By Anders Brownworth, Jon Durfee, Michael Lee, and Antoine Martin
WASHINGTON, USA – Permissionless blockchains, which support the most popular cryptocurrency networks like Bitcoin and Ethereum, have shown that it is possible to transfer value without relying on centralized trusted third parties, something that is new and remarkable (although perhaps most clearly useful for less developed financial markets). What makes permissionless blockchains able to transfer value without relying on a small number of trusted third parties is the combination of several components that all need to work together. The components themselves are not particularly new, but the combination of these components is more than the sum of its parts. In this post, we provide a high-level overview of these components and how they interact, taking Bitcoin as an example.
Payments and trust
Payment systems traditionally operate with a small number of economic agents playing key roles, such as making sure only authorized transactions take place. This requires participants to trust these agents to perform their duties. This trust is bolstered by law, rules, and regulations that provide incentives for the agents playing these central roles not to misbehave. Nevertheless, bad behavior, while rare, cannot be ruled out.
Cryptocurrencies, such as bitcoin, were designed to operate using a different model of trust, one that does not rely on law and regulations. Instead, they rely on two features, open access and a set of incentives to induce good behavior. Open access, or “permissionlessness,” is necessary to eliminate the risk that a gatekeeper could exert power to their advantage. Mechanisms that offer incentives aimed at limiting misbehavior by participants in the system are necessary since open access permits anyone to opt into any role in the system. While each feature is important on its own, we argue that the combination of the two is necessary for a permissionless blockchain to work as intended. We now describe each feature in more detail.
Open access or permissionlessness
In permissionless blockchains, anyone can use the payment system without restrictions. In addition, anyone can choose to take part in any of the roles necessary for the operations of the payment system. If any role that is essential to operations has limited access, then select groups could be positioned to gatekeep those roles and misuse their power.
An important role in a blockchain is forming new blocks. Miners perform that role in Bitcoin. They validate transactions, for example by checking that the same bitcoin is not spent more than one time and add new blocks of valid transactions to the blockchain. Because of open access, anyone can choose to be a miner and Bitcoin has a large number of them. Validation is an important function in any payment system. If the validator, or validators, cannot be trusted, they could authorize transactions that are not valid or exclude transactions that are, whenever favorable, at the cost of other participants.
Permissionless blockchains are transparent, meaning anyone can observe the ledger, know the state of accounts, and see all transactions. It is necessary that the state of the ledger be visible to everyone, so that anyone who wants to propose a new block can verify that proposed transactions are indeed feasible. Transparency also facilitates the detection of attempts to tamper with the ledger.
If governance of the blockchain is too centralized, so that a small number of individuals can make decisions that affect the entire system, then the system is more vulnerable. Bitcoin does not have a single owner. Governance is quite open as anyone can propose a change to the core code through a Bitcoin Improvement Proposal (known as a BIP). However, implementing changes to the core codebase of Bitcoin is restricted to a small number of developers, who could make changes that favor them. That said, any changes these developers make must be accepted by the miners and other nodes in the system to have influence. In this way, the participants of the network also contribute to deciding what rules are executed.
A set of incentives for good behavior
In addition to open access, permissionless blockchains require appropriate economic incentives to make sure all system participants behave in a way that supports the blockchain. In cryptocurrency projects, these incentives are achieved through the issuance and usage of a “base-layer token,” such as bitcoin (Bitcoin, with an upper case ‘B’ typically refers to the system while bitcoin, in lower case, refers to the asset).
To give miners an incentive to attempt to add new blocks to the blockchain, successful miners receive a combination of newly issued bitcoin and fees included by the sender of bitcoins. These rewards give miners the incentive to conduct their work honestly because they only retain value, and thus contribute to profits, if the block is broadly accepted.
An alternative model of trust
While both permissionlessness and economic incentives are important on their own, the interplay of both is necessary to implement a new model of trust. We illustrate the importance of this interplay in the case of Bitcoin miners.
To minimize the need to trust a single or small set of miners, Bitcoin relies on a consensus algorithm that attempts to distribute block creation. Would-be miners compete to solve a mathematical problem resulting in a solution approximately every ten minutes. Anyone can attempt to solve the problem. This mathematical problem has the property that it can only be solved by trial and error. However, once the solution is found, it is easy for anyone to verify that the solution is indeed correct. The probability of being the first to guess the answer is roughly proportional to the processing power at the disposal of the potential miner.
To understand why distributing validation reduces the need for centralized trust, imagine that you are concerned that a referee could be corrupt and affect the outcome of a sports match you care about. One way to reduce the potential impact of a corrupt referee would be to have a large group of potential available referees. Every thirty seconds or so, a new referee could be put into the game. The new referee’s officiating is reviewed by the large group of potential referees. If the large group of referees agrees with the officiating, the game continues; if not, the game is restarted at the previous state. With such a process, the impact any given referee can have on the game is very small and, so, the incentive to misbehave is also small.
To sum up
The key innovation of permissionless blockchains, such as Bitcoin, is not any specific technology. The technological and cryptographic primitives underlying Bitcoin were known for years before the cryptocurrency was created. It is the combination of open access and incentives for good behavior that allows the system to operate with a radically different trust model. This raises important questions for permissioned distributed systems, which rely on a trusted third party. These systems give up on the key innovation of Bitcoin and, so, must bring some other benefit to the table. While a number of benefits has been claimed, such as programmability or the ability to operate twenty-four hours a day, seven days a week, year-round, there is little evidence that these benefits are unique to permissioned distributed ledger systems in practice, at least so far.