Assessing values for multiple and conflicting uses of freshwater in the Canterbury region

Abstract

Freshwater resource management can be challenging as policy makers need to consider the well-being from many uses that includes environmental, social, financial and cultural elements. Generally, there is a lack of information of how to balance these different well-being and what is the related value of improving water quality as this is not always reflected in the market. This study assessed how Canterbury residents value and trade-off multiple attributes of freshwater use by applying Discrete Choice Modelling (DCM) methodology. DCM is a commonly applied non-market valuation method that involves the respondents making multiple trade-offs to measure willingness-to-pay (WTP). In this, freshwater resources are described in terms of their characteristics, or attributes, such as ecological quality, job opportunities or recreational use. The freshwater attributes were chosen to reflect the four elements of well-being, and unlike previous studies of freshwater in New Zealand, this study includes a Māori cultural-specific attribute. This is important as it allowed consideration of cultural values to be included alongside other values in water allocation. A DCM survey was applied to reflect issues on Canterbury rivers with a sample of the general public including Māori. The aim was to provide information on public’s preferences for the freshwater management and the different elements of well-being by providing estimates of welfare measures for changes in river attributes. The marginal WTPs were estimated separately for each freshwater attribute. For the policy scenarios the estimates for attribute levels were combined to show impacts of different irrigation scenarios on employment, environmental, recreational and cultural values of water. This is relevant for the current policy debate in Canterbury involving land use intensification, such as the Central Plains Water (CPW) irrigation scheme. A particular focus of this objective was to identify user groups and test for differences in preferences between them. Also important to this was the inclusion of a Māori cultural attribute. The results show that Canterbury residents were willing to pay for improvement in all the attributes included in the DCM; however, not all attribute levels were significant. The estimated WTP ranges from up to: $182 (increase in rates per year) for improved water quality and habitat, $59 for improved swimming water quality, $57 for the above average cultural quality and $45 for 173 more jobs in the region as a result of additional irrigation. Results suggest that Cantabrians value the cultural attribute, with Māori valuing this attribute more as indicated by the significant ethnicity covariate. The attribute ranking by the different user groups show little significant difference although the values differed for a few. In addition, compensating surplus (CS) values were calculated to four scenarios that involved changes from the current amount of irrigation. These scenarios were (1) increase in irrigation with reduced water quality; (2) increase in irrigation with maintained water quality; (3) increase in irrigation with improved water quality; and (4) a contrasting scenario of reduction in irrigation with improved water quality. Scenarios (3) and (4), while unlikely, resulted in CS up to positive $30 million (rates a year aggregated across the region). More realistic scenarios (1) and (2), that included increase in irrigation, resulted in CS negative $41 million if water quality is reduced and positive $10 million if water quality is maintained. In addition to above, there were two other objectives. The first of these focused on the citizen versus consumer framing effect where different motivational point-of-views impact on respondent’s preferences. A split-sampling approach was applied where the respondents in one survey were asked to adopt a consumer’s point-of-view and the respondents in another survey were asked to adopt a citizen’s point-of-view. A separate analysis was undertaken to compare for differences in welfare measures under alternative valuation frames. The results from the convolution test (Poe et al., 2001, 2005) show no sensitivity for the survey framing. Overall, this result suggests that either point-of-view could be used which is important as it has been argued that people are more likely to adopt the citizen point-of-view in environmental valuations. The second of these objectives focused on the role of complexity in choice experiments defined by the difficulty in making choices amongst options in each choice set. The aim was to test for fatigue effects related to this measure of complexity, and whether this detrimental effect could be ameliorated by an experimental design that ordered choice sets. Another separate analysis was undertaken to test these effects including estimation of the scale parameter and the convolution test. Results show no evidence for fatigue nor that the order in which the choice sets are presented result in different WTP estimates. Thus although choice experiments can be complex, the future studies can benefit from the information that fatigue may not cause inconsistency or that the choice sets can be shown in varying orders. In conclusion, this study adds to the non-market valuation (NMV) literature of freshwater and cultural valuation in New Zealand. The results show that the Canterbury residents are concerned about the state of the region’s rivers including all the elements of well-being. This formal identification of values provided from various perspectives as well as the exploration of how the future irrigation scenarios can impact on society’s well-being are useful information to Canterbury water management. In addition, those applying DCM in practice can benefit from the exploration of survey framing, complexity, fatigue and the choice set order while it was shown that, overall, the results in this study were insensitive for these effects.