Fishing for knowledge.
Although it may come as a surprise to persons unacquainted with fisheries economics, traditional theory assumes away the problem of finding fish.(1) The point of this paper is to begin the development of an economic theory of fishing that emphasizes the problem of producing knowledge about the location of fish. When the fisheries production problem is viewed from this perspective it tends to resemble the intellectual property problem much more than the traditional common property problem.(2) The central question becomes how to maintain the incentives for the production of new knowledge in the face of a highly impaired system of property rights for knowledge. Although fisheries may seem, at first glance, to be far removed from the problems of, for example, research and development, I would assert that the fundamental problem is the same and that many of the institutional and behavioral responses predicted by this theory and found in fisheries have strong parallels elsewhere in the economy.
In the context of the fishery, the specific problem addressed here is this: The ocean is a very large, complex, and rapidly changing environment. No individual fisherman acting alone could hope to acquire the experience necessary to establish the regularity or predictability required for its successful exploitation. This means that knowledge of the ocean must be acquired and exchanged by a large number of individuals. However, when combined with the "rule of capture" - that is, the rule that whoever catches the fish owns them - the need to collectively acquire and exchange information creates conflicting incentives. The rule of capture creates very strong incentives for the individual acquisition of new knowledge (through either its production or trade), but it also leads to equally strong incentives for the withholding, distortion, and strategic use of information - incentives that tend to reduce the traded value of new knowledge. One would expect, consequently, a tendency toward underinvestment in the acquisition of new knowledge and, more important, in compensating response, attempts to establish institutions ( i.e., contractual forms) that reduce the impairment in the exchange of knowledge and, thereby, also the underinvestment (Williamson 1985). How this occurs in fishing communities and the effects upon the structure and performance of fishing fleets is the subject of this paper.
II. A BRIEF OUTLINE OF THE PAPER
The fisherman's learning problem and how this problem appears to be handled by fishing communities is discussed. The discussion then turns to a model of the search process. This model is developed in three stages that correspond to environments in which contracting for the exchange of information becomes increasingly more difficult. The first stage of the model has two important assumptions that lead to an environment in which the exchange of information is relatively problem free:(a) opportunism(3) in the trading of information is assumed absent and(b) the ability to withhold valuable information is assumed present. It is concluded in this initial version of the model that it would be economically efficient for the individual fisherman to coordinate with a larger and larger group of fishermen the lower the abundance of the searched-for fish.
The conclusions and even the graphics of this version of the model are very close to Buchanan's "An Economic Theory of Clubs" (1965) or the recent work by Clark and Mangel (1984, 1986); unlike either Buchanan's or Clark and Mangel's theories, the remainder of the model is concerned with the problems of cooperative search under conditions that make "contracting" for the exchange of information difficult.
When the circumstances in the model are changed to reflect conditions in which information trading is subject to the problems of opportunism (but is still withholdable) the behavior of rational fishermen changes dramatically. In order to mitigate the effects of opportunism, rational fishermen form cooperative groups whose membership remains relatively stable over time. Significantly, this accommodation to opportunism has the effect of impairing the efficiency of search, especially when fish are scarce.
In its third and last version, the model is changed to reflect conditions under which opportunism is present and the ability to withhold valuable information is absent. Under these circumstances the additional problem of free riders or imitators arises; crucial to the preservation of the value of acquired knowledge (and the incentives for its continued acquisition) is the ability to delay or completely forestall imitation. Successful strategies depend crucially on the duration of time over which acquired knowledge retains its value. In those circumstances where the lifetime of valuable knowledge is short, both deception (for the purpose of delaying imitation) and cooperation (for the purpose of more quickly exploiting acquired knowledge and thereby shortening the period over which imitation might be successful) are employed to maintain the incentives for the acquisition of new knowledge. When the lifetime of valuable knowledge is long, the success of cooperative/deceptive behavior declines and tends to be replaced with the establishment of property rights of one sort or another.
The implications of the model for the efficient organization of and appropriate policy for fisheries with different search characteristics are examined in the last section.
Finally, I should post a warning for the reader. This paper assumes that fishermen perceive no relationship between marginal changes in their individual or collective fishing effort and/or the efficiency of that effort and the size of future stocks of fish. Consequently, "stock externality" considerations are assumed not to enter into their individual or collective decision making. Furthermore, the paper assumes that fishermen's perceptions are correct, at least, with regard to the immediate term effects. The paper does not rule out the possibility that there may be circumstances in which there is a relationship between fishing effort and future stock size. If such a relationship exists, there may also be a relationship between the information problem addressed here and the conservation of stocks. This possibility is discussed in Section VI.
III. THE PROBLEM OF FINDING FISH
Imagine yourself 100 miles offshore in a 70 foot boat, it's three a.m., winds are 25 - 30 knots, seas are running 5 to 8 feet, the temperature is below freezing, the boat is beginning to ice, the seafloor is 600+ feet below you. You are looking for bottom-dwelling fish that are between 2 and 3 feet long. The fish tend to form loose patches that change location from day to day and season to season. The area of reasonable search encompasses thousands of square miles. There are a number of species that bring a valuable return in the market; many others bring none. You are aided by a position locater, charts, and a depth recorder that is of some help identifying fish swimming off the bottom and the characteristics of the bottom itself.
Consistently successful search under these circumstances is extremely difficult to carry out. Fishermen try to minimize the search problem by looking for regularities in the movement and location of fish. But the number of observations necessary to establish regularity is far too large for any single individual to acquire. Additionally, regularity in broad seasonal patterns (migration routes, feeding and spawning areas, etc.) is only approximate and only economically useful under conditions when a stock is abundant or sedentary. When stocks are relatively scarce and at all times for certain fast-moving stocks, the search problem requires that knowledge of a patch must be locationally precise and timely.
Fishing communities resolve this problem in a variety of ways depending upon the kind of information involved. An extensive anthropological literature on information networks in fishing and hunting communities documents some broad regularities (see, e.g., Andersen and Wadel 1972; Lofgren 1972; and Gatewood 1984). At the risk of oversimplifying: Information of value to hunters and fishermen can be categorized, following the convention in the ecology literature, in two ways: coarse- and finegrained information. Coarse-grained information is defined as information about the long-term regularities observed in the resource environment. Fine-grained information is defined as the idiosyncratic and ephemeral information related to the immediate location and movements of animals and fish. Coarse-grained information tends to be widely and freely dispersed: it is the subject of family and community discussions; it is often observed directly by young members of the community simply because they are a part of the production process. In short, coarse-grained information has the characteristics of a free good. Fishermen are even willing to give this kind of information to inquiring academics.
Fine-grained information, on the other hand, is dispersed very selectively. The reason for this different treatment of fine-grained information is that it determines the probability of success or failure especially, but not only, in times of scarcity. From the economic point of view, selective withholding and dispersal of fine-grained information is of interest because the circumstances of exchange of this information affect strongly the incentives to acquire new knowledge about the location of fish and, in both the long and the short run, the relative efficiency of individual boats and the overall efficiency of the fleet.
The problem of maintaining the incentives for the acquisition and dispersal of fine-grained information would be nearly trivial if information could be traded like any other commodity; unfortunately, information does not trade well. Markets for information are extremely impaired. It is difficult to know what is being bought; an adequate description of the information is often the information itself. It is difficult to place a value on information without first knowing what it is. Sellers cannot convey exclusive rights to information, and because of language and other "bounded rationality" problems it may be difficult to convey unambiguously the information without demonstration. For example, the texture of the water surface indicative of one and not another pelagic species may be almost impossible to describe; repeated observation may be the only way to learn to distinguish, say, herring from menhaden. Consequently, fine-grained information is rarely, if ever, subject to market exchange.
In circumstances frequently encountered, fishermen appear to solve the problem of acquiring and distributing fine-grained information with various systems of reciprocal barter among groups whose membership remains relatively stable over time. This method of trading information tends to maintain strong incentives for the acquisition of new knowledge and at the same time tends to coordinate and disperse information in a way that provides few benefits to free riders. Why and under what conditions fishermen use this contractual form and its impact on the organization and efficiency of fishing is the subject of the model that follows.
IV. CLUB FORMATION
A. When Contracting is Easy
Consider a fishery with the following characteristics:
* Fish are found in aggregations, called
patches,(4) that move erratically and at
varying speeds in the immediate term
(i.e., the fine-grained attributes of the
environment are assumed to change
* At any point in time, the location of
one patch is unrelated to the location
* However, the location of all patches
follows broad patterns of seasonal
migration (i.e., coarse-grained attributes
are relatively stable over time).
* A boat can be catching or it can be
* Fishermen can coordinate their collective
fishing (catching and searching)
activities through the formation
of cooperative groups called clubs.
* Information exchange within clubs
takes place through the reciprocal
barter of information.
* Club size can be varied over the
* The price of each fish species is assumed
* The operating costs of fishing (fuel,
nets, bait, etc., but not the transactions
cost of coordinating search information)
for an individual boat are
assumed constant over time.
* The transactions costs of coordination
do not include the costs of opportunistic
behavior; that is, within
the club fishermen are assumed not
to engage in the strategic withholding,
distortion, or release to outsiders
of information valuable to the activities
of the group; and
* Knowledge acquired about the location
of fish can be withheld from
Under these circumstances the objective of a non-cooperating fisherman can be assumed to be simply the maximization of catching time. The fisherman who contemplates trading information, on the other hand, must find a way to balance the potential benefits and costs of cooperation.
Consider a fishery with average patch size, y, and typical single trip boat capacity, c. If the technology of harvest and the schooling characteristics of fish are such that the fisherman who first locates and exploits a patch finds that he has the capacity to completely exhaust the patch on his first locating trip, i.e., if
c/y > 1,
the value of his acquired knowledge is similarly exhausted and there is no basis for entering into an exchange. If on the other hand,
c/y < 1,
then upon the completion of his locating trip the fisherman finds himself in possession of valuable knowledge. In this case a choice must be made; the fisherman may either retain this knowledge for his exclusive use (i.e., through repeated trips to the same patch) or he may engage in reciprocal barter with other fishermen.
The immediate costs of barter (i.e., of club membership) depend upon:
* the catching capacity of the fishermen
with whom the exchange takes
place (i.e., in the club);
* the number of fishermen in the club;
* the size of the patch;
* the expected duration of the knowledge
of the patch (i.e., when will the
patch move or be exhausted by fishing?); and
* the transactions costs of arranging a
reciprocal barter arrangement.
For a given patch size and duration, the locating fisherman will bear no cost so long as the number of fishermen with whom the newly acquired knowledge is exchanged does not exceed a threshold (where the capacity of the fishermen in the club exceeds the size of the patch for the period of time for which the knowledge is valid). Beyond that threshold, the costs of sharing in the form of foregone catches (referred to later as "catch reduction costs," and see Figure 1) will rise in proportion to the capacity of the club for the expected duration of the patch (i.e., until the patch is either exhausted or moves to a new, unknown location).
A member of a club must also take into account the transactions costs of coordinating the flow of valuable fine-grained information. In an all channel peer group,(6) transactions costs increase as the square of the number of members of the group. Modifications of peer group organization can lead to some lesser rate of increase but, in general, peer group transactions costs can be expected to rise more than proportionately with the addition of new members (Figure 1).
Summarizing the cost side of the two policies: a policy of secrecy imposes no additional costs on the fisherman; he is faced simply with constant operating costs. The fisherman who exchanges information is faced with the additional transactions costs of coordination and the possible reduction of his own catch in the short run, that is, rising marginal costs as the size of the club is increased (for a given stock size).(7)
The benefits of the two policies are as follows: Fishing alone confers no benefits other than the assurance that a located patch will not have to be shared with others. The benefits of fishing as a member of a club arise principally from the increase in future revenues (or the equivalent, the increase in future time spent catching) which is a function of stock size and the number of fishermen in the club.
If the fishermen in the club are equally skilled and stand an equal chance of finding a patch, then, for any given stock size, future revenues will be increased (search time will be reduced and catching time increased) in declining proportion to the number of fishermen added to the club. For example, over any short period of time two equally skilled fishermen engaged in search are twice as likely to find fish as one fishing alone. For such cooperating fishermen the more frequent discovery of fish means that more time is spent catching and less time searching than if they were fishing alone. Consequently, two cooperating fishermen will not spend (together) twice as much time searching as will a single fisherman and, over an extended period of time, will not locate twice as many patches but will encounter some amount more than the single fisherman. The same will be true, but to a greater extent for a third or any additional number of club members (Figure 2). In short, the marginal benefits of coordination can be expected to decline as new members are added to the club.(8)
Equating the marginal benefits and costs of club size, an optimal club size for the given stock size is derived (Figure 3). For any given stock size the individual fishermen in a club will find it worthwhile to include or exclude other fishermen until the marginal cost of the last fisherman is equal to the marginal benefit that fisherman confers on the club.
Consider now the effect of exogenous changes in stock size. Figure 4A shows total revenue and cost curves for large, small, and single member clubs over a range of stock sizes. As stock size increases (or decreases) the probability of any fisherman finding a patch also increases (or decreases) as does the amount of time spent catching. Consequently, with more (or less) time spent catching and less (or more) spent searching, catches can be expected to rise (fall) at a rate less than proportional to the increase (decrease) in the size of the stock. For fishermen in a club, catching time and total revenue will always exceed catching time and total revenue for fishermen working alone (see Figure 4A).
This extra revenue is obtained only by bearing the transactions costs of coordinating the club. If coordination costs do not rise more rapidly than the increase in revenues, larger club sizes can compensate for declines in stock size. In this circumstance, fishermen would find it to their advantage to move progressively from individual to small to large group fishing practices as stock size declined ("the locus of most profitable club size" in Figure 4A). However, at some point, as stock abundance declines, coordination costs will exceed revenues and the incentive to increase club size disappears (stock size A in Figure 4A). Further declines in abundance might be expected to be accompanied by an end to directed fishing for that species.(9)
In summary, if fishermen were not faced with the problems of opportunism, efficiency considerations would provide strong incentives for variations in the amount of cooperation as stock size changes. At large stock sizes, the marginal benefit of a club would tend to be outweighed by its costs. Declining stock size would increase the benefits of coordinated search and lead to the formation of progressively larger clubs. However, there would be diminishing returns to club formation; the costs of large clubs would eventually exceed their benefits and at some positive stock size neither individual nor coordinated search would be profitable.(10) There would be, in effect, an optimal club size for each stock size (Figure 4B).
The importance of the optimal club size idea is to indicate efficiency induced incentives that are present in the fishery. These incentives may not be fully capturable in real fisheries but they are still operable and important to the organization of fisheries production. Opportunism, considered in the next section, is one of the principal reasons to expect the, the least, partial frustration of these incentives.
B. When Opportunistic Behavior Occurs
The difficulties inherent in the exchange of valuable information can give rise to individual behavioral responses that are not consistent with the collective interests of cooperating fishermen. For example, in his dealings with other club members, each individual has strong incentives to distort, withhold, and deceive for strategic advantage. Each person in the club can gain an individual advantage by withholding information about a patch he has located; he can provide false information; he can provide partial information; he can provide no valuable information; he can "sell" information acquired from other members of the club to fishermen who are not members of the club. In short, the circumstances of club formation create a fertile ground for the growth of behavior that has earned the label "opportunism" in the economics literature (Williamson 1985).
Members of a club beset with opportunistic behavior would quickly find that the net benefits of club membership were seriously eroded and perhaps even turned negative because of such behavior. If one member of the club "sold" information beyond the boundaries of the club, other members would face higher and uncompensated catch reduction costs. If one member provided false, partial or distorted information, or withheld information, the value of reciprocation would diminish for the other members. The only way to deal with this kind of behavior is to set up procedures that monitor, verify, and enforce the information exchanges of each individual in the club. But such procedures can be very costly. As with simple coordination costs, one would expect the costs of monitoring, verifying, and enforcing to rise more than proportionately with an increase in club size. This suggests that opportunism creates a tendency for smaller clubs than might otherwise be preferred.
The anthropological literature referred to previously and observation also suggest that stability of club membership, as well as small numbers, may be necessary for the minimization of the special kinds of transactions cost that arise with opportunism: (a) building accurate expectations of other club members behavior, (b) minimizing language or communications difficulties, and (c) maintaining an approximate balance in the reciprocal exchange among club members. All of these problems are minimized by the creation of a small, stable, long-lived club. Specifically:
1. A small club with stable membership gives rise to frequent, repetitious exchange among all members of the club and allows for the ex post verification of other club members' behavior. In effect, time makes it possible for experience rating and sorting to be carried out and trustworthy relationships to be developed. Trust based on a historical pattern of behavior reduces the need and expense of timely verification of new information.
2. Small stable clubs also increase the capacity of club members to acquire information regarding the personal and environmental context in which information exchange takes place. The often ambiguous nature of available information and the peculiarities of individual articulation and transmission of that information can, in the absence of contextual knowledge, lead to costly misunderstandings and the erosion of trust. Familiarity with the personal and environmental context minimizes such problems.
3. A small club also increases the likelihood that a group of equally skilled fishermen can be assembled. Equality of skill reduces considerably the problems of maintaining approximate balances in the exchange of valuable information. In a club with widely differing skill levels, a balance in the exchange of information can only be maintained if there is some mechanism for metering the value of information provided to each fisherman so that it bore some correspondence with the value of the information provided by that fisherman. The ability to do this is highly improbable, for if information could be metered in this way it would become a commodity like any other.
For these reasons fishermen can be expected to resort to small numbers cooperative arrangements in order to obtain the benefits of coordination without the high costs of opportunism. The hazards and costs of cooperation in the face of potentially opportunistic behavior tend to be mitigated by repetitious, reciprocal exchanges. Each exchange taken by itself constitutes a unilateral transfer of information but the repetition of exchange over time allows for ex post verification, experience rating, sorting, and the gradual building of trust. Small numbers increase the frequency of exchange among members of the club and, as a result, strongly contribute to the feasibility of monitoring and verification.
Repetitious exchange among a small group also reduces the problems of fully articulating often ambiguous information. The problem of ambiguous information is not so much that mistakes can be made but that it introduces numerous occasions for opportunistic behavior, either of which could seriously erode the trust necessary for the efficient functioning of the club. Equality of skill among club members also improves the ability to monitor and verify and, more important, considerably reduces the need to meter the value of the exchanges of information among club members.
In short, the special transactional problems posed by opportunistic behavior lead to what Williamson (1985, 61) calls the "great transformation"; that is, the tendency for potentially large number contracting/bidding situations to be replaced by small numbers or bilateral exchange relationships.
Although small clubs may be superior to large clubs, it is clear that numerous occasions for their failure and dissolution are, nevertheless, continually present. It is not unreasonable to suppose that the marked persistence of kinship groups in fisheries production is closely related to the problem of opportunism. Kinship groups, typically, do not require a strict balance of accounts in reciprocal barter and, consequently, do not have to adhere to the equality of skill requirement; kinship groups are by definition long-lived and for that reason do not have to engage in costly trust building; kinship groups carry a large part of the organizational/transactions costs of coordination as a normal part of family organization and the sanctions they may employ are potentially much more serious to the individual than those available to a group of unrelated individuals.
If clubs can sustain themselves in the face of these difficulties, they are in a position to maintain incentives for the acquisition of new knowledge by assuring, through internal trade, the value of at least part of that knowledge to its discoverer. Nevertheless, the small, stable club solution to the problem of opportunism is not costless. Such clubs incur the obvious costs of coordination; but more important, they forfeit the potential benefits of optimally sized groupings at those times when large club size is appropriate. Imperfect as this kind of arrangement might be, it may be the only practical way to maintain the socially desirable incentives for the acquisition and dispersal of new knowledge about the resource.
C. When Valuable Information Cannot Be
It is often the case that the simple act of exploiting valuable information reveals that information. In the fixed gear inshore lobster fishery, for example, it is difficult for a fisherman to work an area without revealing his presence; the boat is easily observed while setting and hauling gear; the gear itself is marked with buoys, and the success of a boat's fishing is generally well-known locally. Imitation by less skilled fishermen, in the absence of explicit forestalling steps, is easily accomplished. In dragging operations, identifying gear is not left in the water but a boat's activities are often easily observable; although less easily accomplished, imitation by the less skilled or knowledgeable is likely.
At first glance, it would seem that the ability to imitate would lead to a situation in which little or no advantage would accrue to fishermen who generate new knowledge about the resource; imitation would rapidly erode the tradeable value of newly discovered knowledge and the incentive to acquire still more new knowledge would disappear. Whether this is the case or not, depends upon the circumstances of the fishery. In general,
* the longer the duration of the value of
knowledge about a patch, the greater
will be the ability to imitate;
* the greater the predictability contained
in coarse-grained information
(about migration routes and timing,
for example), the greater the likelihood
of imitation because the location
of skilled fishermen is more predictable;
* the more widespread and reliable the
information about the current and
past relative catch success of a boat,
the more likely will be imitation; and
* the smaller a boat's capacity relative
to the size of a patch, the more likely
will be imitation.
Since skilled (i.e., locating) fishermen can be depended upon to recognize the losses they incur because of the imitative actions of other, less skilled, fishermen, it is reasonable to expect that they will take steps to preserve for themselves the value of their new knowledge to the extent possible. The possible steps that they might take fall into two categories: strategic behavior and the establishment of formal or informal resource rights.
By strategic behavior what is meant is behavior designed to delay, distort, or make ambiguous information upon which imitation might be based. The purpose of strategic behavior is to delay or forestall imitation, thereby retaining for the discovering fisherman the value of the acquired knowledge for purposes of trading or his own use. Rights creation, whether formal or informal, is also assumed to be for the purpose of preventing imitation. It is assumed that both strategic behavior and rights creation are costly.
The ability to pursue or develop strategies that completely forestall or significantly delay imitation is likely to be greatest in a rapidly changing environment. Given any lag in the response of imitators there is relatively less value associated with imitation. An extreme example of this is the blue fin tuna harpoon fishery. Fishing is directed at individual fish. If there is any grouping of the fish, the act of harpooning one, or even attempting to harpoon one, causes the group to disperse almost instantly. In this case, the rapid change in the fishes' location makes imitative behavior relatively valueless and there is no need for strategic behavior designed to delay imitation. In the context of the information exchange problem, there is also no tradeable knowledge that might be retained by delaying imitation.
In fisheries in which there is less rapid dispersal of patches or in which the location of one patch is related temporarily to another, there is greater value to imitative behavior and also, of course, greater value in strategic behavior that delays imitation. In other words, the temporary withholding of information made possible by strategic behavior creates circumstances in which the retained information may be traded or reserved for the fisherman's own use. In the high seas swordfish and albacore fisheries, for example, patches are often widely dispersed and tend to move fairly rapidly; nevertheless, when a patch is located it is often possible to remain fishing on the patch for an extended time. In these instances, strategic behavior designed to forestall imitation is not very costly for the simple reason that observation by potential imitators is difficult and as a result imitation is not easily accomplished. Furthermore, if a single boat's capacity over the duration of fishing on the patch is generally less than the yield from the patch, cooperation, or club formation, has the effect of reducing the time over which imitators need to be delayed (as well as conveying the benefits of reciprocal exchange). Put differently, a club can rapidly deplete a patch. It can, thereby, retain within its membership the value of the newly acquired knowledge and, at the same time, reduce (but not eliminate) the extent and the costs of strategic behavior necessary to deter imitators. In short, cooperation and strategic behavior together tend to increase the probability that locating fishermen will realize the value of their acquired knowledge. By doing so, these kinds of behavior maintain the incentives for the continued acquisition of new knowledge.
In situations where the rate of change in the location of patches is less rapid, the potential benefits of strategic behavior are neither greater nor less than in more mobile fisheries; however, the costs of strategic behavior increase and the probability of its success declines. In fisheries such as oysters, mussels, and clams the sedentary characteristic of the resource generally means that the duration of knowledge about the location of a patch is rather long (months or years) and the observation (by potential imitators) of fishing on the patch rather easily accomplished. In these situations, and provided a single boat or a small club does not have the capacity to exhaust a patch quickly, it is unlikely that strategic behavior can substantially delay the entrance of imitators. Consequently, the locating fisherman loses much of the tradable or own-use value of acquired knowledge. Even worse, the involuntary dispersal of information is likely to extend to a number of boats far in excess of the number whose capacity might be necessary to exploit the patch without giving rise to reductions in the catch of the locating fisherman.
Consequently, in these kinds of situations the inability to use either strategic behavior or cooperation to prevent the dissipation of potential rents accruing from acquired knowledge of the resource generates very strong incentives for the creation of artificial barriers to entry - e.g., territories as in the lobster fishery and property or lease rights such as in mussels and oysters. From the perspective of this theory, the purpose of such artificial barriers (property or quasi-property rights) is to maintain the coupling between private and social incentives in the same way that cooperation and deception promote collective efficiency and maintain private incentives for the acquisition of new knowledge in mobile fisheries. From a more conventional perspective, cooperation and deception in mobile fisheries operate toward the same ends as property rights in sedentary fisheries. The next section of the paper turns to a more explicit consideration of the organizational patterns that might be expected in fisheries.
V. THE ORGANIZATION OF
In the examples drawn upon so far, the importance of the mobility characteristics of the stock have been emphasized. The model suggests that other characteristics of the fishery may strongly influence the economic feasibility of, and the particular shape of, the contractual forms required for information exchange. Earlier a number of variables important to the formation of an optimal club were mentioned. These variables were the duration of knowledge of a patch (or the mobility characteristics of a species), the capacity of the boat, typical trip length, and typical patch size. One might add to this list other factors such as the impact of harvest technology on the involuntary transmission of information among fishermen, the competitive structure of the ex-vessel market (monopsonistic, etc.), product market characteristics (fresh, frozen) and so on. Each of these factors is likely to affect the costs and benefits and, consequently, the optimal size of clubs. This suggests that the industrial organization of fleets will vary according to the particular characteristics of each fishery. For example:
* One would expect strong incentives for cooperation in a fleet fishing on highly mobile stocks aggregated in patches larger than the capacity of any boat in the fleet and located far from the fleet's port of origin. In this kind of fishery, the benefits of coordinated search would tend to be relatively high and catch reduction costs to be relatively low. One would expect, of course, relatively small and stable groups of cooperating fishermen and a strongly disadvantaged position for an independent, noncooperating fisherman. The albacore and swordfish fisheries would appear to fit this hypothesis.
* In contrast, highly mobile stocks, characterized by patch sizes smaller than typical boat capacity and in which the immediate location of one patch is unrelated to the location of other patches, might be expected to generate fleets with little or no sharing of information. Small patch size relative to boat capacity and the unrelatedness of one patch to another removes all the tradable value from newly acquired knowledge (Thorlindsson 1988). For example, there are times of the year when the herring fishery in New England appears to fit these characteristics; however, there are other times of the year when the location of patches is related and, consequently, one would expect a tendency for information sharing to occur (see Orbach 1977 for a discussion of this effect in the high seas tuna fishery).
* In a groundfish(11) fishery, conducted on a relatively open abyssal plain such as in southern New England, patch size tends to be large and the density of fish is not great. Patches tend to move relatively slowly and in ways that correspond to known migration patterns. Under these circumstances the search problem is not extremely difficult and, in addition, the ability to withhold information from other fishermen is only moderate (see footnote 5). As a result, one would expect only a modest advantage to cooperative behavior and, to the extent possible, relatively intense strategic games. In the same fishery, but conducted instead in Northern New England on boulder strewn, rocky bottom, with occasional open mud and sand plains, the behavior of the fish and the nature of the search problem changes. The variability of this environment causes fish to congregate in smaller and more difficult to find locations. Consequently, one would expect greater incentives for cooperative search.
* In relatively sedentary fisheries in which it is difficult to withhold information from other fishermen, such as the valuable inshore lobster fishery, one would expect strong incentives for the discouragement of imitation. But the long duration of the value of knowledge about the location of patches would also lead one to expect the defense of that knowledge through cooperation or deception alone to be relatively costly and not likely to succeed. In this instance, the logical extension of strategic behavior is the use of force (Acheson 1972, 1988) for the purpose of establishing territories.(12) In the lower valued crab fishery, with the same environmental characteristics and effective patch size, or in a non-complex environment in the lobster fishery, one would expect little tendency to engage in costly strategic behavior and weak or non-existent incentives for the acquisition of new knowledge of the resource.
In summary, because of the large number of variables affecting the acquisition and exchange of information and the large variation in the circumstances of each fishery, one would expect the institutions or contractual methods devised to maintain the incentives for the acquisition of new knowledge to lead to unique organizational arrangements in each fishery. Depending upon the particular circumstances, the costs and success of these arrangements can also be expected to vary.
VI. A BRIEF NOTE ON RESOURCE
Although this paper is not concerned primarily with resource management issues there are two points that need to be raised.
First, traditional common property theory depends upon an exclusively technological analysis of social efficiency. For example, when purely technological criteria are applied, a fishery with an annual (or average annual) catch of X, pursued by boats each with an annual capacity of Y, is deemed to be socially efficient when the number of boats, n, times their capacity is equal to the annual catch in the fishery (i.e., when X = nY). If the number of boats exceeds the technologically determined efficient number (i.e., X [is less than] nY), the fishery would be described as over-capitalized and the situation taken to be evidence confirming the traditional hypothesis of excessive entry in common property fisheries.
When the problem of search and the kinds of institutional arrangements required to sustain the incentives for search are introduced into the equation of social efficiency, that same circumstance (i.e., X [is less than] nY), can no longer be taken as unambiguous evidence of the social inefficiency of the fishery and, of course, the converse is equally true. Even from a purely technological perspective, search introduces the need for extensive knowledge about the configuration and changes in the location of the resource and the capabilities for successful search before judgments about the social efficiency of a fishery can be made. More important, the institutional problem of sustaining the incentives for search introduces the need to assess the efficacy of the search-related institutions that are likely to arise in each fishery. Neither the technological nor the institutional aspects of the search problem are likely to make the social efficiency question amenable to back-of-the-envelope calculations or to broad rules of thumb.
Following Williamson (1985), I would assert that the only reasonable approach to the assessment of the social efficiency of any particular fishery must involve a deep immersion in its peculiar detail. From a policy perspective this means not only understanding the efficiency functions of existing institutions, but also undertaking an assessment of the search/information/institutional effects of proposed policy changes. (Although this paper is about the harvesting sector, clearly the same advice applies to similar kinds of information problems and institutional solutions in both factor and product markets.) Put differently, I suggest that what are viewed often as "mere sociological details," can and should be interpreted as efficiency-related institutional accommodations to a particular set of information problems in either the harvesting or market sectors.
The resource rents question provides an alternative perspective on the issue of social efficiency. The conventional economic theory of open access, common property fisheries (or other resources) concludes that entry into the fishery will proceed to the point where average profits are driven to zero and all resource rents dissipated. This conclusion is explicitly dependent upon the assumption of homogeneity among the vessels or people exploiting the fishery. The theory presented here is clearly inconsistent with the assumption of homogeneity. The institutional and behavioral accommodations to the problem of acquiring new knowledge predicted by this theory strongly imply differential success (catch) rates among fishermen. This will be true for any of the institutional solutions suggested by the theory: Those fishermen unable to maintain membership in clubs will be disadvantaged relative to those in clubs; the general requirement for equality of skills among club members suggests that there will be differential rates of success among clubs; the fishermen who for one reason or another become imitators can be expected to have success rates that differ from the fishermen they are imitating; those fishermen who are able to establish territories or formal property rights can be expected to benefit at the expense of those who do not.
When there are differential abilities to solve and benefit from the problem of acquiring new knowledge, an entry and exit process very different from conventional theory is implied. One would not expect to find relatively similar profit pictures for all boats in the fishery; rather, one would expect to find a wide distribution of profits among fishermen.(13) Fishermen who are better able to solve the knowledge problem will be doing very well while other fishermen might be going bankrupt. If an individual's ability to solve the knowledge problem persists over time, one would expect a relatively stable population of more skilled fishermen and a population of less skilled fishermen with a fairly high rate of turnover. Entry into the fishery is not likely to be made so much on the basis of expected profits, but more on the expectation of being able to solve, in one way or another, the knowledge problem. Exit, especially after only a few years in the fishery, is likely to be the result of an inability to solve the knowledge problem. Exit, especially after only a few years in the fishery, is likely to be the result of an inability to solve the knowledge problem. Nevertheless, since the severity of the knowledge problem is negatively correlated with stock size, one would still expect exit and entry to be correlated with changes in stock conditions.
In terms of rents, the persistence of differential success among fishermen suggests that the core of successful fishermen will capture a large part of the resource rents available from the fishery even at times when many other fishermen are going bankrupt.(14) Put differently, exit from the fishery or low incomes for some fishermen cannot be taken as evidence that resource rents have been dissipated; the condition of the marginal fisherman is not likely to be indicative of the condition of the entire population of fishermen.
Second, the circumstances required for successfully dealing with the problems of opportunism in information sharing, suggest a major difference in the efficiency of search with and without the presence of opportunism. When the possibility for opportunistic behavior is present, the rate of change in the size of cooperative clubs must be slow. Changes in the biological environment are likely to proceed at a much faster rate than the rate at which fishermen can develop the trustworthy relationships necessary for cooperation. As a result, the possibility of opportunistic behavior is likely to impair fishermen's ability to form groups whose size is optimal for the current abundance of the stock (Figure 5). Clubs may be too large at times of abundance and too small at times of scarcity. They are likely to be optimal (in terms of a narrow technological definition of search efficiency) only by coincidence. When stocks are scarce, the problems of opportunism lead to club sizes below the level that would be optimal for search. Therefore, fleet and individual efficiency is below what it might be when stocks are scarce. Consequently, profitable search can be expected to cease at a larger stock size than it might otherwise.
In fisheries in which there is a relationship between current and future stock size, i.e., a stock-recruit relationship as conventionally defined, the inability to adapt search to current stock size may have an important effect upon the conservation of stocks and appropriate management policies. In a multiple species fishery, one would expect the increasing costs of search with declining stock size to lead to a termination of fishing at population levels higher than would be the case in a single species fishery. This can be expected because the relative efficiency that accrues to cooperative (as opposed to individual) fishing would lead to the earlier detection of alternative profitable stocks. Given elastic demand for each species, the efficiency of information sharing would lead fishermen to switch away from declining stocks earlier than otherwise (Wilson et al. 1989). This suggests that it is appropriate for management to reinforce those factors that influence the decision to direct fishing effort away from declining populations and toward ones of greater abundance. It also suggests that single species management that effectively prohibits switching may build in a tendency to fish populations to lower levels than would be experienced in a multiple species fishery.
Although it is not within the scope of this paper to delineate the other factors that might reinforce this search cost/switching effect it is worthwhile to mention at least a few. The overall elasticity of demand for fish and the elasticity of substitution are clearly important because of their impact on fish prices. The higher both of these elasticities, the larger the population at which one would expect switching to take place. Given the often very rigid contract and product distribution systems in fisheries, both these elasticities may be influenced by public policy.
The number of exploited species in the system is also important; the greater the number, the larger the population at which switching can be expected to take place simply because there is a greater probability of finding a more abundant (and profitable) species. Royce (1988), for example, writes about the Japanese tendency to "fish for fish" rather than for particular species. Public policies toward foreign boats fishing within the fisheries zone for species not exploited by the domestic fleet might be altered to encourage exploitation by the domestic fleet. In almost all countries permission for foreign fishing is granted on the basis of the size of the stocks of (domestically) unexploited species. Cognizance of search cost/switching effects suggests that the status of domestically exploited stocks should form the basis for the decision about foreign fishing. If domestic stocks are in good shape, fishing by foreign boats may be acceptable. If domestic stocks are in poor shape, so-called over-the-side sales by domestic boats to foreign processor boats may be preferred both for the obvious economic benefits and also for the biological benefits. Other factors that are likely to influence the population size at which switching takes place are the age of first capture, appropriately timed and located closed areas, and the technological capabilities of boats - all of which may be influenced by public policies.
In short, these search cost/switching effects suggest that the allocation and the efficiency, or effectiveness, of fishing effort may be subject to public policy in many cases. In multiple species fisheries especially, search cost/switching effects tend to create systematic linkages between all species of the fishery. Many of the policies just discussed are, in one form or another, policies that are employed in a variety of fisheries, but they tend to be employed only with regard to their impact on a single species. For example, age of first capture or minimum size regulations are common to many fisheries; however, the basis for their implementation is almost without exception based upon the growth and/or maturity characteristics of each individual species. Search cost/switching effects suggest new reasons for implementing old policies and, in some cases, new policies for old problems.
Economically efficient fishing requires the continuous acquisition of new knowledge about a complex and changing resource. Individuals cannot accomplish this alone but must depend upon the knowledge of others and a means for distributing that knowledge. Because new knowledge is valuable and not easily traded, incentives arise for a combination of strategic and cooperative behavior, the purpose of which is to preserve the value of new knowledge for those who acquire it (and, thereby, the incentive for its continued acquisition) and to deny its value to those who might be termed free riders in the search for valuable knowledge. The exact ways in which strategic and cooperative behavior are employed and the extent to which they preserve the incentive for acquisition of new knowledge can be expected to vary widely depending upon the particular circumstances of each fishery. Consequently, one should find that the ways in which fisheries are organized and the kinds of resource management policies that might effectively couple private and social incentives will vary widely. Unlike traditional fishery theory with its uniform prescription of a property rights policy, this view of the fishery suggests a much more eclectic approach to fisheries management (Wilson 1982). Additionally, as discussed just above, cognizance of search cost effects suggests a range of alternative policies that might positively reinforce the conservation aspects of search costs.
In a more general context, the theory developed here suggests that the dilemma posed by Arrow (1962) and Demsetz (1969) with regard to the need to maintain the incentives for the acquisition of new knowledge while at the same time assuring its dispersal throughout society can be, and often is, addressed through the creation of institutions and strategic behavior. The kinds of institutions and strategies that might prove effective can be expected to vary according to the circumstances, especially as they relate to the lifetime of valuable knowledge.
The model suggests that the difficulties of contracting for the exchange of information will lead to contractual adaptations that have the purpose of minimizing that impairment. Although these adaptations tend to maintain the incentive for the acquisition of new knowledge, they are costly and tend to erode the benefits of, and incentives for, that acquisition. Depending upon the characteristics of the situation, e.g., especially the duration of the value of knowledge, these contractual adaptations can be expected to be more or less successful but always to be aimed at minimizing the problems of information exchange. When the costs of voluntary contractual arrangements or behavioral responses exceeds their benefits one would expect a shift toward reliance upon privately enforced rights, e.g., territories, or formal property rights. The particular kinds of contractual forms that arise and, consequently, the organization of production in each situation will also depend upon the unique circumstances of that market or industry.
(1) In the traditional theory of fisheries economics, fish are assumed to be either randomly or evenly distributed in the ocean. This is mathematically convenient because it leads to a situation in which catch rates are then determined by a combination of fishing effort and the abundance of the stock. If any two of the variables - effort, catch rate, and abundance - can be measured directly, the third can be estimated.
(2) Both the intellectual and common property problems are characterized by an impaired system of property rights; both tend to emphasize the uncoupling of social and private incentives and, for that reason, a "less than optimal" social outcome. In the common property problem this uncoupling leads to short-run over-production and long-run depletion or destruction of the resource. In the intellectual property problem, the uncoupling leads to the short- and long-run underproduction of intellectual resources (Arrow 1962; Demsetz 1969).
(3) By opportunism is meant behavior that may withhold, distort, or otherwise manipulate information for self-interested strategic advantage.
(4) The term patch implies discrete aggregations of fish. An alternative approach that might be more appropriate for many species, especially groundfish, would be to consider an irregular spatial distribution that varied from dense to sparse aggregations. This approach would imply a slightly different view of successful fishing. Fishing would not involve a catch or don't catch situation; rather the process would become one of search leading to differential catch rates. Successful search would be characterized by high catch rates, unsuccessful search by low catch rates.
(5) The short term is defined here as a period of time shorter than the time necessary for intergenerational effects to take place in the biological environment.
(6) An all channel peer group is defined as one in which all information transmitted to one member of the group is transmitted to all other members.
(7) Catch reduction costs might also be a function of the current status of the stock. In times of less abundance fishermen might spend more time catching on any located patch than they might during periods of abundance when the probability of successful search is higher.
(8) For the outsider who might be admitted as, say, the fourth or fifth member of the club, the marginal benefits of admission are quite high. One would expect, consequently, club members to extract this potential value through admission fees of one sort or another. Although they may exist, I am not aware of any situation in which admission fees are actually charged.
(9) The end to directed fishing, of course, does not mean the end of all fishing mortality for a species. By-catch mortality, especially in a trawl fishery, will continue.
(10) Viewed from a slightly different perspective, the analysis suggests that the transactions costs of cooperation would cause the stock size at which fishing became unprofitable to be larger than it might otherwise be.
(11) The term groundfish refers to fish who live on or near the bottom of the water column. Cod, flounder, and haddock are typical groundfish species.
(12) It is interesting, however, that even where cooperative territories are enforced and where the ocean bottom is not uniform, fine-grained information is still relatively important, probably because entry into the cooperating gang is not restricted to the point where it is at or near the optimal group search size (Wilson 1977). Given the high value of the catch and the high catch reduction costs arising from larger than optimal group size, one would expect (and one observes) a willingness on the part of individuals within the group to undertake relatively costly strategic games designed to deceive other members of the group. As Acheson (1988) notes, the combination of cooperation in the defense of the territory and deception in fishing creates often very interesting group dynamics.
(13) For example, there are approximately 200 scallop boats operating in the New England-Mid Atlantic area. The Executive Director of the New England Fisheries Council estimates that 25 of these 200 boats account for 70% of the total catch. Part of that record is attributable to the size of the boats and the number of days spent at sea, but these factors are, over the long term, highly correlated with the skill of the captain and crew.
(14) It does not suggest, however, that the long-term resource rents (that might be available if fish populations were relatively stable and subject to controls that would generate maximum economic yields) would be preserved.
Wilson is professor, Department of Agricultural and Resource Economics, University of Maine.
Research for this paper was supported by the Sea Grant Program of the Universities of Maine and New Hampshire. I would also like to thank the following people for their very helpful comments: John Sutinen, Gardner Brown, Ralph Townsend, Ronald Turner, James Acheson, David Wihry, and two anonymous referees.
Acheson, James. 1972. "The Territories of the
Lobstermen." Natural History 81 (Apr.):60-69. ___. 1988. The Lobster Gangs of Maine.
Hanover, NH: University Press of New England. Andersen, Raul, and C. Wadel. 1972. North Atlantic
Fishermen: Anthropological Essays on
Modern Fishing. St. John's: Institute of Social
and Economic Research, Newfoundland
Social and Economic Papers No. 5. Arrow, Kenneth. 1962. "Economic Welfare and
the Allocation of Resources for Invention."
In The Rate and Direction of Inventive Activity.
Economic and Social Factors, 609-25,
NBER. Princeton: Princeton University
Press. Buchanan, James. 1965. "An Economic Theory
of Clubs." Economica 32, (Feb.): 1-14. Clark, Colin, and Mark Mangel. 1984. "Flocking
and Foraging Strategies: Information in an
Uncertain Environment." American Naturalist
123:626-41. ___. 1986. "The Evolutionary Advantages
of Group Foraging." Theoretical Population
Biology 30:45-75. Demsetz, Harold. 1969. "Information and Efficiency:
Another Viewpoint." Journal of Law
and Economics 12 (Apr.): 1-22. Gatewood, John. 1984. "Cooperation, Competition,
and Synergy: Information Sharing
Groups among Southeast Alaska Salmon
Seiners." American Ethnologist 11, issue
2:350-70. Lofgren, Orvar. 1972. "Resource Management
and Family Firms: Swedish West Coast Fishermen."
In North Atlantic Fishermen, eds.
C. Wadel and R. Andersen. St. John's: Institute
of Social and Economic Research, Newfoundland
Social and Economic Papers No.
5. Orbach, Michael. 1977. Hunters, Seamen, and
Entrepreneurs: The Tuna Seinermen of San
Diego. Berkeley: University of California
Press. Royce, William. 1988. "Comments on Japanese
Fisheries Management." In Fisheries Science
and Management: Objectives and Limitations,
ed. Warren Wooster. New York:
Springer-Verlag. Thorlindsson, Thorolfur. 1987. "The Skipper
Effect in the Icelandic Herring Fishery." Human
Organization 47:199-212. Williamson, Oliver. 1985. The Economic Institutions
of Capitalism. New York: The Free
Press. Wilson, James. 1977. "The Tragedy of the Commons - A
Test." In Managing the Commons,
eds. Garret Hardin and John Baden.
New York: The Free Press. ___. 1982. "Economical Management of Multispecies
Fisheries." Land Economics 58
(Nov.): 417-34. Wilson, James, P. Kleban, S. McKay, R. Townsend,
and J. French. 1989. "Economic and
Biological Benefits of Interspecies Switching
in a Simulated Chaotic Fishery." Forthcoming
in the proceedings of the Symposium
Peches Artisanales (July 1989) or Department
of Economics, University of Maine.
PHOTO : FIGURE 1 THE COSTS OF A CLUB AND FISHING ALONE
PHOTO : FIGURE 2 CATCHING TIME AND CLUB SIZE
PHOTO : FIGURE 3 THE COSTS AND BENEFITS OF CLUB SIZE
PHOTO : FIGURE 4A COSTS, REVENUES, AND SURPLUS FOR DIFFERENT CLUB SIZES AT DIFFERENT STOCK SIZES
PHOTO : FIGURE 4B OPTIMAL (PROFIT MAXIMIZING) CLUB SIZE VARIES WITH STOCK SIZE
PHOTO : FIGURE 5 CLUB SIZE WITH AND WITHOUT OPPORTUNISM
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||economic theory of fishing|
|Author:||Wilson, James A.|
|Date:||Feb 1, 1990|
|Previous Article:||Intergenerational resource rights, efficiency, and social optimality.|
|Next Article:||OCS leasing policy and lease prices.|