Who Controls the Bitcoin Blockchain?

Introduction

The Bitcoin network is decentralized, and its governance and liveness (progression of the blockchain) are upheld by a symbiotic relationship between nodes and miners. Miners ensure the network moves forward, producing new blocks and processing valid transactions. Nodes maintain the rules of the ledger by validating all transactions and incoming blocks. Together, these parties support the decentralized blockchain through the Nakamoto Consensus.

This article explores the nuance behind “who” controls the blockchain. Historically, after the 2017 SegWit soft fork, users (nodes) maintained more control of the network, while miners were considered to have less power. However, as Bitcoin matures and control over funds becomes more centralized, new players and complexities may shift the power dynamics of the network.

Despite the changing landscape, Bitcoin was built to withstand change, an attribute that keeps the network decentralized and resistant to manipulation.  

Key Takeaways:

• Nodes collectively influence the evolution of the network, but not all nodes are equal.
• Economic nodes have greater influence over the network but without meeting the 95% acceptance threshold, changes to Bitcoin are unlikely to occur.
• If a contentious fork transpires, it could be detrimental to all parties, making the incentives and resulting game theory very important.
• A chain split poses an even greater risk as security and economic activity are effectively divided between the two chains.
• An economic majority still needs someone to trade with to give their bitcoin value. Therefore, centrally controlled economic nodes’ influence is not as strong as many might believe.
   

Nodes, Miners, and the Mempool

From a 10,000-foot view, it may seem like nodes simply uphold the network’s rules by validating blockchain data and miners create hashes directed at finding new valid blocks. However, just as every hashing machine (miner) is not equal, each node is not a direct equivalent to the other. Together, miners and nodes comprise the network, unable to survive without the other.

Nodes

Nodes are computers running Bitcoin software, connected with one another, creating the Bitcoin network. Nodes use the Gossip Protocol to relay transactions, blocks, and historical data to one another. Most nodes connect to multiple peers (other nodes) and self-regulate these connections, verifying that new and existing data complies with their current ruleset.

Nodes are the primary source of truth on the blockchain. If an invalid transaction is sent, it typically will not make it past the first node it tries to relay through. Importantly, each node can run its own local ruleset known as the node’s policy. 

A node’s policy may lead it to reject transactions that are considered valid under Bitcoin’s consensus rules. The recent debate around Knots and BIP-110 highlights this feature. Knots nodes have elected to reject transactions that contain more than 83 bytes of data in a particular part of the transaction.

However, without a consensus rule change, these transactions circumvent Knots’ BIP-110 rules by relaying transactions to non-BIP-110 nodes or by submitting transactions directly to mining nodes.¹ The blocks containing these transactions were validated by Knots’ nodes until their mandatory signaling period.²

Miners

Miners are the liveness component of Bitcoin that progresses the blockchain forward. Miners search for a random hash that satisfies the network’s next block requirement (difficulty). Each new block is “chained” to the previous one, creating an immutable history that goes back to the genesis block. 

Without miners, the blockchain stalls. New transactions sit in the mempool, never to be mined. A halted blockchain is fundamentally useless.

Miners specialize in hashing and thus rely on their own node to compile, validate, and announce new blocks.

Mempool

The mempool is the waiting room for new transactions. As illustrated in the BIP-110 example, where some nodes reject otherwise valid transactions, each node maintains its own independent local mempool. In other words, the mempool is not a singular large waiting room but a collection of thousands of customized rooms that other nodes can selectively query and relay. 

Through the Gossip Protocol, transactions propagate across the network of mempools. Eventually, they find their way to a miner’s node. Miners then typically process transactions according to the fee-to-weight ratio, ordering the timeline of transactions and maximizing their profits.³ 

Power Dynamics Between Nodes

Nodes on the network can generally be grouped into three categories. Sovereign individual nodes, economic nodes, and miner nodes. Each one introduces a different type of influence over the network. 

Even within a single category, nodes can direct different types of influence within the network. Additionally, the node categories may overlap with one another, deriving influence from multiple categories. For example, a sovereign individual or exchange (economic node) could run their own mining operation, bringing hash rate under their influence.

Sovereign Individual Node: A node owned by the “sovereign individual.” These nodes typically validate their owner’s transactions and UTXOs (unspent bitcoin). They are extremely important to these individuals but do not carry the same amount of influence when it comes to network consensus because they are not typically responsible for validating large amounts of value. 

Example: A personal node at home.

Economic Node: An economic node is a node operated by a larger entity or group of individuals. It is one that validates a relatively large amount of value. These nodes carry the influence of the bitcoin validated through them as they typically “speak” for their user base.

Example: An exchange or bitcoin custodian node.

Miner Node: The Miner Node is owned by a miner or pool of miners. These nodes validate new transactions to be mined and directly relay newly found blocks to the network. They store unconfirmed transactions waiting to be mined directly in their mempool. These nodes carry the influence of the hashing dominance associated with the miner or pool. For example, a 30% dominant miner has more influence within the network than a pool with 5%.

Example: Foundry USA mining pool.

Voting

Contrary to a common misconception, Bitcoin is not a simple “one node, one vote” system. The influence of each node can be determined by estimating the number of users and the funds it directly validates. For example, one node that validates $100 million worth of transactions per month is essentially equivalent to the hundreds of sovereign individual nodes that collectively validate a similar dollar amount. 

A non-economic node is functionally equivalent to a miner operating with little hash power. Both lack the ability to meaningfully influence the progression of the chain.

Exchanges running their own nodes command powerful economic influence because they aggregate the voices of their user base and can point them in the direction they see fit. Yet, even highly influential participants can be left behind or disbanded if they misjudge where the broader economic consensus will settle.⁴

An exchange that rejects a widely accepted upgrade may end up supporting an obsolete chain, while one that backs an unpopular change may find itself operating on an illiquid fork of the original network. Influence matters, but consensus ultimately determines which chain retains economic relevance. 

Decentralized Governance’s Natural Resistance to Change

The default state of the network is unchanging. Bitcoin is highly resistant to change because its governance is widely distributed. To avoid chain splits (a situation where two competing chain tips represent one blockchain), a threshold of implementation has been formally set at 95%.⁵ In other words, any changes should aspire to reach 95% acceptance before implementation. Anything less could lead to negative outcomes for users and the network.

David vs. Goliath

While voting power aligns with the amount of funds a specific node validates, it does not mean economic nodes hold all the power to make changes to the network. The Nakamoto Consensus is much more nuanced, and for broad consensus to be met, each node is influential. 

The Blocksize War of 2015-2017 provides a great example of the Nakamoto Consensus in action. In 2015, a debate began over how to scale the network. After two years without a resolution, the “New York Agreement” was introduced and signed by 58 companies, including exchanges, wallet companies, and mining pools, all aligned to implement a hard fork. This group represented 20.5 million wallets and roughly 83% of mining power.⁶ 

Despite the significant support from large economic- and miner-nodes, Bitcoin Core developers, along with many sovereign individual node runners, did not signal intent to upgrade to the new rules. In broad terms, it was the developers and sovereign individual nodes against a consortium of big corporations and miners (hash power).

The physical hashing power was eventually brought to heel as a relatively small group of nodes threatened to implement SegWit without the hard fork. Ultimately, the hard fork would create an incompatible blockchain not recognized by nodes that did not directly opt-in to the update. 

In other words, the New York Agreement hard fork would create a new version of Bitcoin with a new name not recognized by existing infrastructure (users and exchanges), risking a loss of network effects that had accrued to the Bitcoin network over its lifetime. A chain split poses an even greater risk as security and economic activity are effectively divided between the two chains.

From this example, it becomes clear that even the most influential participants are powerless without broad consensus, underscoring the durability of Bitcoin’s governance model and the 95% activation threshold. 

Technological Adoption

From a technical standpoint, the current rules represent the status quo. If Bitcoin Core introduced any rule changes, nodes could simply refuse to upgrade (do nothing), allowing the network to continue without changes. New versions of Bitcoin’s software require nodes to opt-in through an update, keeping major Bitcoin changes conservative, occurring only rarely and with broad agreement.

Bitcoin’s resilience ensures nodes cannot be arbitrarily governed or changed by a single party. Resistance is a fundamental part of a decentralized protocol. Ultimately, it enables the network to quell hostile economic changes without broad consensus. 

Anything less than an overwhelming 95% acceptance threshold for both soft forks and hard forks could disrupt the entire ecosystem. Therefore, improvement proposals spend years campaigning for acceptance and feedback. Any change must ultimately be recognized and adopted by the network, otherwise the proposer risks forking into a new chain.^7,8,9,10

Economic Adoption

From an economic perspective, no single holder can unilaterally redefine Bitcoin’s rules. A fork only has value if users, businesses, exchanges, and other market participants choose to recognize and support it. Therefore, even a majority holder cannot impose new rules without the support of the market (economic adoption) because a chain that nobody uses or values is economically irrelevant. 

Every network needs economic adoption as well as technological adoption. Anyone can fork the blockchain, but they cannot compel users, nodes, exchanges, or investors to assign value to that chain.

The Bitcoin network is fundamentally comprised of its users, economic nodes, and market participants. Bitcoin’s governance is distributed throughout those participants. Consistent with Metcalfe’s law, the value of broad participation increases nonlinearly as support for a proposal grows. In fact, the 95th percentile of adopters is arguably more consequential than the first 50% because of the binary nature of the threshold.

In other words, users collectively drive the direction of the network, with the influence of each additional adopter increasing as the network approaches the consensus threshold. While a small minority cannot dictate the network’s direction, the final adopters often have outsized influence as they determine whether support crosses the activation threshold.

Furthermore, a single economic node may not effectively influence other nodes lest they fork off and become irrelevant. For example, even a single entity or person that controls a sizable percentage of all 21 million bitcoin cannot direct all the other nodes. 

Calculating Influence

An important aspect of mining is the coinbase reward. The coinbase is the network’s payment to miners for submitting a new valid block. This reward cannot be spent until an additional 100 blocks are added to the chain. Therefore, miners are highly incentivized to follow a single chain tip, primarily the one they think will continue to exist. Mining a valid block on the wrong chain requires the same amount of computational input (electricity) but results in zero payment. The delayed coinbase payment keeps miners from abandoning the network in favor of an unproven chain or contentious change.

Imagine a scenario in which a small minority of users choose to implement an incompatible change to Bitcoin. The change causes a fork in the blockchain, producing “Bitcoin Legacy” and “Bitcoin Changed.” The miners are incentivized, not forced, to stay on the Bitcoin Legacy chain as they are unlikely to collect any rewards on “Bitcoin Changed” when it eventually withers out.

Alternatively, imagine a user majority led change. Whether it is a soft fork or hard fork, miners are incentivized to build on the chain most likely to succeed. However, as the probability of success converges across the chain tips, the risk associated with choosing between them increases and could lead to chaos within the ecosystem. 

The SegWit (BIP-141) soft fork provides a useful example. Miner signaling (orange) was slow at first as most miners wanted to implement a different version called SegWit2x. Note that signaling quickly climbed as miners learned they would be creating a new Bitcoin blockchain if they proceeded with a hard fork. 

The looming user activated soft fork (UASF) compelled miners to rally around the upgrade. Had the miners maintained their SegWit2x implementation (a hard fork), an economic fight between “Bitcoin Legacy” and “Bitcoin SegWit2x” would have ensued. 

For miners and users alike, diverging from the network's economic consensus could create significant disruption. Historically, chains that deviated from the economic majority and failed to attract the economic majority have generally become minority chains with significantly lower market valuations than the dominant chain.^{11,12,13,14,15}

Remember, if miners do not collectively progress a chain, the chain may stall and become worthless. Alternatively, if nodes reject blocks from miners, miners will never receive payment via the block reward. 

The dynamic between these two groups maintains a secure and decentralized governance structure. If one threatens to overrun the other, they risk forking into an obsolete version of themselves.

Unilateral Exit

The next point of contention is whether scaling introduces market-driven centralization of economic nodes. As Bitcoin scales globally, the convenience and usability advantages offered by various scaling solutions may encourage new participants to adopt derivative or custodial forms of bitcoin rather than self-custodied, on-chain bitcoin. If left unchecked, the proportion of users who directly verify, hold, and transact on Bitcoin’s base layer will fall.

Whether it be fees, usability enhancements, or network effects from bitcoin banks and exchanges, a large portion of bitcoin may eventually live behind an oligopoly of economic nodes. As users concentrate around major economic nodes, a small group of operators can effectively speak for millions, concentrating influence and economic consensus in the hands of a few institutional operators.

Unilateral exit is one of Bitcoin’s most important safeguards. Users of higher-layer systems should be able to return to the base layer without seeking approval from an intermediary. 

Unilateral exit, or the ability to withdraw funds without permission, is a cornerstone of trust-minimized Bitcoin scaling. Users can reclaim their funds to the base layer through the protocol itself rather than through the discretion of a third party. In contrast, the bitcoin held with banks and exchanges is ultimately subject to the custodian’s approval and ability to process them.

However, the ability to withdraw funds from these custodians acts as an important check on their power and influence. Users can withdraw their bitcoin from an economic actor that goes against their interests.

For example, if an exchange signals support for a protocol change or fork that its customers oppose, those customers can move their funds elsewhere, removing the economic weight and influence they contribute to that institution. In this way, users retain the ability to withdraw both their assets and their support. 

Conclusion: Bitcoin’s Consensus is an Example of Spontaneous Order

Historically, Bitcoin’s natural resistance to change has empowered its user base but did so without handing control to any single economic actor. 

There are many changing variables and incentives involved in calculating decentralized governance. The various theoretical outcomes change with each additional assumption, leading many to oversimplify governance. However, the framework that may best capture the essence of Bitcoin’s governance and consensus structure is economist Friedrich Hayek’s concept of spontaneous order.

Spontaneous order describes a system where complex coordination emerges from many individuals following local rules and incentives without a central planner.

Bitcoin was built around incentive structures that, to date, have proven remarkably resilient to manipulation and greed. In fact, self-interest is a key component to maintaining and securing the network. Node operators want to secure their bitcoin, and miners want to collect fees and block rewards. This overlapping dynamic all but guarantees that Bitcoin’s users are incentivized to decentralize the ruleset by running personal nodes and miners will continue to secure and process new transactions according to consensus rules to increase their revenue. 

Furthermore, if users cannot find someone willing to trade with, their bitcoin is worthless. This simple truth ensures that everyone acting within their best interests continues to drive value and security to a singular blockchain, Bitcoin.

Get in touch to learn more about Bitcoin and how the network’s consensus mechanism may provide value to bitcoin investors.