Examining the Graph Network’s GRT Token Economy

  • Commentary
  • October 3, 2020

The Graph, a blockchain indexing and querying project, recently unveiled a relatively high-level overview of its upcoming GRT token economy, set to debut alongside The Graph Network’s mainnet before the end of the year. The Graph allows for the indexing and querying of blockchain and other open network data, including Ethereum and IPFS. Data is pulled via APIs and the use of GraphQL, an API query language. For now, the project operates as a centralized service, yet, the upcoming mainnet will allow for anyone to index and curate network data and be compensated with the GRT token for doing so. Blockchain clients are not suitable for pulling data directly therefrom and require an indexing service such as The Graph. However, the Network’s supposed unique value proposition is that it allows for dApps and businesses to pull useful data without any reliance on centralized servers, such as Infura.

Under the token economy, indexers and curators will stake the GRT token for the right to perform indexing and curating services, earning compensation through a combination of inflationary block rewards and querying fees, or, in the case of curators, solely through fees. Bearers of the GRT token will be able to earn a portion of indexers’ compensation by delegating their GRT tokens to indexers, while users—which in most cases will be applications—pay indexers and curators for their services with the token. For further details of actor roles and interactions with the token economy, we refer readers to The Graph’s original post.

While, in many ways, GRT represents the popular hybrid utility-payment model, it does introduce some unusual and perhaps entirely novel mechanisms. For one, the use of a rebate pool strives to incentivize indexers to tune their GRT stake in proportion to the amount of query fees that they generate through the use of a Cobbs-Douglas Production Function, an econometrics function that attempts to relate an actor’s capital contribution to their labor contribution. While not explicitly referenced, it appears that in addition to this incentivization, the pool will also act as a velocity sink of sorts, collecting GRT from query fees, and theoretically providing price support to the token.

In addition, the Network implements a withdrawal tax on curators and delegators when they wish to withdraw some or all of their GRT stake. While the exact rate is yet unspecified, any amounts raised will be burnt in addition to any rewards that are not claimed from the aforementioned rewards pool. Aside from countering the inflationary pressure from rewards, the tax, according to the post, is in part designed to encourage curators and delegators to make decisions that help the Network. However, without any more details, this explanation is quite unsatisfactory. Notwithstanding this, a direct tax represents a significantly more aggressive control mechanism on token holders and delegators than is typically seen in cryptoeconomic designs, where time-mediated unbonding periods are the norm.

Beyond these mechanisms, the governance of the upcoming Network raises some concerns. The economic parameters of the token, such as its rate of issuance, are determined not by GRT token holders, but by otherwise undefined “independent technical governance.” While direct token-mediated on-chain governance certainly has its problems, The Graph’s design choice seemingly puts the interest of token holders at odds with governance, and without control over GRT’s economic levers, token owners are subject to a significant degree of unpredictability, thus risk, concerning their asset’s supply dynamics.