Bob Fay is the managing director of digital economy at the Centre for International Governance Innovation (CIGI) and co-chair of the IARIW-CIGI conference on the valuation of data.
Data has been referred to as the new oil, the oxygen of AI and even the new plutonium. No matter the analogy, it’s clear that data is broadly deemed to have value. That’s as it should be. Data is at the heart of our intangibles-based economy. It’s central to innovation and new technologies such as generative AI. And it’s the foundation of all social media platforms. Data is, quite literally, transforming existing businesses and creating new ones as we speak. It is the principal driver of economic growth and wealth in this era.
Surely something this valuable must have a well-defined method of valuation? But no. That is not the case. And this needs to change, urgently.
This is not to say there haven’t been efforts in the right direction. Statistics Canada produced experimental estimates in 2019 in the range of $200-billion for Canada alone. To put that in perspective, Statscan has estimated the value of Canada’s oil reserves at about $300-billion. Others point to the market capitalizations of so-called big data companies such as Google, Amazon, Apple and Meta, which are in the trillions of dollars (though the book values of their tangible assets of these firms are in the billions, the difference being the inferred value of data.)
But the truth is that even accountants, whose job it is to value assets, including intangible ones, haven’t yet figured this out.
Importantly, the value of data to a firm is not necessarily its value to society. When data is aggregated, it can reveal highly beneficial new insights – for example, new treatments in health care, improvements to traffic flow, tracking pandemics, a better understanding of climate change and the like.
Conversely, aggregated data can be used for practices that are invasive and socially harmful, including the erosion of personal privacy, greater economic concentration and less competition, the promotion of disinformation, electoral interference and even genocide.
These factors shed light on why no ideal, standardized methodology has yet emerged to measure data’s value either to the individual, to a firm or to society; it depends on its use in a particular context. This context, in turn, is framed by governance – the rules and regulations that determine how data can be used, which of course vary across jurisdictions. But it also reflects a lack of laws and regulation, or the inability to enforce them, especially around uses of personal data.
This leads to very important policy implications for Canada.
First, there are prenotification rules when a merger takes place – should revenues or assets be above a certain threshold, they must be reviewed by the federal Competition Bureau. If the purchaser is a foreign company, it also must also comply with the Investment Canada Act. In neither case is the value of data assets currently included in that assessment. Yet acquiring the data may be the main purpose of the acquisition, with implications for the protection of personal privacy, competition and even national security. This is an enormous regulatory gap.
Second, private-sector data is essential for the crafting of sound policy. Yet firms may be unwilling, unsure or incapable of sharing that information, despite the fact it could be used to benefit society. To unleash that value of data, firms need clear intellectual property rules, regulations and standards related to data use and reuse, as well as clear rules on data mobility. This, in turn, requires sound privacy legislation, including in the areas of deidentification and/or anonymization, plus interoperability.
Third, given the overlapping ways in which data permeates the economy, policy making must be adaptive. Although digital technologies touch on personal data and privacy, public health, competition and consumer protection, policy making typically treats each of these areas separately, creating fragmented responses and overlapping regulations. That’s why a whole-of-government approach is required – one that must include all orders of government. To add to this complexity, policy structures must also consider what is happening beyond Canada’s borders.
Fully unleashing the value of data – for individuals, for firms, for society – requires sound governance, at the regional, national and international levels. Integral to that process is determining how we set data’s value. This should be the starting point of every current discussion about data governance.