BIS Export License Report Leaves Delay Dates to Applicants
Subject to available appropriations, the first export control transparency report is due to Congress by August 19 without a required decision-date field. It reaches only the Act's narrow covered group. Exporters need their own delay ledger.
Primary lensExport controls
Sub-topicLicensing regime
Evidence base10 records used
Use caseExport-control exposure
The first report under the Maintaining American Superiority by Improving Export Control Transparency Act is due to Congress no later than August 19, 2026, subject to available appropriations. It will report on end-use checks, license applications, and other authorization requests involving a narrow class of listed entities. It is not designed as a general measure of export-license delay.
The missing field is simple. requires the date an application was submitted and the decision, but not the date of that decision. It also omits the date BIS registered a completed application, the periods excluded from agency processing time, the age of matters still pending, and the agency that controls the next step. Without those fields, a reader cannot reconstruct either the regulatory processing clock or the full commercial wait.
Exporters should not wait for the report to answer a question it was not written to answer. A live case needs its own delay ledger. That record should separate submission from registration, identify each excluded interval and interagency handoff, preserve every status contact, and place commercial events such as quote expiry, capacity release, and customer requalification on a second timeline. A long commercial wait does not by itself prove that BIS missed a legal deadline. The point of the ledger is to show what happened, who owned the next action, and which clock was still running.
The statute does not require a decision date
The statute calls the new product a transparency report, but its purpose is transaction visibility for Congress. For each covered application or other authorization request, Commerce must provide the applicant, a brief item description, any relevant ECCN and reason for control, the end user and location, a value estimate, the decision, and the submission date. End-use checks receive their own date, location, and result. Aggregate statistics are also required.
That is a useful record of what the government considered. It can show which controlled items were requested for which covered entities and how BIS resolved those requests. Yet the minimum statutory fields do not show how long a completed review took. A reported decision has no required decision date beside it. A submission date alone supplies the start of an elapsed-time question without supplying the end.
The same design leaves pending inventory unclear. The law asks for each application and a decision with respect to it, but it does not prescribe a pending status, an age band, or a reason why no decision has been reached. BIS may choose to provide more. It may publish a richer aggregate appendix or add dates that Congress did not specify. As of August 12, however, the first deadline has not arrived and the public record does not establish what voluntary fields the agency will include.
Disclosure is narrower still. Case-level fields fall under ECRA's withholding rule in 50 U.S.C. 4820(h) and are not required to be posted, although the Secretary retains national-interest release authority. Aggregate statistics are carved out of that particular rule, but the law directs the report to the House Foreign Affairs Committee and Senate Banking Committee. It does not itself direct BIS to post the report. The Secretary must also omit information that may jeopardize an ongoing investigation. A congressional report can therefore improve oversight without becoming a public queue monitor.
Coverage begins with a two-part listed-entity test
The word transparency also risks overstating coverage. Public Law 119-34 does not reach the full BIS licensing docket, every China transaction, or every application for a sensitive technology. It defines a covered entity through two conditions that must both be met.
The entity must be located in a country listed in the fifth column of Country Group D under supplement 1 to part 740. It must also appear on either the Entity List in supplement 4 to part 744 or the Military End-User List in supplement 7. An unlisted customer is outside the statutory definition even when the item, destination, or end use triggers a demanding license review under another provision of the EAR. A listed party outside that country group is also outside this report.
That boundary makes sense if Congress wants transaction-level visibility into licensing and checks for a defined set of national-security risks. It makes the report a poor denominator for the wider delay debate. A company cannot compare its entire licensing portfolio with the report's aggregate figures unless it first removes every case that fails the same two-part test. Even then, published aggregate detail may not be sufficient for a reliable comparison.
The narrow definition also prevents a common analytical error. The report's first aggregate could rise or fall because the covered lists changed, application mix changed, or the relevant transactions changed. None of those movements would by itself show that the whole licensing system became faster or slower. Coverage has to be held constant before trend claims begin.
Submission is not registration
The report's chosen date does not match the event that starts the familiar 90-day framework. Under EAR section 750.4, the 90-calendar-day period starts when BIS registers a completed application. A submission can arrive earlier. If an application is deficient, BIS may decline to register it or return it without action. A corrected resubmission is treated as a new application when it is registered.
This distinction is not clerical. Suppose a seller submits through SNAP-R, answers a completeness question two weeks later, and then receives a registration notice. The new statutory report requires the first date. The regulatory count uses the later event. The seller's customer may have started waiting before either one, when the order became contingent on authorization.
Processing time can also exclude real elapsed days. Section 750.4 identifies periods for applicant-agreed delay, additional information, pre-license checks, government-to-government assurances, consultations, multilateral review, congressional notification, and specified satellite approvals. Executive Order 12981 supplies the interagency review and escalation architecture behind that rule.
Those exclusions do not mean nothing is happening. A pre-license check may be necessary to test a proposed end user. A foreign-government assurance can be material to the license judgment. Interagency consultation can expose a risk that the original submission did not resolve. The legal framework removes specified intervals from processing time because the case is not in an ordinary agency-decision posture. The customer's wait continues all the same.
The result is two valid but different measurements. BIS can report a processing interval under its rules. A seller can report the time during which revenue, production capacity, or a customer relationship remained contingent. Neither figure refutes the other. Trouble begins when one is presented as if it answered both questions.
Substantive transaction diligence remains a separate file. Traverse's BIS foundry due diligence analysis shows why item and party evidence can be complete or incomplete independently of queue age. A well-aged case is not necessarily a well-supported application.
The 62-day average leaves live cases outside the denominator
The BIS FY 2025 Annual Report says the agency processed approximately 30,500 license applications in calendar year 2025 with an average processing time of 62 days. BIS describes that result as below a 90-day benchmark and notes that the year included a 43-day government shutdown.
The figure is the best current official headline measure, but its denominator matters. It describes applications processed during the year. It does not state the age distribution of applications that remained pending when the measurement period closed. A small group of old unresolved matters can coexist with a lower average for completed work because those live matters have not entered the completed-case calculation.
Average also compresses the shape of the completed cohort. It does not reveal the median, the 90th percentile, destination, ECCN, reviewing-agency path, excluded time, or how returned-without-action matters were timed. The annual report is candid about the reported measure. The mistake would be to turn its 62 days into an individual service promise.
That is why the August 19 report does not close the measurement gap merely by adding transactions. Its required data still lack the decision date needed to calculate a completed-case interval. They lack a status and age field for live inventory. They use submission, while the EAR clock starts at registration. More rows without the right dates create a larger transaction record, not a delay distribution.
Transaction volume does not identify the bottleneck
GAO-25-107431 separately found that BIS workload metrics did not capture the time or staffing required for the work. Reviewing agencies also lacked ready access to all relevant information spread across unclassified and classified systems. Officials told GAO they had sometimes recommended denial to trigger dispute resolution, where more information became available, which could increase process time.
Those findings explain why a transaction count cannot identify the bottleneck in a live case. They do not establish which agency delayed any particular application. Public Law 119-34 requires no referral date, stage owner, information-request date, or dispute status. An applicant cannot recreate the classified review, but it can preserve dated status contacts and the questions received so that an unexplained interval remains visible.
Congress has proposed the missing delay fields elsewhere
The distinction between transaction transparency and delay measurement is visible in another bill. H.R. 8289, the BIS Licensing Efficiency Act of 2026, was introduced in April and ordered reported by the House Foreign Affairs Committee. It has not become law.
The proposal uses a different data design. It would call for quarterly counts by listed application status, average and median time from first submission to a communicated decision, counts pending for at least 90 days, and reasons for delay. It would also require a notice when no decision had been communicated within 120 days. The Congressional Budget Office estimate for H.R. 8289 confirms that the committee ordered the bill reported and summarizes the proposed 90-day objective and 120-day notice.
Those provisions create no present applicant right. As of August 12, the Congress.gov status page for H.R. 8289 lists no enactment, and the latest cited action is the committee's April 22 order to report. The bill says the Secretary should seek to adjudicate within 90 days. Its field choices are still instructive. Congress knows how to ask for a median, a pending tail, a current status, and a reason for delay when that is the policy objective. Public Law 119-34 asks a different question.
The two measures should remain separate even if H.R. 8289 advances. Its proposed interval begins at first submission, while section 750.4 begins at registration and excludes specified periods. A future congressional performance metric could therefore measure end-to-end administrative elapsed time without becoming the regulatory clock. One number would still not do every job.
The applicant delay ledger
The practical response is a record that can be reconciled with both frameworks. The following matrix compares what the enacted report must contain with what the EAR measures and what an applicant needs to retain. It is a case-management tool, not a substitute for BIS status information or legal advice.
Event or question
Public Law 119-34 report
EAR processing treatment
Applicant delay ledger
Commercial commitment begins
Not required
Outside the section 750.4 count
Record quote date, order condition, capacity hold, customer milestone, and owner
Application submitted in SNAP-R
Submission date required for covered cases
Does not by itself start the 90-day period
Preserve receipt, version, applicant, item, end user, and transaction value
Completed application registered
Not required
Starts the 90-calendar-day period
Record registration notice and date separately from submission
Additional information requested
Not required
Time awaiting information can be excluded
Record request, response, responsible party, and counted or excluded treatment if known
Covered pre-license check or government assurance
Date, location, and result are required for a covered end-use check. A government assurance is not separately required
Specified periods can be excluded
Record notice, country contact, open date, close date, and status source
Interagency referral or dispute
Not required
Governed by section 750.4 and Executive Order 12981
Record status contact, known reviewing stage, open question, and next follow-up date
Decision communicated
Decision required but decision date is not
Ends the case subject to the applicable route
Preserve decision, communication date, conditions, and commercial disposition
Application remains pending
No required status or age band
Depends on registration and exclusions
Age the regulatory and commercial clocks separately and document uncertainty
The ledger should not manufacture dates the applicant cannot know. If BIS has not identified an excluded interval, mark the field unknown. If STELA shows only a general status, retain that wording and the retrieval date. The BIS licensing page directs applicants to SNAP-R for submission and STELA for monitoring. A precise record of what those systems showed is more defensible than a reconstructed timeline after a customer has left.
Commercial milestones belong in the same file because they explain exposure without being mistaken for agency time. A capacity reservation can expire while a pre-license check is open. A quote can require repricing even if the application is eventually approved. Commercial consequences can continue while an excluded interval is running. These are transaction consequences, not findings that the agency acted unlawfully.
How to read the first report
When the first report reaches Congress, start with the dates. A decision date would permit elapsed-time analysis, while pending ages would expose the unresolved tail. Registration and exclusion fields would allow comparison with section 750.4. Detailed aggregates would show whether comparisons across entity class, destination, item, and outcome are stable.
An answer of no would not mean the agency failed to comply. It would mean the report performed the narrower task Congress enacted. The appropriate response would be to use it for transaction and end-use-check oversight while looking elsewhere for queue measurement.
An answer of yes could materially improve congressional oversight. Voluntary decision dates would permit a submission-to-decision calculation for completed covered cases. They would improve the public record only if BIS lawfully released them or used them in public aggregate statistics. Pending-age fields could expose the unresolved tail on the same terms. Registration and exclusion fields would make comparison with section 750.4 possible. Even then, the covered-entity definition would prevent generalization to the whole BIS docket.
The first report should also be read beside GAO's unresolved institutional questions. A faster average can mask costly handoffs, while a larger workforce may leave Export Administration staffing unchanged. Added transaction detail still says little about stage-level timing. Each measure has a denominator and a scope. Oversight improves when those limits are stated, not when every number is pushed into one score.
The August 19 test
The statutory deadline creates a useful near-term checkpoint. It does not create a public release date, an individual decision deadline, or proof that a particular application has aged improperly. The reporting duty is also expressly subject to appropriations. Those limits should appear beside any claim about compliance after August 19.
For applicants, the immediate decision is simpler. Do not use the coming report as a reason to postpone case reconstruction. Preserve the submission receipt, registration notice, information requests and responses, known excluded events, STELA checks, status correspondence, and final communication. Put customer and capacity milestones on a second clock. Assign an owner and next follow-up date to every open interval.
Public Law 119-34 will give Congress a new view of controlled transactions involving covered entities. It may prove valuable for oversight of end users, item mix, values, decisions, and end-use checks. The law did not require the dates that turn those records into a delay measure. Until the first report supplies them voluntarily or Congress enacts a different field design, the most reliable license-delay record will still be the one an applicant builds case by case.
Free account
Keep reading with a free account.
Today's analysis is open to everyone. A free account opens the full archive and full tool output. No card required.