Seafarer abandonment falls to a flag, port and home state chain with no fixed time limits — in 2025 the ILO/IMO database recorded 410 new cases, 185 noted as having no obligatory financial guarantee

What is happening?

Since 2017-01-18, the 2014 amendments to the Maritime Labour Convention, 2006 have required every flag state bound by them to make sure its ships are covered by a financial security system that pays for repatriation, living support and outstanding wages if a shipowner abandons the crew, with evidence of that cover carried on board. As of 2024-12-20 the Convention had been ratified by 108 states representing more than 96.6 percent of world shipping tonnage.

Reported cases on the joint ILO/IMO database moved unevenly after that date — 55 in 2017, down to 40 in 2019, then higher every year from 2020 onward. No source ties those movements to the amendments. For calendar year 2025 the IMO Legal Committee, at its 113th session in April 2026, noted 410 new cases involving more than 6,000 seafarers, and noted that in 185 of those cases there was no obligatory financial guarantee. It described the 2025 total as far exceeding every previous year on record, and reported a further 103 cases logged between 2026-01-01 and the session.

What happens after a case is reported is set out not in the Convention but in joint ILO/IMO guidelines adopted in December 2022. Those guidelines pass responsibility down a chain — the flag state first, then the port state, then the state of which the seafarer is a national, the home state — using the word should at each step, with no number of days attached to any step. The guidelines state of themselves that they are not legally binding and not subject to ratification or supervisory mechanisms.

Whose problem is this?

The guidelines themselves note that the great majority of the shipping industry operates without any threat of abandonment. The problem sits in the minority of cases where an owner walks away, and in how long the seafarers then wait while three kinds of state work out which of them acts.

Where does this problem end?

Two value questions sit next to this problem and are set aside by name. The first is what level of compensation is adequate — the guidelines give as an example that a guarantee may not cover everything when more than four months of wages are owed, and this document reports that wording without judging whether the cover is enough. The second is which ship registries are legitimate, a debate this document does not enter.

What is the state now, and what should it be?

Now

What it should be, in the words of the instruments themselves

The Convention sets the standard: each flag state shall ensure that a financial security system is in place for ships flying its flag, so that abandoned seafarers are repatriated and paid. The guidelines add that a flag state notified of a case should set the shipowner and the security provider an expeditious time frame to comply, and that every state should develop national standard operating procedures for each role it may hold. A case counts as closed on the database only when the ILO receives clear notice from the original informant that every seafarer who wanted to go home has been repatriated and all outstanding pay has been settled.

No source opened here sets a target for how many cases should occur or how quickly a case should close. The 113th Legal Committee asked the joint working group to consider creating step by step procedures with specific timelines and trigger dates — a request, not a target that yet exists.

How big is it?

410 new cases in calendar 2025, involving more than 6,000 seafarers, with 185 noted as having no obligatory financial guarantee. That is the figure the IMO Legal Committee recorded at its April 2026 session.

The earlier years, each with its own reference date, show the direction. The database report dated 2024-12-20 gives 12 to 19 cases a year for 2011 to 2016, then 55 in 2017, 44 in 2018, 40 in 2019, 85 in 2020, 95 in 2021 and 109 in 2022. It gives 142 for 2023, with 62 noted as having no obligatory financial guarantee, and 310 for 2024 up to 2024-12-20, with 144 so noted.

Between the entry into force on 2017-01-18 and 2024-12-20, 874 cases were reported. The units matter here: a case is not a ship and not a seafarer, and the number of seafarers inside the 185 cases of 2025 is not given anywhere. No count of all ships or seafarers covered by financial security was opened, so these figures cannot be turned into a rate.

Under what conditions does it arise?

1. The obligation is national and the ship moves between nations. The duty to require financial security rests on each flag state for its own ships, while abandonment is usually discovered in a port of another state, and the seafarers are nationals of a third. Each of the three holds part of the leverage and none holds all of it.

2. The response chain is written as a sequence of fallbacks without a clock. The flag state should act, and if it does not the port state should, and if neither does the state of nationality should. Because no step carries a number of days, there is no fixed moment at which the next state in line knows the previous one has failed and its own turn has begun.

3. The chain lives in a non-binding instrument. The binding part of the regime is the requirement to have financial security. The part that says who does what once security fails to respond is guidance that no supervisory body reviews.

4. Some cases may sit outside the binding requirement. Of the 874 cases reported between 2017-01-18 and 2024-12-20, 35 involved flag states that are party to the Convention but had not yet sent the ILO their acceptance of the 2014 amendments; 15 of those were resolved and 20 remained disputed or unresolved. In 2022, 15 new cases involved flag states that have not ratified the Convention at all.

5. Counting depends on someone reporting. A case enters the database when port authorities or other concerned parties report it to the port state authority and the ILO. The database report states that accuracy is critical, which means the counts can move with reporting effort as well as with events.

What has been tried?

Every step since 2017 has taken the same form — urging states to ratify, to report and to apply guidance. The one proposal that would put numbers on the chain, the timed procedures, was at the stage of a request to consider as of the April 2026 session.

What was found?

Why is it still unsolved?

Coordination failure — the regime asks three kinds of state to cover for each other in sequence, and gives none of them a moment at which it must act.

The binding piece of the 2014 amendments works on the ship before anything goes wrong: a flag state requires cover and the ship carries proof of it. Once an owner has stopped paying, however, the question is no longer whether cover exists on paper but who makes it respond, and that question is answered in guidance rather than in the Convention. The guidance hands the task to the flag state, then to the port state if the flag state does not act, then to the state of nationality if neither does. Each handover depends on the previous state failing, but nothing defines when failure has happened. A flag state that is slow and a flag state that has stopped looking produce the same record, and the port state cannot tell from the outside which one it faces.

The second part is that the 185 cases noted as having no obligatory financial guarantee in 2025 sit at exactly the point the chain was written for. When no guarantee responds, repatriation passes to the flag state, then to the port state, then to the state of nationality, again without a number of days. The sources do not break down those 185 cases, so this document cannot say whether they involved flag states outside the Convention, flag states that had not accepted the amendments, or cover that had lapsed.

The third part keeps the pattern stable. The bodies that meet — the Legal Committee, the tripartite working group — can urge, welcome and ask others to consider. Timelines and trigger dates entered the record in April 2026 as something the working group was asked to think about, which leaves the chain as it was for every case already on the database.

What observation would mean it is solved?

Candidates — (a) the joint ILO/IMO working group adopts procedures with specific timelines and trigger dates for each state in the chain (b) the number of new cases per year noted as having no obligatory financial guarantee falls for several consecutive years (c) the share of database cases that reach resolved status, and the time they take to get there, improves and is published.

(a) alone is paper. Procedures with dates would still sit inside a non-binding instrument unless they are carried into the Convention or into national law, and a date nobody supervises is a date nobody is late for.

(b) alone can mislead in both directions. The counts rest on reporting, and the Legal Committee has itself been urging better reporting, so a rise can mean more cases or more cases being found, and a fall can mean fewer abandonments or less reporting. The database report says the link between the post-pandemic surge and any cause is unclear.

(c) alone measures only what is reported. A case counts as resolved only when the original informant confirms that seafarers are home and paid, so resolution time says nothing about cases never reported. The three have to be read together, with (b) checked against reporting effort.

What is it connected to?

Fills with researchthe application of Convention financial security to fishing vessels, the separate IMO work on the fair treatment and criminalisation of seafarers, seafarers stranded by regional security situations, and the voluntary ship registration guidelines of 2026. Relation type and evidence grade were not confirmed in this round.

What these sources do not say

• What lies behind the 185 cases of 2025. The IMO wording does not separate cases where a bound flag state had no cover in place, cases where the flag state was not bound because it had not ratified or had not accepted the 2014 amendments, and cases where on-board evidence existed but the cover had lapsed or did not respond. Only partial pointers exist for earlier years — 15 new cases in 2022 under non-ratifying flags and 35 cases from 2017 to 2024 under flags that had not yet accepted the amendments. The sources therefore do not support saying that any flag state breached its obligation.
• How many seafarers were in those 185 cases. The more than 6,000 figure covers all 410 cases of 2025.
• Any link between the response chain and the number of cases. No source attributes the rise in reported cases to the way the flag, port and home state chain is built or to its lack of fixed time limits. The chain governs what happens after a case is reported, not whether a case occurs.
• Whether the rise reflects more abandonment or more reporting. The database report says accuracy is critical and that the link to the post-pandemic period is unclear, and it does not separate the two.
• What guarantees actually paid and how long cases took to close. Neither amount nor duration appears in the ILO and IMO material opened here.
• Whether the 2025 and 2026 decisions changed anything. No source says what effect the calls of the 112th and 113th sessions had on case numbers.
• Why the 2022 figure differs. The database report gives 109 new cases for 2022 and the 2022 guidelines give 119; neither mentions the other.
• A denominator. No count of ships or seafarers under financial security was found, so no abandonment rate is given.
• The positions of shipowner organisations and of individual flag states. This document names no flag state and no shipowner, so no individual position was sought; the positions available are those of the ILO and IMO themselves, set out in the guidelines and the committee records, including the statement that most of the industry operates without the threat of abandonment and the fact that shipowner representatives make up a third of the working group that wrote the guidelines.

See the evidence

How this was assembled. The two PDFs — IMO document LEG 112/4(a) and the 2022 ILO–IMO guidelines — were read directly, the former from a copy hosted by a national maritime administration rather than on the IMO server. The IMO meeting summary for the 113th session and the two IMO press briefings were opened directly and are the only sources for 2025 and 2026 figures. Where sources overlap they agree on the entry into force date, on 2024 more than doubling 2023, and on the direction of the yearly counts. Where they disagree the disagreement is left visible: 109 or 119 new cases in 2022. Standard A2.5.2 is described through the wording the 2022 guidelines quote from it, since the text of the amendments was not opened directly. The annex of LEG 112/4(a) lists cases by ship, and none of those identifiers is carried into this document. This is a Path A output (research-based definition), so observation_refs is empty and provenance_mode: press-derived.

People affected

Estimated range Not derivable

The reason and what is missing are listed under “What is missing” below