Risk Management In Fisheries

RISK MANAGEMENT AND INSURANCE OPTIONS FOR FISHERIES RESOURCES INTRODUCTION Risk pervades modern day fisheries management where uncertainty, variability, scarcity and multiple objectives are common terms. Two major drivers underlie this: the inability to predict the behaviour of complex socio-ecological fishery systems (Charles 1998) and a transition of ocean ecosystems away from frontier settings of untouched and unlimited resources to scarce resources and conflicting goals (Hanna 1997). In short, fisheries management is the business of making trade-offs in a complex, unpredictable and variable world. Risk management methods provide pragmatic means of navigating this increasing complexity by systematically identifying and coping with risk. Fisheries science has reacted to the mounting challenges to successful management, for example by making recommendations to incorporate risk and uncertainty into the decision-making process (Ludwig et al., 1993; Rosenberg and Restrepo 1994), however risk management as a pragmatic and proactive framework remains underutilized. Risk, risk management and the Fisheries/Aquaculture context A first step in exploring risk management is to examine risk in fisheries and its causes. Risk is an intuitive concept to humans; everyone deals with it on a daily basis in making decisions. Any risk involves three underlying components (based on concepts from: Athearn 1971; Crockford 1991; Rowe 1994; Kangas and Kangas 2004): •

a variable state of the world,

•

imperfect knowledge on the state of the world, including in the future, and

•

a desired state of the world.

In short, a risk entails the ideas of variability, uncertainty and loss, leading to the following definition: a chance of adverse effects from deviations from expectations. Note

that a risk is a possibility of a bad thing happening, whereas a realized risk is an actuality, i.e. an adverse outcome has transpired. Using the definition suggested above, there are myriad risks in fisheries management and their identification is of critical significance. Uncertainty is widely regarded to be pervasive in fisheries (FAO 1995; Charles 1998; Weeks and Berkeley 2000; Harwood and Stokes 2003), and risks can be identified simply by following the sources of variability and uncertainty as these drive deviations from expectations. Table 1 outlines some common risks affecting the functions of different parts of a fishery system. Multiple risks can be associated with a function. Table 1: A partial list of risks affecting the functions of components of a fishery system Biological resource

Management agencies

Function Biomass Production Habitat provision Biodiversity, genetic diversity storage Nutrient/chemical cycling Climate regulation Recreational/cultural opportunity provision

Regulate and allocate harvest Protect habitat Collect data and perform research Stock enhancement Enforcement and compliance

Fishermen, fishing Harvest communities, Process fishing Market industry

2

Example Risks Stock depletion Habitat degradation Pollution Exotic species introductions Climate change Natural disasters Disease Genetic stock structure changes Species interactions/ecosystem effects, e.g. trophic cascades Failure to achieve social benefits goals through overly cautious harvest Failure to achieve conservation goals through overly aggressive harvest Funding changes Catch fluctuation Price fluctuation Cost fluctuation Changes in rights to resource use Personal injury Equipment failure Employment loss

Risk management is a loose term for the general process of identifying, characterizing and reacting to risk. Dorfman (2008) offered a straightforward definition, ‘the logical development and implementation of a plan to deal with potential losses.’ Crockford (1991) offered a more comprehensive definition, focused on corporate management, but equally applicable to a natural resource system: ‘the identification, measurement, control and financing of risks which threaten the existence, the assets, the earnings or the personnel of an organization, or the services it provides.’ The focus in either case is the pragmatic goal of minimizing the effects of unpredictable variability. Risk management comprises two stages (Fig. 1). In the first, risks are identified and characterized. Then in the treatment stage, they are dealt with (Crockford 1991; Outreville 1998). The treatment phase can be broken down into three avenues for handling risk: •

avoid,

•

transfer or

•

retain risk.

In avoidance, management decisions are made to forego risky prospects, for example by deciding not to develop a new resource.

3

Figure 1: A risk management framework. Rounded boxes are entry and exit points into the risk management process. Rectangles, diamonds and ovals denote actions, decisions and chance events, respectively.

In transfer, risks can be shifted in whole or in part to another entity that is better able to bear them, for example through insurance or by sharing risk over a pool of individuals. If the decision is made to retain risk, several options are available. Risk control attempts to reduce the likelihood and/or the magnitude of an adverse outcome. Note that some risks are uncontrollable, such as natural disasters. Another option is to finance risk by making preparations to absorb realized losses. Alternatively, a decision can be made to ‘do nothing’ and absorb realized risks with no preparation. This encompasses cases where extant risks go unidentified, or when no action is taken on identified risks citing a lack of sufficient information to understand the problem. A final measure to handle retained risks is risk recovery, where efforts are taken to improve the response time of a system after a realized adverse outcome.

4

RISK MANAGEMENT AND INSURANCE OPTIONS FOR FISHERIES RESOURCES 1). Avoiding risk: the Precautionary Principle Perhaps, the most straightforward method of managing risk is to avoid it; prevention vs. treatment. Sharing similar components with decision analysis and risk assessment, the general process of risk avoidance involves identifying the risks associated with a proposed action, followed by cost and benefit analysis. If potential costs outweigh the benefits, then decide against the activity. Risk avoidance opportunities in fisheries management are case-by-case specific. In some instances, risky activities can be spatially or temporally separated. It is usually difficult if not impossible to predict the true costs and benefits of a proposed management action. The Precautionary Principle provides a guiding framework for policy creation under risk avoidance. In its pure form, the Principle states that no action should be taken until evidence demonstrates that it is harmless (Foster et al., 2000). Taken too literally, it would result in failures to achieve socioeconomic goals of ocean resource management by being overly conservative in foregoing catch and employment. Paradoxically, by being too conservative in avoiding risks, the overall chance of management failure increases and risk management is incomplete. This has led to a levelheaded practical implementation of the Principle, called the Precautionary Approach (Garcia 1994; FAO 1995, 1996). Recognizing the balance between resource use and conservation, the Approach maintains the heart of the Principle, but interprets it in a manner that is open to some risks associated with resource use. 2). Managing risk through transfer Successful fisheries management requires consideration not only of biological sustainability, but also of economic sustainability for resource users (Hilborn 2006, 2007). Towards that end, insurance policies which transfer risk or the vulnerability to environmental and economic variability away from individual producers provide a business-oriented component to dealing with uncertainty in fishery systems. Insurance can be defined as a financial arrangement that redistributes the costs of unexpected losses (Dorfman 1978). The key idea is that risk can be transferred to

5

someone who is better able to bear it, moving towards Pareto efficiency (Ahsan et al., 1982). Pareto efficiency is often used as an economic target for policy, describing a situation where no one can be made better off without making someone else worse off. The transfer of risk to another party comes with a payment for riskbearing services, an insurance premium. Risk management through insurance is not widespread in fisheries, however, three forms have been implemented to varying degrees of success: •

personal health and safety,

•

asset, and

•

production and market insurance.

Insurance instruments for fisheries in developing countries are less common (Hotta 1999). Individual insurance policies, where the financial arrangement concerns one insured and an insurance body, are appropriate when loss events are randomly distributed throughout a population. Alternatively, group insurance policies, where the financial arrangement concerns a collection of insureds and an insurance body, provide opportunities for managing risks that simultaneously affect an entire fishery, such as a weak salmon run. Group risk management in fisheries is important due to the large scales at which biological, environmental, and economics processes operate, including aggregated fish populations, large water masses, and marketwide price changes, inter alia. 3). Controlling risk: diversification and the portfolio effect The goal of risk management through diversification is to take advantage of probabilistic properties to both reduce the likelihood and severity of a loss by constructing a bundle of assets, a portfolio. The term asset is general and can refer to species, fish stocks, income sources, or financial securities, inter alia. Portfolio theory focuses on the selection of assets to create a bundle that provides the greatest expected performance, say catch or annual income, at the least variation about the expected performance. Diversification and portfolio theory rely on two phenomena: statistical averaging and correlations amongst portfolio components. Statistical averaging is the effect that a sum of random variables, such as catch value, has lower variance than the individual variables themselves, contributing to portfolio performance stability (Doak et al., 1998; Tilman et

6

al., 1998). The effect can operate both with statistically independent or correlated assets. The second component of diversification plays out when assets’ performances are not independent. Discrepancies in correlations amongst asset returns are exploited to reduce variability about an expected performance. This is the effect taken advantage of by a fish farmer who stocks a mix of warm- and cool water adapted fish to stabilize harvest in the face of unpredictable weather. 4). Controlling risk: price risk management Price variability is a primary component of revenue variability, a major source of risk for fishery participants and the broader fishing industry. This form of variability is also termed marketing risk, where the transformation of production activities to financial reward occurs under uncertainty (RMA 1997). Marketing risk can make operating a business difficult and can negatively affect inter-temporal resource allocation decisions (Larson et al., 1998), for example the financing of new fishing gear or processing facilities. To see the relationship between price and revenue risk, consider the following simple relationship in log space: V (ln R ) = V (ln P.Q) = V (ln P ) + V (ln Q) + 2Cov(ln P, ln Q) where V() is variance, Cov() is covariance, and R, P, and Q, are revenue, price and catch. In cases where production has a strong effect on prices, called an elastic price response, some natural buffering of price variability occurs where low (high) production results in higher (lower) prices, or the covariance term is negative. Natural buffering is not a given, however, as price is a complex function of both endogenous and exogenous factors, such as the supply of substitutes. Local fisheries are often price takers, i.e. local catch is too small relative to total market supply to move prices. Several price risk management strategies are applicable to fisheries: (i) marketing timing strategies, (ii) forward contracting and futures, and (iii) enterprise integration. i). Marketing timing: strategies involve spreading out the sale of products on cash markets over time, providing a natural buffer against price variability.

7

ii). Forward contracting and futures are a means for buyers and sellers of fish to remove price uncertainty by locking into an agreement to sell or buy at a given price at a later time. A future is a standardized, tradable contract to deliver agreed upon quantities of graded product at a later date for a specified price. They can be used to manage price risk in two ways. First off, future contracts held to delivery provide a means of reducing price uncertainty to zero, however, with some possibility of one side of the exchange not fulfilling their end of the bargain, or counter party risk. Second, futures provide a method of hedging price movements in the case where contracts are not held until delivery. Hedging is taking a position in two or more markets such that a loss in one market can be offset by a gain in another. iii). Enterprise integration takes two forms: vertical and horizontal. Vertical integration internalizes different production stages such as catching and processing fish in one operation. By internalizing transactions within an operation, the vertically integrated producer avoids some market transactions and their associated price uncertainty. Horizontal integration entails the consolidation or cooperation of many similar firms at the same stage of production, for example the formation of a marketing cooperative for salmon harvesters. 5). Risk financing: buffers A risk that is not avoided, transferred or controlled is then retained. Two options exist for retained risks: do nothing, or prepare to bear a possible loss, i.e. to finance the risk, as it is termed in business. Risk financing typically refers to financial preparedness, but it can include any investment to absorb losses from realized risks such as foregoing some catch to maintain conservative harvest limits. Importantly, risks that go unidentified are by default retained. A successfully managed fishery might not identify an oil spill as a risk; however, if shipping occurs in the area, the risk still exists. Given the prevalence of uncertainty and variability in fisheries and the difficulty in identifying all risks that need management, financing is an important strategy to deal with unexpected events, or surprises. Buffers are the primary tool to finance risk in fisheries management. They may take several forms: (i) harvest-related buffers that prescribe conservative catch or effort restrictions that can accommodate unexpected biological shocks, (ii) area or temporal

8

closures, or (iii) more directly related to livelihoods, financial reserves. In addition to calling for conservative harvest limits, the FAO’s statement of the Precautionary Approach for the management of capture fisheries calls for marine reserves as buffers, quoting ‘to limit risks to the resource and the environment, use area closures, which are relatively quick to implement and are easily enforceable (FAO 1996). NEED FOR FISHERIES AND AQUACULTURE INSURANCE: ADVANTAGES AND CHALLENGES FOR DEVELOPING COUNTRIES Problems and Constraints Fish production is an inherently risky business, in which fishermen and fish-farmers face a variety of risks associated with weather and natural phenomena, variations in production, prices and income, human-error, mechanical failure and accidents from a variety of causes beyond their control. Fishing differs from other types of economic activity in a greater dependence on nature than is the case for other businesses. Fishing must face difficulties arising from weather phenomena, sea conditions, and currents, while disease, parasites, salinity levels and water quality affect aquaculture production. The volume of catches is strongly affected by the state of fish stocks and their migration, in the case of pelagic species. Production costs, such as fuel and other inputs, interest rates, etc. can also fluctuate. In addition to production risks, fishermen also face uncertain market conditions, including dramatic declines in the price of fish when catches are abundant. Being a highly dangerous occupation, the health and personal safety of those engaging in fishing is frequently at risk. The high level of risk and above average repayment problems associated with fisheries, compared to other branches of economic activity, make financial institutions reluctant to lend for fisheries. The prevalence of risk in fisheries is not new and fishermen, fish-farmers, fisheries cooperatives, and banking institutions have developed mechanisms adapted to local conditions for coping with such problems. These traditional mechanisms of risk management are in some cases still adequate. Governments also often intervene by providing emergency assistance and tax relief to fishing communities in the wake of natural disasters. Such relief can be provided on an

9

emergency basis, but is not feasible, either financially or politically, on a long-term basis, since preferential treatment for fisheries is unlikely to be acceptable to other sectors. Fisheries insurance can promote the sustainability of the productive capacity of fishermen and fish farmers without provoking criticism of this type. Relatively small premiums can provide fishermen and fish-farmers with the compensation they need to cover the loss of and damage to vessels, gear and production or catch. In some cases governments find it worthwhile to provide partial subsidies for premiums and for administrative costs of insurance programmes; some governments also support re- insurance schemes. The major risks confronting fishermen can be divided into two categories: a) Asset risks, production and management risks, and market risks: Asset risks include loss of or damage to fishing vessels, equipment, and gear and aquaculture installations, as a result of natural disasters. Production and management risks involve the loss of catch, production failure and fish disease. Market risks relate to changes in the prices of outputs and inputs, as well as increases in interest rates. b) Personal and health risks: Personal and health risks include accidents at sea and death and job related illnesses. In some countries the goal of insurance programmes is not merely to provide coverage for insurable risks, but also to foster mutual assistance and co-operation. Small-scale fishermen and fish-farmers are more likely to be reached by programmes of this type, although the financial viability of the schemes may be less certain. In other cases providing insurance is left entirely to the private sector, limiting coverage to those groups better able to support the full cost of programme operation. A number of difficulties and problems often arise with respect to the introduction of fisheries insurance. The most important of these are the following: 1) Limited financial resources; 2) Exclusion of small-scale fishermen and fish-farmers from insurance; 3) Lack of well-established village institutions, such as co-operatives, to act as insurance agents; 4) Lack of legal framework for fisheries insurance; 5) Difficulty in promoting insurance policies, designing sustainable insurance programmes and co-ordinating the work of the agencies concerned;

10

6) Difficulty of covering credit projects under insurance schemes; 7) High cost of premiums and providing appropriate levels of benefits; 8) Speed and transparency of claims settlement; 9) Lack of trained personnel; 10) Lack of understanding of the value of insurance; and 11) Lack of reliable actuarial data for establishing premium rates. Advantages and Benefits The advantages and benefits provided by fisheries insurance vary from scheme to scheme. However, in general, the principal benefits of fisheries insurance are: 1) Protecting fishermen and fish-farmers against accidents and natural hazards beyond their control; 2) Providing basic compensation for the loss of or damage to fishing vessels, gear and catch (or harvest), thus contributing to stabilisation of incomes within the fisheries sector; 3) Reducing the risk to financial institutions, which providing credit to fishermen and fishfarmers, in relation to fisheries credit; 4) Reducing the risk for fishermen and fish-farmers in investing their own resources in the adoption of new technologies and acquiring improved equipment; 5) Fostering mutual assistance and co-operation among fishermen, fish-farmers and their organisations; 6) Reducing the unpredictable burden on government of providing emergency assistance in the wake of natural disasters; 7) Promoting stability in fishery enterprises and contributing to the general welfare of fisheries communities; 8) Stabilising the contribution of the fisheries sector to national economy. PLANNING AND

PRACTICES

OF FISHERIES

AND

AQUACULTURE

INSURANCE Fisheries insurance programmes can be either voluntary or compulsory. Which alternative is best for achieving high levels of participation and adequate coverage of risks at reasonable costs depends on the social, economic, and cultural orientation of the

11

communities the programmes are designed to serve. For a scheme to be compulsory, some level of government involvement is necessary. Compulsory schemes have a number of advantages: (i) A high degree of participation, a key requirement for the success of insurance schemes, is essential; (ii) Administrative costs can be pro-rated over a larger number of policies, thus reducing the cost of the system to users and of subsidies to government. There are a number of steps in the establishment of a compulsory system. Legislation must be enacted, as a pre-condition for making the system mandatory. It is necessary to create awareness, on the part of fishermen and fish-farmers, of the benefits and obligations of the system and to take their views into account in designing and implementing the system. Financial institutions can also help encourage greater use of insurance by requiring that fisheries assets (and the lives and health of fishermen) be insured as a condition for providing loans. Linking a compulsory system to credit or marketing operations can be an effective tool for assuring compliance with mandatory coverage. When the production is relatively stable, objections to a compulsory system may be reduced, since premiums are likely to be low and thus not be a major burden on the payment capacity of fishermen, which is often low. Under this scenario, participation in a voluntary scheme might be rather low. Government Involvement and Private Sector Insurance Government intervention can be justified on the following grounds: (a) Potential losses are difficult to estimate with any degree of precision to allow for coverage under a private insurance scheme; (b) Monetary losses due to uncertainty in the volume of catch or production and in the price of fish sold would have a serious impact on the well-being of small fishermen with no alternative sources of income. Private sector insurance works well with large commercial fisheries companies and aquaculture producers of high value species. However, such schemes do not provide for the needs of many of prospective customers almost always requires some degree of financial support from government. Criteria for Determining Insurance Coverage

12

Fisheries insurance does not have to cover all elements of fishing or fish-production. During the initial years of operation, an insurance programme may limit coverage to those areas most critical to the success or failure of the types of enterprises participating. Coverage can later be expanded gradually to cover other areas as the agency implementing the programme acquires the skills and the experience needed to expand coverage to items of lesser importance to these enterprises or to areas which will encourage participation by enterprises not initially targeted by the programme. The choice of particular insurance items to cover varies from one locality to the next even within the same country. In general terms, the choice of items for coverage should be guided by certain criteria, the most important of which are the following: (i) Coverage of an item should contribute to improving food security or nutrition. (ii) It should contribute to the overall viability of fisheries or aquaculture production and enhance national economic and social stability. (iii) Programme design should be such that it does not preclude or discourage participation by a wide range of small-scale fishermen and fish-farmers. (iv) Actuarial data on losses and casualties covering a sufficient period of years should be available to allow accurate establishment of appropriate rates for premiums. (v) Availability of the infrastructure necessary to keep damages and losses in production, marketing and credit to a minimum. (vi) A high percentage of fishermen and fish-farmers should have access to and be using credit. (vii) The losses resulting from items to be covered should have a significant impact on the income of fishermen and fish-farmers. DESIGN OF FISHERIES AND AQUACULTURE INSURANCE SCHEMES Premium Standards Premiums should be based on sound actuarial calculations, using available records on incomes, damages and losses on insured fishermen and fish-farmers. Premiums should be set at high enough levels to cover average indemnities, administration costs and, where appropriate, a contribution towards building up financial reserves. In order to be actuarially sound, a fishery insurance programme should be based on a large volume of comparable statistical data regarding incomes of potential policyholders,

13

and losses and damages experienced in the past. This is required for the calculation of premium rates and indemnities to be paid. Ideally, such data should cover an extended period of time, showing fisheries and aquaculture yields and their variation from season to season and year to year. Data on damage and losses should be classified according to the sources of the losses and damages, discriminating between the various sources, such as those attributable to fishermen’s neglect on handling gear, fish-farmers’ poor management of fish ponds, lack of proper fish handling, etc. Moreover, such data need to be further broken down by types of fishing and aquaculture. In many developing countries, inaccuracy, incomplete coverage or complete lack of such statistical data is a serious problem. In some cases it may be necessary to delay implementation of fishery insurance programmes until a reasonable volume of accurate data has been accumulated. Even where the data is available, they need to be analysed for insurance purposes. Many countries do have some sets of data which, though not entirely satisfactory, can be useful in developing estimates of sufficient accuracy for the initial planning of a pilot programme, with adjustments made during the course of initial operations based on more accurate and more complete data. Pilot Project Operation There probably is no single best way to conduct an experiment of this type and no standard timeframe for a pilot insurance programme. The best course is often to run the experimental programme long enough to permit the drawing of meaningful conclusions and to observe client reactions to its operations. The effects may emerge gradually, whether they are positive or negative. The pilot project should be confined initially to a single item; whichever one is most important to fishermen’s income and national interests. This and other items selected for coverage should be those for which there are reasonably good records for past production and losses, and which are relatively simple to insure. It is also wise to conduct the pilot project in a limited geographical area, where the conditions under which fishermen operate are as nearly homogeneous as possible. Experience gained from pilot operations would provide highly relevant and reliable data on which to expand the service the programme provides and to assist in the design of larger scale operations to cover other areas of the country or to serve other types of customers.

14

Awareness Building Fisheries insurance, like other forms of insurance, is oriented towards the future. In most cases it provides no immediate and tangible benefits unless and until, at some time in the future, a covered loss occurs. For that reason, it is often difficult for fishermen and fishfarmers to appreciate the value of a fishery insurance programme. They have difficulty in understanding the promise of possible benefits at some uncertain future time in exchange for immediate outlays of money and effort in the form of premiums and adherence to a variety of conditions of coverage, some of which, such as carrying safety equipment onboard, may require further expenditure. It is likely, therefore, that considerable time and effort will have to be devoted to educational campaigns among the fishermen to achieve their co-operation with an insurance programme. A variety of media should be used, including public meetings at local and national levels, brochures, newspapers, radio, television, and the Internet. Fishermen’s organisations can be useful in generating support amongst their members and in encouraging them to participate in insurance schemes. Fostering Trained Personnel Fisheries insurance is a technical field. Its effective operation requires a thorough understanding of the complexities of fisheries and aquaculture operations. The lack of properly trained personnel, not only at village level, but also at provincial and national levels, poses a serious problem. Different expertise is required at each level. At the top administrative level, there is a need for expertise in the over-all design of the insurance programmes, taking into account the integration of insurance components into national fisheries development plans. At a technical level, terms and conditions of coverage need to be established and actuarial aspects incorporated into programme design. At the local level, there must be extension officers or personnel who can effectively explain the insurance system to fishermen and fish-farmers and are able to make clear to an unsophisticated audience the complexities of the claims settlement function. Unfortunately, insurance matters are normally not dealt with by extension officers who have direct contacts with fishermen and fish-farmers. REFERENCES Ahsan, S.M., Ali, A.A.G. and Kurian, N.J. (1982) Toward a theory of agricultural insurance. American Journal of Agricultural Economics 64, 520–529.

15

Athearn, J.L. (1971) What is risk? The Journal of Risk and Insurance 38: 639 – 645. Charles, A.T. (1998) Living with uncertainty in fisheries: analytical methods, management priorities and the Canadian groundfishery experience. Fisheries Research 37: 37 – 50. Crockford, N. (1991) Risk Management. Witherby, London. Doak, D.F., Bigger, D., Harding, E.K., Marvier, M.A., O’Malley, R.E. and Thomson, D. (1998) The statistical inevitability of stability-diversity relationships in community ecology. The American Naturalist 151, 264 – 276. Dorfman, M.S. (1978) Introduction to Insurance. Prentice Hall, Englewood Cliffs. Dorfman, M.S. (2008) Introduction to Risk Management and Insurance. Prentice Hall, Upper Saddle River. FAO (1996) Precautionary Approach to Capture Fisheries and Species Introductions. FAO Technial Guidelines for Responsible Fisheries, No. 2. FAO, Rome, 54 pp. FAO (Food and Agricultural Organization of the United Nations) (1995) Code of Conduct for Responsible Fisheries. FAO, Rome, 41 pp. Foster, K.R., Vecchia, P. and Repacholi, M.H. (2000) Risk Management: science and the Precautionary Principle. Science 288, 979 – 981. Garcia, S.M. (1994) The precautionary principle: its implications in capture fisheries management. Ocean and Coastal Management 22, 99 – 125. Hanna, S.S. (1997) The new frontier of American fisheries governance. Ecological Economics 20, 221 – 233. Harwood, J. and Stokes, K. (2003) Coping with uncertainty in ecological advice: lessons from fisheries. Trends in Ecology & Evolution 18, 617–622. Hilborn, R. (2006) Fisheries success and failure: the case of the Bristol Bay salmon fishery. Bulletin of Marine Science 78, 487 – 498. Hilborn, R. (2007) Moving to sustainability by learning from successful fisheries. Ambio 36, 296–303. Hotta, M. (1999) Fisheries Insurance Programmes in Asia: Experiences, Practices and Principles. Food and Agricultural Organization Fisheries Circular, No. 948, FAO, Rome, 58 pp.

16

Kangas, A.S. and Kangas, J. (2004) Probability, possibility and evidence: approaches to consider risk and uncertainty in forestry decision analysis. Forest Policy and Economics 6: 169–188. Ludwig, D., Hilborn, R. and Waters, C. (1993) Uncertainty, resource exploitation, and conservation: lessons from history. Science 260, 17 – 36. Outreville, J.F. (1998) Theory and Practice of Insurance. Kluwer Academic, Boston. Rosenberg, A.A. and Restrepo, V.R. (1994) Uncertainty and risk-evaluation in stock assessment advice for U.S. marine fisheries. Canadian Journal of Fisheries and Aquatic Sciences 51: 2715 – 2720. Rowe, W.D. (1994) Understanding uncertainty. Risk Analysis 14: 743 – 750. Tilman, D., Lehman, C.L. and Bristow, C.E. (1998) Diversity-stability relationships: statistical inevitability or ecological consequence? The American Naturalist 151, 277–282. Weeks, H. and Berkeley, S. (2000) Uncertainty and precautionary management of marine fisheries: can the old methods fit the new mandates? Fisheries 25: 6 – 15.

17