Last year, the good people from MRAG (with whom I have worked in the past) asked to collaborate on this study PEW contracted them to do, and it has finally been published.
doing good PSM is the challenge, not just signing the agreement
Of course, PSM is an area of fundamental interest for me, and I have worked (with different levels of intensity) in 3 of the 5 ports analysed (Majuro, Bangkok and Montevideo), so I provided my technical inputs but also local knowledge on whom to contact as to get on-site knowledge.
Of course, it is good to read the whole document for the details of each port! Yet I will quote the analysis, main messages and recommendations:
For most case study ports, the introduction of PSMs has had relatively little impact on the natural dynamics of port usage
For four of the six case study ports examined here, the implementation of the PSMA (or PSMs) appeared to have relatively little impact on port usage by foreign fishing fleets. This is largely because the choice of port for foreign fishing vessels is primarily driven by commercial and other imperatives, which tend to overwhelm any possible ‘disadvantage’ associated with stronger port controls. For example, in the case of Bangkok, the number of port visits changed little between the before and after periods, largely because Bangkok is the main hub for the processing of canning-grade tuna from the WCPO and demand from canneries remains strong.
Carriers continue to use Bangkok despite stronger port controls being introduced in 2016 because ‘that’s where the canneries are’. While alternative ports for tuna processing exist within the region, these don’t have the same dominant market position of Bangkok. In the case of Majuro, the number of port visits is largely influenced by the distribution of fishing effort within the WCPO purse seine fishery. As an attractive port for transhipment, Majuro will be used preferentially when fishing effort is concentrated in the central and eastern parts of the WCPO (typical during El Nino periods), but will see fewer port visits when fishing effort is concentrated in the western parts of the WCPO (typical during La Nina periods). The introduction of stronger PSMs in 2017 will have little influence in the context of these more dominant macro trends. For Montevideo, the number of Chinese squid jig vessels and Panamanian flagged carriers declined across the study period, although these trends commenced prior to the implementation of the PSMA and were most likely driven by unrelated factors. Chinese industry sources advised that the trend away from visits to Montevideo was mainly driven by lower squid catches in the SW Atlantic, combined with an increase in the number of Chinese- controlled carriers (flagged primarily to Panama and China), directives from squid companies to preferably use Chinese ports (to reduce the risk of crew absconding in foreign ports) and an increasing domestic demand for squid meaning more squid were transhipped at sea and transported directly back to China. Other carrier visit trends by flag occurred primarily as a result of reflagging, and thus were also unrelated to PSMA implementation.
In other cases, the lack of a strong ‘signal’ in foreign fishing fleet behaviour around the time of PSMA implementation may be also partly driven by the incremental implementation of PSMs over time. For example, Namibia is a member of a number of RFMOs that each have PSM-related CMMs. Many of these require similar (or identical) measures to the PSMA and were implemented well before Namibia acceded to the PSMA in 2017. To that end, any adjustment in foreign fishing fleet behaviour may have happened well before practical implementation of the PSMA. This lack of a clear signal associated with incremental implementation is also likely to be true of other ports, with few going from ‘0 to 100’ on PSMs overnight – most PSMs are introduced incrementally over time as capacity builds and industry is allowed periods to ‘get used to’ new arrangements.
For many ports it may also be true that the lack of a clear change post-PSMA is driven by a view amongst the vast majority of foreign fishing vessels that they consider themselves compliant with relevant fisheries laws and therefore have nothing to fear from entry into PSMA ports.
Broadly, these results should provide confidence to most States considering acceding to the PSMA that the Agreement can be implemented without ‘scaring off’ foreign fishing fleets and forgoing important economic activity associated with FFV port visits.
But, in some cases, new PSMs can have a big impact
The most notable example of a PSM-related impact amongst our case study ports was the response to Peru’s requirement to install and operate an approved VMS system, as well as submit VMS data for the previous six months, for all vessels wishing to enter Peruvian ports post-2020. In this case, the introduction of the VMS requirement led to the complete abandonment of Chimbote port by the Chinese high seas squid jig fleet – which had used Chimbote since 2017 as an important port for hull maintenance with up to 129 visits annually – as well as a significant decline in the use of Callao port by both Chinese and Korean squid jig fleets. While the VMS requirement was introduced in the same year as the COVID pandemic, advice from Chinese industry sources confirmed that opposition to the VMS requirement was the main driver of behaviour. Our analysis indicates that these vessels continued to fish in the same areas, but either remained at sea during our study period (with catch likely transhipped at sea) or returned to Chinese ports for maintenance. In the case of the Korean jig fleet, the VMS requirement contributed to a substantial change in fishing pattern. In this case, industry advice indicates that the VMS requirement, together with low CPUE in the south east Pacific, led to the fleet fishing solely on the Atlantic side of South America (abandoning the typical ‘squid route’ seasonal migration from the Atlantic side to the Pacific side for the second half of the year). Interestingly, other foreign fleets to whom the VMS requirement did not apply – e.g. the tuna longline fleet – continued to use Peruvian ports (principally Callao) in the same manner throughout our study period.
The VMS requirement was originally introduced in response to concerns around high seas vessels switching off AIS and fishing in Peru’s EEZ. While the extent to which the VMS requirement contributed is unknown, independent analyses have highlighted changes in Chinese jigging fleet behaviour in the period after the requirement was introduced (GFW, 2020; 2021b). In particular, there were fewer vessels with AIS irregularities leading to confusion over either identity or location in 2021 compared to 2020, and perhaps more importantly the fleet appeared to have introduced a self-enforced buffer zone adjacent to the Peru and Ecuador (Galapagos) EEZs. While 2020 AIS data showed the fleet fishing right up to the EEZ boundary, the closest AIS record for 2021 was 10nm and 150nm from the Peru EEZ and Ecuador (Galapagos) EEZ boundaries respectively.
In addition to what could be considered positive developments, a number of unintended consequences may have resulted. For example, the exit of the Chinese jigger fleet from Chimbote has reportedly had a significant impact on the local economy for whom vessels making up to 129 visits annually at an average of 12 days each made an important contribution. We understand this has led to a level of opposition to the VMS requirement from local business owners. Some analysts have speculated the choice to remain at sea for longer periods amongst the jig fleet also risks a higher level of labour rights abuses.
There is limited evidence of a shift from PSMA designated to non-designated ports
One of the possible consequences of implementing the PSMA was a shift by fishing fleets from PSMA- designated to non-designated ports (either non-designated ports in PSMA Parties, or ports in States not Party to the PSMA) in order to avoid stronger port controls (or to avoid any additional administrative burden). This was particularly the case for fleets with higher risk of involvement in IUU fishing. The available evidence from the case studies indicates that broadly this hasn’t happened. As discussed above, in most cases fishing fleet behaviours around port usage have been driven by natural commercial dynamics around the port/fishery, with relatively few changes that could be explained primarily through the introduction of PSMs. In some cases, this may have influenced in part by relatively few inspections being undertaken (e.g. Montevideo).
When changes did occur – for example, the exit of Chinese and Korean flagged squid jiggers from Peruvian ports following the introduction of the VMS requirement in 2020 – the results were mixed. Some vessels appeared to remain at sea for longer, others returned to home ports (particularly during the height of the COVID pandemic), while others called in to both PSMA and non-PSMA ports. While the return to home ports, where PSMA measures do not automatically apply to domestic vessels, could theoretically reduce inspection coverage of these vessels, it is not yet clear whether these are longer term trends.
Of course, our case studies sample only a very small fraction of PSMA-designated ports internationally andeach port has its own dynamics. On that basis, it’s possible that fleets in other ports may have responded differently, although the risk of vessels shifting to non-designated ports should only reduce in future as more States accede to the PSMA.
There was limited impact on the operational aspects of port usage – e.g. time in port, time at anchor
Another possible consequence of the introduction of stronger port inspection regimes and the requirement for advance entry request under the PSMA was a change in the operational efficiency of using ports – for example, extending the duration of port visits and time at anchor waiting for port entry approval. The available evidence from the case studies indicates no substantial change in the operational aspects of port usage following PSM implementation. Broadly, time in port and wait times at anchor remained relatively constant in the periods before and after PSM implementation (with the exception of time spent at anchorage in Bangkok).
This, again, is likely largely driven by the natural operational dynamics of the port (e.g. the time taken for a carrier to fill up hasn’t changed post-PSMA), and also by the considerable efforts of port States (for all of whom the economic activity associated with port usage is very important) to make the PSMA process as efficient as possible. For example, in several cases (e.g. Bangkok, Majuro) port States have developed electronic platforms, accessible online, to manage port entry requests and data submission requirements as well as coordinate local agencies.
In the one port for which there was some evidence of change – Bangkok – the median length of carrier visits to port was shorter and wait times at anchor were longer. There are a range of non-PSMA factors that could contribute to this result (e.g. smaller average transhipment volumes resulting in faster unload times), although it is also plausible that wait times at anchorage are slightly longer because vessels now have to wait for approval before entering port, while port visit duration may be shorter partly because onshore processes (e.g. inspection teams, unloading logistics) have had slightly longer to get themselves organised. Nevertheless, Thailand’s DoF noted that unload volumes are the key driver of time in port, and because data on these is not available pre- and post-PSMA they cautioned against drawing strong conclusions around the impacts of PSMA processes on the duration of time in port.
AIS should be complemented by other data sources and local knowledge
The emergence of publicly-available AIS data together with sophisticated analysis through ‘big data’ platforms in recent times has offered unprecedented insights into global fishing patterns and fleet behaviour. In this study, AIS has proven an extremely useful tool to examine the practical impacts of the PSMA. However, AIS coverage remains patchy for some fleets and areas, particularly in the earlier years of our datasets. To that end, supplementing AIS data with other sources of information – e.g. publicly available port visits records, transhipment volumes – has been important to gauge the completeness of the AIS dataset as well as fill in gaps where necessary. For most ports, AIS data appeared reliable enough to be a reasonable reflection of actual trends in fishing fleet behaviour, although for Montevideo we substituted AIS data for a groomed version of official port records as our primary dataset.
The other essential element to ensuring best use of AIS data in studies such as this is to use local knowledge. In practice, there are a myriad of commercial and operational factors which influence fleet behaviour, port usage and PSM application across FFVs which may not be evident from trends in AIS data alone. To that end, involving local experts and stakeholders to help interpret any trends is essential in gaining an accurate understanding of the factors driving fleet dynamics.
RECOMMENDATIONS
Much of the analysis in this study focused on changes in fleet behaviour in response to the implementation of PSMs and the underlying drivers, rather than examining the effectiveness of the PSMs themselves. To that end, our recommendations below focus on observations and learnings that should assist the conduct of similar studies in future, as well as other similar work that might usefully be undertaken.
Monitoring of PSM impacts – for countries interested in implementing the PSMA (or stronger PSMs) and keen to monitor any practical changes in fishing fleet behaviour, AIS offers a potentially valuable and cost-effective dataset. A framework of indicators similar to those used here can be tailored to examine changes in the main issues of interest (e.g. changes in total numbers of vessel visits, changes in flag States, changes by vessel type, duration of vessel visits). One of the main challenges in these types of ‘before and after’ analyses is that PSMs are often implemented incrementally and the point at which key measures ‘start to bite’ is not always well-recorded. To that end, countries interested in analysing changes in fleet behaviour post-PSM/PSMA implementation should maintain clear records of key dates of implementation so that trends may be analysed in the appropriate context.
Complement AIS with other available data – As discussed above, the increasingly widespread use of AIS technology across the global fishing fleet and the emergence of big data platforms including Global Fishing Watch offer previously unavailable opportunities to track and analyse patterns in fishing fleet behaviour. The value of these insights should only improve as AIS data coverage and technology improves over time. Nevertheless, this study highlights that AIS data coverage is better in some fleets, areas and ports than others and can be patchy, particularly the further you go back in time. To that end, future studies using AIS data to analyse changes in fishing fleet behaviour should complement AIS with other relevant data (e.g. fisheries/port agency official records of port activity) to gain insights into the completeness of the AIS dataset and asses the confidence with which conclusions can be drawn. Caution should be exercised when AIS is the only available dataset.
Use of local experts – The factors driving port usage by fishing fleets are complex and often dynamic. While AIS data is potentially extremely useful in identifying and analysing trends in fleet behaviour, the drivers underlying trends may not be evident from AIS data alone. To that end, stakeholders knowledgeable in the dynamics of the port should be engaged to assist in the interpretation of results.
Monitoring global implementation of the PSMA – in addition to examining trends in fleet behaviour at the individual port level, AIS data has potential utility in monitoring implementation of the PSMA at the global level. For example, AIS data could be used to track the proportion of port visits by FFVs to designated ports under the PSMA versus non-designated ports (either ports in non-party States, or non-designated ports in party States). Similar trends could be tracked for other relevant indicators (e.g. proportion of visits by hold capacity) or geographic region (e.g. FAO areas, continents). Clearly there would need to be confidence that the underlying data provided an accurate representation of actual visits, but assuming this was the case, indicators of PSMA implementation might be published in relevant publications including the FAO’s periodic State of World Fisheries and Aquaculture reports.