Previous Page Table of Contents Next Page

Schrödinger’s TAC - superposition of alternative catch limits from 2003 to 2006 under the South Tasman Rise orange roughy arrangement between Australia and New Zealand

A. Serdy
Sea Law, Environmental Law and Antarctic Policy Section
Department of Foreign Affairs and Trade[276]
R.G. Casey Building, John McEwen Crescent, Barton ACT 0221, Australia
<[email protected]>


Quantum mechanics normally has little to do with either the science, the management or the law of fisheries, although measurement of the size of a stock by fishing it could be seen as a partial application on the macroscopic scale of Heisenberg’s Uncertainty Principle, in which measurement is impossible because the very act of observation changes the quantity being observed. The decision on the future total allowable catch (TAC) for the South Tasman Rise orange roughy fishery made by Australia and New Zealand in 2003 does however call to mind Schrödinger’s famous thought experiment illustrating the probabilistic nature of the subject. A cat is sealed in a box with a radioactive substance whose decay may or may not kill it. As long as the box remains sealed, the cat is said to be simultaneously in two superposed states of being alive and dead, until its state is "resolved" to one or the other by opening the box to see whether it has survived.

It is just such a superposition that has been achieved by the 2003 TAC decision (Appendix I), serving simultaneously the ends of conservation and rational exploitation of the South Tasman Rise orange roughy stock by in effect having two alternative catch limit pathways, one leading downwards to a TAC of 200 t, sufficient only for a bycatch or at most a one-trip fishery, and the other maintaining catch limits at up to 1 300 t should the stock in fact still be in a relatively healthy state. The embedded decision rules, taking as their input empirical evidence in the form of success or failure of fishing operations early in the season, gradually resolve the superposition in favour of one of the possible TAC "states". Before explaining how this came about, however, a brief recapitulation of the history of the regulation of this fishery is in order.


It is not necessary to go far into the history of the South Tasman Rise orange roughy fishery and its regulation, because the Australia-New Zealand 2000 Arrangement for this fisheries regulation[277], together with matters surrounding its management, have been expertly analysed by Molenaar (2001).

The South Tasman Rise is a large, submerged plateau forming part of Australia’s continental shelf - the significance of which will become clear below - rising to less than 1 000 m from the surface. It lies partly within and partly beyond the outer limit of the Australian Fishing Zone (AFZ)[278]. Its general shape can be inferred from the bathymetric contours in Figure 1.

In September 1997, significant aggregations of orange roughy were discovered on this feature and a fishery rapidly developed, with Australian vessels landing over 1 500 t by the end of the year. As the bulk of these fish were taken outside the AFZ, the fishery also attracted vessels from the New Zealand deepwater trawl fleet. During 1997, the two fleets made a total catch of over 2 000 t. Mindful of the earlier rapid depletion of the nearby Lord Howe Rise orange roughy stock, the two countries negotiated a formal Arrangement[279] to take effect from 1 March 1998 to establish what was thought to be a precautionary TAC of 2 100 t for orange roughy within a defined area of high seas encompassing the known fishery, based and apportioned on the verified 1997 catch of the two countries (1 669 t Australia, 431 t New Zealand). The defined area adjoins the AFZ lying south of Tasmania between 146 º 30 'E and 150 ºE and is bounded to the south by the parallel of latitude 48 º 30 'S, as illustrated in Figure 1 above.

For a number of reasons that essentially boil down to the usual problems of allocation and which therefore need not detain us further[280], it proved more difficult than expected to reach accord on a TAC and national shares for 1999, but the two Governments reached an informal understanding so as to allow their fleets to resume fishing from 1 March 1999. Officials subsequently settled on an increased TAC of 2 400 t, shared 75 percent (1 800 t) to Australia and 25 percent (600 t) to New Zealand, but before it could be reduced to writing for the two ountries’ Fisheries Ministers to approve, the New Zealand fleet’s catch grew to more than 1 600 t for lack of legal capacity to limit the catch until the instrument was concluded[281].

The South Tasman Rise, showing bathymetric contours and the defined High Seas Area

As an interim measure the two Ministers, by an exchange of letters, closed the fishery for the remainder of the season, though Australian operators, who had retained a small portion (60 t) of their allocation to allow for orange roughy bycatch when targeting oreos, were permitted to target oreos until the orange roughy quota was filled.

In 2000 Australia and New Zealand entered into a new Arrangement (the 2000 Arrangement) for the 2000-2001 fishing year and beyond - which has continued in effect to the present day - with the TAC again being set at 2 400 t and apportioned 75 percent to Australia and 25 percent to New Zealand[282]. For the first six years, however, New Zealand also agreed to repay 640 t of its notional overcatch by reducing its actual catch limits over seven years - 100 t a year for six years and 40 t in the seventh year[283].

Despite considerable searching and fishing effort on the South Tasman Rise during the 2000 season, only 830 t (790 t Australia, 40 t New Zealand) of orange roughy were landed. Subsequent catches have been even lower (Caton (Ed.) 2003): 188 t in 2001, though fishing effort (expressed in terms of number of tows) was down by roughly the same proportion, and 103 t in 2002[284]. While environmental factors cannot be excluded as a cause, the precautionary approach to fisheries set out in Annex II to the Agreement for the Implementation of the Provisions of the United Nations Convention on the Law of the Sea of 10 December 1982 relating to the Conservation and Management of Straddling Fish Stocks and Highly Migratory Fish Stocks (hereinafter the UN Fish Stocks Agreement)[285] dictates that it should be presumed to reflect a decline in resource size. The low catches in recent years suggest that even though the two Governments thought they were being precautionary when they set their TAC at 2 400 t, the maximum sustainable yield (MSY) may well be appreciably smaller than that, and the remaining catch available for the fishdown phase[286] may also be quite limited. The TAC was therefore reduced to 1 800 t in 2002. In addition, consequent on New Zealand’s acceptance that there was in fact one stock straddling the AFZ boundary, Australia’s domestic management zone for the fishery was extended to include the area of the South Tasman Rise lying within the AFZ.


The further fall in catch in 2002 prompted the Governments to reduce the TAC again more sharply in 2003. The Australian industry had funded surveys of orange roughy aggregations during the winter-spawning seasons in 2000 and 2001. Scientific advice obtained by the industry suggested that the data could be interpreted as showing that the stock retained a fish-down capacity for some years at 900 t per annum, with MSY around 300 t. There is a belief in some sections of the industry that the movements of the stock are not necessarily governed by a yearly cycle; the large aggregations found in 1997-1999 could, on this view, be expected to return at some point in a few years, none the worse for the fishing that had already taken place. While this hypothesis is considered in most quarters to be improbable, there is potential for wastage of the resource should it actually be correct[287], and the two Governments thought it reasonable to make some provision for that contingency, as described below.

Ministers agreed in 2002 that the current annual management cycle should be reviewed with the aim of establishing a three-year cycle. Their 2003 decision implementing such a cycle is formally an adjustment of the TAC in accordance with Paragraph 4 of the 2000 Arrangement, though such is its significance that it is all too easy to slip into the error of describing it as an "amendment".

The contingent nature of the TAC under this adjustment operates in two ways - one within a season and the other between seasons. Viewed in one way, the decision starts with a TAC for 2003 of 800 t and maps out a path for reducing it in equal steps to 200 t in 2006. Viewed another way, the decision offers a means for keeping the TAC at 1 300 t for this and the next three seasons. There is also a fixed number of possible intermediate TAC outcomes.

In the first season there are only the two states of 800 t and 1 300 t - and the parallel with Schrödinger’s cat holds. It holds in future seasons too, even though the number of possible TAC levels grows to four in 2004, six in 2005 and seven in 2006. This is because, by the time each season starts, enough information will be known to reduce the number to two.

How is this done? The base case is that, starting from 800 t in 2003, the TAC falls in 2004 and each subsequent season by 200 t, until in 2006 it reaches 200 t. This reduction is however postponed for a year, if in the season just ended the trigger mechanism has been activated and the catch exceeds the original TAC by at least 100 t[288]. It is this which explains the proliferation of possible TAC levels. That is, seen from 2003, in 2004 the initial TAC will be either 800 t or 600 t, and the triggering of a further 500 t from either starting point will generate another two possibilities (1300 t and 1100 t respectively). Similarly, in 2005 there are again these four possibilities plus (if the initial 2004 TAC was 600 t) 400 t and 900 t, making six. A seventh is then added in 2006 (200 t, if the initial 2005 TAC was 400 t).

The extra 500 t is an adjustment made to the TAC mid-season if a large aggregation of orange roughy is demonstrated to have returned to the South Tasman Rise in any of the next three seasons, by three eighths of the TAC (i.e. 300 t, 225 t or 150 t, depending on whether the TAC at the start of the season is 800 t, 600 t or 400 t) being taken in a number of consecutive observer-verified tows (including tows with nil catch) small enough to yield a mean catch per tow of 2 t or more. If this condition is met, the additional tonnage can become available for that season, but this is neither automatic nor immediate. Rather, the increase takes effect only once 75 percent of the original TAC (600 t, 450 t or 300 t respectively) has been or appears likely to be caught. This ensures that the additional tonnage does not become available unless it is likely to be needed, which may not be the case if the catch rate subsequently falls (It would also call into question the higher abundance it initially suggested.[289]) The mechanism is deactivated for the 2006 season if the operation of the decision as a whole results in a 200 t TAC in that season, so that there are seven possibilities for that season’s TAC, not eight.

Finally, the 2003 decision provides for review of the TAC in 2006. Under the 2000 Arrangement, the default position if the parties are unable to reach accord on a TAC is that the previous season’s TAC applies again, i.e. the TAC for the 2007 season would be the same as that of 2006 determined under the 2003 decision[290].

The contingent TAC appears to be without precedent, although there are examples of contingent national allocations depending on the size of future TACs. One such is a 1996 ICCAT Recommendation on bluefin tuna in the Western Atlantic[291] (ICCAT 1997, p. 82), which fixed quotas in absolute terms for the relevant years, but went on to set out two alternative quota-sharing ratios among Canada, Japan and the United States depending on whether the quota for 1999 was between 2 350 t and 2 660 t, or above 2 660 t.


Another possible service that the South Tasman Rise orange roughy fishery has done the international fisheries law community lies in raising, though not really answering, the question of the burden of proof as to whether a stock is a straddling stock, a matter on which both the United Nations Convention on the Law of the Sea[292] (LOSC) and the UN Fish Stocks Agreement are silent.

Though not incontrovertible, all available scientific evidence indicates that the South Tasman Rise orange roughy fishery is based on a single discrete stock that straddles the AFZ boundary. A joint Australian/New Zealand study during 1998 examined stock structure through fish samples from summer commercial catches and from three winter (spawning season) research cruises. Genetic analyses carried out by the National Institute of Water and Atmospheric Research (New Zealand), and otolith microchemistry studies by the Commonwealth Scientific and Industrial Research Organisation (Australia) found no differences between South Tasman Rise fish inside and outside the AFZ, indicating a common straddling stock. The principal recommendation of a joint scientific workshop to discuss research findings held in Wellington in December 1998 was that the South Tasman Rise fishery for orange roughy should be managed as a single discrete stock.

But what if New Zealand had continued to deny that the fish of the same species taken on either side of the AFZ boundary belonged to the same stock?[293] The 1998 Arrangement took an indirect approach to the onus of proof, contemplating (Paragraph 23) that a wider negotiation would follow "if as a result of the collaborative scientific work undertaken under [the 1998 Arrangement] the preponderance of the evidence indicates that the orange roughy stock on the South Tasman Rise is a straddling stock occurring both within the AFZ and in the adjacent high seas area". This peculiar formulation both assumes through the use of the singular that there is a single stock and yet hints that the onus lies on the state asserting that the stock in question straddles the outer limit of its exclusive economic zone (EEZ) to prove this.

If the matter were ever litigated under the dispute settlement provisions of the LOSC or the UN Fish Stocks Agreement, it would seem that the state asserting that there is a single stock straddling its EEZ boundary must either prove that element of its case or argue as a preliminary point that the onus of disproving it should fall on the denying state.[294]

A possible argument that the asserting state might use in this situation is that once fish of the same species are found on both sides of an EEZ boundary in its vicinity, a presumption that they belong to the same stock straddling the boundary is more in keeping with the precautionary approach to fisheries management mandated by Article 6 and Annex II to the UN Fish Stocks Agreement. The actual text of these provisions is, however, not of much assistance here. On the one hand it may be precautionary to presume that fishing on one side of a line will have some effect on what is possibly the same stock on the other side of the line. On the other hand, two smaller stocks will each be more vulnerable to overfishing than a single larger stock that is an amalgam of the two, so that precautionary managers should presume that a stock which may straddle a boundary is in fact two separate stocks until the evidence suggests otherwise. What little literature there is on the question appears to prefer the latter approach (Molenaar 2001).

There is relatively little state practice on the question. The best examples, though not cast in terms of the onus of proof as such, may be the opportunistic stances taken in this regard by the states that subsequently became parties to the 1994 Convention on the Conservation and Management of Pollock Resources in the Central Bering Sea.[295] Before this Convention was negotiated, the United states took the view that the pollock found throughout the Bering Sea constituted a single stock, while Japan maintained that there were three separate stocks, the range of one of them corresponding with the high-seas area in the middle of that Sea to which the Convention relates. If correct, the Japanese view would mean that LOSC Article 63(2) would be inapplicable, leaving the United States with no automatic right to participate as a coastal state in the negotiation of the Convention, and none under Article 118 either, as its vessels had not fished in the high-seas area. Though there were practical and political reasons for not seeking to exclude it from the negotiation, the fact that the United States did so participate, and subsequently became party to the Convention, would support the contention that the onus of proof was on the party denying that a stock straddles an EEZ boundary. One factor pointing the other way is that under the Convention the assessment of the biomass of pollock in the high-seas area may in some circumstances be done by reference to a specific area in the vicinity of Bogoslof Island lying within the United States EEZ, clearly implying that the stock moves between the high-seas part of the Bering Sea and that area, to which the United States had extended domestically the moratorium on high-seas fishing.[296]

Approximate extent of South Tasman Rise extending beyond Australia's EEZ

A truly precautionary test could be formulated as "Which assumption when used as a basis for conservation measures, should its falsity later become evident, will be less likely to have damaged the stock(s)?", but this is a question that may be best answered by scientists on a case-by-case basis, and does not lend itself readily to distillation into a legal rule.


During the 1999 moratorium four vessels flagged to third states caught significant amounts of orange roughy just outside the AFZ, desisting only as a result of diplomatic pressure by Australia on the flag states, citing the duty of cooperation in Article 63, Paragraph 2 and Articles 117 and 118 of the LOSC (Molenaar 2001, Staples 2002). Although this fishing was prompted by knowledge that orange roughy were there for the taking, recent catch statistics provide much less of an incentive for third-state vessels to make the long journey to the South Tasman Rise purely on speculation.[297] What if a similar incident were now to recur, however?

It does not seem open to serious question that the 2000 Arrangement is a subregional arrangement within the meaning of Article 8, Paragraph 3 of the UN Fish Stocks Agreement. The third-party provisions of the 2000 Arrangement indeed speak of admitting new participants on the basis of a "real interest", although what realistic expectation they can have of gaining a share of the TAC is open to question, particularly if it ultimately falls under the 2003 decision to 200 t, which as noted above could be tantamount to a ban on targeting orange roughy. At any rate, it would seem reasonable to assume that Australia and New Zealand would both prefer third states not to enter the fishery, at least until such time, if ever, as there is strong evidence that the MSY is a great deal more than 300 t a year. New Zealand could be expected to have the additional concern that, to the extent that any of Australia’s catch is taken within the AFZ, Australia would have a cogent argument based on LOSC Article 116(b) that any reallocation in favour of third states should come out of the high-seas share, disproportionately affecting New Zealand.

The proposition advanced in this section is that a ban on fishing altogether for demersal species could be enforced unilaterally by Australia by virtue of the South Tasman Rise being part of Australia’s continental shelf. Since 1994 the outer limit of Australia’s continental shelf has been defined by reference to Article 76 of the LOSC[298], which makes the depiction of that limit on a map difficult, given that Australia has yet to make its submission in this regard to the Commission on the Limits of the Continental Shelf - though by November 2004 Australia will have done so if it adheres to the original 10-year rule in Article 4 of Annex II to the LOSC[299]. A rough idea of the shelf’s extent can, however, be gained from material prepared by Australia’s National Oceans Office in relation to the draft South-East Regional Marine Plan (Figure 2), which shows that it encompasses the whole of the area covered by the 2000 Arrangement with a great deal to spare.

Part XII of the LOSC allows coastal states to take measures to prevent damage to fragile ecosystems found on their continental shelves.[300] Orange roughy aggregations tend to occur on seamounts and are typically exploited by methods which involve nets coming into contact with - and damaging - the bottom. Seamounts are features of great biological diversity, which has led to calls for a moratorium on trawling on such elevations.[301] It is submitted that it would be open to a state to prohibit activities that cause, or could reasonably be expected to cause, damage to the environment of the seabed of its continental shelf. As the coastal state’s jurisdiction over fisheries does not extend beyond the EEZ, however, it will be severely constrained in the design of its measures. In particular, it would be dangerous for the prohibition to be directed solely against fishing vessels. Still less would it be possible for the prohibition to extend only to foreign vessels or activities. Such a measure would in effect be claiming an exclusive right for the coastal state to damage the marine environment, which would be a perversion of the purpose of Part XII of the LOSC and contrary to Article 227.[302] In a practical sense, the blunt instrument of a total ban may be all that Part XII allows.

Accordingly, a coastal state wishing to impose less restrictive measures tailored to the operation of fishing vessels, particularly those intended to place its own vessels at an advantage, would need to find justification for these, if it can, in the fisheries provisions of the LOSC or in the UN Fish Stocks Agreement. The exclusive jurisdiction that the coastal state has over sedentary species on its continental shelf under the LOSC Article 77, Paragraph 4 may not be sufficient. A ban only on foreign fishing, which would entail a claim by the coastal state to the exclusive right to damage its own sedentary species fisheries, seems not as obviously contrary to principle as the same claim in respect of the seabed environment, but only the brave would predict with confidence that it would withstand a challenge based on the abuse of rights limb of the LOSC Article 300. For one thing, its practical effect would be to reopen the controversial question of whether demersal fish species are included within sedentary species, clearly settled in the negative by the 1958 Convention on the Continental Shelf[303], whose Article 2, Paragraph 4 is in almost identical terms. Banning demersal fishing in a well defined location on the continental shelf outside the EEZ may be defensible where it threatens a known sedentary species fishery, but would be much less so in respect of a wide area on the off-chance that it may contain sedentary living resources.


It is perhaps one of the ironies of international fisheries law and management that, although developed to settle a disagreement about fish that may well no longer be there, the Australia-New Zealand South Tasman Rise orange roughy arrangement of 2000, together with aspects of the decisions taken under it, can nonetheless serve as a model for other straddling stocks. The superposed alternative catch limits in the 2003 decision on TAC in particular can be seen as one of the first attempts to manage a fishery in accordance with Annex II to the UN Fish Stocks Agreement. Although not expressed in terms of target and limit reference points - too little is known about the stock for managers to be able to set these with any confidence - the decision is replete with reference points for "trigger[ing] pre-agreed conservation and management action" for which Paragraph 4 of that Annex calls. This innovative "resolve-as-you-go-through-feedback" approach could become a standard way of accommodating opposing views on abundance and desirable catch outcomes dependent on these. It also serves as a neat illustration of the fact that, while the precautionary approach to fisheries cannot completely eclipse the coastal state’s obligation under LOSC Article 62, paragraph 1 to promote the optimum utilisation of its EEZ resources, it does now cement utilisation into a rung in the hierarchy of norms that is subordinate to the one occupied by conservation.

Where the fishery takes place in waters superjacent to an extended continental shelf, Part XII of the LOSC may provide the coastal state with a blunt instrument to close the fishery altogether in order to protect the seabed environment, but does not extend the ordinary regulatory powers available to it in its EEZ to the entire fishery, at least unless all reasonable efforts to cooperate with other states have failed.

If nothing else, the events on the South Tasman Rise during the 1999 catch limit hiatus show that, however imperfect management measures may be, they are far better than not having any. While it would be tempting to conclude that no management decision rule should henceforth be regarded as complete unless it includes a deadlock-breaking formula for deriving one season’s TAC and national allocations from the previous season in the absence of a decision to the contrary, it may be futile to insist on this in circumstances where to do so would prevent the adoption of such a rule in the first place.

Finally, a lesson from the ad hoc moratorium that followed the hiatus is that, irrespective of the legal opposability of such a measure to third states, it is of no practical value unless they know of it. Because the moratorium was not widely publicised, the flag states of the vessels that fished the South Tasman Rise during its currency could reasonably say that they did not know the area was closed to fishing (and one did so: see Willing 2002, page 6)[304]. Although owing a duty of cooperation to states already managing a straddling stock under Articles 63, 117 and 118 of the LOSC and now also Article 7, paragraph 1 of the UN Fish Stocks Agreement, as a practical matter the only way for them to prevent fishing by their vessels, which is surely preferable to having to rein them under pressure once fishing has begun, is if the measures with which they are expected to cooperate are brought to their attention in good time.


Burke, W.T. 1984. Highly Migratory Species in the New Law of the Sea. Ocean Development and International Law. 14(3): 273-314.

Caton, A. (Ed) (2003). Fisheries Status Reports 2003: Resource Assessments of Australian Commonwealth Fisheries. Bureau of Rural Sciences, Canberra.

Holt, S.J. 1979. "Marine Fisheries" in E. Mann Borgese (Ed.) Ocean Yearbook 1 (Chicago: University of Chicago Press), 58-82.

International Commission for the Conservation of Atlantic Tunas 1992. Report for biennial period, 1990-91 Part II (1991).

International Commission for the Conservation of Atlantic Tunas 1997. Report for biennial period, 1996-97 Part I (1996).

Kaye, S.B. 2001. International Fisheries Management (The Hague; London: Kluwer Law International).

Molenaar, E.J. 2001. he South Tasman Rise Arrangement of 2000 and Other Initiatives on Management and Conservation of Orange Roughy. International Journal of Marine and Coastal Law. 16(1): 77-124.

Serdy, A. 2004. One Fin, Two Fins, Red Fins, Bluefins - Some Problems of Taxonomy and Nomenclature Affecting Legal Instruments Governing Tuna and Other Highly Migratory Species. Marine Policy. 28(3): 235-247.

Staples, D. 2002. Management of Shared Fish Stocks - Australian Case Studies. In Papers Presented at the Norway-FAO Expert Consultation on the Management of Shared Fish Stocks - Bergen, Norway, 7-10 October 2002 <>.

Willing, J. 2002. Arrangement between the Government of Australia and the Government of New Zealand for the Conservation and Management of Orange Roughy on the South Tasman Rise. In Papers Presented at the Norway-FAO Expert Consultation on the Management of Shared Fish Stocks - Bergen, Norway, 7-10 October 2002. <>.


The 2003 TAC decision

Extract from letter dated 25 June 2003 from the Hon Pete Hodgson MP, New Zealand Minister of Fisheries, to Senator the Hon Ian Macdonald, Australian Minister for Fisheries, Forestry and Conservation

[The Minister begins by rehearsing the recent history of the management of the South Tasman Rise (abbreviated in the letter to STR) since the conclusion of the 2000 Arrangement. He continues:]

The limited catches taken on the STR over the last two seasons and previous scientific advice suggest a declining orange roughy stock and has [sic] given cause for concern. In order to responsibly manage the stock based on scientific advice while creating some flexibility should the fish return in large quantities, we have provided a two-tiered approach for the next three years. I propose that the following arrangements apply.

The TAC for the 2003/04 fishing season will be set on 14 July 2003 at 800 tonnes, an overall reduction of 1000 tonnes. If the fish do not return in large quantities the TAC will be incrementally decreased annually to 200 tonnes in 2006-07. If the fish return in large quantities there will be an additional 500 tonnes of Trigger TAC available to industry. A detailed description of the TAC reduction and Trigger TAC mechanism is presented in Appendix II. The TAC will still be divided between Australia (75%) and New Zealand (25%) as stated in the 2000 [Arrangement].

These management arrangements will apply unless we mutually decide that a review of these arrangements is appropriate in accordance with paragraph 4 of the 2000 Arrangement. In addition, I propose that officials ensure information is exchanged as frequently as is necessary to achieve the above in a timely way. Both countries shall also prepare and exchange, by 1 March each year, a consolidated annual report detailing fishing information and scientific research from the previous fishing season.

In any event, I propose that officials from both our countries meet during the 2006-07 season, to decide on management arrangements for the next three-year cycle. In accordance with our mutual objective to retain flexibility, a future arrangement should take account of best science.

[The Minister goes on to suggest measures that a future arrangement might include, before concluding:]

I look forward to receiving confirmation from you that this proposal is acceptable to the Australian Government and that this letter, together with your reply to that effect, will constitute an understanding between the Governments of New Zealand and Australia. In that case, I ask that Australian officials notify the Fisheries Division of the United Nations Food and Agricultural Organisation on behalf of both countries of the management arrangements contained in this understanding.

Yours sincerely


Hon Pete Hodgson
Minister of Fisheries

Extract from reply dated 14 July 2003 from Senator the Hon Ian Macdonald, Australian Minister for Fisheries, Forestry and Conservation to the Hon Pete Hodgson MP, New Zealand Minister of Fisheries

[The Minister refers to his New Zealand counterpart’s letter and continues:]

I am particularly pleased that the proposal put forward by officials recognises the special circumstances of this orange roughy fishery. I know that from the perspective of the Australian industry, the proposal for a longer-term arrangement is appreciated as it allows industry some certainty in managing their fishing operations in a fishery where the ecology of the target species is not clearly understood.

[...] I support the introduction of these arrangements from Monday 14 July 2003. I also note that that these arrangements will apply unless we mutually agree to review them and that officials from both countries will, in any event, meet during the 2006-07 season to decide upon management arrangements for a further three years.

I therefore confirm that the arrangements as detailed in your letter and its two enclosures are acceptable to the Australian Government and that your letter, together with this response, constitute an understanding between the Governments of Australia and New Zealand.

[a paragraph is omitted on the arrangements to apply beyond 2006-07, the Minister suggesting that they be negotiated in view of the state of the fishery at that time]

As requested, I will arrange for Australian officials to notify the United Nations Food and Agricultural Organization of the management arrangements contained in this understanding.

Yours sincerely


Ian Macdonald


Proposal for the Total Allowable Catch of Orange Roughy on the South Tasman Rise


Total allowable catch (TAC) will be set at 800 tonnes (represents a reduction of 1000 tonnes)[305]. An additional 500 tonnes of catch will be available in the form of a Trigger TAC if the trigger conditions are met[306].


TAC will be set at either 800 tonnes or 600 tonnes depending on the level of catch in the previous season[305]. An additional 500 tonnes of catch will be available in the form of a Trigger TAC if the trigger conditions are met[306].


TAC will be set at 800 tonnes, 600 tonnes or 400 tonnes depending on the level of catch in the previous seasons[305]. An additional 500 tonnes of catch will be available in the form of a Trigger TAC if the trigger conditions are met[306].


TAC will be set at 800 tonnes, 600 tonnes, 400 tonnes or 200 tonnes depending on the level of catch in the previous seasons[305]. An additional 500 tonnes of catch will be available in the form of a Trigger TAC if the trigger conditions are met[306]. Parties will meet during 2006 to develop [an] arrangement for the next three years.

Total Allowable Catch (TAC) of Orange Roughy on the South Tasman Rise

[276] The standard disclaimer is made. The author records here in addition his gratitude to Erik Jaap Molenaar and Richard Tilzey for helpful comments on earlier drafts; any remaining errors of fact, law or judgement, however, are solely his own.
[277] Confusingly bearing the same title as its 1998 predecessor cited in Note 3, this Arrangement is reproduced as an appendix in Molenaar 2001.
[278] Although they are not completely identical, for present purposes the term "Australian Fishing Zone" may be used interchangeably with the Australian Exclusive Economic Zone (EEZ); for the precise definition see s.4(1) of the Fisheries Management Act 1991 (Cth).
[279] Arrangement between the Government of Australia and the Government of New Zealand for the Conservation and Management of Orange Roughy on the South Tasman Rise, obtainable from the website of the Australian Department of Foreign Affairs and Trade: <>.
[280] See the slightly conflicting accounts from each side of the Tasman in two papers presented at the Norway-FAO Expert Consultation on the Management of Shared Fish Stocks (Staples 2002; Willing 2002). Some of the views attributed to Australia in the latter, however, appear to mistake the Australian industry’s wishes for Government policy. In particular, Australia did not claim "exclusive rights to catch and manage the orange roughy fishery on the high-seas area of the South Tasman Rise" (Willing 2002, page. 3). Rather, Australia pointed out that, in line with its (since vindicated) view that the stock did straddle its EEZ boundary, by Article 116(b) of the United Nations Convention on the Law of the Sea fishing of the stock in the adjacent high-seas area is subject to the rights, duties and interests of Australia as the coastal State, as provided in Article 63(2). The primary obligation of States concerned under the latter provision is to seek to agree on measures necessary for the conservation of the straddling stock. Should such agreement not be possible despite their best efforts, it may be surmised that Australia - or indeed any coastal State - would find attractive the view of Professor Burke that the coastal State may proceed to regulate the entire fishery, though not enforce its laws beyond 200 miles (Burke 1984). Now that the UN Fish Stocks Agreement is in force, however, there is room for argument over whether, as among its parties, this would be supplanted by the compulsory availability of provisional measures through the dispute settlement procedures in Article 7(5). It is in this sense that the passage "Australia claims the manage the orange roughy fishery as a straddling stock" (Staples 2002, p. 4; note that the claim is not described as "exclusive") should be understood. Note also that Willing’s Figure 2, distinguishing diagrammatically among straddling stocks depending on whether their biomass is mostly inside the EEZ, mostly outside the EEZ or approximately evenly distributed across the boundary, has no basis in law: all are equally captured by Article 63(2). This is not to say that the distinction lacks practical significance; in the extreme case, if the coastal State does not actually fish the stock within its EEZ, because its vessels find the catch rates on the high seas better, and has no, or insufficiently specific, measures in place to benefit from the non-undermining rule of UN Fish Stocks Agreement Article 7(2), this could render hollow the legal advantage of Article 116(b)
[281] States’ differing domestic legislative capacities to regulate unilaterally their own fishing on the high seas can produce an anti-precautionary effect on bargaining power. The position at the time was that Australia did, but New Zealand did not, have power at domestic law to regulate the fishing activities of its nationals on the high seas in the absence of an agreement or understanding with another country: Fisheries Management Act 1991 (Cth), s.8; Fisheries Act 1996 (NZ), s.297(1)(o) - see now however s.297(3) of the latter, as amended by the Fisheries Act 1996 Amendment Act No 2 1999 (NZ). Thus, Australia was able to hold its catch to its 1800-tonne quota in 1999 even in the absence of a renewal of the 1998 Arrangement, while the New Zealand authorities could do nothing to prevent the overcatch despite strong entreaties to the industry at Ministerial level. Had New Zealand sought to exploit this imbalance, it would have put Australia in the difficult position of having to seriously contemplate taking an anti-precautionary step of its own in order to restore the bargaining equilibrium. Where two countries are less closely attuned in terms of culture and fisheries management philosophy, it is easy to see how this can lead to undesirable outcomes.
[282] Paragraph. 3 of the 2000 Arrangement.
[283] Although this represents little more than half of the notional New Zealand overcatch had the negotiated limit been in effect, it is more than a year’s worth of quota. As far as the author is aware, this proportion is unsurpassed in international fisheries practice by way of settlement of a disagreement about overfishing. The repayment entailed displacement of the default rule in Paragraph. 7 of the 2000 Arrangement, which sets up a system resembling the board game Monopoly. The stock itself acts as the bank, from which Australia and New Zealand collect their respective equivalents of $200 (Australia 75 percent of the TAC, New Zealand 25 percent) every time they "pass Go" (1 March), with their actual catch limit for the season adjusted to reflect overcatch or undercatch in the previous season unless the Parties jointly decide otherwise: see Paragraphs 7 and 8. They did so decide in the case of the 640 t, recording it in a further exchange of letters. The stringent standard set by Australia and New Zealand in Paragraph 7, whereby a party’s "annual catch limit" takes account of any overcatch in the previous season by debiting against its quota one tonne for each of the first hundred tonnes of excess catch, and two tonnes for each tonne of excess catch thereafter, is decidedly more demanding than, say, the practice that has developed in the International Commission for the Conservation of Atlantic Tunas (ICCAT), starting with the "Recommendations (made in 1991) for Enhancement of the Current Management for Western Atlantic Bluefin Tuna" (Annex 7 to Proceedings of the Twelfth Regular Meeting of the Commission, Madrid, November 11-15 1991), where overcatch is repaid not in the following year but the subsequent one, initially with no penalty rate and subsequently for some stocks with a 25 percent penalty: "if the catch of [a relevant State] exceeds its annual or biannual scientific monitoring quota, then in the biannual period or year following reporting of that catch to [the Commission], that [State] will reduce its catch to compensate in total for that overage" (ICCAT 1992, page. 67; emphasis added); for the 25 percent penalty, which is not automatic and applies only if quota is exceeded during two consecutive management periods, see the "Recommendation by ICCAT Regarding Compliance in the Bluefin Tuna and North Atlantic Swordfish Fisheries" (Annex 5-14 to Tenth Special Meeting of the Commission, San Sebastian, November 22-29, November 1996) (ICCAT 1997, page. 95). Paragraph 8 in effect provides that if the overcatch formula yields a negative annual catch limit, the party concerned must refrain from fishing for as long as necessary to work off its "debt" to the stock. This too bespeaks a more serious determination to prevent overcatch than any other fisheries instrument known to the author.
[284] The entire catch was taken in both years by Australian vessels, the New Zealand fleet apparently deterred by the small quota remaining for them after factoring in the annual 100-tonne repayment to the stock.
[285] 2167 United Nations Treaty Series 3.
[286] The fishdown phase is described in the following terms on the Orange Roughy Management Company’s website <> (visited on 7 October 2003): "Once a new fishery has been located and assessed, catch limits are set at a level that will reduce the biomass size down to the BMSY level [i.e. that yielding the MSY]. This involves reducing each stock down to around 30 percent to 44 percent of B0 [virgin biomass] through the setting of catch levels that are above sustainable levels during the fishing down transition. Once the stock size is estimated to have reached BMSY, the [catch limit] is set at a lower level to maintain catches at long-term sustainable levels." In theory the rate at which a stock is reduced from B0 to a biomass approaching BMSY should not affect the stock, though if the annual catch in this phase significantly exceeds the MSY, then the necessary subsequent reduction to a sustainable level may be politically difficult for managers to achieve, particularly if a level of fishing capacity for the fishdown phase has been permitted to emerge that represents overcapacity once this phase is over.
[287] The concept of waste has been called into question on the basis that the nutrient value of fish that die of old age is returned to the ecosystems as they decompose (Holt 1979, page. 81). It may be surmised that this recycling would be faster in shallower waters.
[288] The intertwining of the decision rules may mislead hurried readers of the decision who go no further than this to form the impression that the parties are inviting and rewarding overcatch. This is assuredly not the case. For postponement of the reduction, the TAC must first have been increased by 500 t under the trigger rule, so that the true threshold is undercatch of less than 400 t - the rationale being that greater undercatch would be evidence that the abundance may not after all be as high as the fulfillment of the trigger rule would have suggested.
[289] This is consistent with the rationale outlined in the previous footnote.
[290] Paragraph 4 provides a mechanism to "vary" the TAC, with the necessary implication that it continues unvaried from year to year unless the mechanism is activated. It will not have escaped the parties’ notice in drafting this clause that, had they included a similar provision in their 1998 Arrangement, the problems that beset them in 1999 could have been avoided.
[291] "Recommendation by ICCAT to Establish a Scientific Monitoring Quota for Bluefin Tuna in the Western Atlantic for 1997-1998" (Annex 5-4 to Tenth Special Meeting of the Commission, San Sebastian, November 22-29, 1996)
[292] 1833 United Nations Treaty Series 3.
[293] One short-term consequence of New Zealand’s initial denial was that Australia felt no obligation to advise New Zealand of its management arrangements for the orange roughy fishery within the AFZ. This was because New Zealand would have a legitimate interest in knowing these, as set out in Article 7(7) of the UN Fish Stocks Agreement ("Coastal States shall regularly inform States fishing on the high seas in the subregion or region...of the measures they have adopted for straddling fish stocks and highly migratory fish stocks within areas under their national jurisdiction."), only if it accepted that the stock did indeed straddle the boundary. In the abstract, a coastal State in this position will also need to take into account the risk that adhering too rigidly to this stance might suggest a lack of confidence in its scientific case. As noted above (text following note 10), however, in this instance Australia’s measures applying in the high-seas area of the South Tasman Rise were extended to the AFZ in 2001.
[294] Note that there is no comparable problem with highly migratory species, because Article 64 of the LOSC creates (in Annex I) a closed list of these species, although there may be some uncertainty at the margins (described in Serdy 2004).
[295] For the text of this Convention see (1995) 34 International Legal Materials 67. The remainder of this paragraph draws on Kaye’s recent analysis of the negotiations (Kaye 2001, at 323ff).
[296] Japan’s view (shared by China) that, while the moratorium continued, the coastal States should take appropriate conservation measures in their EEZs (Kaye 2001, at 350n) must either have rested on the assumption that there was but one stock - inconsistently with its former position - or been an attempt to call the United States’ bluff. The fact that the United States did not do so until urged does not, however, necessarily undermine its own one-stock stance; it is explicable as an application of coastal State priority as reflected in the UNCLOSC Article 116(b).
[297] If the tonnage taken by these vessels was at the higher end of the range in the anecdotal reports (Staples 2002) and the subsequent low catches reflect depletion of the stock, then there may in fact be a cause-and-effect relationship between them.
[298] Seas and Submerged Lands Act 1973 (Cth), ss 3(1) and 12 (as amended by the Maritime Legislation Amendment Act 1994 (Cth)).
[299] The Eleventh Meeting of States Parties to the UNCLOSC decided in May 2001 to extend the deadline to May 2009 for all States that would otherwise have faced an earlier deadline: Decision regarding the date of commencement of the ten-year period for making submissions to the Commission on the Limits of the Continental Shelf set out in Article 4 of Annex II to the United Nations Convention on the Law of the Sea, UN doc SPLOS/72 (29 May 2001).
[300] Although the focus of Part XII is pollution, and the general provision (Article 208(1)) is expressed in terms of coastal States "adopt[ing] laws and regulations to prevent, reduce and control pollution of the marine environment arising from or in connection with sea-bed activities subject to their jurisdiction...", Article 194(5) provides that measures taken in accordance with this Part "shall include those necessary to protect and preserve rare or fragile ecosystems as well as the habitat of depleted, threatened or endangered species and other forms of marine life."
[301] This was for example a theme common to the conclusions presented by many of the working groups formed at the recent Workshop on High-Seas Biological Diversity, Cairns, 16-20 June 2003, and is the first short-term option listed in the Summary Record of Discussion and Suggestion for a Way Forward, available online at <>.
[302] This Article provides that "In exercising their rights and performing their duties under this Part, States shall not discriminate in form or in fact against vessels of any other State."
[303] 499 United Nations Treaty Series 312.
[304] Constructive notice can be achieved for measures contained in a treaty by prompt registration of the treaty under Article 102 of the Charter of the United Nations, but no equivalent procedure is available for instruments of less than treaty status. The 2000 Arrangement therefore provides for the Parties to lodge it with the FAO for circulation to all the latter’s Members (see Paragraph 35), as does the 2003 TAC decision.
[305] TAC Rules
The total allowable catch (TAC) will initially be set at 800 tonnes in 2003-04. The TAC in following seasons will be reduced by 200 tonnes each season until reaching 200 tonnes in 2006-07 at the end of this Arrangement.
The preceding paragraph will apply, unless the total catch taken from the entire area of the STR by vessels authorised by either country to operate on the STR (‘combined total catch’) by 1 February is greater than or equal to 100 tonnes above the TAC, in which case the TAC will not be decreased in the following fishing season.
[306] Trigger TAC Rules
A Trigger TAC of 500 tonnes will be available if the average catch rate has reached two tonnes per tow. The average catch rate will be calculated from a sample of three eighths of the original TAC for the season taken from consecutive, observer-verified tows, including any tow that produces zero tonnes of catch. Only observed tows undertaken by vessels authorised by either country to operate on the STR will be taken into account.
The Trigger TAC will be made available from the date on which the combined total catch reaches, or is likely to reach, 75% of the original TAC for the season. A Trigger TAC will not be available for the 2006 season if the TAC is set at 200 tonnes

Previous Page Top of Page Next Page