December 9, 2008

Mr. Daniel Furlong
Mid-Atlantic Fishery Management Council
Room 2115 Federal Building
300 South New Street
Dover, DE  19904-6790

Dear Mr. Furlong:

At its December, 2008 meeting, the Mid-Atlantic Fishery Management Council (the “Council”) will make a number of decisions that, taken together, will determine the shape of the recreational summer flounder fishery in 2009, and could determine the fate of that fishery not only next year, but well into the future.  Out of all of such decisions that the Council must make, the question of whether to continue the failed policy of setting regulations on a state-by-state “conservation equivalency” basis or replace it with coastwide, or perhaps mandatory regional, management measures is paramount.  As Coastal Conservation Association New York (“CCA NY”) will demonstrate in the remainder of this letter, “conservation equivalency” is a concept based on a false premise, which is impossible to adequately administer, fails to protect the summer flounder stock and causes significant hardship to anglers and angling-related businesses along most of the coast.  Through the use of a single year’s questionable data, “conservation equivalency” has created a windfall for a single state, while harming those in other jurisdictions. 
It is time for a new approach. 

I
State-by-state “conservation  equivalency” is a demonstrably failed approach to summer flounder management

            Recent data from the Marine Recreational Fisheries Statistics Survey (“MRFFS”) indicates that overfishing was again widespread in the recreational summer flounder fishery.  By the end of August, 2008, every state with such a fishery, other than Delaware, Virginia and North Carolina, had already exceeded its allocation
In a 2007 letter to W. Peter Jensen, then Chairman of the Council, Dr. William T. Hogarth, former Assistant Administrator for Fisheries for the National Oceanic and Atmospheric Administration and former director of the National Marine Fisheries Service (“NMFS”), stated that “while the concept of allowing states to craft management measures designed to achieve the necessary recreational harvest reduction to be equivalent to Federal measures is both appealing and has strong merits, it has not served the rebuilding efforts well.”   Patricia Kurkul, the Northeast Regional Administrator for NMFS, echoed Hogarth’s sentiments, stating that “The 2007 state-by-state conservation equivalency management measures have not been effective in constraining harvest.”   Data shows that the 2008 conservation equivalency management measures resulted in overfishing that was at least as widespread as it was in the previous year.
            Managers now have years of real-world experience that demonstrate the practical difficulty—and perhaps the practical impossibility—of properly constraining recreational harvest through the conservation equivalency approach.  Given both that experience and NMFS’ realization that state by state management is inadequate to constrain recreational harvest, it would be irresponsible for the Council to continue to recommend such an approach to summer flounder management.  Instead, it is time for the Council to recognize that, as Dr. Hogarth suggested, “new approaches” are needed, to “better ensure that the recreational targets are not exceeded” in 2009 and beyond. 

II
The current state-by-state approach is not supported by reliable data

            It has been said that one cannot build a solid house on a poor foundation, and that notion reflects the fatal flaw of the conservation equivalency approach.  Under state-by-state management, “states are required to adjust effort to achieve landings proportional to their landings from 1998, as reported by MRFSS”, yet MRFSS was never meant to be employed in such a manner.  Doing so is a gross misuse of MRFSS data.
            MRFSS, while the only available estimate of recreational summer flounder harvest, contains serious flaws.  Those flaws have been recognized by the National Research Council, which, in  2006, issued a report critical of that survey.   Unfortunately, managers still use MRFSS inappropriate purposes, including the estimation of harvest at geographical scales finer  than the coastwide scale at which MRFSS was originally intended to operate.   Using a single year of MRFSS data to allocate recreational summer flounder harvest among the states goes far beyond the intended purpose and the capabilities of the MRFSS product.
            The recreational fishing community has generally been very critical of MRFSS, and cannot reasonably argue that any allocation based on a single year of MRFSS data is valid.  Comments made by Capt. Tony Bogan, a frequent spokesman for the for-hire industry in the northern mid-Atlantic, are typical: “The MRFSS numbers are unusable and totally flawed…They admit that there is a standard of error in them that is almost 20 percent either way—over or under.  The trouble is management officials will use these numbers to the last fish and the last pound as if there were no error factor.”   Capt. Bogan, who has served as a member of the Council, has expressed particular disdain for any conclusion drawn from a single year of MRFSS data.  “…MRFSS is trying to do its job,” he noted.  “…It’s a trend analysis and it shows trends in the recreational fishery.  Well, a trend is not one year.  A trend is more than one year…[emphasis added]“   Ray Bogan, counsel for the United Boatmen of New Jersey and New York, has unequivocally stated that “MRFSS is miserably broken” with respect to the determination of single-year harvest in a single state.
            If the survey lacks the requisite accuracy to be reliably used for quota management or to accurately determine the annual recreational harvest in each state, then a single year of MRFSS data also lacks the requisite reliability to be used to determine each state’s proportionate share of the summer flounder fishery.  Without a reliable means to determine states’ shares, the Council’s only reasonable action is to adopt a single, coastwide management measure for summer flounder.

III
Recent landings patterns contradict the assumptions underlying the state-by-state approach

            The assumption underlying the current state-by-state allocation is that each state’s proportional share of the 1998 recreational summer flounder harvest represents what such state would normally harvest if it fished under the same set of regulations as every other state.  Thus, the Council and the Atlantic States Marine Fisheries Commission (“ASMFC”) agreed that states could be permitted to set whatever combination of size limit, bag limit and seasons they believed would best benefit their local fishery, so long as they remained within their allotted share of the recreational summer flounder harvest.  That decision appeared sound at the time that it was made, but experience has shown that, in practice, it has not achieved the desired result, neither adequately constraining harvest nor maintaining each state’s historical share of recreational landings.  The difference between the theoretical results of conservation equivalency and the observed failure of that approach can probably be attributed to two primary factors.

A
Current patterns of local abundance differ markedly from those of 1998

            The first problem arose out of the inherent inaccuracy of MRFSS data when applied on anything less than a coastwide basis, an inaccuracy exacerbated by the decision to project a single year’s MRFSS estimates to predict each state’s harvest years into the future, with no consideration given to year-to-year differences in weather or the apparent change in the local abundance of summer flounder as the stock rebuilds, a change that has made the proportion of the recreational summer flounder harvest landed in each state in recent years very different from the proportion landed in each state in 1998.
            In 1998, each state allegedly landed the following proportion of the summer flounder harvest:
                                    New Jersey                 39.09%
                                    New York                   17.63%
                                    Virginia                       16.69%
                                    Rhode Island                 5.66%
                                    North Carolina              5.60%
                                    Massachusetts               5.49%
                                    Connecticut                   3.75%
                                    Delaware                       3.14%
                                    Maryland                       2.95%

For purposes of the following discussion, such landings will be rounded off to the nearest 0.5% and the landings of Massachusetts, Connecticut and Rhode Island  combined, so that the landings of New Jersey are deemed to be 39.0%, New York 17.5%, Virginia 16.5% and New England 15.0%.
            In the years since 1998, states adopted various regulatory schemes designed to constrain harvest within their allotted shares.  Each year, as harvest figures were compiled and state regulations adjusted in response, patterns began to develop.  New Jersey regulations were consistently among the most liberal on the coast, while those of its immediate neighbor to the north, New York, were usually the most stringent.  Regulations in Virginia and New England generally fell somewhere between those two extremes.   Since each state’s original allocation was based on a time when regulations were uniform along the entire coast, one would assume that states which developed stricter regulations would land proportionately fewer fish than states with more liberal rules.  However, that was not what occurred.  Instead, the proportionate shares of the harvest broke down as follows:

            State                Theoretical       2001    2002    2003    2004    2005    2006    2007
           
            New Jersey     39.0%              36.5%  30.0%  35.5%  37.5%  30.5%  36.6%  35.4%
            New York       17.5%              17.5%  23.5%  37.0%  21.0%  34.0%  22.2%  26.0%
            Virginia           16.5%              23.0%  21.5%  11.0%  13.0%  13.0%  19.2%  15.8%
            New England  15.0%              13.5%  16.5%  11.5%  20.5%  15.5%  16.8%  13.0%

Thus, if the intent of the Council and ASMFC was to perpetuate the proportionate share of the recreational summer flounder harvest that each state landed in 1998, it failed completely.  The only year in which the four major harvesting areas remained somewhat close to their assigned proportionate share of recreational landings was 2001.  Even then, Virginia’s high landings skewed the relationship; after 2001, the change was even more striking.  Despite having some of the most liberal regulations on the coast, New Jersey was never able to land the 39% of the overall recreational harvest that it landed in 1998; conversely, New York, although fishing under the most restrictive regulations on the coast, never saw its recreational landings near 17.5% of the harvest after 2001; instead, they ran consistently high, more than doubling its theoretical share of the harvest in 2003 and nearly doubling it again in 2005.  Recreational harvest in Virginia and the New England states fluctuated on either side of the 1998 share for those regions, with no clear pattern emerging.
            Such patterns suggest that there is no similarity between today’s fishery and that which existed in 1998, when the stock rebuilding process had just begun.  The harvest pattern may well change again, as an expanding flounder population continues to move north and east along the coast.   The dynamics of the summer flounder fishery are constantly subject to change, and only a coastwide or, at the least, a regional management approach will have the inherent flexibility to account for both long-term trends and transient patterns in the fishery. 

B
State-by-state management introduces additional sources of error into the data

            The other problem created by conservation equivalency is that the use of MRFSS data to predict the future harvest of anglers in any particular state allows numerous opportunities for error to enter the calculations, which errors can compound as each state derives its own set of regulations.  Perhaps the most obvious source of such error arises out of the calculation of season length.  Typically, each day within any two-month “wave” is assigned a value representing the total harvest estimate for the two-month period, divided by the number of days in the wave.  That results in a satisfactory estimate so long as the entire wave is included within the open season.  However, once a wave straddles a season opening or closure, error necessarily intrudes.  Using the northeast (New Jersey to Massachusetts) fishery as an example, far more summer flounder are caught in the last two weeks of June than are caught in the first two weeks of May, and far more are landed in early September than in late October.  Yet, for purposes of setting seasons, no allowance for such intra-wave variations are made, and all days of the wave are given equal weighting.  Thus, a season which begins in late May or early June, and ends sometime in September may, on paper, be adequate to constrain harvest but, in reality, is almost certain to lead to overfishing due to the greater daily harvest which occurs at the end of Wave 3 and beginning of Wave 5, when compared to those waves overall.  By eliminating state-by-state management, and going to a coastwide management regime embodying no closed season, such source of overfishing is effectively foreclosed.
            Coastwide management also limits the impact of local abundance of summer flounder and local weather conditions.  It is inevitable that, somewhere on the coast, unusually good weather will coincide with an abundance of summer flounder, and the resultant harvest will push the “fortunate” state well above its annual allocation.  Similarly, other areas will, for reasons of poor weather or a local scarcity of fish, take lower than average numbers of fish in a given wave or even in an entire season.  While such underharvest creates no management problems in the year that it occurs, it does create an issue in the succeeding year, when weather and the availability of fish return to more typical levels and the affected state, now fishing under relatively liberal regulations because of the previous year’s low harvest, substantially exceeds its allocation.  Coastwide regulations, because they are not affected by issues of local abundance or access, remove this source of error from the specification setting process, and makes over-all recreational harvest easier to predict.
            Finally, one must recognize that theoretical calculations of harvest do not take into account patterns of human behavior and other “what if?” factors that can only be identified in retrospect.  For example, it is not unusual for a state to craft sets of alternative management measures, all of which, on paper, adequately constrain harvest.  Some alternative may pair a smaller size limit with a painfully short season, while others impose seemingly punitive size limits but allow anglers to fish throughout the traditionally productive months.  Yet reducing the minimum size may encourage anglers formerly discouraged by an “impossibly high” limit to re-enter the fishery, increasing harvest despite the shortened season; or, at a time when fisheries managers are succeeding in increasing the population size and extending its age and size composition, raising the size limit may merely compel anglers to fish a little longer or in somewhat different areas in order to ultimately land “legal” fish, and might not have the intended effect of significantly lowering landings.  Finally, state fisheries managers do not operate in a vacuum, but are influenced by their constituents, who collectively possess a deep understanding of the quirks of their local fisheries.  Such constituents will reliably urge, and often convince, state managers to elect an alternative that, regardless of any theoretical conservation equivalency, will in reality maximize harvest and probably exceed the state allocation.

IV
State-by-state allocation, as currently administered, unjustifiably benefits anglers and businesses in some states at the expense of those in others

The Magnuson-Stevens Fishery Conservation and Management Act establishes a number of National Standards which govern fisheries management in federal waters.  National Standard Four states that
Conservation and management measures shall not discriminate between residents of different States. If it becomes necessary to allocate or assign fishing privileges among various United States fishermen, such allocation shall be (A) fair and equitable to all such fishermen; (B) reasonably calculated to promote conservation; and (C) carried out in such manner that no particular individual, corporation, or other entity acquires an excessive share of such privileges.
The current state-by-state allocation of the summer flounder harvest arguably violates that national standard in practice, as it causes unnecessary and unjustifiable hardship to anglers and angling-related businesses in some states while allowing anglers and businesses in neighboring states to enjoy a far less restricted fishery. 
            Under the current management approach, we see a single state receive nearly 40% of the entire recreational allocation, more than twice the summer flounder allocated to anglers in any other jurisdiction on the coast.  As a result of receiving such a generous share of the harvest, that state could set its recreational bag limit at 8 fish, with an 18-inch minimum size and a reasonably long, May to September season, while its neighbors, collectively, are allowed barely more than 20% of the harvest, and must impose far stricter regulations.  It would be disingenuous to suggest that such an arrangement does not provide a real competitive benefit to for-hire vessels and tackle shops (along with support businesses such as diners and gas stations) in the state with the more liberal regulations, and that it does not do harm to businesses in those states which, as a result of the conservation equivalency approach, must enact much more restrictive regulations in an attempt to adequately constrain harvest.
            If there was a valid conservation purpose underlying such disparate impacts, they might be justifiable.  However, no such benefits exist, as the conservation goals of the management plan may be addressed more effectively through the application of coastwide or, at the least, mandatory regional management measures. 
            One of the most commonly-heard objections to coastwide regulations is that some states “must” be able to impose a smaller size limit, because larger summer flounder are unavailable in their waters.  However, an examination of 2008 data (throughWave 4) demonstrates that, with few exceptions, all states probably have the ability to produce relatively large summer flounder, and that any disparity in the average size of fish landed is most likely a product of differing size limits.  Such data has been assembled in the following table.

                                    State                            Average weight

                                    Connecticut                             3.71 lbs.
                                    Delaware                                 4.09 lbs.
                                    Maryland                                 2.58 lbs.
                                    Massachusetts                         2.80 lbs.
                                    New Jersey                             2.84 lbs.
                                    New York                               4.42 lbs.
                                    North Carolina                        1.55 lbs.
                                    Rhode Island                           3.93 lbs.
                                    Virginia                                   3.48 lbs.

            North Carolina stands out with an unusually low average size, which arguably can be attributed to fishery conditions unique to that state.  Otherwise, there is no clear geographic pattern.  Instead, it appears that, if people have an opportunity to keep smaller fish, they will, but if the size limit is increased, anglers will usually manage to find sufficient larger fish to harvest a state’s full allocation.  For example, New Jersey has a relatively small 18” minimum size and a low, 2.84 pound average size of fish landed.  However, New York, the state to its immediate north, has both the highest minimum size, 20 ½ inches, and the highest average size of fish landed, 4.42 pounds; while Delaware, the state to its immediate south, has both a 19 ½ inch size limit and the second-largest average size, 4.09 pounds.  Similar “brackets” exist around other states at the lower end of the average size range, suggesting that the lower average weights, at least in states other than North Carolina, are almost certainly due to smaller minimum sizes rather than any relative scarcity of larger fish.
            The uniformity of the average sizes of the summer flounder retained in the various states makes intuitive sense, as the summer flounder managed by the Council represent a single stock.  Particularly in northeastern states, where vessels from neighboring states often fish side by side, the fact that anglers are fishing on the same stock of fish should be reflected in the applicable regulations.  Under the current management scheme, widely divergent bag and size limits can apply to fish caught on opposite sides of a single channel.  On its face, such divergence appears pointless, and as a practical matter, it can cause serious economic harm to angling-related businesses in the more restrictive state, create law-enforcement difficulties and result in anglers’ estrangement from the management system. 
            In order to end the inequitable and unjustifiable disparity in regulations between neighboring states, and to better permit all anglers and angling-related businesses an equal opportunity to benefit from the summer flounder resource, the current state-by-state management approach must be abandoned, and a coastwide or regional approach adopted.

V
Conclusion

            There is little if any justification for continuing to allocate recreational summer flounder landings on a state-by-state basis.
            The initial basis for the current allocation, MRFSS estimates of each state’s harvest in 1998, has been effectively discredited by the National Research Council’s criticism of MRFSS generally, and more particularly by its conclusion that the MRFSS methodology is inadequate for accurately estimating harvest at finer (i.e., state-level or smaller) geographic levels.  User-group criticism of MRFSS, while not based on a sophisticated analysis of the survey, generally supports the National Research Council’s findings.
            Even if the 1998 harvest estimates were valid, subsequent experience has demonstrated that allocations based on such estimates are not applicable to the summer flounder fishery that exists today.  New Jersey, a state receiving nearly 40% of the entire recreational harvest, more than twice the allocation of any other state, cannot regularly harvest its entire allocation despite enjoying some of the least-restrictive regulations on the coast.  Other states, best illustrated by New York, are allocated far fewer fish, and fish under very restrictive regulations, but still land far more summer flounder than are allotted to them, as a changing fishery creates its own de facto allocation scheme that defies managers’ best efforts to force it into the Procrustean bed of 1998 data.
            Trying to ignore the changing character of the recreational summer flounder fishery and force it to conform to an obsolete harvest allocation has had a serious adverse impact on anglers and angling-related businesses in a number of states.  Instead of sharing the summer flounder resource as equitably as practicable among all recreational users, the state-by-state allocation scheme has created a windfall for anglers and businesses in states with large allocations, while simultaneously depriving anglers in neighboring states that received lower allocations from equal access to such resource.  Such unequal access was not created to meet a conservation imperative, or because of a state’s previous failure to adopt acceptable regulations, but merely because landings patterns have changed since 1998.
            CCA NY thus requests that the Council recommend, and that NMFS adopt, regulations that abandon the current state-by-state allocation in favor of a coastwide regulatory scheme that is more appropriate to MRFSS capabilities, recent landings patters and traditional notions of equity.
            Thank you for considering CCA NY’s views on this matter.

Sincerely,

s/Charles A. Witek, III
State Chair