OECD Review of Fisheries 2025

Healthy and productive fish stocks are essential for an economically, socially and environmentally sustainable fisheries sector. A fish stock is considered healthy when the population is large enough that the risk of collapse in the short term is small (i.e. it is above limit thresholds often defined in terms of biomass). The health of fish stocks is not only impacted by fishing, but also by a range of natural factors (such as climate) and other economic activities (such as recreational fishing) and pollution. The goal of sustainable fisheries management is to ensure that stocks are not only healthy but also productive; in other words, the catch value or volume can be maximised under sustainability constraints.

Unhealthy fish stocks are bad for both the marine environment and the profitability of fishers, leading to negative impacts on the communities in which they live. Further, unhealthy fish stocks will also result in lower food production, which leads to negative impacts on food security, particularly in vulnerable coastal communities.

Regular stock assessments are fundamental to fisheries management and essential for ensuring decisions are based on scientific evidence. Ideally, stock assessments should accurately model the population dynamics of a stock and adequately model all the key natural processes that define the status of the stock, including natural mortality, recruitment and growth (Punt, 2023[1]). However, the accuracy of most stock assessment is limited by the availability of resources and data. As such, it is rarely possible to conduct “ideal” assessments. Instead, fisheries managers must balance the need for accuracy against available resources and the socio-economic importance of the stock. Consequently, the ability of any given stock assessment to accurately diagnose the state of the stock varies, with some assessments having higher uncertainty than others (Edgar et al., 2024[2]).

Climate change will exacerbate these issues, with long-term warming trends, short-term weather events (e.g. marine heatwaves) and ocean acidification having significant impacts on the distribution and abundance of stocks (IPCC, 2019[3]) (for more information see Chapter 4). The increased variability in fish stocks from year to year and changing ecosystems only increase the importance of conducting regular accurate stock assessments for effective fisheries management.

Accuracy of stock assessments is important because it directly impacts the decisions fisheries managers take. Recent evidence suggests that overfished stocks tend to have less accurate stock assessments, with a systematic bias towards overestimating the underlying stock biomass (Edgar et al., 2024[2]). Further, stocks of lower value species or in warmer waters are also more likely to have positively biased assessments, suggesting the resources available for assessment and the technical challenges to assessment (stocks in warmer waters are more likely to be part of multispecies stock complexes) also play an important role. While the direction of causality is not yet clear – i.e. are the stocks overfished due to inaccurate stock assessments or are the assessments inaccurate due to the stocks being overfished – better stock assessments are associated with healthier stocks, further underlining the important role they play in fisheries management systems.

The Food and Agriculture Organization (FAO) estimates that, globally, the proportion of fish stocks outside biologically sustainable levels increased from 10% in 1972 to 37.7% in 2021, driven primarily by overfishing (FAO, 2024[4]). These declines in the health of fish stocks are still ongoing and the proportion of overfished stocks globally has increased 2.3% since 2019. However, these global numbers mask important regional trends, with some regions (e.g. Mediterranean and Black Sea and North West Pacific) having a much higher proportion of overfished stocks than others. Importantly, some fish stocks are much larger than others, so despite 37.3% of stocks being overfished, only 23.1% of capture production came from overfished stocks in 2021 (FAO, 2024[4]). Finally, these figures are based on a fixed list of 445 reference aggregated stocks from which global and region totals are extrapolated. The geographical scale of the published figures does not highlight local nuances, making it difficult to translate them into actionable policy recommendations in individual countries.

The OECD approach to stock status attempts to complement the FAO’s global approach by collecting and publishing individual stock status data for the assessed stocks harvested by the countries and territories covered in this report where the results of the assessments are publicly released or reported to the OECD. For this edition of the Review of Fisheries, stock assessment results were received or identified from public sources for 1 623 stock assessments across 31 OECD Members and non-Members.1 Data are presented at an aggregated and a national level, with respect to both health limits and productivity targets, as explained in Box 3.1.

The database underpinning the analysis in this chapter has been compiled from a range of sources, including country submissions, publicly available documents from regional fisheries management organisations/arrangements (RFMO/As)2 and data downloaded directly from the International Commission for the Exploration of the Seas. It is likely there are data sources that have not been included, because they are either not publicly available or difficult to access. As such, the chapter illustrates both the most up‑to‑date understanding of stock status in the countries and territories covered in this report and also the extent of publicly available information on the status of the resources on which their fisheries sector relies.3

Of the 1 623 assessed stocks reported in the database, 62% were reported to be healthy (i.e. above their limit reference points), 15% were unhealthy and for 23%, the health status was not able to be determined (Figure 3.2a). Moreover, 31% of assessed stocks were reported to be meeting productivity targets (e.g. maximum sustainable yield or maximum economic yield), 19% were reported not to meet such targets, in 21% of the stocks the status was undetermined with respect to productivity and in 29% the assessments did not report status with respect to productivity targets (Figure 3.2b). For 50% of the assessed stocks, the productivity status is thus unknown, which complicates the interpretation of the productivity results. The stocks where the productivity status is unreported largely come from Australia where currently the assessments only report the status with respect to health limits, not productivity targets (Roelofs et al., 2024[5]).

Looking at stocks for which the health status could be determined (henceforth “conclusively assessed” stocks) – i.e. removing stocks with an undetermined health status – shows that 81% of them were healthy and 19% unhealthy. In other words, one in five conclusively assessed stocks is unhealthy.

The majority (73%) of healthy stocks were also meeting productivity targets, where both the health and productivity statuses were known.4 Thus, applying this ratio to all healthy stocks means that an estimated 59% of conclusively assessed stocks were also meeting productivity targets. Where healthy stocks are below the levels that allow for optimal productivity, fishing can lead to below optimal incomes, or below optimal production volume, and in both cases increased greenhouse gas emissions (Parker et al., 2018[6]). This highlights the importance and the potential of improved fisheries management to optimise food production or fisher incomes and reduce greenhouse gas emissions, even for healthy stocks (Bastardie et al., 2022[7]).

The proportion of conclusively assessed stocks that are healthy (81%) is significantly higher than the global average of 62% of sustainable stocks (i.e. stocks that are underfished or maximally sustainably fished) reported in the FAO State of Fisheries and Aquaculture 2024. This is likely driven, in part, by the different stocks included in each data set and the different methodologies applied.5 Another possible explanation for this difference is that fish stocks are healthier where scientific management is possible thanks to rigorous stock assessments, which highlights the importance of science-based management in maintaining global fisheries resources. Further, OECD Members tend to assess more of their stocks than non-Members (see below). Therefore, the OECD database contains a subset of stocks that is healthier than the global average. Conversely, the 19% of conclusively assessed stocks which are not healthy is likely caused by a combination of inappropriate management (either historic or ongoing), such as setting catch limits above scientific advice, ineffective enforcement and environmental factors which are outside the control of fisheries managers.

However, there are some caveats for interpreting these data. First, the database treats all stocks equally, but their relative contribution to national fisheries landings varies considerably. It is not currently possible to calculate the contribution of each stock to production at the country level, so these numbers do not necessarily indicate the extent to which a country relies on healthy or unhealthy stocks (and productive and unproductive stocks).

Second, the current status of a stock cannot be taken as an indicator of the quality of the effectiveness of fisheries management. The status of a stock in any given year is the outcome of fishing pressure, environmental factors and pollution from other economic activities, the impacts of which can be large (e.g. in the case of marine heatwaves) and difficult to predict. Further, an unhealthy stock may already be subject to management and rebuilding plans. Thus, the data should only be interpreted as representative of the current situation of resources, not of current management effectiveness at a national level.

Finally, some assessments cannot determine the health status of some stocks. Because the size of fish populations is not directly observable, stock assessments rely on mathematical models to estimate the size of the underlying fish population. These models are informed by data collected by fishers and during scientific surveys. An undetermined status occurs when there are insufficient data for the models to accurately estimate population sizes. A lack of data can have several different causes, for example scientific surveys may be delayed due to bad weather or a lack of resources, or in shared stocks, due to problems sharing data between different fisheries authorities. In some cases, the biology of the species or the ecosystem can create issues for effective assessment, for example Norway lobster are notoriously difficult to assess quantitatively (Aguzzi et al., 2022[8]). In these cases, more qualitative stock assessments, combing data from various sources to create an expert assessment, can be used to guide management decisions and can be effective for ensuring stocks are not overfished (Punt, 2023[1]).

There is considerable variation at the country level in the health status of conclusively assessed stocks (Figure 3.3). New Zealand (92%), Argentina and Korea (89%) have the highest proportion of conclusively assessed healthy stocks. There is also considerable variation in the proportion of assessed stock where health could not be determined, with the highest proportion found in Brazil (100%), United Kingdom (45%) and Ireland (44%). Importantly, stocks with an undetermined health status were found in all countries and territories.

There are also large differences in the number of fish stock assessments across countries (the number of which is reported in Figure 3.3). For example, Australia reports 476 stocks that were assessed at a domestic level and the United States 307 stocks. At the other end of the scale, Brazil only reported four and Peru six.6 These differences in the number of assessments highlight differences in the structure of the fishing sectors: some countries exploit more stocks and therefore need to conduct more assessments. But it also likely reflects the differing levels of investment in stock assessment capacity (Chapter 7), along with the difficulty and cost of conducting certain assessments. For example, multispecies stock complexes, which are common in warmer waters, can be difficult to assess, in part because established stock assessment methods were designed for single species stocks. In countries where the assessed stocks only account for a small share of production, the reported statistics provide a limited understanding of the status of underlying resources.

Finally, not all stocks are the same size, with some making a much larger contribution to national landings than others. Therefore, how representative the numbers are will only be apparent when stock status and harvest data can be reconciled at the stock level. This will also allow understanding what proportion of landings in a particular country is from healthy stocks – a key element of sector sustainability and resilience.

With respect to productivity, Korea (89%), Latvia (83%) and Peru (80%) have highest proportions of conclusively assessed stocks meeting productivity targets (Figure 3.4).7 In total, 21 Members and non‑Members have more than 50% of conclusively assessed stocks meeting productivity targets. Generally, conclusively assessed stocks tend to meet productivity targets as well, further highlighting the benefit of basing management decisions on scientific evidence.

The proportion of assessed stocks that are healthy has increased from 57% in 2019 to 62% in 2024 while the proportion of assessed stocks that are unhealthy has declined from 19% to 14% (Figure 3.5). The proportion of stocks with an undetermined status with respect to health remained stable (23%). Hence, considering only conclusively assessed stocks also shows a positive trend, slightly more pronounced: the proportion of conclusively assessed stocks that are healthy increased from 75% to 81% from 2019 to 2024 while the proportion of those that are unhealthy declined from 25% to 19%. While this indicates that the health of assessed fish stocks is likely improving over time, this interpretation must be made with caution given the changes to the stocks in the underlying database between iterations.

The majority of stocks (60%) have been in the database every year since its creation (in 2019). However, there is significant turnover for stocks moving in and out of the database (approximately 23% every iteration) (Figure 3.6). There are three main mechanisms for the turnover of stocks in the database. First, if a stock is assessed that was not previously assessed, it will be added to the database. For example, this can happen if more resources are made available and a country conducts more assessments or if a new stock becomes commercially valuable due to climate change and therefore warrants assessment when it previously did not. Second, the stocks in the database can change if there are changes to what is considered a stock worthy of its own assessment. In these cases, the old stock will not appear in the next iteration of the database and instead a new stock or stocks may be added. Finally, the database is limited to stocks that have been assessed in the previous ten years. If a stock has not been assessed within that time frame, it will be removed from the most recent iteration of the database.

The improving health status of assessed stocks over time is further supported by a closer analysis of the stocks that appear in multiple years of the database. Between 2019 and 2022, the status of 59 stocks changed from unhealthy to healthy and 36 stocks went the other way (healthy to unhealthy) (Figure 3.6). Further, between 2022 and 2024, the health of 26 stocks improved and 14 declined. In both periods, the number of stocks becoming healthy, therefore, dominates. Our understanding of stock health is also improving over time as more stocks move from undetermined to either healthy or unhealthy (51 stocks in 2019-22 and 29 stocks in 2022-24 respectively) than the contrary (30 stocks in 2019-22 and 24 stocks in 2022-24 becoming undetermined). Even accounting for the turnover of stocks in the database, evidence suggests that investing in data collection and stock assessment is paying dividends in the form of improved understanding and improved health of stocks.

The identification of trends with regards to productivity is less clear. The proportion of stocks meeting productivity targets has declined from 35% in 2019 to 31% in 2024 (Figure 3.7). However, the proportion of stocks not meeting productivity targets has also declined, from 22% in 2019 to 19% in 2024. Interpreting productivity status statistics is complicated by the stocks where the status with respect to productivity is not reported (29% in 2024 and largely from Australia). Removing these stocks and limiting the calculation to only stocks where information on productivity targets is reported results in the proportion of stocks meeting productivity targets increasing slightly from 43% in 2019 to 44% in 2024. Therefore, where data are available, there is a slight increase in stocks meeting productivity targets, but more data will be required to understand if this is a lasting trend or an artefact of the data. As with the health status, the turnover of stocks in the database is an important caveat when interpreting these results. The very small improvement in the proportion of stocks meeting productivity targets could indicate that management plans are not sufficiently cautious or simply that total allowable catches have routinely been set at levels in excess of management advice (Carpenter et al., 2016[9]; Winter and Hutchings, 2020[10]).

Analysis of trends from the OECD stock assessment database suggests that the health of assessed stocks is slowly improving and that assessed stocks are healthier than the global average. This aligns with recent research showing the links between accurate assessments and healthy stocks (Edgar et al., 2024[2]). Corresponding evidence for the improving productivity of stocks is still limited, notably because of missing information. However, this should not detract from what is a broadly positive message about the status of assessed stocks, which highlights the key role stock assessments play in effective fisheries management systems.

This chapter highlighted significant data gaps and priorities for research. First, the capacity to analyse the relative importance of assessed stocks should be developed to understand the share of landings volume and value coming from healthy stocks, both at the aggregate and the country level. This would allow for a more nuanced conversation on the status of stocks and what that means for the resilience and the productivity of fisheries sectors in both OECD Members and non-Members. Attempts are being made to improve reporting at a national level in the context of tracking progress towards SDG 14.4.1 (on sustainable fisheries), but reporting is limited.8

Second, to better leverage stock assessments to improve the sustainability and productivity of the sector there is a need for increased research and development to better refine existing stock assessment methodologies and develop new assessments for difficult to assess stocks, in particular low-cost and low data methodologies that are applicable to multispecies stock complexes. This could be the focus of development assistance for sustainable fisheries.

Finally, increasing investment in stock assessments and data collection more generally can help ensure, where possible, that all the commercially important stocks are assessed on a regular basis (at a time interval appropriate to the biology of the species), and that stock assessments report against both health (limit reference points) and productivity (targets) status.

[8] Aguzzi, J. et al. (2022), “Advancing fishery-independent stock assessments for the Norway lobster (Nephrops norvegicus) with new monitoring technologies”, Frontiers in Marine Science, Vol. 9, https://doi.org/10.3389/fmars.2022.969071.

[7] Bastardie, F. et al. (2022), “Reducing the fuel use intensity of fisheries: Through efficient fishing techniques and recovered fish stocks”, Frontiers in Marine Science, Vol. 9, https://doi.org/10.3389/fmars.2022.817335.

[9] Carpenter, G. et al. (2016), “Landing the blame: The influence of EU member states on quota setting”, Marine Policy, Vol. 64, pp. 9-15, https://doi.org/10.1016/j.marpol.2015.11.001.

[2] Edgar, G. et al. (2024), “Stock assessment models overstate sustainability of the world’s fisheries”, Science, Vol. 385/6711, pp. 860-865, https://doi.org/10.1126/science.adl6282.

[4] FAO (2024), The State of World Fisheries and Aquaculture 2024, Food and Agriculture Organization, Rome, https://doi.org/10.4060/cd0683en.

[3] IPCC (2019), IPCC Special Report on the Ocean and Cryosphere in a Changing Climate, Pörtner et al. (eds.), Intergovernmental Panel on Climate Change, https://www.ipcc.ch/site/assets/uploads/sites/3/2019/12/SROCC_FullReport_FINAL.pdf.

[6] Parker, R. et al. (2018), “Fuel use and greenhouse gas emissions of world fisheries”, Nature Climate Change, Vol. 8/4, pp. 333-337, https://doi.org/10.1038/s41558-018-0117-x.

[1] Punt, A. (2023), “Those who fail to learn from history are condemned to repeat it: A perspective on current stock assessment good practices and the consequences of not following them”, Fisheries Research, Vol. 261, p. 106642, https://doi.org/10.1016/j.fishres.2023.106642.

[5] Roelofs, A. et al. (2024), Status of Australian Fish Stocks Reports 2024, Fisheries Research and Development Corporation, Canberra.

[10] Winter, A. and J. Hutchings (2020), “Impediments to fisheries recovery in Canada: Policy and institutional constraints on developing management practices compliant with the precautionary approach”, Marine Policy, Vol. 121, p. 104161, https://doi.org/10.1016/j.marpol.2020.104161.