Public Blockchains: Open but Costly
Public blockchains, Ethereum, Bitcoin, Solana, are open networks where anyone can participate as a node, submit transactions, and read the ledger. There is no gatekeeper deciding who can join. Security comes from economic incentives: validators stake capital or spend computational resources, and the protocol rewards honest behavior while punishing cheating.
The strengths are genuine. Public chains offer censorship resistance, no government or corporation can unilaterally shut down the network. They provide the strongest form of data integrity because the cost of attacking the network scales with its size. And they allow permissionless innovation, anyone can build applications on top without asking for access.
But for enterprises, the drawbacks are significant. Transaction data is visible to everyone by default. Throughput is limited, Ethereum processes roughly 15 to 30 transactions per second on the base layer, compared to Visa’s 65,000. Transaction costs are unpredictable; during peak demand in 2021, Ethereum gas fees reached over $100 per transaction. And regulatory compliance becomes complicated when your data lives on a network operated by anonymous validators in unknown jurisdictions.
Private Blockchains: Controlled but Centralized
Private blockchains are operated by a single organization. Only authorized participants can read or write data. The organization controls who runs nodes, who can submit transactions, and what the consensus rules are. Hyperledger Fabric deployments within a single enterprise are the most common example.
Performance is dramatically better. Without the overhead of public consensus mechanisms, private chains can handle thousands of transactions per second. Data privacy is straightforward, the operator controls access. Regulatory compliance is simpler because the data stays within known jurisdictions and known entities.
The uncomfortable question that follows is: why use a blockchain at all? If one organization controls every node, controls membership, and controls the rules, you have effectively built a replicated database with extra complexity. The immutability guarantees are only as strong as the controlling organization’s willingness to enforce them. Nothing stops the operator from rewriting history if they control all the nodes.
Private blockchains make sense in narrow scenarios, when a single organization has multiple divisions that need tamper-evident shared records, or when regulatory requirements mandate an immutable audit trail that the blockchain structure naturally provides. But calling them “blockchains” can be misleading about the trust properties they actually deliver.
Consortium Blockchains: The Middle Ground
Consortium blockchains are where most enterprise action is happening. A defined group of organizations, a banking consortium, a supply chain network, an industry association, jointly operates the network. No single party controls it. Membership is permissioned, but governance is distributed among the participants.
This model preserves the core value proposition of blockchain: shared, tamper-resistant records between parties who do not fully trust each other. Unlike a private chain, no single participant can unilaterally alter the ledger. Unlike a public chain, participants are known entities bound by legal agreements, which simplifies compliance and enables much higher throughput.
R3 Corda was explicitly designed for this model. Its architecture is optimized for point-to-point data sharing within a consortium, where not all participants need to see all transactions. Only the parties involved in a specific transaction validate and store it. This addresses a major concern in industries like banking, where institutions cannot share customer data with competitors even on a shared ledger.
Comparing What Actually Matters
When evaluating these three models, the marketing materials will give you comparison tables with checkmarks. Here is what actually matters in practice.
Data confidentiality. If your transactions contain sensitive business data, pricing, quantities, customer information, public chains are problematic. Even with encryption, metadata leaks information. Private channels on Hyperledger Fabric or Corda’s point-to-point architecture handle this. Public chains require layer-2 solutions or zero-knowledge proofs, which add complexity.
Governance. Who decides on software upgrades, dispute resolution, and membership changes? On a public chain, governance is community-driven and slow. The Ethereum DAO hack in 2016 led to a contentious hard fork that split the community. On a consortium chain, governance is defined by a legal framework that all members sign. Messy, political, but structured. On a private chain, the operator decides, simple but potentially arbitrary.
Regulatory fit. Regulated industries, healthcare, financial services, government, face data residency requirements, right-to-deletion mandates under GDPR, and KYC/AML obligations. Consortium and private chains can be designed to comply with these requirements. Public chains, where data replication is uncontrollable and deletion is architecturally impossible, create genuine compliance challenges.
Cost predictability. Consortium and private chains have fixed operational costs, server infrastructure, maintenance, and development. Public chain costs fluctuate with network demand. For budgeting purposes, telling your CFO that transaction costs could triple next month because of an NFT drop on the same network is a hard sell.
The Hybrid Reality
In practice, the lines are blurring. Projects like Polygon, Avalanche subnets, and Hyperledger Besu allow organizations to run permissioned networks that can anchor their state to a public chain for additional security guarantees. You get the performance and privacy of a consortium chain with the censorship resistance of a public chain as a backup.
Ernst & Young’s Nightfall protocol takes a different approach: it runs private transactions on the public Ethereum network using zero-knowledge proofs. Participants get the public chain’s security without exposing their transaction details. It is technically elegant, though operational complexity is higher than a pure consortium deployment.
The right choice depends on your specific trust requirements, regulatory environment, and the power dynamics between participants. If one organization dominates the ecosystem, they will push for private. If competitors need to collaborate, a consortium is usually the answer. If censorship resistance is a genuine requirement, not just a nice-to-have, public chains are the only option. Most enterprise projects land on consortium, and for good reason.