Decentralization by Design

By Dr. Leemon Baird & Mance Harmon

There is currently much discussion of decentralization of blockchains, and the implications for trust, control, and whether their native coins should be viewed as securities or commodities within regulatory frameworks. So it is useful to think through why decentralization is desirable, and what a useful definition of it would be.

Decentralization is foundational to blockchains and Distributed Ledger Technology (DLT) networks, yet it is often misunderstood. Regulation, public trust, and utility depend on how network influence is distributed and how transparently decisions are made.

In this paper, we will provide an overview of a decentralized network governance model that was designed to prevent malicious behavior, maintain regulatory compliance, and sustain secure operations at scale.

Purpose of decentralization

If a blockchain or online service of some kind is controlled by a central entity, then that entity has the power to shut it down at any time. That entity has the power to modify it in ways that are damaging to users. That entity has the power to secretly plan such shutdowns or malicious actions, and to benefit from the asymmetric information they have about those plans. On the other hand, if control is spread across many entities, which can only together vote for such actions, and if no single entity could do such things, then the result is a system that is more trustworthy, less likely to damage users, and less likely to create information asymmetries. This spread of control is called “decentralization”, and is critical for protecting users of such systems.

Definition of decentralization

Given this purpose, the most useful definition of decentralization is that control is not vested in any individual person or company, but is instead distributed among many actors who are unlikely to collude, and that this distribution is transparent. Control can be considered as a matter of technical implementation, or as a matter of governance. Blockchain systems generally spread control, in the technical sense, among participants in the network according to rules set by code. From the governance perspective, how the network code is updated, and who controls that process, is often far more opaque. In either the technical or governance sense, centralized control exists if a single actor can unilaterally shut down the network, change the ledger, or otherwise act maliciously to their own benefit and the detriment of all other users of the network. Such centralized control should be prevented by the system itself, rather than just relying on the goodwill or legal obligations of any single actor. Decentralization must also be transparent. To truly be decentralized, it isn’t enough to prevent a single actor from taking control; it must also be possible to verify that no single actor can take control.

This was explained well in the paper “Defining decentralization: it comes down to control” [1]. That paper argues that the essence that distinguishes a digital asset from being a commodity or a security is not whether there are ongoing efforts to develop the asset it represents, but whether control of it is centralized or decentralized. For example, a token issued by Amazon allowing access to AWS cloud services might qualify as a security, because a single company, Amazon, could decide to shut down AWS at any time. But bitcoin would be more of a commodity, because there is no single entity that can shut the network down or modify it. It is, instead, governed by the collective actions of a large group of independent individuals.

Setting the benchmark

As the industry matures, it must raise the bar for what constitutes “decentralized”. It is no longer enough to eliminate single points of failure. Networks must actively demonstrate governance that is transparent, resilient, and resistant to capture. This requires not just decentralized infrastructure, but accountable, verifiable processes. Hedera illustrates how these principles can be implemented, achieving decentralization with clear, responsible governance.

Hedera’s decentralized governance

The Hedera network is an example of decentralized governance of a blockchain or DLT. It is currently governed by more than 30 independent organizations, including Fortune 500 companies, top-ranked universities, and leading web3 companies. They act and vote independently to govern a council and the Hedera network, rather than delegating decisions to a single actor. Each of them is well known, with a reputation to protect. They are all independent. And by governance design, each council member always has an equal vote. The companies are in different industry sectors, the universities have different specialties, and they are located in different countries and jurisdictions, spread across every continent except Antarctica.

Control of the network is exercised through on-chain actions of the Council Members that is visible to the world for all code updates, changes to pricing, additions or removals of new voting members, additions or removals of new nodes, and transfers of treasury. Each such action happens with an on-chain signature (representing a vote in favor of the transaction) that is visible to the world, and is forever recorded immutably [2]. Attacks outside such voting would require collusion. But to implement any planned collusion, the malicious actors would have to vote in favor of the malicious plan through an on-chain signature, rendering any collusion between two such entities visible almost immediately, destroying their reputation. That makes it unlikely. It is even more unlikely for collusion to occur among one third of them. And without collusion of at least one third, it is impossible to shut down the network, or violate immutability, or misbehave, such as by allowing double spends.

Technical discussions for features to implement (Hedera Improvement Proposals, HIPs) are published in real time on GitHub, similar to how it is done for other blockchains, such as Ethereum or Bitcoin, with participation in those discussions coordinated through Linux Foundation Decentralized Trust — an independent organization. In addition, the offline discussions, such as for higher-level governance matters, have their minutes published [3], which is not done by those blockchains. This is extremely decentralized governance, with more transparency than is common in most blockchains.

Hedera’s decentralized network

Hedera also has decentralized the deeper technical aspects of the network. The network of nodes (computers) is also decentralized. Nodes are run by some council members, and no individual can shut them all down, or modify all of them. In addition, these entities have always voted to admit only nodes that are spread across many different systems. This is necessary to be truly decentralized. For example, a blockchain consisting of only nodes running as AWS instances would not be truly decentralized, because a single company, Amazon, would have the technical ability to shut them down, or to install malicious code on them. There might be a contract making it illegal, but it would be physically possible. That is why network decentralization requires that the nodes be spread out among a diverse range of infrastructure, including cloud service providers.

Similarly, it would not be decentralized if the majority of nodes were anonymous nodes that were all secretly being run by a single actor, such as North Korea, which would allow it to control the network. It would also not be decentralized if nodes could only join when permissioned by a single actor. The largest bitcoin mining pools each have centralized permissioning to join the pool, by banning anonymous nodes, requiring KYC, and allowing them to join only when the centralized pool allows it. Hedera currently has neither anonymous nodes nor centralized permissioning. Its nodes would fall under both the Decentralization Research Center’s definition of “permissionless nodes” [4] and a16z’s definition [5], because the voters are not under common control. This could also be called “decentralized-permissioned nodes”. New nodes are added when approved by the on-chain vote of these independent organizations, rather than by any single actor. The voters are the Council members, and they currently agree to only allow nodes to join that are known, are spread out, and are each the responsibility of a council member.

In addition, all of the code is open source, as project Hiero, under the Linux Foundation Decentralized Trust. Hedera is the first public layer 1 blockchain or DLT to contribute its entire codebase to a neutral, third-party foundation. This also helps to increase trust and transparency in the code development process. Moreover, code updates are ultimately implemented on the Hedera network only after an on-chain vote to approve them. In all these ways, Hedera is decentralized and transparent.

Building a trusted future

Decentralization is the foundation that allows users to trust blockchains and DLTs. There must be decentralization of both the hardware and the governance. This must be structural, verifiable, and resilient, by design. Developers need to build on technology that is trustworthy because it is decentralized. Enterprises need network governance that is decentralized, transparent, and verifiable. Policymakers have a role to play in recognizing responsible decentralization as the cornerstone of digital trust. Only by holding ourselves to these standards can we achieve the future of the open, transparent, and trustworthy digital world.

[1] Miles Jennings, a16zCrypto, Andreeson Horowitz, Defining decentralization: It comes down to control, https://a16zcrypto.com/posts/article/defining-decentralization-control/ (Februrary 13, 2025).

[2] Each operation such as changing the code, changing prices, adding/removing nodes, moving treasury, or adding/removing signers is accomplished by a transaction, with multilevel multisig. The transaction must be approved by a majority of the council members, where each approval must be in the form chosen by each organization. That is typically a 1-of-3 signature of any 1 of the 3 representatives that the organization chooses. These transactions, with their approval signatures, are visible through chain explorers such as any of the mirror nodes or https://hashscan.io.

[3] https://hedera.com/council/minutes

[4] Decentralization Research Center, Designing Policy for a Flourishing Blockchain Industry, https://thedrcenter.org/wp-content/uploads/2025/02/DRC-Designing-Policy-Final.pdf, (February 2025).

[5] Andreesen Horowitz, SEC Crypto Task Force Written Input, https://www.sec.gov/files/ctf-input-andreesen-horowitz-2025-03-13.pdf (March 13, 2025).

Decentralization by Design was originally published in Hedera Blog on Medium, where people are continuing the conversation by highlighting and responding to this story.