A digital platform is a place online where people can buy and sell goods, services, and assets. These platforms range from well-known brands such as Uber and Amazon to cryptocurrency platforms such as Ethereum and Aave.

Over the last decade, a growing number of these digital platforms have begun issuing their own currencies, which they often call tokens.1 On some platforms, tokens can be used for certain transactions. Some platforms also issue tokens for voting rights and other governance-related activities. In their paper, “Token-Based Platform Governance,” Joseph Abadi of the Philadelphia Fed and Markus Brunnermeier of Princeton University evaluate these digital platform innovations. Specifically, the authors created a model to examine whether the introduction of tokens for transacting and governance can make digital platforms more efficient. They accomplish this examination by comparing three broad types of platforms: traditional digital platforms, which don’t issue tokens; platforms that issue tokens for transacting but retain a shareholder governance structure; and platforms that issue tokens for transacting and participation in governance.2

Traditional platforms are owned and governed by shareholders who act to maximize the platform’s stock price. Users typically have no financial stake in the platform and do not play a role in its decision-making process. A common concern raised by critics of traditional platforms is that their shareholders may attempt to increase profits by exploiting users — by, for example, charging inefficiently high prices for the platform’s services or selling user data to third parties.

When a digital platform issues tokens, users can purchase the tokens for certain transactions, such as to receive discounts on trading fees or to rent out computational resources. Because these tokens can increase in value, they can provide a platform with an additional source of financing. Users will pay a higher price for tokens when shareholders implement policies (relating to, for example, software upgrades and fee structure) that benefit users.

Still other platforms issue tokens for transaction purposes, the right to vote, and other activities traditionally reserved for shareholders. When ownership is decentralized in this way, both users and investors receive tokens. Users hold tokens to transact on the platform or to utilize its services. Investors, meanwhile, set aside tokens as collateral — a process known as “staking” — and receive financing in the form of dividends. All token holders have the right to vote on a platform’s policies.

This decentralized framework fundamentally alters the relationship between users and the platform. Indeed, governance tokens embody the spirit of decentralization that inspired the development of cryptocurrencies in the first place. Still, voting rights are proportional to token holdings, and thus they give certain groups a disproportionate influence on a platform’s policies.

To understand the impact of token issuance and decentralized ownership on the efficiency of digital platforms, Abadi and Brunnermeier developed a general model of platform governance that captures the characteristics of a traditional platform, a platform that issues tokens for transacting but maintains a traditional governance structure, and a platform that issues tokens for transacting and governance. They measure efficiency in terms of the level of user participation and investment on a platform.

They find that a traditional platform is less efficient because “shareholders may exercise the firm’s market power to extract rents [that is, excessive payments] from users.” Specifically, when a platform increases its fees above the marginal cost of processing transactions, it lowers users’ participation on the platform as well as transaction volumes.3

Issuing tokens that offer a claim on a platform’s services, the authors find, helps align shareholders' policy preferences with those of users. Much depends, however, on whether shareholders can commit to these policies in advance. If shareholders are able to commit, then issuing tokens can lead to a more efficient outcome. Users will be willing to pay a greater price to purchase tokens if the platform passes policies that benefit them, and shareholders can issue tokens to users at a higher price and increase their rate of return. However, if the shareholders can’t commit to maintaining these policies, “this mechanism no longer aligns preferences," Abadi and Brunnermeier write. "After selling tokens to users, investors will again be tempted to extract rents from them.”

Decentralized governance, they show, overcomes the potential commitment problem associated with token issuance. If shareholders cannot commit in advance to policies that benefit the platform, the same efficient outcome can be achieved by issuing a token that offers both transactional services and an ownership share. This ownership share includes both voting rights and cash flow claims on the platform. “Both constituencies care about maintaining the token’s value,” they write, “which partially aligns policy preferences even without commitment and thereby limits rent extraction regardless of which group controls the majority of voting power.”

Abadi and Brunnermeier’s paper builds upon the existing literature by analyzing various ways tokens can be used — for transaction service claims, voting rights, and cash flow claims. Their findings provide novel insights into the optimal design of tokens, which can be used to enhance the welfare of both users and investors.

  1. The views expressed here are solely those of the author and do not necessarily reflect the views of the Federal Reserve Bank of Philadelphia or the Federal Reserve System.
  2. Cryptocurrencies and tokens are two different kinds of digital currency. Cryptocurrencies are used to make payments for goods and services. Tokens are used to conduct business on a digital platform and/or to have a vote in that platform’s decision-making.
  3. Within these groupings, the platforms vary in their token offerings and governance structure.
  4. For more on this topic, see Michael Magill, Martine Quinzii, and Jean-Charles Rochet, “A Theory of the Stakeholder Corporation,” Econometrica, 83:5 (2015), pp. 1685–1725.