OFAC lists most new UN targets first, yet about one in five UN entries is not on its lists

UN sanctions list vs OFAC: is screening against OFAC enough?

The UN Security Council Consolidated List “includes all individuals and entities subject to measures imposed by the Security Council” [1]. On 6 October 2026 it had 1,010 entries: 736 individuals and 274 entities, spread across 14 regimes. Countries are expected to freeze these targets' assets “without delay” [8], which FATF defines as “ideally, within a matter of hours” of a UN designation [9].

The UK and EU lists are built to include the UN list; our guide to OFAC vs UN vs EU sanctions lists explains how each list is made. The UK Sanctions List carries all 1,010 entries: 991 carry the UN reference number, 17 more match on name with a UN source or the UN date, and the last 2 Iran entries sit under the UK's own Iran nuclear regime. The EU's consolidated file, dated 22 September 2026, carries 986 (97.6%) on our rules. OFAC's lists are different. OFAC publishes its own designations under US law, and we found 817 of the 1,010 UN entries (80.9%) on its SDN or non-SDN lists. Adding 10 matches we judged probable but not certain gives 827 (81.9%).

Matching across lists is not trivial, because OFAC records carry no UN reference number. We used exact name matches (in any word order, with aliases) checked against birth years, then reviewed by hand every borderline case. We also drew three random samples of entries that were still unmatched at the time, 95 in all, and searched them by hand against every OFAC name. They turned up 3 entries that were on OFAC's lists after all and 1 possible one: 4 of 95 (4.2%, 95% CI 1.6% to 10.3%). We corrected the 3 confirmed cases. Applied to the 183 entries still unmatched, that rate suggests a handful more, perhaps 3 to 19, so true OFAC coverage is likely between about 81% and 84%, not far above our figure. Every UN entry was found on at least one of the three other lists.

Which UN regimes are missing from OFAC's lists?

The gap is concentrated. Of the 193 UN entries we did not find on OFAC's lists, 102 are under the Taliban regime and 41 under the Iran regime, whose UN measures were re-imposed in 2025 [13]. Only 38 of 140 Taliban entries (27.1%) and 80 of 121 Iran entries (66.1%) were found. Ten of the 193 are probable matches we could not confirm, three of them Taliban entries with common names. Libya follows with 20 of 31 (64.5%). Among the smaller regimes, none of the 10 Guinea-Bissau entries is confirmed on OFAC's lists (one is a probable match), and none is on the EU's. By contrast OFAC carries 98.1% of the DPRK entries, 95.9% of Iraq entries and 94.9% of Al-Qaida and ISIL entries.

Some absences look like recent decisions. The names of nine unmatched UN entries appear in OFAC SDN removal notices since 2009: six in 2026, during OFAC's review of outdated entries [10], one in 2025 and two in 2012. Name tracing cannot prove identity, so treat these as likely, not certain. For most of the others, the data show only that the entry is not on OFAC's lists today, not why. US law may reach some of these persons in other ways, and this study does not assess that.

The European gaps are smaller and different. The EU file lacks all 10 entries of the UN's Guinea-Bissau regime, 3 older Libya entries, 3 Iran entries, and 8 entries the UN listed between April and July 2026 that were still missing from the EU file of 22 September 2026, 62 to 147 days later. The UK list has every UN entry.

Who lists first: the UN, the US, the EU or the UK?

For timing we look only at the 207 UN entries listed since 1 January 2017, where all four lists carry a usable date for most entries. For each one we compare the UN listing date with the date the same target first appeared on each other list: OFAC's SDN publication date, the earliest EU Official Journal date on the EU record, and the date designated shown on the UK list.

The US usually goes first. Of the 196 such entries on OFAC's lists, 165 (84.2%) were on the SDN list before the UN listed them, a median 142 days earlier. Only 10 (5.1%) appeared on the same day and 21 (10.7%) afterwards. In practice, for most new UN listings, a firm screening against OFAC was already screening against the name.

The UK list mostly shows the UN's own date. For 167 of the 207 entries (80.7%) the UK date designated equals the UN date, and 197 of the 198 entries the UK had not listed earlier were dated within three days. UN listings take effect in UK law by reference to the UN list [5], so this field may record legal effect rather than the day the FCDO updated its published list. OFSI's guidance states that the FCDO aims to update the list “within one working day for all new UN and UK listings” [6].

The EU is slower. Of the 199 entries on the EU list, 16 (8.0%) carry an EU date earlier than the UN listing. 9 (4.5%) carry an Official Journal date on the same day as the UN listing, 79 (39.7%) two to seven days later, 70 (35.2%) eight to 30 days later and 25 (12.6%) more than 30 days later.

How long does the EU take to implement UN sanctions?

UN listings reach EU law through Council or Commission acts published in the Official Journal. The Council's own guidelines say the EU “should aim to have the necessary implementing legislation in place without delay and within 30 days of the adoption of the UNSC Resolution at the latest” [7]. The EU's consolidated file does not carry the date each entry entered EU law, so we used the earliest Official Journal publication date attached to any part of the EU record.

For the 183 UN entries listed since 2017 that the EU had not already listed itself, the median gap is 8 days, and 86.3% fall within 30 days. Only 9 (4.9%) carry the same date as the UN listing, and 48.1% fall within a week. Measured against FATF's “within a matter of hours”, most new UN listings arrive in the EU's legal texts about a week late. FATF's own 2022 review of 59 mutual evaluation reports concluded that many countries take too long to transpose UN designations, but it classified report text rather than measuring dates [11]. CSIS has measured lags between US and EU or UK designations made under their own programmes [12]. In our search we found no published per-entry measurement of the UN-to-EU gap.

From 2011 to 2024 the median gap by year of UN listing ranged from 4 days (2016) to 28 days (2013), with 27 to 27.5 days in 2023 and 2024. The 12 entries the UN listed in 2025 and 2026 have a median gap of 0 days, which may reflect faster transposition or simply how those records were built; the sample is too small to say. We leave out UN listings before 2011: the earliest surviving date on many older EU records, notably the 2001 Al-Qaida and Taliban listings, is years after the UN date, which points to rebuilt records rather than real delays.

Two caveats are important. The earliest date on an EU record is an upper bound: if the EU later rewrote every part of a record, the original date is lost. That is the likely explanation for 10 DPRK entries from 2017 and 2018 whose earliest surviving date is two to seven years after the UN listing. And the EU file we used is dated 22 September 2026, so entries the EU adopted after that date are not counted.

Is the same sanctioned target spelled the same way on every list?

For the 808 UN entries found on all of the UK, EU and US lists, we compared what each list publishes. The volume of names is similar: the median entry has 3 Latin-script spellings on the UN list, 3 on OFAC's and 3.5 on the EU's. The spellings themselves are not the same.

For 417 of the 808 entries (51.6%), at least one spelling the UN publishes does not appear on the OFAC record, even after ignoring case, accents, punctuation and word order. For 412 (51.0%), OFAC publishes a spelling the UN does not. The UN's primary name, after the same normalisation, appears among OFAC's names for 590 entries (73.0%). For the other 218, about a quarter, OFAC does not publish the UN's main name in any form we could match. Because our matching needed at least one shared name, these figures if anything understate how different the lists are.

Dates of birth mostly agree. Of 540 matched individuals where both the UN and OFAC give a birth year, 20 (3.7%) share no year at all; in 10 of them the nearest years are one year apart. Our matching rules only kept such conflicts when the primary name matched, so the true rate of disagreement may be higher. A rule that demands an exact birth year agreement would miss these people when screening one list's data against the other's.

What this means for compliance teams

1. Check which UN regimes your lists actually cover. If you screen only against OFAC, test your list set against the UN Taliban, Iran and Guinea-Bissau regimes in particular, where 102, 41 and 10 UN entries were not found on OFAC's lists. Our overview of global sanctions lists shows what each list covers.
2. Do not assume a UN listing is in EU law the same day. The median EU record shows an Official Journal date 8 days after the UN listing. If your obligations follow the UN list directly, screen the UN list itself, not only the EU file. See EU fintech sanctions screening and the EBA restrictive measures guidelines.
3. Screen every published spelling from every list you rely on. Half of shared targets have a UN spelling that OFAC does not publish. Merging lists by name and keeping one record's aliases loses spellings. A watchlist screening API should screen each list's own names.
4. Allow for birth-date disagreements between lists. 3.7% of shared individuals have no birth year in common between the UN and OFAC. Check that a birth-date mismatch lowers a match score rather than discarding the match; our matching methodology describes how Verifex weighs date of birth with other identity evidence.
5. Record when each list file was generated. The EU file we downloaded on 6 October was dated 22 September and lacked eight UN entries listed from April to July 2026. Your evidence should show the generation date of every list you screened against. For the lists Verifex loads, see source health.

Methods

Data. All files were downloaded on 6 October 2026. We also used OFAC's Recent Actions notices (see VR-2026-04) to trace whether unmatched UN entries had ever been added to or removed from the SDN list.

Matching rules. UK: the UN reference number on the UK record (991 entries). For the rest, a shared name on a UK record whose designation source is the UN (13) or whose date equals the UN date (4). The last 2 were found by hand under a UK autonomous regime. EU: the UN reference number on the EU record (83 entries); otherwise a shared name on an EU record whose designation date equals the UN listing date (558); otherwise a shared name on an EU record under the matching EU programme (343). Two more were matched and one judged probable by hand. US: a shared name (primary or alias, word order ignored, apostrophes removed, and bracketed names of two or more words used as extra aliases) on an OFAC record of the same type. For individuals, both records must share a birth year when both give one. Where they do not, the match is kept only if the shared name is the primary name on one side, and it is flagged.

Hand review. Every multiple match, every birth-year conflict, every candidate from a looser search (name similarity, partial names, rare shared name tokens with a shared birth year) and every hit from tracing OFAC notices was reviewed by one analyst and recorded as match, probable or reject with a reason. The decisions are published as the three codebook files under Data and code.

Absence checks. Three random samples of entries unmatched at the time (25, 40 and 30 entries; seeds 7, 20261007 and 42) were searched by hand against all OFAC names. The first two found 3 confirmed misses, which were corrected. The last, drawn from the final 183 unmatched entries, found 1 possible miss (a trading-house name that differs from an OFAC-listed name by one letter; we kept it as a reject because the names differ).

Precision check. A random sample of 60 final automatic US matches (seed 7001) was checked by hand: 59 correct and 1 uncertain, none wrong (precision at least 98.3%, 95% CI 91.1% to 99.7%). All 19 automatic matches made under the birth-year-conflict rule were reviewed individually; 18 were kept and 1 rejected.

Timing. UN date: LISTED_ON. US date: SDN publication date of the matched OFAC record (validated in VR-2026-03). EU date: the earliest publicationDate of any regulation summary on the EU record, an upper bound. UK date: DateDesignated, which may be a legal-effect date. Timing uses UN entries listed from 1 January 2017.

Statistics. Coverage figures are full counts at the snapshot. The hand-check rates come from random samples and carry Wilson score 95% intervals.

This study describes public data and general regulatory context. It is not legal advice; confirm your obligations with a qualified adviser.

Limitations

• Matching without shared identifiers can make errors in both directions. The precision check found no false matches in 60 (one uncertain), and the absence checks found 4 misses in 95 sampled unmatched entries (4.2%, 95% CI 1.6% to 10.3%), three of which were corrected.
• “Not on OFAC's lists” does not mean “not restricted under US law”. Other US authorities may apply. This study measures list contents only.
• The EU file is dated 22 September 2026, two weeks older than the other files, so very recent EU changes are not counted.
• EU timing uses the earliest surviving date on a record, which is an upper bound on when the entry entered EU law. Some long gaps (10 DPRK entries) are probably record rewrites, not real delays.
• The UK date designated may record when a UN listing took legal effect, which by law is the UN date, rather than when the FCDO list was updated. The UK figures are therefore not a measure of operational speed.
• US timing compares the UN date with the first SDN date of today's record. If OFAC removed and re-added a target, the later date is used.
• Iran regime entries kept their original 2006 to 2012 UN dates after the 2025 re-imposition, so they are outside the timing window and excluded from Figure 4.
• Every hand-review decision was made by one analyst; the decisions are published below so they can be checked.

Questions and answers

Data and code

Every value plotted in Figures 1 to 5 is in the figures file below. The coverage file gives, for each UN regime, how many entries we found on each list, with and without probable matches. The three decision files are the hand-review codebook: for each reviewed UN reference number, the record chosen, the decision and a short reason. Reasons describe evidence and never name individuals. No customer data was used.

• jsonFigures 1 to 5: every plotted value
• csvCoverage by UN regime and list
• csvHand-review decisions: OFAC
• csvHand-review decisions: EU
• csvHand-review decisions: UK