Evidentiary Infrastructure Lowers the Cost of Trusted Relationships

Commercial digital-timestamping services have operated continuously since the mid-1990s, providing tamper-evident proofs of when electronic documents existed and in what form. The underlying cryptographic architecture was published in 1991 by Stuart Haber and W. Scott Stornetta at Bellcore in a paper titled How to Time-Stamp a Digital Document in the Journal of Cryptology.

The architecture hashed client documents into a growing chain of cryptographic commitments, making it computationally infeasible for any party, including the timestamping service itself, to alter the apparent creation date of a document. This was evidentiary infrastructure as a commercial product.

Three decades before public-ledger architectures recast similar primitives as instruments for trustless economic relationships, the same underlying mechanism was being sold to corporate clients. It was marketed as a way to make existing records more legally defensible in dispute, audit, and litigation contexts.

The conventional crypto-native phrase, 'don't trust, verify,' frames cryptographic verification as a replacement for trust. Most commercial and institutional relationships continue to function differently, however. They run on trust supported by reputation, repeat interactions, courts, and insurance, with verification operating as a cheaper supplement.

Evidentiary infrastructure, in this configuration, lowers the joint cost of maintaining trusted relationships.

Two Proverbs and Two Poles

The historical reference point for trust-plus-verification regimes is the Intermediate-Range Nuclear Forces Treaty signed in 1987 by Ronald Reagan and Mikhail Gorbachev. At the signing ceremony, Reagan recited a Russian proverb in English: trust, but verify.

The line, as discussed by the Royal United Services Institute, captured a verification regime that was sampling-based and intrusive but never total. Inspectors visited declared facilities, counted weapons, and used the inspection regime to deter cheating rather than to demonstrate every transition.

The crypto-native phrase 'don't trust, verify' is the deliberate inversion. Early public-ledger architectures expressed this by treating trust as the variable to be minimized, publishing and replaying every state transition for every participant. The subsequent development of privacy networks and zero-knowledge virtual machines has since complicated this picture, demonstrating that verification can be decoupled from full transparency, but the foundational logic remains the substitution of cryptographic proof for interpersonal trust.

The cost of these early, fully transparent verification regimes included the loss of privacy at the transaction level, the operational overhead of maintaining a fully replicated ledger, and the difficulty of evolving the system except through social consensus among holders. The two phrases mark the spectrum's poles.

The first treats trust as the substrate and verification as audit. The second treats verification as the substrate, minimizing reliance on interpersonal trust. Most of the commercial economy operates closer to the first pole.

The literature on transaction-cost economics, developed primarily by Oliver Williamson at the University of California, Berkeley, provides the analytical framework for understanding why.

The Joint-Cost Mechanism

Williamson's work on the economic institutions of capitalism, published in 1985 and recognized in 2009 by the Nobel Memorial Prize in Economic Sciences, argues that economic organization is shaped by the cost of contracting. Every commercial transaction carries costs that arise from negotiating, monitoring, and enforcing the agreement.

These costs determine how economic relationships are structured. Where transaction costs are high, firms tend to integrate vertically or use richer governance mechanisms. Where transaction costs are low, market transactions predominate.

Evidentiary infrastructure operates on this cost dimension across three channels. In the ex-ante phase, the existence of an audit trail reduces the cost of negotiating agreements because both parties can anticipate cheaper dispute resolution.

During the interim of the relationship, the infrastructure lowers monitoring costs by providing a continuously available record of system state. In the ex-post phase, when disagreements arise, the resolution path through litigation, regulatory examination, or contractual arbitration becomes faster and less expensive for each party.

The joint-cost framing matters because it explains which commercial verification arrangements survive over time. Where the cost of recordkeeping is bilateral, with both parties bearing infrastructure costs and capturing dispute-cost reductions, the arrangement is durable.

Where the cost is one-sided, with one party bearing the burden while the other captures the benefit, the arrangement is unstable and eventually re-negotiated or abandoned. The distinction predicts which evidentiary regimes scale and which remain niche, and the prediction is consistent with the regulated categories in which evidentiary infrastructure has become a standard commercial offering.

A Mature Commercial Category

Evidentiary infrastructure is a long-established category in regulated commerce, with commercial logic that predates and stands apart from distributed-ledger applications of the same primitives. SEC Rule 17a-4, the broker-dealer recordkeeping regulation under the Securities Exchange Act of 1934, has required electronic records to be preserved in tamper-evident form since 1997.

In 2022, as the SEC documented, the rule was amended to allow either a write-once, read-many format or an audit-trail alternative. This alternative permits the recreation of an original record if it is altered, overwritten, or deleted.

The same architecture appears across other regulatory frameworks. The HIPAA Security Rule at 45 CFR §164.312(b) requires covered entities to implement audit controls that record and examine activity in systems containing electronic protected health information.

The Food and Drug Administration's Drug Supply Chain Security Act, enacted in 2013, mandates package-level traceability and interoperable electronic data exchange among pharmaceutical trading partners. FDA's 21 CFR Part 11, in effect since 1997, governs electronic records and signatures across pharmaceutical and medical device manufacturing operations.

Underpinning many of these regimes is a single technical standard. IETF RFC 3161, published in 2001, specifies the Time-Stamp Protocol used by trusted timestamping authorities to provide proof of existence for digital data at a specific time.

The protocol is the basis for commercial timestamping services that serve litigation discovery, patent priority disputes, financial recordkeeping requirements, and other applications where the evidentiary value of an electronic record depends on its temporal anchor.

Section 404 of the Sarbanes-Oxley Act of 2002 reaches the same architecture from a different direction. As described by the Public Company Accounting Oversight Board, Section 404 requires every public company to issue an annual report on the effectiveness of its internal controls over financial reporting, attested by an independent auditor.

The internal-controls regime that emerged in the wake of Sarbanes-Oxley produced extensive demand for documentation, audit trails, and tamper-evident recordkeeping across corporate finance functions, in a form that the auditor could test and verify against the underlying records.

Federal records management follows a parallel pattern. The National Archives and Records Administration publishes Universal Electronic Records Management Requirements that identify baseline business needs for managing electronic records across federal agencies, derived from existing statutes and NARA regulations.

Federal agencies have been required to transition to fully electronic recordkeeping, with vendors self-certifying to the Universal ERM Requirements as a condition of inclusion on the General Services Administration schedule. The architecture across these regimes shares the same general structure: tamper-evident retention, audit-trail reconstruction, defined retention periods, and inspectability by external authority.

The broader category includes commercial cloud append-only ledger products, audit-log services integrated with major enterprise software platforms, and standalone notarization services. None of these regimes invoke trustlessness as a design goal; rather, they treat verification as a supplement to existing trust mechanisms.

Each is a commercial product designed to lower the cost of dispute resolution, regulatory examination, or litigation exposure in a relationship that already runs on named counterparties, reputation, and contractual recourse. The categories grew through the 1990s and 2000s as standalone commercial businesses, anchored by regulatory mandates in some industries and by litigation-defensibility demand in others.

The Notarization Lineage

The intellectual foundation of these systems is the Haber and Stornetta paper from 1991, which proposed computationally practical procedures for digital timestamping that made backdating and forward-dating infeasible even with the collusion of the timestamping service itself. The architecture used cryptographic hash chains and required no trust in any single party, including the timestamping authority.

The paper preceded the public-ledger architectures that would later adapt these primitives for consensus. The conceptual problem Haber and Stornetta addressed was notarization: verifying that a document existed in a specific form at a specific time, rather than coordinating among pseudonymous parties without recourse to a central authority.

The commercial implementations of the protocol, beginning in the mid-1990s, served traditional notarization clients who had named counterparties and reputational stakes in the underlying records. The chronological priority matters because it places digital evidentiary infrastructure within a longer commercial-notarization lineage.

The work of certifying that something was the case at a specific time is older than computers, and was well-served by various pre-digital arrangements including notaries public, registered mail, and corporate-records affidavits. The 1991 paper digitized that work without changing its purpose.

The architectures derived from this lineage continue to function as notarization tools, with cryptography replacing the wax seal and the courthouse clerk.

Boundary Conditions

The trust-primary framing holds wherever the rider's cost is bilateral and both parties capture benefits from cheaper dispute resolution. The thesis fails in three identifiable cases, each representing a different failure mode of the joint-cost calculation.

The first case is one-sided cost. Some evidentiary infrastructure burdens one party while delivering most of the benefit to the other. Workplace monitoring systems often fit this pattern, with the employer capturing dispute-resolution benefits while the employee bears the privacy and compliance burden.

Supplier-side audit requirements imposed by large buyers can have a similar structure, in which the supplier funds the recordkeeping infrastructure and the buyer benefits from cheaper procurement review. Where the cost-benefit allocation is one-sided, the verification regime is sustained by power asymmetry rather than by joint efficiency, and it is unstable when that asymmetry shifts.

The second case is signaling distrust. Introducing formal verification into a relationship that previously ran on reputation can itself communicate a degraded view of the counterparty. This effect appears in some commercial relationships where requesting documentation or audit access disrupts established norms.

The cost of the verification regime in such relationships includes the relational cost of having introduced it, which in close commercial networks can exceed any dispute-resolution savings the infrastructure would have produced.

The third case is surveillance asset. Recorded evidence intended for dispute resolution can be repurposed for surveillance, behavior monitoring, or competitive intelligence. The same audit log that lowers the cost of compliance examination can become a tool for managerial oversight or competitive analysis that the relationship never bargained for.

Where the recorded data carries surveillance value to one party, the verification infrastructure becomes asymmetric in a different sense, distorting the joint-cost analysis even when the immediate dispute-resolution function appears bilateral.

The Distributional Question

The analytical core of the thesis is therefore distributional. Which commercial and institutional relationships sit in the verification-justified class is an empirical question that varies by industry and by relationship type.

Regulated industries where examination is recurring (financial services, healthcare, pharmaceuticals, food safety) tend to make the joint-cost math straightforward because both parties anticipate the audit and price it into the relationship. High-stakes commercial relationships with named counterparties and substantial defection payoffs benefit similarly.

Many smaller commercial relationships, particularly those running on repeat-game reputation in dense local markets, derive minimal benefit from formal verification and can be destabilized by it. The criteria that determine which class a relationship falls into include the presence of regulatory examination, the magnitude of defection payoff, the cost of dispute resolution in the absence of records, whether the counterparties are named or pseudonymous, and whether the parties expect repeated future interactions.

The continuing growth of evidentiary infrastructure as a commercial category, evident in the SEC's 2022 modernization, in regulatory expansion in pharmaceuticals and healthcare, and in the broader commercial timestamping and audit-log markets, suggests that the verification-justified class is expanding. The growth proceeds along the commercial logic of joint-cost reduction in trusted relationships, with regulatory mandates accelerating adoption in industries where the cost-benefit math was already favorable.

Two questions follow from this. The first concerns adjacent commercial categories that are not yet under regulatory mandate but might benefit from the same joint-cost analysis. Construction documentation, contract manufacturing quality records, environmental compliance attestations, and insurance underwriting evidence all share the structural features that make evidentiary infrastructure attractive: named counterparties, recurring disputes, and litigation exposure of meaningful magnitude.

The second concerns the technical form that infrastructure takes as it expands. The 2022 SEC modernization, by permitting an audit-trail alternative to the older write-once, read-many standard, suggests a regulatory willingness to accept less rigid implementations as long as the recreation property is preserved.

The architectural space between formal immutability and reliable reconstruction is the practical territory in which most expansion will occur. The design choices in that space will shape whether the infrastructure remains bilaterally cost-reducing or drifts toward the one-sided, signaling, or surveillance failure modes identified above.

Sources

• Haber, Stuart, and W. Scott Stornetta. "How to Time-Stamp a Digital Document." Journal of Cryptology, Springer, 1991.
• Dolzikova, Darya. "The Role of Verification in the Intermediate-Range Nuclear Forces Treaty Dispute." Royal United Services Institute, 2019.
• Williamson, Oliver E. "The Economic Institutions of Capitalism: Firms, Markets, Relational Contracting." Free Press; abstract via Social Science Research Network, 1985.
• United States Securities and Exchange Commission. "Amendments to Electronic Recordkeeping Requirements for Broker-Dealers." United States Securities and Exchange Commission, 2022.
• Legal Information Institute, Cornell Law School. "45 CFR §164.312: Technical Safeguards (HIPAA Security Rule)." Cornell Law School, 2003.
• United States Food and Drug Administration. "Drug Supply Chain Security Act (DSCSA)." United States Food and Drug Administration, 2013.
• Adams, C., P. Cain, D. Pinkas, and R. Zuccherato. "Internet X.509 Public Key Infrastructure Time-Stamp Protocol (TSP), RFC 3161." Internet Engineering Task Force, 2001.
• Public Company Accounting Oversight Board. "The Costs and Benefits of Sarbanes-Oxley Section 404." Public Company Accounting Oversight Board, 2005.
• National Archives and Records Administration. "Universal Electronic Records Management Requirements." National Archives and Records Administration, 2023.