M-19-21 closed the archive’s door to paper and left open the processes that made it
OMB and NARA’s June 2019 memo set dates for when the National Archives would stop taking paper. It only encouraged changes to the forms, signatures and mail that kept producing paper. The deadline later moved, and the hard work stayed upstream.
Looking Back revisits past weeks one at a time, up to the most recent, and asks what it meant for the banks, health systems and public agencies we serve. The series was written in 2026, with the benefit of hindsight, and each article says which week it looks back on.
Two-minute summaryThe dek and the first sentence of each section, quoted from the article
OMB and NARA’s June 2019 memo set dates for when the National Archives would stop taking paper. It only encouraged changes to the forms, signatures and mail that kept producing paper. The deadline later moved, and the hard work stayed upstream.
- The week On Friday, June 28, 2019, the Office of Management and Budget and the National Archives and Records Administration issued a joint memorandum to the heads of executive departments and agencies, M-19-21, Transition to Electronic Records.
- The intake gate The memo’s firmest commitment was about NARA itself.
- The storage exit The second directive was concrete: close agency-operated records storage facilities and move inactive temporary records to Federal Records Centers or commercial storage that meets NARA’s requirements.
- The paper sources Paper in government does not appear on its own.
- The proof M-19-21 required that permanent records be managed and transferred with appropriate metadata.
- Objections, and where the argument is weakest One objection holds that M-19-21 did reach business processes, because it encouraged their transition and tied itself to a management agenda about electronic government.
- What it meant by type of organization Departments with large field networks and public-facing programs could expect many paper sources and large backlogs.
- With hindsight The 2022 deadline did not hold.
The week
On Friday, June 28, 2019, the Office of Management and Budget and the National Archives and Records Administration issued a joint memorandum to the heads of executive departments and agencies, M-19-21, Transition to Electronic Records. It was signed by Russell T. Vought, then Acting Director of OMB, and David S. Ferriero, then Archivist of the United States. It rescinded two earlier memoranda, M-12-18 on managing government records and M-14-16 on managing email, and consolidated their requirements.
The memo placed itself inside the President’s Management Agenda of March 2018 and the June 2018 government reform plan. That plan had proposed moving federal business processes and recordkeeping to a fully electronic environment and ending NARA’s acceptance of paper records by December 31, 2022. M-19-21 set out the steps to meet that goal. It said the government spent hundreds of millions of dollars and thousands of hours a year creating, using and storing records in analog formats, and that paper processes burdened the public and trapped data where it could only be extracted manually.
The memo directed agencies to do two things: ensure that all federal records are created, retained and managed in electronic formats with appropriate metadata, and plan to close agency-operated storage facilities for paper and other analog records, moving those records to NARA’s Federal Records Centers or to commercial storage. It applied to all agencies, whether or not covered by the Chief Financial Officers Act. It also noted that agencies had been required to manage all email records electronically since 2016 and were expected to continue doing so.
The deadlines were specific. By December 31, 2019, agencies were to manage all permanent electronic records electronically to the fullest extent possible. By December 31, 2022, they were to manage all permanent records electronically with appropriate metadata, manage temporary records electronically or store them in commercial facilities, and close their own records storage facilities. After December 31, 2022, NARA would no longer accept new transfers of analog records to the fullest extent possible. From January 1, 2023, agencies would have to digitize permanent analog records before transferring them. Agencies also had to keep a Senior Agency Official for Records Management at the Assistant Secretary level or equivalent and an Agency Records Officer holding NARA’s certificate in federal records management.
The memo gave NARA until September 30, 2020 to update its regulations and guidance on electronic recordkeeping, digitization and transfer, and to set up a process for exceptions. It gave the Office of Personnel Management until December 31, 2020 to update position classification standards for archivists, archives technicians and records and information management staff to include electronic records work. The memo also required agencies to inform all personnel annually of their records management responsibilities, to make sure every record was covered by a NARA-approved schedule, and to update schedules as business practices moved to electronic workflows. One more detail needed correcting: the memo referred to an exception process under a paragraph 2.2 that did not exist. NARA’s September 2020 guidance called this a numbering error and said the content sat in the final paragraph of section 2.1.
M-19-21 is easy to read as a technology mandate: go paperless by 2022. Read closely, it was a records management mandate with a much narrower set of levers. It controlled what NARA would accept, what format records had to be in when they were transferred, and where inactive records could be stored. It did not control the business processes that create paper in the first place. On those, the memo encouraged agencies to consider cost-effective opportunities to transition related processes to an electronic environment. Encouragement is not a deadline.
Our argument is that M-19-21 closed the archive’s door to paper while leaving open the processes that kept producing it, and that this mismatch may help explain why the deadline moved and where the remaining work ended up. We break the memo’s effect into four parts that do not overlap and together cover it. First, the intake gate: NARA’s acceptance of records and the format of transfers. Second, the storage exit: the closure of agency-run storage and the move of inactive paper to other facilities. Third, the paper sources: the forms, signatures and public channels that generate paper every day. Fourth, the proof: the metadata, digitization standards and validation that let an agency show its electronic records are complete and authentic. Two further sections test the argument against objections and look at what it meant for different kinds of organization.
The memo bound federal agencies under the Federal Records Act. NARA later confirmed that its requirements apply to federal records that contractors create or receive. State and local governments were not covered, though many buy from the same vendors and face similar choices. Agencies should check their own obligations with counsel, especially where a statute or regulation requires a paper original.
The intake gate
A deadline set at the end of the life cycle
The memo’s firmest commitment was about NARA itself. After the deadline, NARA would stop accepting new analog transfers, to the fullest extent possible, and permanent records would have to arrive in electronic form with metadata. The lever acts at the very end of a record’s life cycle. A permanent record may be created decades before it is transferred to the National Archives. Changing the format of transfers in 2023 does little to change how a record created in 2019 is captured, unless the agency also changes the capture.
The gate still mattered, because it changed the cost of doing nothing. Before M-19-21, an agency could keep producing permanent paper records and transfer them to NARA as paper when they came due. After the gate closed, any permanent paper record still in agency hands would have to be digitized to NARA’s standards before transfer. That converts a routine paper transfer into a future, paid digitization project. Agencies that understood this had an incentive to stop producing permanent paper, or to transfer existing paper before the gate closed.
The deadline as a budget signal
Deadlines in federal management memos can also work through budgets. A dated requirement gives a chief information officer or a chief financial officer a reason to fund work that would otherwise lose to more visible priorities. M-19-21 gave records management a date and tied it to the President’s Management Agenda. For agencies that used it, the memo was an argument in budget formulation: here is a cost we will face if we do not digitize at the source, and here is the date it starts.
The signal was strongest for the records office and the storage exit, where the work was clearly scoped. It was weaker for process redesign, because process redesign competes with mission spending inside program budgets, and the memo set no date for it. An agency budget office reading the memo literally would fund boxes and scanners before it funded new forms and signature systems. Leaders who wanted the second outcome had to make that case themselves.
Records already in the pipeline
The memo protected records already in the federal system. Permanent records accessioned into NARA or transferred to Federal Records Centers before the deadline did not need to be digitized. NARA would continue to store and service them until their scheduled disposition date, and then accept the permanent ones in their original analog format. That created a rational strategy for agencies with large paper backlogs: move eligible paper into the Federal Records Centers before the deadline, and let it age out under existing rules.
NARA later said as much. When it issued its digitization standards in 2023, it responded to commenters worried that accepting only digital records would delay transfers by saying it expected agencies to transfer overdue permanent paper records rather than pay to digitize them. It noted that M-19-21 itself allowed agencies to transfer permanent paper to the Federal Records Centers before the deadlines. In other settings that might be called a loophole. Here it was the design: clear the backlog through the existing door, then keep new paper from accumulating behind it.
What “to the fullest extent possible” did
The phrase recurs through the memo. NARA’s September 2020 bulletin explained that it comes from 44 U.S.C. 2902(6), which directs the Archivist to emphasize the transfer of records in digital or electronic form to the greatest extent possible. The bulletin said the phrase recognized that agencies might need exceptions to transfer a limited number of analog records after the deadline. The gate had a side door. The question for each agency was whether its paper fit through it, which depended on the exception process.
The storage exit
Closing agency-operated facilities
The second directive was concrete: close agency-operated records storage facilities and move inactive temporary records to Federal Records Centers or commercial storage that meets NARA’s requirements. After the deadline, temporary analog records newly eligible for transfer had to go to commercial storage. This part of the memo was about real estate, contracts and logistics as much as records management. A closed storage room frees space and ends a cost. It also requires an inventory of what is in the room, a schedule for each series, and a plan for boxes that turn out to hold records no one can identify.
For agency operations staff, this is a moving project with a long tail. Boxes must be indexed before they move, because a box that arrives at a commercial facility without an index is hard to find again. Schedules must be current, because a records center needs to know when each box may be destroyed. Retrieval must be planned, because programs still need some inactive files. And the commercial facility must meet NARA’s records storage requirements, as the memo said; we would also want the contract to give the agency control over retrieval, transfer and destruction.
Transfers become a routine operation
Under a paper regime, transfers to NARA were periodic events: boxes packed, forms signed, trucks dispatched. Under an electronic regime, a transfer is a data operation. Files must be exported from the system that holds them, packaged with metadata, checked for completeness and sent through NARA’s electronic systems. That work repeats every time a series reaches its transfer date. We would expect agencies that build it as a repeatable process, with tested export routines and named owners, to spend less on each transfer than agencies that improvise each one.
Records with no schedule
Many storage rooms are likely to hold some records that no approved schedule covers. They may be from a program that ended, an office that was merged, or a series no one ever scheduled. They cannot be destroyed without authority, and moving them elsewhere without knowing what they are only moves the problem. Clearing a facility therefore forces scheduling work that may have been deferred for years. M-19-21 required that all records be covered by a NARA-approved schedule, so this was a requirement as well as a practical need.
Scheduling can take time, because each schedule needs NARA’s approval. An agency that discovers unscheduled series late in a facility closure will find its timeline set by the approval process, not by the moving company. The sensible order is to survey the facility, identify unscheduled series and submit schedules first, then plan the move around their approval.
Cost moves, it does not vanish
Moving paper out of agency space does not by itself make it cheaper to own; it makes the cost visible in a different line. Storage elsewhere still costs money, whether through a records center or a commercial contract. Commercial storage adds a procurement, a contract to manage and a vendor to oversee. For temporary records, the cost ends at disposition. For records that keep arriving because the processes that create them keep running, the cost has no end. Here the storage exit meets the paper sources: closing the room only works if fewer boxes come in.
Temporary records and digitization
The memo allowed agencies that received an exception to keep producing and storing temporary records in analog form. For everyone else, temporary records had to be managed electronically to the fullest extent possible. Digitizing temporary paper, and then destroying the paper, needs standards and authority. Both already existed in 2019: NARA’s 2023 rule records that standards for digitizing temporary records had taken effect on May 10, 2019, weeks before the memo, and NARA’s 2024 rule says digitized source records had been covered by GRS 5.2, Transitory and Intermediary Records. NARA later added a dedicated schedule, GRS 4.5 Digitizing Records, in June 2023, and a 2024 direct final rule told agencies to use it when disposing of temporary records that had been digitized. The authority is conditional. For permanent records, the 2023 rule said that agencies that do not meet its standards do not have the authority to dispose of the source records. An agency that scans without meeting the standards keeps the paper as well as the images, which doubles its storage rather than halving it.
The paper sources
Where paper comes from
Paper in government does not appear on its own. It is produced by specific steps in specific processes. NARA’s own staff, writing in 2024, named some of them: documents printed to receive a wet-ink signature, forms filled out online and then printed and mailed to an agency, and agencies that still accept comments on regulations in paper form. Each of those steps sits in a program office, a public-facing service or a legal requirement, not in the records office. M-19-21 gave the records office deadlines and gave the program offices encouragement.
That allocation explains a pattern we would expect to see in many agencies. The records officer could plan the storage exit and prepare for the intake gate. But the records officer could not redesign a benefit application, change an enforcement procedure that required a signed original, or alter a statute that required a paper filing. Those changes needed program owners, the chief information officer, counsel and sometimes Congress. Without them, the records office would keep receiving paper and keep needing to store or scan it.
Signatures and identity
The wet-ink signature is a stubborn paper source because it carries legal weight. Replacing it means deciding what electronic signature method is acceptable for each transaction, how the signer’s identity is established, and how the signature and its evidence are preserved. For internal approvals, an agency can decide alone. For signatures from the public, the agency must also consider accessibility, equity of access and the strength of identity proofing. Each choice has an owner outside records management.
NARA’s 2024 guidance on forms pointed in the same direction. It said it would likely be more cost effective to replace wet-ink signatures with electronic or digital signatures, replace printed and mailed forms with web forms, and capture data from any remaining paper forms into a recordkeeping system. Once the information is captured, NARA said, the paper form should be considered an input record and can be disposed of under GRS 5.2, Item 20. It is a records management outcome that only a business process change can deliver.
Email and messages were already electronic
One class of records had already crossed over. The memo noted that agencies had been required to manage all permanent and temporary email records electronically since 2016. Email showed what the rest of the program would need: a capture method that works without user effort, retention rules applied by role or account, and the ability to find and produce records at scale. Where those worked for email, they offered a template for other electronic records. Where they did not, they were a warning about what electronic records without discipline look like.
Messaging beyond email was the next frontier. M-23-07 later committed NARA to issue guidance, by June 30, 2023, extending its Capstone approach to all types of electronic messaging, including ephemeral and encrypted messages. That request confirms the general point. Each new channel through which officials do business is a new source of records, and each needs a capture method designed at the start rather than retrofitted.
Public channels
Some paper arrives because the public sends it. Agencies accept mailed applications, letters and comments, and some are required to. A memo about records cannot stop the mail. What an agency can do is offer an electronic channel that is easier than paper, and digitize what still arrives at the front door rather than at the archive. Scanning at intake, with the image becoming the record and the paper destroyed under an approved schedule, moves the paper problem to the start of the process, where it is cheapest to solve.
Intake scanning has its own requirements. The scanning must meet NARA’s standards if the records are permanent. The images must be indexed so they can be found. And the process must handle records with intrinsic value, which may need to be kept in original form. Those requirements are real but bounded, and they are far cheaper to meet at intake than decades later at transfer.
Funding the redesign
Process redesign costs money in places records budgets do not reach. A new electronic form needs development, accessibility testing, identity verification and a back-end system to receive it. A new signature method needs policy, legal review and technology. A new intake scanning operation needs equipment, staff and quality control. Those costs fall on program budgets and the chief information officer’s portfolio. The savings, fewer boxes and less storage, show up in the records program and the facilities budget. When costs and savings land in different budgets, a business case needs someone senior enough to see both.
The memo’s senior agency official for records management, at Assistant Secretary level or equivalent, is one candidate for that role. The official has the rank to bring program, technology and finance leaders together. Whether the role is used that way depends on the agency. In our view it is the best place in the structure the memo created to connect records deadlines with process funding.
Electronic channels must work for everyone
Moving the public to electronic channels carries its own obligations. Some people cannot use an online form easily, because of disability, connectivity, language or limited access to devices. An electronic channel should be accessible to people with disabilities, and an agency that removes a paper option must consider who loses access. NARA’s bulletin recognized this in its first exception ground: cases where replacing analog records with electronic systems would be burdensome to the public. Agencies making that case had to show the added burden in time, effort or resources.
The workable answer for many services is a mixed model. The electronic channel becomes the default, assisted channels remain for those who need them, and paper that still arrives is digitized at intake. The record is then electronic regardless of how the person chose to engage. That model keeps the public’s options open while still closing the paper stream before it reaches storage.
The exception process as a map of paper
The exception process is where the paper sources become visible. NARA’s bulletin set three broad grounds for exceptions: cases where replacing analog records would be burdensome to the public; cases where, after a comprehensive analysis, the cost would exceed the benefit; and cases where analog records cannot be replaced for another reason, including statutory or regulatory barriers. It asked agencies to submit a single request covering all the records series they believed needed exceptions, with a business case that could include digitization costs, reference costs, legal analysis and time estimates.
For a chief information officer, that request is a useful document whether or not NARA grants it. It is an inventory of where the agency still produces paper, why, and what it would cost to stop. It names the statutes and regulations that require paper, which is the list an agency would need to seek legislative or regulatory change. An agency that treated the exception request as a planning tool, rather than a compliance form, would come out with a roadmap.
The proof
Metadata as a condition of transfer
M-19-21 required that permanent records be managed and transferred with appropriate metadata. Metadata is the part of electronic recordkeeping that is easy to defer and expensive to recreate. A scanned file with no description, date, creator or series is a record in law but useless in practice. NARA’s later guidance made the consequence explicit. In January 2025, its staff reminded agencies that NARA may not approve proposed accessions that lack metadata, and that electronic finding aids must list every file in a transfer so NARA can confirm each one is received.
That requirement reaches back into the systems where records are created. If a case management system does not capture the creator, date and series of each document, the agency will have to reconstruct them at transfer. If a shared drive holds permanent records mixed with temporary ones and personal files, someone must sort them. The cheapest metadata is the metadata captured when the record is made, which means the requirement lands on system owners and application design as much as on records staff at the end.
Digitization standards and validation
Converting paper to digital records that can replace the paper requires standards. NARA issued its rule on digitizing permanent records on May 4, 2023, effective June 5, 2023. It added standards that agencies must meet to dispose of source records, and metadata requirements for transferring permanent digital records. NARA said it drew on its experience of microfilm conversion, requiring regular checks of imaging equipment, precise calibration and diligent quality control. It warned that quick and cheap digitization could produce digital records that met neither the agency’s business needs nor NARA’s archival needs.
The rule also required validation. Agencies must not dispose of source records until they validate that they have followed the digitization and quality control requirements, and they must have an applicable records schedule covering the disposal. NARA said its standards and validation requirements would help inspectors general and chief information security officers review agency programs. Digitization, on that view, is an auditable process. An agency that scans without recording what it checked, when, and against which standard cannot later prove that its digital copy is a faithful replacement.
Schedules that match the new workflows
M-19-21 told agencies to make sure every record is covered by a NARA-approved schedule, and to update schedules as business practices move to electronic workflows. The 2023 rule added a requirement to review, every five years, all schedules that are ten years old or older. Schedules written for paper describe paper files. When a process moves online, the records it creates may be different in content, volume and retention value. An out-of-date schedule leaves electronic records either kept too long, at cost, or destroyed too soon, at legal risk.
Security of the electronic record
An electronic record is only as trustworthy as the system that holds it. Proof that a record is complete and unaltered depends on access controls, audit logs, integrity checks and backups. Those are security controls, owned by the chief information security officer and system owners. NARA’s 2023 rule said its validation requirements would help chief information security officers and inspectors general review agency programs. That places records integrity inside the security program’s scope.
The practical step is to treat records repositories as systems that hold high-value information, with integrity monitoring and access review to match. A digitized permanent record that replaces destroyed paper has no paper behind it. If it is altered, lost or encrypted by an attacker, there is no paper to fall back on. Moving to electronic records shifts the risk from fire, flood and misfiling to the familiar list of cyber risks, and the security program must plan for that shift.
People who can do the work
The memo recognized that the transition needed people with different skills. It told OPM to update classification standards for archivists, archives technicians and records and information management staff to include electronic records work. It required agency records officers to hold NARA’s certificate and agencies to train all personnel annually on their records responsibilities. Those provisions are easy to overlook beside the deadlines, but they address the workforce that would carry the change.
Electronic records work draws on data management, systems configuration, metadata design and digital preservation. A records office staffed for paper may need retraining or new hires to work with system owners as peers. Annual training for all staff matters too, because in an electronic environment every employee who saves a file or sends a message creates records. The workforce provisions were the memo’s acknowledgment that records management had become a technical discipline.
Who owns the proof
The memo assigned clear roles: a senior agency official for records management at Assistant Secretary level, with direct responsibility for compliance, and an agency records officer with NARA’s certificate. Those roles own the program. They do not own the systems where records are made. Proof of an electronic record’s completeness depends on system configuration, access controls, audit logs and retention settings, which belong to the chief information officer and system owners. Effective proof needs those owners to accept records requirements as system requirements, written into design documents and contracts.
Objections, and where the argument is weakest
The memo did address business processes
One objection holds that M-19-21 did reach business processes, because it encouraged their transition and tied itself to a management agenda about electronic government. That is fair: the memo framed itself as part of a government-wide modernization agenda. But encouragement and deadlines produce different behavior. Records offices had dates; program offices had a suggestion. Our argument is about where the binding pressure fell, and it fell on the records office and on NARA.
Paper was already a small share
A second objection holds that the bulk of records were already born digital, so the remaining paper was a residual problem. NARA’s own staff wrote in 2024 that over the past thirty years most government records were initially created on computers. That supports the objection, and it sharpens our point. If records are born digital and then printed for a signature or a mailing, the paper is an artifact of process design. The fix lies in the process. A deadline at the archive does not touch it.
The pandemic, not the design, moved the deadline
A third objection holds that the delay came from COVID-19, not from any flaw in the memo. M-23-07 did cite pandemic delays when it moved the dates. In September 2020, NARA had said the 2022 targets were not imminent and did not need adjusting at that point. Both are part of the record. We do not claim to know how much of the delay each cause explains. It seems plausible to us that the pandemic disrupted most the in-person and paper-handling steps, such as wet-ink signatures and mailed submissions, that the memo had left to encouragement; we have no data that measures this. NARA’s own 2024 guidance still named those steps as places where paper was being produced.
Electronic records bring their own risks
A fourth objection would say that pushing hard on electronic records trades a known risk for an unknown one. Paper is slow and costly, but it does not vanish in a ransomware attack, and it does not depend on file formats remaining readable. That objection is serious. It argues for doing the electronic transition well, with security, format planning and validation, rather than for keeping paper. NARA’s own standards address format and quality. The security program has to address the rest, as we noted above.
Where the argument is weakest
Our argument would be weaker if the intake gate alone had changed behavior upstream. It may have, in agencies where leaders used the deadline to drive process change. The memo made that possible. We would still say the design relied on leaders choosing to do so, rather than requiring it, and that agencies without that leadership had a path to meet the letter of the memo by moving paper rather than stopping it.
What it meant by type of organization
Large departments with field operations
Departments with large field networks and public-facing programs could expect many paper sources and large backlogs. For them, the storage exit was a major logistics project and the exception request a long document. The useful sequence was to inventory paper sources by program, prioritize the ones that produce permanent records, and fund process redesign for those first. Records officers needed a seat in program modernization decisions to make that sequence happen.
Small and independent agencies
Smaller agencies typically have fewer records but also fewer staff. The memo applied to them in full. For them, shared services and government-wide contract vehicles for electronic records management were the practical route. M-23-07 later gave NARA until December 31, 2023 to finalize standards and requirements for electronic records management solutions under its Federal Electronic Records Modernization Initiative, which the memo describes as NARA’s government-wide strategy for procuring records management solutions and services. A small agency is a buyer of such solutions, not a builder.
Agencies with statutory paper requirements
Some agencies operate under statutes or regulations that require a paper original, a physical signature or a specific format. For them, the exception process was the main route, and the useful work was building the evidence for it: which provision requires paper, what it covers, and what change would remove the requirement. NARA’s bulletin asked agencies to identify the specific statute or regulation in their requests. That list is also the agenda for any legislative or regulatory proposal the agency might make to modernize the requirement.
Agencies replacing core systems
Agencies in the middle of replacing a case management, grants or benefits system had an opportunity the memo did not name. A new system is the cheapest moment to build electronic capture, metadata and retention into the design. It is also the moment when old records must be migrated or retired, which raises the question of whether they are complete and properly scheduled. An agency that wrote M-19-21’s requirements into the system’s requirements, acceptance tests and contract would get electronic records as a by-product of the modernization. An agency that left records to a later phase would face a second project, on a live system, with less money.
The same logic applies to cloud migrations and software-as-a-service adoption. When records move into a vendor’s platform, the agency must still be able to apply retention, place holds, export with metadata and transfer permanent records to NARA. Those capabilities belong in the procurement evaluation. They are hard to negotiate after award.
Contractors
Because the requirements apply to federal records that contractors create or receive, contractors that run programs or systems for agencies inherited them. NARA’s bulletin, citing 36 CFR 1222.32, said agencies should ensure that all contracts and agreements address records management obligations. For a contractor running a call center or a claims operation, that means electronic capture, metadata and transfer readiness are contract requirements, and the contract should say so in testable terms.
Banks, health systems and state agencies
Organizations outside the federal government were not bound by M-19-21. They meet similar choices under their own rules, for example retention rules from financial regulators or state records laws. The structural point carries over. A retention or archive requirement controls the end of a record’s life; the business process controls its birth. A program that funds the archive and leaves the process alone will keep paying to store, scan and search paper it did not need to create.
With hindsight
The 2022 deadline did not hold. On December 23, 2022, OMB and NARA issued M-23-07, Update to Transition to Electronic Records, signed by OMB Director Shalanda D. Young and Acting Archivist Debra Steidel Wall. It said agencies may have been delayed during the COVID-19 pandemic, reaffirmed the goal and moved the dates to June 30, 2024. By that date, agencies had to close their own storage facilities. After it, they had to transfer permanent records in electronic form with metadata and send newly eligible temporary analog records to commercial storage. From July 1, 2024, agencies had to digitize permanent analog records before transfer. M-23-07 also listed exceptional intrinsic value in the original format among the grounds for a limited exception.
NARA’s new regulations for permanent records came later than the September 30, 2020 date in M-19-21. Its rule on digitizing permanent records was published on May 4, 2023 and took effect on June 5, 2023. A rule on digitizing temporary records, incorporating the General Records Schedule for digitizing records that NARA issued in June 2023, was published on May 30, 2024, to take effect on August 28, 2024 unless adverse comment was received. M-23-07 also committed NARA to guidance extending its Capstone approach to electronic messaging, including ephemeral and encrypted messages, and for guidance on records in collaborative working environments.
In the 2023 rule, NARA also required agencies to review, every five years, all records schedules ten years old or older. Answering commenters who feared that accepting only digital records would delay transfers, it said it expected agencies to transfer overdue permanent paper records rather than pay to digitize them, noting that M-19-21 let agencies move permanent paper to the Federal Records Centers before the deadlines.
As the 2024 deadline approached, NARA’s staff wrote that the new digitization rules should not be used to continue paper-based business processes. They named wet-ink signatures, printed online forms and paper comments as examples of paper that all-digital workflows should eliminate, and encouraged agencies to transfer permanent analog records to the Federal Records Centers before June 30, 2024. In July 2024 they followed with guidance on moving forms to fully digital workflows. In January 2025 they reminded agencies that NARA may not approve accessions lacking metadata.
NARA’s 2025 reminder also showed what a compliant transfer now involves. It pointed agencies to a file-listing tool that records each file’s name, path, size, extension, date modified and a SHA-256 hash, so that every file in a transfer can be accounted for. It noted that digitized records and Capstone records need additional metadata fields. Those are system and data tasks, far from the boxes and storage rooms of 2019. We have not found a government-wide count of how many agencies met the June 30, 2024 requirements in full, so we do not state one.
What held up from June 2019 was the direction and the gate. NARA set standards for digital replacements and said that, after June 30, 2024, it would accept permanent records only in digital formats with required metadata, outside limited exceptions. What took longest was the part the memo left to encouragement. The guidance NARA was still issuing in 2024 and 2025 was about forms, signatures and metadata: the processes where records are born. That is where the 2019 memo had the least grip, and where the remaining work sat.
Spotted an error? Tell us and we will check it and correct the article where needed.