected. Second, were the case to lead to novel data-related findings, they would concern intellectual property rights, especially copyright which so far has not been addressed in investor-State arbitration, rather than data as such. The tribunal might however assess data sharing requirements in a manner that the case would have broader implications for data policy-making.
Shedding Light on Digital Economy’s ‘Data’?
So far, neither international investment agreements nor investor-State arbitration case law explicitly address whether data is an asset that is protected as an investment. Against this background, already since the initial stages of Einarsson v. Canada, its potential to break new ground or provide insights about types of data in a digital economy context has been discussed.
But can Einarsson v. Canada be expected to address matters that would be essential for claims related to social media platforms or cross-border data flows? To start with, the data in Einarsson v. Canada is not exclusively digital: the most recent submission requirements for GSI’s by Canadian authorities referred to by the claimants (para. 45) consisted in a CD copy with a file alongside a paper copy. More important even is that the kind of data serving as a building block to the digital economy often is data ‘as such’ (i.e. not fulfilling specific legal requirements such as those for copyright). That type of data raises different legal challenges from the data in Einarsson v. Canada:
- Could a property claim be established by a claimant over ‘data-as-such’? For example, even if social media companies were to hold assets protected as investment, would that include user data, when current legal developments tend to enhance users’, not companies’, position towards user data’s property status?
- Tribunals might apply the Salini criteria less rigidly than to the extreme of a checklist where each of the boxes need to be ticked off, or not resort to this ‘test’ at all. But even then, could data obtained through an app starting to get used by many people in one territory, without any specific effort directed towards that market, qualify for protection as an investment (see here, p. 377)?
- How could a territorial link be established when, as pointed out amongst others by UNCTAD, the digital economy implies that a business may provide a service while being ‘asset-light’ in that territory? In significant contrast to the link with intellectual property rights of the data in Einarsson v. Canada, discussions of ‘data-as-such’ and other digital assets’ fulfilment of the territorial requirement resort to analogies with certain financial assets of elusive territoriality, such as those in Abaclat and others v. Argentina (see e.g. here, p. 563-567).
- And last, would claims about data make ratione personae concerns by tribunals particularly wary of forum-shopping and treaty-shopping particularly salient, in view of the complexity of e.g. social media companies’ business structure?
Such questions caution against assuming that data-related findings in Einarsson v. Canada might shed light on how tribunals would assess claims in relation to social media companies’ user data, which exemplifies ‘data as such’.
Conclusion
Einarsson v. Canada holds the potential to clarify important aspects about the protection of claimants’ data as an investment and related intellectual property rights. An analysis of the data at stake in the case, as regards property, protection as an investment, territorial link, and establishing ratione personae, nonetheless reveals that expectations should not be raised too high. Indeed, it should not be expected that the Einarsson v. Canada tribunal will venture into unchartered territory so much that it returns therefrom with insights applicable to data that is typical in a digital economy context. Rather, the case’s relevance in relation to data should be approached not in view of the suggestive word ‘data’, but the actual legal implications of different types of data as assets.
________________________
To make sure you do not miss out on regular updates from the Kluwer Arbitration Blog, please subscribe here. To submit a proposal for a blog post, please consult our Editorial Guidelines.
