The world is drowning in data. Why don't we trade it like on a stock exchange?

The digital revolution is flooding the world with data. By 2025, an astounding 175 zettabytes – that’s 175 trillion gigabytes – of new data is expected to be generated annually, a fivefold increase on 2018. Much of that data will end up sitting idle on servers and hard drives –unused, unanalyzed, and, perhaps worst of all, unshared.

Data is valuable only when it is in the right hands: One person’s meaningless statistic can be another person’s gold. If that other person is, say, an epidemiologist, a climate scientist, or an emergency worker in a natural disaster, the value of shared data can be measured in lives.

The World Economic Forum has identified this issue and in response created the Data for Common Purpose Initiative (DCPI), working with a multistakeholder community to unlock data for common purposes while protecting and respecting individual privacy rights. Luckily, some organizations are starting to wake up to the importance of making data more widely accessible. But there is still a long way to go.

One idea for promoting the use of data and data-driven insights that is gaining traction is data exchanges: platforms where information, or the right to access certain information under certain conditions, can be traded in an open, efficient and accountable way. The obvious analogy is financial markets. Instead of buying and selling stocks, bonds or cattle futures, participants in data exchanges would trade information collected in a wide range of fields, from healthcare to manufacturing.

This concept is currently being explored and developed in India, Colombia and Japan as a mechanism for achieving each country’s respective nationwide digitization strategies. In late 2021, the Colombian government, together with the Centre for the Fourth Industrial Revolution Colombia and DCPI collectively devised a framework for data exchanges to be deployed in the country as part of a broader digitalization effort.

The Centre for the Fourth Industrial Revolution Norway is also collaborating with DCPI on further advancing the Centre’s ship emissions tracking initiative and leveraging the insights for climate-related action. The Data Marketplace Service Providers (or DMSPs) have also been proposed as likely primary operators and managers of data exchanges by the World Economic Forum.

One key consideration to note is the unique nature of data as an asset. It can gain value from a mosaic of multi-iterated combinations and recombinations with other datasets to produce a different insight each time, and therefore distinct value. Traditional and even in large part alternative assets do not have these properties.

So what exactly are the similarities and differences between data exchanges and more established financial exchanges like the New York Stock Exchange or Nasdaq? The answers matter for everything from regulating data exchanges to getting them up and running in the first place.

Say you own a share in a company, but decide to sell it. Poof – when the transaction closes, the share is no longer yours. It belongs to another person, and that person alone. But data is different. It is copyable, which means you can sell or give away data while keeping it for your own use, too. Data can be used by many “owners” at once.

That creates different challenges for data trading than for trading stocks or bonds. Financial exchanges keep track of who owns what. Their concern is exclusivity. Data exchanges, as conduits for infinitely copyable assets, will be more concerned with authenticity. Is this data real? Has it been corrupted or tampered with since it was originally collected?

To become trusted platforms for the use of data, data exchanges will need to incorporate anti-falsification measures such as strong authentication protocols and blockchain-based tracing and verification tools. Trust, much like financial markets, is the bedrock for making the platform function optimally. One aspect of this is by reinforcing and creating a strong foundation for trust, data sellers and buyers are more likely to participate. A byproduct of this is greater liquidity and less volatile trading. This in turn will inspire additional participants to join the platform, which will create even more liquidity and smoother trading, and so on and so forth.

In a world without stock exchanges, every investment in a company would require drawing up a complex contract from scratch. How much of the business is the investor buying? What dividends does she have a right to receive? How will his opinions about the way the company is run be communicated to managers, and weighed against the opinions of other owners?

Stock exchanges simplify all of that. On a stock exchange, standardized stock certificates replace bespoke contracts. They lay out all the rights and responsibilities that a contract would have, but make investing and trading easier.

Similar tools could and should be used in data exchanges. The buyer in a transaction facilitated by a data exchange would receive the right to access and use certain data, under certain conditions. The details could be enumerated in a tradable “data certificate” rather than a complex contract, reducing transaction costs and enhancing liquidity. Higher liquidity makes it easier for prices to be discovered, and at the same time stabilizes the business of the exchange. As data trading evolved and became more commonplace, standardized products and trading formats might emerge, as they have in financial markets. That could make things even easier for users and further promote the exchange of data.

One reason financial exchanges are useful is they help people understand what things are worth. Stock prices are public, so buying shares on an exchange takes away the fear that someone else might be getting vastly better deal.

Could data exchanges do something similar for information? Data is notoriously hard to value, but exchanges would lend a degree of transparency to the market and help track supply and demand. There would be limits, though. Assessing the underlying value of data – as opposed to its current market price – would require the development of valuation models that do not currently exist. Without them, potential data traders might still stay away.

Such models do exist for financial assets, which are typically assessed in terms of the present value of their expected future cash flows. A legion of analysts exists to do the math and advise investors. Data could in theory be valued the same way, but the practice is still in its infancy. Data valuation also cannot be narrowly understood in the context of narrow buy-and-sell models. Rather, it needs to also include data’s progenitive nature and the implication that it has on the contextual value of data and various datasets that each provide a unique view, and therefore, a unique value.

Financial exchanges need trust to function, and trust requires effective regulation. Securities regulations seek to ensure transparency and prevent market manipulation and other abuses. In places like Japan, financial regulations have been expanded to cover new kinds of assets such as cryptocurrencies.

There is no equivalent of the US Securities and Exchange Commission (SEC) for data trading, nor are there specific laws governing the practice. As long as data are bought and sold privately, there may be no need for disclosure rules or penalties for insider trading. But an equitable and successful public market would require such regulations, as well as a system of enforcement. As in finance, exchanges open to the general public might need tighter rules and stricter enforcement than those restricted to professionals.

A major difference between financial and data exchanges is the role of privacy protection, which is a critical factor in data transactions. Any regulatory regime governing data-exchange operators would need to address the privacy concerns of both market participants and the public.

It’s one thing to set up a data exchange; it’s another thing to get people to use it. Markets are attractive when there is plenty of liquidity – that is, lots of buyers and sellers are actively trading (as we mentioned before). But how do you appeal to potential traders when nothing exists yet? Particularly when it comes to data sellers (lots of businesses are eager to buy data), providing clear, convincing incentives will be key. That includes financial rewards but also trusted governance, the kind that can only be created through the efforts of a wide range of public and private-sector stakeholders.

The world’s rising tide of data can be used for good. But first we need to share it.