Scrimgeour, FrankScobie, GrantMoyle, Brendan James2026-11-092026-11-091995https://hdl.handle.net/10289/18760The loss of biodiversity, as measured by extinctions of species, is presumed to be occurring at a rate that poses a serious danger to the human and natural world. Species conservation is now, a public policy issue. However, a major obstacle to achieving efficient conservation policy, is the difficulty in attributing value to species. This research develops a methodology, that could be used to allocate resources among recovery programmes for endangered species. An economic approach is used because the resource allocation dilemma is an economic problem. The situation is analogous to an investment decision, where costs are borne now in order to generate uncertain future benefits. The research identifies where market failure may be present. If property rights to a species can not be defined, market failure may occur because of the public good nature of species. This is likely for species that are ecologically important. If property rights can potentially be defined, but are not actually defined, then market failure may occur. However, in this case, species are not public goods. Inefficiencies also occur as a result of government failure. It is difficult for public agencies to gather accurate value-information among the candidates for recovery work. Voting and lobbying by environmentalists are an unreliable guide to the relative values of endangered biota. The agency also may face incentives to lower the efficiency of recovery work. The subjective basis for the neoclassical axiology is explained. The ability of subjective value theory to account for market behaviour, chiefly that voluntary exchange is a win-win game, accounts for its acceptance over objective value theories. The use of environmental ethics is rejected on the grounds that such value theories are not operational. This point is conceded in some of the environmental ethics literature. The scope of the environment, comprising things both animate and inanimate, both holistic and individualistic and both, sentient and non-sentient presents a major obstacle. In addition, the problem of subjective reduction is difficult to overcome. Finally, there is the teleological problem. It is not possible to know a priori which species are more vital for the well-being of ecosystems in the future. Several approaches to valuing endangered species are considered. The first is the Contingent-Valuation-Method. Such an approach has little use for deriving relative values for biota. The technique depends on being able to describe the good in question accurately to subjects. However, the technique does not identify the problems of time and uncertainty. Thus, species can not be described accurately. It is a partial equilibrium approach. It is also subject to bias and embedding problems. The second approach is the Safe-Minimum-Standard approach. The technique argues that preservation should always occur so long as the costs are not too high. It does not identify what constitutes too-high costs. There is an explicit bias towards preservation despite the low odds that any particular species will yield future economic benefits. The third approach is to derive objective criteria to rank species. In New Zealand, this approach is manifested in the Species-Priority-Ranking-System. This approach sums seventeen unweighted ordinal numbers to produce a cardinal sum. This is mathematic nonsense. The approach is vulnerable to subjective reduction, given the choice of criteria has a strong effect on rankings. An implicit bias towards rare species, rather than species of high biological value is noted. The research instead adopted a hypothetical market approach based on game theory. This was determined to more accurately capture the nature of the problem facing a species manager. In the game, subjects must allocate a fixed budget among twenty-four candidates for recovery work. Each species is described in terms of ecological value, genetic value, taxonomic group and cost. Time to extinction is correlated to cost, the more expensive a candidate is, the less time it is expected to persist. Species are not named but are coded to eliminate mega-fauna effects. The subject plays against nature which generates a strategy randomly. The strategy may be favourable to recovery work or unfavourable. Players are scored on being able to produce a net improvement in the status of the candidates. Several efficient solutions exist, which permits preferences for some species (eg. birds) to be pursued. Money payoffs were not used in the trial market, only the real market game. The bidding data was analysed using the TOBIT, LOGIT and PROBIT statistical techniques. Players shifted their strategy towards an optimising one during their real games. In the trial games, over-investment in very costly, risky candidates was observed. Relative to the Contingent-Valuation-Method, the game theory approach produced data of comparable statistical significance. Its weakness lies in its cost and self-selection bias of subjects. The research identified costs as being the dominant variable affecting bidding behaviour. The decision among the sample, to indulge in preferences for taxonomic groups such as reptiles, was only observed when the marginal dollar was being allocated. Thus, as the budget was exhausted, only then did players allocated their marginal dollar on the basis of biological variables. This reflected the desire to save as many species as possible. The deterrent effect of costs was corroborated with a study of real bidding behaviour at the Otorohanga Kiwi House. Public choice problems in New Zealand were also discussed. The bureaucratic goal of budget maximisation produces a bias towards high profile species, often of low biological value. Finally principal-agent problems and a low ability to monitor field staff, also have the potential to limit the effectiveness of conservation expenditures. This is illustrated with examples from the kakapo recovery programme.enAll items in Research Commons are provided for private study and research purposes and are protected by copyright with all rights reserved unless otherwise indicated.Thesis