The recent literature on local collective actions has, on the one hand, emphasized its specific advantages over individual actions or over publicly organized actions, and on the other hand, it has raised several interesting questions that cannot be answered unambiguously at a theoretical level and that therefore wait for an empirical answer. The factors concurrent with the emergence, the success and the sustainability of local collective actions should clearly be studied on a case-by-case basis before any generalization, given the absence of a clear theory identifying them and their precise role. Wealth inequality is probably the most concrete example (and one with very high stakes in terms of policy recommendations ! ) of a variable sometimes favorable and sometimes detrimental to the success of local collective actions (Baland & Platteau, 1997, Oxford Economic Papers, 49).

The case studies that we introduce below analyze collective actions undertaken by coastal fishing communities in Senegal and in Japan under a common focus: what is the link between the market structure (Cournot or Bertrand competition, one-sided or two-sided oligopolies,...) and profitable opportunities of collective action on the supply side at a local level ? The following reasoning leads us to a working hypothesis : a specific possibility of reallocating rents is given by the particular market structure ; if participants to the collective action have different characteristics, this opportunity goes hand in hand with (re)distributive issues ; individual traits will therefore heavily bear on the individual willingness to participate, and on the sustainability of the collective action as whole. Our hypothesis is that the relevant individual traits will differ according to the market structure. In other words, market structure determines (part of) the role of heterogeneity in local collective actions. This hypothesis can be tested either on the basis of a large number of case studies or, more partially, if the attention paid to market structure variables generates clear priors (to be tested) on the role of heterogeneity. As a first step, we adopted the second methodology.

Before turning to the description of the case studies, it should be noticed that their respective authors designed the data gathering process (although a questionnaire was administrated by independent enumerators in the African case) and interacted frequently in the frame of the CRED (Centre de Recherche en Economie du Developpement, University of Namur, Belgium).

The case study conducted in Senegal focuses on the design of self-regulations in four fishing communities in the Northern coast between Dakar and Saint-Louis. The considered regulations forbid overfishing in different ways in relation with the fishing technology. Line fishermen organized themselves to allow only a fixed amount of output to be landed and sold on the shore. Fishermen using purse seines are allowed to go fishing only once a day (given the difficulty of controlling the precise amount of output caught by the seine, fine-tuned output regulation would be more difficult here).

The study aims at assessing the relative success of the different regulations and at understanding the motivations underlying them, the problems undermining them and the conditions under which they can be reproduced.

The field work began in 1996 and a questionnaire was designed in January 1997. One of the authors participated in March & July 1997 in the data gathering process, which was mainly performed by independent enumerators in five villages (from North to South : Saint-Louis, Kayar, Yoff, Soumbédioune, Hann), the last one being a control group where no regulation was either operating or intended.

The economic context faced by the fishermen is composed of three elements : supply conditions, market structure (demand side) and incentive considerations.

Supply conditions are marked by a high degree of uncertainty over the catch, especially for line fishing. Another point is also the alleged increasing scarcity of marine livestock. Fishermen particularly blame a recent international convention allowing foreign industrial boats to exploit the national resources.

The market is based on intermediation between fishermen and exporters that address the final demand abroad. The local currency was devaluated in 1994 by 50%. Beside this major, one-shot shock, world prices are almost fixed, but intermediaries (called "mareyeurs") used to capture a large rent thanks to a mutual agreement not to buy at the world price. These coalitions are observed in all five villages under scrutiny, albeit with different degrees of success (in Kayar, a limited number of major intermediaries could seize the largest part of the rent). Self-regulation by the fishermen is essentially aimed at recapturing the rents appropriated by the intermediaries and also at a more careful intertemporal management of the marine livestock given its recent alleged scarcity.

On the incentive side, a free-rider problem is raised by the regulation : if prices are high, each fisherman has an incentive to catch an additional amount of output, regardless of the detrimental effect on prices perceived by other fishermen. However, the landed output is quite easily observable by a supervision committee operating on the beach and enforcing fines if necessary.

A key event occurred in Kayar in 1994, when the intermediaries tried to lower prices further on the very day of an exceptionally good catch. A lot of fishes were not sold and were consequently wasted. Fishermen decided to go on strike in order to retaliate. During the strike, they decided to limit supply in the future and they appointed a supervision committee. The news spread and self-regulation emerged in other villages.

Although we wait for more data concerning the actual change in fish prices and in traded quantities on the shore, it seems that the commercial impact of the regulations was considerable (fishermen speak of a 100% to 300% rise in prices). Risk on output is increased for big owners who cannot compensate a low catch on one boat by a high catch on another boat anymore. Members of the supervision committee in Kayar report a number of infractions to the rules, but they keep enforcing fines. On the other hand, the regulation was abandoned in Yoff and Soumbédioune for reasons to be determined (the determinants of support to the regulation is the main topic of our empirical analyses in process).

The convergence of commercial and biological effects expected by the fishermen is certainly a major reason why self-regulation was possible and spontaneous. The motivation of fishermen who supported the regulation is clearly linked with profit opportunities. Because of the existing market structure, these fishermen have undoubtedly a collective action potential. Yet their awareness of the biological effects of their present fishing behavior seems to be only tenuously related to their willingness to participate in the collective action.

Another major finding of the study is that support of the collective mechanism is highly differentiated according to individual characteristics of fishermen. Heterogeneity in wealth, working capital and ability to cope with the (activity-specific) risk on output quantities translates into differential support for self-regulation at the community level.

More detailed empirical analyses will try to show whether there is any interaction between the market structure and the role of heterogeneity.

The Japanese case study compares three groups engaged in small shrimp fishing in two communities in the Toyoma prefecture. One community has been exposed to market forces and therefore groups in this community are highly responsive to commercial demand. The daily bidding process administered by the fishery co-operative enables the groups of fishers to exercise market power (i.e. to capture higher producer surplus by restraining fishing effort). One group in this fishing community has managed to establish and sustain a collective production institution with income sharing arrangement over 30 years. The other group in the same community, on the other hand, failed to maintain a similar arrangement. In the second community, unlike in the first one, the co-operative does not administer bidding but fixes constant unit prices. Consequently, fishers are less influenced by commercial forces and not responsive to market demand. Partial income sharing system, instead of full income sharing, has evolved in the second community.

The Initial Situation

Small shrimp fishing has been operated in both communities in Japan since the 18th century. Fishers have experienced a rapid technological and economic transition over the last 60 years, which has accelerated capital investment in fishing equipment. Rapid economic growth and increased consumption demand in post W.W.II period, innovation of freezing facilities, advanced means of transportation, and advancements in fishing technologies (such as echo-sounder, advanced navigation system, and synthetic net) in the 1960’s have resulted in multifaceted problems of inefficiency and inequity. Notably, intensified investment in the advanced fishing equipment had exacerbated conflict over the limited fishing spots, worsen catch per unit of effort, and deteriorated unit prices. Invention of synthetic material enlarged the size of net, which contributed to intensification of fishing effort. Synthetic net has added risk of net damages and fluctuation of income. In addition, there had been increasing concern about equity in taking advantage of and successfully coping with the technological transition.

All three groups in Japan had instituted a ‘pooling’ (income sharing) system in which gross income and costs related to fishing nets (i.e. costs of construction and repair) were equally shared. Various rationales underlie this institutional innovation : exploiting potential economies of scale (i.e. efficiency gains from work co-ordination, sharing information and expertise, collective net repair, and orderly allocation of fishing spots), restraining total landings in order to get higher average price, and insuring against risks of net damages. Monopolistic market power seemed to be particularly important for resource users to become aware of potential benefits of restraining aggregate effort. Institutional changes have been initiated by ‘harbingers’ who are often the most skilled, productive and respected fishers. It is interesting to note that the most performing fishers played pivotal roles in reaching consensus on the institutional design and in maintaining the collective production mechanisms.

In spite of apparent differences in skill among fishing units, one group has not only managed to maintain income pooling arrangement for more than 30 years but also encouraged the lower skilled to improve their performance. These findings are all the more surprising because they are contrary to the prediction of standard economic theory in which individuals are rational and selfish. It should be noted that differences in individual performances are observable in the studied group, unlike in the case of conventional team-work. Then, why do the systematically more productive fishers have stayed in the income pooling arrangement though they may be apparently better off by staying out of such arrangement? Why have the less skilled fishers not taken easy-ride on the more skilled ones? Results of in-depth interviews offer one possible explanation that considerations of fairness and concerns about relative position in the group, on the one hand, compensated the better skilled with leadership recognition, and, on the other hand, prevented the lower skilled from becoming opportunistic.

Participatory process of organising collective institutions is sensitive to social and economic surroundings. More specifically, the existence of price-elastic demand and monopolistic market power can provide conditions conducive to a process of institutional change for self-management. Policy interventions of market liberalisation and privatisation in developing and in transition economies may require careful evaluation in terms of their impact on self-management initiatives at local level.

Heterogeneity in skill endowment appears to be empirically important determining variable. On the one hand, heterogeneity can give a rise to a leadership role in initiating institutional changes. On the other hand, the social status associated with the leadership role is essential for the sustainability of the equal income sharing pattern (that is not expected in cases with observable differences in skill according to conventional economic theory).