The aim of this thesis is to examine issues pertaining to freshwater quantity in New Zealand agriculture. Currently, freshwater is managed under the Resource Management Act (1991) and the allocation system is essentially on a “first-come first-served” basis – whoever gains access to water rights first blocks out subsequent users, if catchment allocation limits have been reached. One of the recommendations from a government taskforce engaged to look at the issues of freshwater management suggests that market mechanisms or charges could be viable options of demand management. For any such market or price mechanism to work, it is imperative to know the value of irrigation water in New Zealand as well as to understand wider economic ramifications of establishing such mechanisms. Currently, there is relatively little research concerning these issues and this thesis aims to fill this research gap. The methodologies used include an econometric analysis of dairy farm panel data, a stated choice experiment of irrigation consent holders and a computable general equilibrium [CGE] analysis. In addition to filling the research gap, this thesis also aims to provide improvements in each of the methodologies used. Panel data analysis imputes the water demand function without actual data on water through examining the relationship between milksolid production and output-weighted expected payout – a unique dairy price index developed that more accurately reflects farmers’ incentives than the final payout. The choice modelling section pays particular attention to the issue of attribute non-attendance [ANA], eexamining its effects on model outputs and compares methods of data collection on ANA. The main findings of the survey suggest that the majority of farmers would be willing to pay for water instead of facing an abstraction ban. In terms of ANA, the results indicate that not accounting for ANA, particularly if it is due to heuristics or respondent fatigue, may significantly bias the welfare estimates and decrease statistical significance. A novel calibration method is presented to negate the ANA bias. Finally, the CGE modelling work modifies the well-known GTAP model to include water as factor of production, as well as to disaggregate the results to a regional level within the New Zealand economy. The findings indicate strong interconnectedness between sectors and regions, with policies in one region having the potential to affect economic activity and resource demand in other regions.