With markets in real property, personal property, and intellectual property quite cornered, the future-savvy lawyer might consider their cutting-edge cousin, if France’s data-mining tax proposal has its way: what could be termed existential property*, courtesy of Google, Facebook, Amazon, and the like. Or rather, courtesy of their users, whose digitally collected personal data may be wholesale commoditized as a direct source of tax for the French government, according to a recent report on France’s digital economy.
Background: “Google France checked in at Bermuda”
The recommendation to tax data-mining is the latest volley in an ongoing tax battle between France and internet behemoths such as Google (case study) and Apple (case study). Essentially, it has become common practice for these companies to operate with expenses (such as labour) concentrated in high-tax countries in the European Union, such as France and the UK, while routing most of their revenues through “tax havens” such as Ireland, the Netherlands, and Bermuda, thereby avoiding an estimated average of 500 million euros per year in corporate tax, in France alone. The data-mining tax is one of several proposed solutions, following an attempted online advertising tax and controversial Google link tax.
The Rationale: “User added a new job at Facebook, Google, Amazon, and Apple”
The rationale behind the tax recommendation, elaborated upon in Forbes by one of the report’s authors, is as follows: Data plays such an important and ubiquitous role that it may now be considered the “raw material” of the digital economy. Users provide what may soon be literally lifetimes of data in various forms online, whether collected through behaviour-tracking cookies, submitted through tweets and searches, or inferred through analytics. This allows online companies and applications to laser-target users through features and ads, monetizing the collated data. Thus, users themselves provide data that feeds back into the supply-production-distribution-consumption chain, and according to the report’s authors, this turns users into employees whose unpaid labour of providing data produces value for these companies. This user-created value is unaccounted for, and should be taxed.
Implementation: “Facebook added 1 billion friends. Auditor poked Facebook.”
Since international tax law currently fails to account for the geography-heedless nature of user data-based business models, the data-mining tax (which the French government has yet to endorse), is meant as a step towards the report’s proposed international tax law reform. The tax would apply to both international and domestic businesses that regularly and systematically monitor online user behaviours of those in France. Tax rates would depend on various factors: how many users are tracked, the type of data collected, ethical issues, and level of respect for user privacy and control, among others.
Analysis: “@User tweets about #Privacy #ConflictofInterest #Competition and #PublicUtilities”
The idea of taxing data-mining immediately brings a number of issues to mind, the first of which is suggestively indicated by other names for the proposal: some call it a privacy tax or a “predator pays” policy. It may be problematic to create monetary incentives for corporations to respect user privacy, as it essentially commoditizes privacy (or the lack thereof) and may erode higher ideals of respecting privacy for its own sake; perhaps those who warrant the term “predator” should not be made to pay, but should restrain from undue preying altogether. From a practical perspective, the act itself of auditing companies’ practices may involve questionably invasive technological practices, such as deep packet inspection.
Second, tying government revenue to companies’ privacy practices the way this tax would (where less user control means the government levies higher taxes) creates a potential conflict of interest, if the government is supposed to have citizens’ best privacy interests at heart. Moreover, since the data belongs to the user, the labour model underlying the report’s recommendation raises the thought that perhaps users themselves should be paid for it.
Third, the data-mining proposal prompts interesting connections between privacy and competition (or antitrust) law. As demonstrated by cases against Google, Facebook (decision), and Apple, such companies walk a fine line between maintaining a healthy monopoly and engaging in anti-competitive practices. Incentivizing better privacy policies through taxes may put a damper on the endless reach for data to sell to advertisers, while creating room for smaller competitors who more effectively prioritize user privacy and control.
Fourth, turning data-mining into a source of taxation evokes questions about the role of privately owned technological platforms in the public sphere. Whether with Facebook and activism or Google and free speech, such websites at times seem to approach the status of public utilities. The problem is that regardless of sociological status, economically and structurally these companies are wholly private. This unique yet rising combination means that attempts to regulate the driving business model warrant particularly careful scrutiny, and perhaps a conversation about what such sites’ status ought to be.
Finally, it bears remembering, first, that the companies in question are offering free services whose terms people agree to on a voluntary basis (though see the public utility debate linked above). Second, whether or not data-mining becomes taxable, Google, Facebook, et al. already and will continue to monitor and benefit from users’ data regardless. One could then argue that the public may as well take advantage of that fact, in this case via taxation. As the old adage goes, after all, you are what you tweet.
Cynthia Khoo is a JD Candidate at the University of Victoria.
*Term coined for this post, based on the notion that the collected “property” is intangible (unlike real or personal property), yet not necessarily created or thought up (unlike intellectual property), but simply gleaned from users’ data trails as they go about their daily lives on the internet.