Policy referendums in the United Kingdom, New Zealand and Australia : constitutional functions and effects

This thesis examines the theory and practice of 'policy referendums' -- i.e. referendums held on questions which are *not* fundamental constitutional issues. In the wake of the 2014 Scottish independence referendum and the 2016 Brexit referendum, much was written on how referendums on certain constitutional issues could be understood to operate within Westminster-style constitutions, however no major work has considered the conduct of policy referendums. This thesis looks at the conduct of this latter class of votes across the United Kingdom, New Zealand and Australia. It is the first detailed study of the subject-matter. In those places, a surprising number of such policy referendums have been held (or have at least been proposed to be held): where, in New Zealand, dozens of national votes on ordinary policies have addressed topics including the state-ownership of banks and public infrastructure, liquor licensing laws, and the legalisation of assisted suicide. In Australia, policy referendums (or 'plebiscites', as they are called there) have addressed topics including military conscription and marriage equality. In the United Kingdom, too, a wide range of examples exist, from across the 20th century, for the proposed incorporation of policy referendums into the ordinary legislative process. Some of these were reflected in draft laws and, at the level of the United Kingdom's constituent countries, a series of policy referendums have in fact been held. The old assumption is that such policy referendums are little more than political theatre or a mass opinion poll, and have little bearing on the normative architecture of the constitution. It is *that* assumption which is tested in this thesis, through undertaking a series of country-based case studies. The conclusion reached is that, in the uncodified or partially codified constitutional communities here considered, it is not productive to distinguish between 'constitutional' and 'policy' referendums. In flexible constitutions, both exercises can bear upon how power is constructed under the constitution, and both can operate as vectors of informal constitutional change. In the result, we must think more carefully about how and why we engage in popular politics on 'ordinary' matters.